From 47d37e943cb00a312a010b07d9bf800e6f09b906 Mon Sep 17 00:00:00 2001 From: Narendra Date: Tue, 1 Sep 2026 20:08:46 +0530 Subject: [PATCH] Bumping vendored uriparser --- cpp/src/arrow/CMakeLists.txt | 16 +- cpp/src/arrow/meson.build | 14 + cpp/src/arrow/vendored/uriparser/README.md | 5 +- cpp/src/arrow/vendored/uriparser/Uri.h | 1775 ++++++-- cpp/src/arrow/vendored/uriparser/UriBase.h | 377 +- cpp/src/arrow/vendored/uriparser/UriCommon.c | 1191 ++--- cpp/src/arrow/vendored/uriparser/UriCommon.h | 130 +- cpp/src/arrow/vendored/uriparser/UriCompare.c | 274 +- cpp/src/arrow/vendored/uriparser/UriConfig.h | 47 + cpp/src/arrow/vendored/uriparser/UriCopy.c | 248 ++ cpp/src/arrow/vendored/uriparser/UriCopy.h | 76 + .../arrow/vendored/uriparser/UriDefsAnsi.h | 50 +- .../arrow/vendored/uriparser/UriDefsConfig.h | 111 +- .../arrow/vendored/uriparser/UriDefsUnicode.h | 52 +- cpp/src/arrow/vendored/uriparser/UriEscape.c | 718 ++-- cpp/src/arrow/vendored/uriparser/UriFile.c | 387 +- cpp/src/arrow/vendored/uriparser/UriIp4.c | 444 +- cpp/src/arrow/vendored/uriparser/UriIp4.h | 127 +- cpp/src/arrow/vendored/uriparser/UriIp4Base.c | 107 +- cpp/src/arrow/vendored/uriparser/UriIp4Base.h | 56 +- cpp/src/arrow/vendored/uriparser/UriMemory.c | 743 ++-- cpp/src/arrow/vendored/uriparser/UriMemory.h | 92 +- .../arrow/vendored/uriparser/UriNormalize.c | 1448 ++++--- .../arrow/vendored/uriparser/UriNormalize.h | 75 + .../vendored/uriparser/UriNormalizeBase.c | 186 +- .../vendored/uriparser/UriNormalizeBase.h | 50 +- cpp/src/arrow/vendored/uriparser/UriParse.c | 3826 ++++++++--------- .../arrow/vendored/uriparser/UriParseBase.c | 113 +- .../arrow/vendored/uriparser/UriParseBase.h | 52 +- cpp/src/arrow/vendored/uriparser/UriQuery.c | 851 ++-- .../arrow/vendored/uriparser/UriRecompose.c | 1236 +++--- cpp/src/arrow/vendored/uriparser/UriResolve.c | 636 +-- .../arrow/vendored/uriparser/UriSetFragment.c | 174 + .../arrow/vendored/uriparser/UriSetHostAuto.c | 124 + .../arrow/vendored/uriparser/UriSetHostBase.h | 49 + .../vendored/uriparser/UriSetHostCommon.c | 248 ++ .../vendored/uriparser/UriSetHostCommon.h | 74 + .../arrow/vendored/uriparser/UriSetHostIp4.c | 91 + .../arrow/vendored/uriparser/UriSetHostIp6.c | 158 + .../vendored/uriparser/UriSetHostIpFuture.c | 157 + .../vendored/uriparser/UriSetHostRegName.c | 124 + cpp/src/arrow/vendored/uriparser/UriSetPath.c | 343 ++ cpp/src/arrow/vendored/uriparser/UriSetPort.c | 149 + .../arrow/vendored/uriparser/UriSetQuery.c | 172 + .../arrow/vendored/uriparser/UriSetScheme.c | 168 + .../arrow/vendored/uriparser/UriSetUserInfo.c | 180 + cpp/src/arrow/vendored/uriparser/UriSets.h | 174 + cpp/src/arrow/vendored/uriparser/UriShorten.c | 667 +-- cpp/src/arrow/vendored/uriparser/UriVersion.c | 81 + cpp/src/arrow/vendored/uriparser/config.h | 47 - 50 files changed, 11439 insertions(+), 7254 deletions(-) create mode 100644 cpp/src/arrow/vendored/uriparser/UriConfig.h create mode 100644 cpp/src/arrow/vendored/uriparser/UriCopy.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriCopy.h create mode 100644 cpp/src/arrow/vendored/uriparser/UriNormalize.h create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetFragment.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetHostAuto.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetHostBase.h create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetHostCommon.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetHostCommon.h create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetHostIp4.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetHostIp6.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetHostIpFuture.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetHostRegName.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetPath.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetPort.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetQuery.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetScheme.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSetUserInfo.c create mode 100644 cpp/src/arrow/vendored/uriparser/UriSets.h create mode 100644 cpp/src/arrow/vendored/uriparser/UriVersion.c delete mode 100644 cpp/src/arrow/vendored/uriparser/config.h diff --git a/cpp/src/arrow/CMakeLists.txt b/cpp/src/arrow/CMakeLists.txt index b9ff2ffa2c2d..a9c206e023c2 100644 --- a/cpp/src/arrow/CMakeLists.txt +++ b/cpp/src/arrow/CMakeLists.txt @@ -536,6 +536,7 @@ set(ARROW_VENDORED_SRCS vendored/musl/strptime.c vendored/uriparser/UriCommon.c vendored/uriparser/UriCompare.c + vendored/uriparser/UriCopy.c vendored/uriparser/UriEscape.c vendored/uriparser/UriFile.c vendored/uriparser/UriIp4.c @@ -548,7 +549,20 @@ set(ARROW_VENDORED_SRCS vendored/uriparser/UriQuery.c vendored/uriparser/UriRecompose.c vendored/uriparser/UriResolve.c - vendored/uriparser/UriShorten.c) + vendored/uriparser/UriSetFragment.c + vendored/uriparser/UriSetHostAuto.c + vendored/uriparser/UriSetHostCommon.c + vendored/uriparser/UriSetHostIp4.c + vendored/uriparser/UriSetHostIp6.c + vendored/uriparser/UriSetHostIpFuture.c + vendored/uriparser/UriSetHostRegName.c + vendored/uriparser/UriSetPath.c + vendored/uriparser/UriSetPort.c + vendored/uriparser/UriSetQuery.c + vendored/uriparser/UriSetScheme.c + vendored/uriparser/UriSetUserInfo.c + vendored/uriparser/UriShorten.c + vendored/uriparser/UriVersion.c) if(APPLE) list(APPEND ARROW_VENDORED_SRCS vendored/datetime/ios.mm) endif() diff --git a/cpp/src/arrow/meson.build b/cpp/src/arrow/meson.build index b5443c28cf84..5e0bbdec06ec 100644 --- a/cpp/src/arrow/meson.build +++ b/cpp/src/arrow/meson.build @@ -112,6 +112,7 @@ arrow_components = { 'vendored/musl/strptime.c', 'vendored/uriparser/UriCommon.c', 'vendored/uriparser/UriCompare.c', + 'vendored/uriparser/UriCopy.c', 'vendored/uriparser/UriEscape.c', 'vendored/uriparser/UriFile.c', 'vendored/uriparser/UriIp4.c', @@ -124,7 +125,20 @@ arrow_components = { 'vendored/uriparser/UriQuery.c', 'vendored/uriparser/UriRecompose.c', 'vendored/uriparser/UriResolve.c', + 'vendored/uriparser/UriSetFragment.c', + 'vendored/uriparser/UriSetHostAuto.c', + 'vendored/uriparser/UriSetHostCommon.c', + 'vendored/uriparser/UriSetHostIp4.c', + 'vendored/uriparser/UriSetHostIp6.c', + 'vendored/uriparser/UriSetHostIpFuture.c', + 'vendored/uriparser/UriSetHostRegName.c', + 'vendored/uriparser/UriSetPath.c', + 'vendored/uriparser/UriSetPort.c', + 'vendored/uriparser/UriSetQuery.c', + 'vendored/uriparser/UriSetScheme.c', + 'vendored/uriparser/UriSetUserInfo.c', 'vendored/uriparser/UriShorten.c', + 'vendored/uriparser/UriVersion.c', ], }, 'arrow_base': { diff --git a/cpp/src/arrow/vendored/uriparser/README.md b/cpp/src/arrow/vendored/uriparser/README.md index b3a219c7f34e..1393cf0c4bb9 100644 --- a/cpp/src/arrow/vendored/uriparser/README.md +++ b/cpp/src/arrow/vendored/uriparser/README.md @@ -17,9 +17,10 @@ under the License. --> -The files in this directory are vendored from uriparser git tag "uriparser-0.9.3". +The files in this directory are vendored from uriparser git tag "uriparser-1.0.2". Include paths fixed using: + ``` -sed -E -i 's:include :include "\1.h":g' src/arrow/vendored/uriparser/* +sed -E -i 's:include "(.*).h":include "\1.h":g' src/arrow/vendored/uriparser/* ``` diff --git a/cpp/src/arrow/vendored/uriparser/Uri.h b/cpp/src/arrow/vendored/uriparser/Uri.h index 9315a12c3fe9..ddbbd26f9e44 100644 --- a/cpp/src/arrow/vendored/uriparser/Uri.h +++ b/cpp/src/arrow/vendored/uriparser/Uri.h @@ -1,4 +1,4 @@ -/* 8cd64b75589a7efa22989ae85f71e620a99e3c44776338f7d3114eacf73d27a3 (0.9.3+) +/* c9d94656d067288e474df19a062d487c736b0fa8517d2ef7bbeb8dcd5a70c05b (1.0.2+) * * uriparser - RFC 3986 URI parsing library * @@ -6,32 +6,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -45,47 +45,41 @@ */ #if (defined(URI_PASS_ANSI) && !defined(URI_H_ANSI)) \ - || (defined(URI_PASS_UNICODE) && !defined(URI_H_UNICODE)) \ - || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) + || (defined(URI_PASS_UNICODE) && !defined(URI_H_UNICODE)) \ + || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* What encodings are enabled? */ -#include "UriDefsConfig.h" -#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +# include "UriDefsConfig.h" +# if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "Uri.h" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "Uri.h" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "Uri.h" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "Uri.h" +# undef URI_PASS_UNICODE +# endif /* Only one pass for each encoding */ -#elif (defined(URI_PASS_ANSI) && !defined(URI_H_ANSI) \ - && defined(URI_ENABLE_ANSI)) || (defined(URI_PASS_UNICODE) \ - && !defined(URI_H_UNICODE) && defined(URI_ENABLE_UNICODE)) -# ifdef URI_PASS_ANSI -# define URI_H_ANSI 1 -# include "UriDefsAnsi.h" -# else -# define URI_H_UNICODE 1 -# include "UriDefsUnicode.h" -# endif - - - -#ifdef __cplusplus +# elif (defined(URI_PASS_ANSI) && !defined(URI_H_ANSI) && defined(URI_ENABLE_ANSI)) \ + || (defined(URI_PASS_UNICODE) && !defined(URI_H_UNICODE) \ + && defined(URI_ENABLE_UNICODE)) +# ifdef URI_PASS_ANSI +# define URI_H_ANSI 1 +# include "UriDefsAnsi.h" +# else +# define URI_H_UNICODE 1 +# include "UriDefsUnicode.h" +# endif + +# ifdef __cplusplus extern "C" { -#endif - - - -#ifndef URI_DOXYGEN -# include "UriBase.h" -#endif - +# endif +# ifndef URI_DOXYGEN +# include "UriBase.h" +# endif /** * Specifies a range of characters within a string. @@ -99,12 +93,10 @@ extern "C" { * @since 0.3.0 */ typedef struct URI_TYPE(TextRangeStruct) { - const URI_CHAR * first; /**< Pointer to first character */ - const URI_CHAR * afterLast; /**< Pointer to character after the last one still in */ + const URI_CHAR * first; /**< Pointer to first character */ + const URI_CHAR * afterLast; /**< Pointer to character after the last one still in */ } URI_TYPE(TextRange); /**< @copydoc UriTextRangeStructA */ - - /** * Represents a path segment within a %URI path. * More precisely it is a node in a linked @@ -114,14 +106,13 @@ typedef struct URI_TYPE(TextRangeStruct) { * @since 0.3.0 */ typedef struct URI_TYPE(PathSegmentStruct) { - URI_TYPE(TextRange) text; /**< Path segment name */ - struct URI_TYPE(PathSegmentStruct) * next; /**< Pointer to the next path segment in the list, can be NULL if last already */ + URI_TYPE(TextRange) text; /**< Path segment name */ + struct URI_TYPE(PathSegmentStruct) * next; /**< Pointer to the next path segment in + the list, can be NULL if last already */ - void * reserved; /**< Reserved to the parser */ + void * reserved; /**< Reserved to the parser */ } URI_TYPE(PathSegment); /**< @copydoc UriPathSegmentStructA */ - - /** * Holds structured host information. * This is either a IPv4, IPv6, plain @@ -132,13 +123,16 @@ typedef struct URI_TYPE(PathSegmentStruct) { * @since 0.3.0 */ typedef struct URI_TYPE(HostDataStruct) { - UriIp4 * ip4; /**< IPv4 address */ - UriIp6 * ip6; /**< IPv6 address */ - URI_TYPE(TextRange) ipFuture; /**< IPvFuture address */ + UriIp4 * ip4; /**< IPv4 address */ + UriIp6 * ip6; /**< IPv6 address */ + URI_TYPE(TextRange) + ipFuture; /**< IPvFuture address +@note +With non-NULL members in UriUriStructA.hostData context, +this text range's pointers must be identical to those +of UriUriStructA.hostText at all times. */ } URI_TYPE(HostData); /**< @copydoc UriHostDataStructA */ - - /** * Represents an RFC 3986 %URI. * Missing components can be {NULL, NULL} ranges. @@ -149,24 +143,23 @@ typedef struct URI_TYPE(HostDataStruct) { * @since 0.3.0 */ typedef struct URI_TYPE(UriStruct) { - URI_TYPE(TextRange) scheme; /**< Scheme (e.g. "http") */ - URI_TYPE(TextRange) userInfo; /**< User info (e.g. "user:pass") */ - URI_TYPE(TextRange) hostText; /**< Host text (set for all hosts, excluding square brackets) */ - URI_TYPE(HostData) hostData; /**< Structured host type specific data */ - URI_TYPE(TextRange) portText; /**< Port (e.g. "80") */ - URI_TYPE(PathSegment) * pathHead; /**< Head of a linked list of path segments */ - URI_TYPE(PathSegment) * pathTail; /**< Tail of the list behind pathHead */ - URI_TYPE(TextRange) query; /**< Query without leading "?" */ - URI_TYPE(TextRange) fragment; /**< Query without leading "#" */ - UriBool absolutePath; /**< Absolute path flag, distincting "a" and "/a"; - always URI_FALSE for URIs with host */ - UriBool owner; /**< Memory owner flag */ - - void * reserved; /**< Reserved to the parser */ + URI_TYPE(TextRange) scheme; /**< Scheme (e.g. "http") */ + URI_TYPE(TextRange) userInfo; /**< User info (e.g. "user:pass") */ + URI_TYPE(TextRange) + hostText; /**< Host text (set for all hosts, excluding square brackets) */ + URI_TYPE(HostData) hostData; /**< Structured host type specific data */ + URI_TYPE(TextRange) portText; /**< Port (e.g. "80") */ + URI_TYPE(PathSegment) * pathHead; /**< Head of a linked list of path segments */ + URI_TYPE(PathSegment) * pathTail; /**< Tail of the list behind pathHead */ + URI_TYPE(TextRange) query; /**< Query without leading "?" */ + URI_TYPE(TextRange) fragment; /**< Query without leading "#" */ + UriBool absolutePath; /**< Absolute path flag, distincting "a" and "/a"; always + URI_FALSE for URIs with host */ + UriBool owner; /**< Memory owner flag */ + + void * reserved; /**< Reserved to the parser */ } URI_TYPE(Uri); /**< @copydoc UriUriStructA */ - - /** * Represents a state of the %URI parser. * Missing components can be NULL to reflect @@ -177,15 +170,14 @@ typedef struct URI_TYPE(UriStruct) { * @since 0.3.0 */ typedef struct URI_TYPE(ParserStateStruct) { - URI_TYPE(Uri) * uri; /**< Plug in the %URI structure to be filled while parsing here */ - int errorCode; /**< Code identifying the error which occurred */ - const URI_CHAR * errorPos; /**< Pointer to position in case of a syntax error */ + URI_TYPE(Uri) + *uri; /**< Plug in the %URI structure to be filled while parsing here */ + int errorCode; /**< Code identifying the error which occurred */ + const URI_CHAR * errorPos; /**< Pointer to position in case of a syntax error */ - void * reserved; /**< Reserved to the parser */ + void * reserved; /**< Reserved to the parser */ } URI_TYPE(ParserState); /**< @copydoc UriParserStateStructA */ - - /** * Represents a query element. * More precisely it is a node in a linked @@ -194,13 +186,58 @@ typedef struct URI_TYPE(ParserStateStruct) { * @since 0.7.0 */ typedef struct URI_TYPE(QueryListStruct) { - const URI_CHAR * key; /**< Key of the query element */ - const URI_CHAR * value; /**< Value of the query element, can be NULL */ + const URI_CHAR * key; /**< Key of the query element */ + const URI_CHAR * value; /**< Value of the query element, can be NULL */ - struct URI_TYPE(QueryListStruct) * next; /**< Pointer to the next key/value pair in the list, can be NULL if last already */ + struct URI_TYPE(QueryListStruct) * next; /**< Pointer to the next key/value pair in + the list, can be NULL if last already */ } URI_TYPE(QueryList); /**< @copydoc UriQueryListStructA */ +/** + * Checks if a URI has the host component set. + * + * @param uri IN: %URI to check + * @return URI_TRUE when host is set, URI_FALSE otherwise + * + * @since 0.9.9 + */ +URI_PUBLIC UriBool URI_FUNC(HasHost)(const URI_TYPE(Uri) * uri); +/** + * Converts an IPv6 text representation into 16 bytes. + * + * Uses default libc-based memory manager. + * + * @param output OUT: Output destination, can be NULL + * @param first IN: First character of IPv6 text to parse + * @param afterLast IN: Position to stop parsing at + * @return Error code or URI_SUCCESS on success + * + * @see uriParseIpFourAddressA + * @see uriParseIpSixAddressMmA + * @see uriIsWellFormedHostIp6A + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(ParseIpSixAddress)(UriIp6 * output, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Converts an IPv6 text representation into 16 bytes. + * + * @param output OUT: Output destination, can be NULL + * @param first IN: First character of IPv6 text to parse + * @param afterLast IN: Position to stop parsing at + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or URI_SUCCESS on success + * + * @see uriParseIpFourAddressA + * @see uriParseIpSixAddressA + * @see uriIsWellFormedHostIp6MmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(ParseIpSixAddressMm)(UriIp6 * output, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); /** * Parses a RFC 3986 %URI. @@ -208,7 +245,8 @@ typedef struct URI_TYPE(QueryListStruct) { * * @param state INOUT: Parser state with set output %URI, must not be NULL * @param first IN: Pointer to the first character to parse, must not be NULL - * @param afterLast IN: Pointer to the character after the last to parse, must not be NULL + * @param afterLast IN: Pointer to the character after the last to parse, must + * not be NULL * @return 0 on success, error code otherwise * * @see uriParseUriA @@ -216,12 +254,11 @@ typedef struct URI_TYPE(QueryListStruct) { * @see uriParseSingleUriExA * @see uriToStringA * @since 0.3.0 - * @deprecated Deprecated since 0.9.0, please migrate to uriParseSingleUriExA (with "Single"). + * @deprecated Deprecated since 0.9.0, please migrate to uriParseSingleUriExA (with + * "Single"). */ -URI_PUBLIC int URI_FUNC(ParseUriEx)(URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast); - - +URI_PUBLIC int URI_FUNC(ParseUriEx)(URI_TYPE(ParserState) * state, const URI_CHAR * first, + const URI_CHAR * afterLast); /** * Parses a RFC 3986 %URI. @@ -236,12 +273,10 @@ URI_PUBLIC int URI_FUNC(ParseUriEx)(URI_TYPE(ParserState) * state, * @see uriParseSingleUriExA * @see uriToStringA * @since 0.3.0 - * @deprecated Deprecated since 0.9.0, please migrate to uriParseSingleUriA (with "Single"). + * @deprecated Deprecated since 0.9.0, please migrate to uriParseSingleUriA (with + * "Single"). */ -URI_PUBLIC int URI_FUNC(ParseUri)(URI_TYPE(ParserState) * state, - const URI_CHAR * text); - - +URI_PUBLIC int URI_FUNC(ParseUri)(URI_TYPE(ParserState) * state, const URI_CHAR * text); /** * Parses a single RFC 3986 %URI. @@ -260,10 +295,8 @@ URI_PUBLIC int URI_FUNC(ParseUri)(URI_TYPE(ParserState) * state, * @see uriToStringA * @since 0.9.0 */ -URI_PUBLIC int URI_FUNC(ParseSingleUri)(URI_TYPE(Uri) * uri, - const URI_CHAR * text, const URI_CHAR ** errorPos); - - +URI_PUBLIC int URI_FUNC(ParseSingleUri)(URI_TYPE(Uri) * uri, const URI_CHAR * text, + const URI_CHAR ** errorPos); /** * Parses a single RFC 3986 %URI. @@ -285,11 +318,9 @@ URI_PUBLIC int URI_FUNC(ParseSingleUri)(URI_TYPE(Uri) * uri, * @see uriToStringA * @since 0.9.0 */ -URI_PUBLIC int URI_FUNC(ParseSingleUriEx)(URI_TYPE(Uri) * uri, - const URI_CHAR * first, const URI_CHAR * afterLast, - const URI_CHAR ** errorPos); - - +URI_PUBLIC int URI_FUNC(ParseSingleUriEx)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + const URI_CHAR ** errorPos); /** * Parses a single RFC 3986 %URI. @@ -311,11 +342,10 @@ URI_PUBLIC int URI_FUNC(ParseSingleUriEx)(URI_TYPE(Uri) * uri, * @see uriToStringA * @since 0.9.0 */ -URI_PUBLIC int URI_FUNC(ParseSingleUriExMm)(URI_TYPE(Uri) * uri, - const URI_CHAR * first, const URI_CHAR * afterLast, - const URI_CHAR ** errorPos, UriMemoryManager * memory); - - +URI_PUBLIC int URI_FUNC(ParseSingleUriExMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + const URI_CHAR ** errorPos, + UriMemoryManager * memory); /** * Frees all memory associated with the members @@ -323,6 +353,10 @@ URI_PUBLIC int URI_FUNC(ParseSingleUriExMm)(URI_TYPE(Uri) * uri, * itself is not freed, only its members. * Uses default libc-based memory manager. * + * @remarks + * Calling on an all-zeros structure (e.g. through memset or calloc) is + * safe.
Calling on an uninitialized structure is not safe. + * * @param uri INOUT: %URI structure whose members should be freed * * @see uriFreeUriMembersMmA @@ -330,13 +364,15 @@ URI_PUBLIC int URI_FUNC(ParseSingleUriExMm)(URI_TYPE(Uri) * uri, */ URI_PUBLIC void URI_FUNC(FreeUriMembers)(URI_TYPE(Uri) * uri); - - /** * Frees all memory associated with the members * of the %URI structure. Note that the structure * itself is not freed, only its members. * + * @remarks + * Calling on an all-zeros structure (e.g. through memset or calloc) is + * safe.
Calling on an uninitialized structure is not safe. + * * @param uri INOUT: %URI structure whose members should be freed * @param memory IN: Memory manager to use, NULL for default libc * @return 0 on success, error code otherwise @@ -344,18 +380,24 @@ URI_PUBLIC void URI_FUNC(FreeUriMembers)(URI_TYPE(Uri) * uri); * @see uriFreeUriMembersA * @since 0.9.0 */ -URI_PUBLIC int URI_FUNC(FreeUriMembersMm)(URI_TYPE(Uri) * uri, - UriMemoryManager * memory); - - +URI_PUBLIC int URI_FUNC(FreeUriMembersMm)(URI_TYPE(Uri) * uri, UriMemoryManager * memory); /** - * Percent-encodes all unreserved characters from the input string and + * Percent-encodes all but unreserved characters from the input string and * writes the encoded version to the output string. - * Be sure to allocate 3 times the space of the input buffer for + * + * NOTE: Be sure to allocate 3 times the space of the input buffer for * the output buffer for normalizeBreaks == URI_FALSE and 6 times * the space for normalizeBreaks == URI_TRUE - * (since e.g. "\x0d" becomes "%0D%0A" in that case) + * (since e.g. "\x0d" becomes "%0D%0A" in that case). + * + * NOTE: The implementation treats (both char and) wchar_t units + * as code point integers, which works well for code points U+0001 to U+00ff + * in host-native endianness but nothing more; + * in particular, using uriEscapeExW with arbitrary Unicode input will + * not produce healthy results. + * Passing UTF-8 input to uriEscapeExA may be useful in some scenarios. + * Keep in mind that uriparser is about %URI (RFC 3986) not %IRI (RFC 3987). * * @param inFirst IN: Pointer to first character of the input text * @param inAfterLast IN: Pointer after the last character of the input text @@ -369,18 +411,25 @@ URI_PUBLIC int URI_FUNC(FreeUriMembersMm)(URI_TYPE(Uri) * uri, * @since 0.5.2 */ URI_PUBLIC URI_CHAR * URI_FUNC(EscapeEx)(const URI_CHAR * inFirst, - const URI_CHAR * inAfterLast, URI_CHAR * out, - UriBool spaceToPlus, UriBool normalizeBreaks); - - + const URI_CHAR * inAfterLast, URI_CHAR * out, + UriBool spaceToPlus, UriBool normalizeBreaks); /** - * Percent-encodes all unreserved characters from the input string and + * Percent-encodes all but unreserved characters from the input string and * writes the encoded version to the output string. - * Be sure to allocate 3 times the space of the input buffer for + * + * NOTE: Be sure to allocate 3 times the space of the input buffer for * the output buffer for normalizeBreaks == URI_FALSE and 6 times * the space for normalizeBreaks == URI_TRUE - * (since e.g. "\x0d" becomes "%0D%0A" in that case) + * (since e.g. "\x0d" becomes "%0D%0A" in that case). + * + * NOTE: The implementation treats (both char and) wchar_t units + * as code point integers, which works well for code points U+0001 to U+00ff + * in host-native endianness but nothing more; + * in particular, using uriEscapeW with arbitrary Unicode input will + * not produce healthy results. + * Passing UTF-8 input to uriEscapeA may be useful in some scenarios. + * Keep in mind that uriparser is about %URI (RFC 3986) not %IRI (RFC 3987). * * @param in IN: Text source * @param out OUT: Encoded text destination @@ -393,9 +442,7 @@ URI_PUBLIC URI_CHAR * URI_FUNC(EscapeEx)(const URI_CHAR * inFirst, * @since 0.5.0 */ URI_PUBLIC URI_CHAR * URI_FUNC(Escape)(const URI_CHAR * in, URI_CHAR * out, - UriBool spaceToPlus, UriBool normalizeBreaks); - - + UriBool spaceToPlus, UriBool normalizeBreaks); /** * Unescapes percent-encoded groups in a given string. @@ -414,10 +461,9 @@ URI_PUBLIC URI_CHAR * URI_FUNC(Escape)(const URI_CHAR * in, URI_CHAR * out, * @see uriEscapeExA * @since 0.5.0 */ -URI_PUBLIC const URI_CHAR * URI_FUNC(UnescapeInPlaceEx)(URI_CHAR * inout, - UriBool plusToSpace, UriBreakConversion breakConversion); - - +URI_PUBLIC const URI_CHAR * + URI_FUNC(UnescapeInPlaceEx)(URI_CHAR * inout, UriBool plusToSpace, + UriBreakConversion breakConversion); /** * Unescapes percent-encoded groups in a given string. @@ -439,13 +485,11 @@ URI_PUBLIC const URI_CHAR * URI_FUNC(UnescapeInPlaceEx)(URI_CHAR * inout, */ URI_PUBLIC const URI_CHAR * URI_FUNC(UnescapeInPlace)(URI_CHAR * inout); - - /** * Performs reference resolution as described in - * section 5.2.2 of RFC 3986. - * Uses default libc-based memory manager. - * NOTE: On success you have to call uriFreeUriMembersA on \p absoluteDest manually later. + * section 5.2.2 of + * RFC 3986. Uses default libc-based memory manager. NOTE: On success you have to call + * uriFreeUriMembersA on \p absoluteDest manually later. * * @param absoluteDest OUT: Result %URI * @param relativeSource IN: Reference to resolve @@ -459,16 +503,14 @@ URI_PUBLIC const URI_CHAR * URI_FUNC(UnescapeInPlace)(URI_CHAR * inout); * @since 0.4.0 */ URI_PUBLIC int URI_FUNC(AddBaseUri)(URI_TYPE(Uri) * absoluteDest, - const URI_TYPE(Uri) * relativeSource, - const URI_TYPE(Uri) * absoluteBase); - - + const URI_TYPE(Uri) * relativeSource, + const URI_TYPE(Uri) * absoluteBase); /** * Performs reference resolution as described in - * section 5.2.2 of RFC 3986. - * Uses default libc-based memory manager. - * NOTE: On success you have to call uriFreeUriMembersA on \p absoluteDest manually later. + * section 5.2.2 of + * RFC 3986. Uses default libc-based memory manager. NOTE: On success you have to call + * uriFreeUriMembersA on \p absoluteDest manually later. * * @param absoluteDest OUT: Result %URI * @param relativeSource IN: Reference to resolve @@ -482,16 +524,15 @@ URI_PUBLIC int URI_FUNC(AddBaseUri)(URI_TYPE(Uri) * absoluteDest, * @since 0.8.1 */ URI_PUBLIC int URI_FUNC(AddBaseUriEx)(URI_TYPE(Uri) * absoluteDest, - const URI_TYPE(Uri) * relativeSource, - const URI_TYPE(Uri) * absoluteBase, - UriResolutionOptions options); - - + const URI_TYPE(Uri) * relativeSource, + const URI_TYPE(Uri) * absoluteBase, + UriResolutionOptions options); /** * Performs reference resolution as described in - * section 5.2.2 of RFC 3986. - * NOTE: On success you have to call uriFreeUriMembersMmA on \p absoluteDest manually later. + * section 5.2.2 of + * RFC 3986. NOTE: On success you have to call uriFreeUriMembersMmA on \p absoluteDest + * manually later. * * @param absoluteDest OUT: Result %URI * @param relativeSource IN: Reference to resolve @@ -507,11 +548,10 @@ URI_PUBLIC int URI_FUNC(AddBaseUriEx)(URI_TYPE(Uri) * absoluteDest, * @since 0.9.0 */ URI_PUBLIC int URI_FUNC(AddBaseUriExMm)(URI_TYPE(Uri) * absoluteDest, - const URI_TYPE(Uri) * relativeSource, - const URI_TYPE(Uri) * absoluteBase, - UriResolutionOptions options, UriMemoryManager * memory); - - + const URI_TYPE(Uri) * relativeSource, + const URI_TYPE(Uri) * absoluteBase, + UriResolutionOptions options, + UriMemoryManager * memory); /** * Tries to make a relative %URI (a reference) from an @@ -536,11 +576,9 @@ URI_PUBLIC int URI_FUNC(AddBaseUriExMm)(URI_TYPE(Uri) * absoluteDest, * @since 0.5.2 */ URI_PUBLIC int URI_FUNC(RemoveBaseUri)(URI_TYPE(Uri) * dest, - const URI_TYPE(Uri) * absoluteSource, - const URI_TYPE(Uri) * absoluteBase, - UriBool domainRootMode); - - + const URI_TYPE(Uri) * absoluteSource, + const URI_TYPE(Uri) * absoluteBase, + UriBool domainRootMode); /** * Tries to make a relative %URI (a reference) from an @@ -565,11 +603,10 @@ URI_PUBLIC int URI_FUNC(RemoveBaseUri)(URI_TYPE(Uri) * dest, * @since 0.9.0 */ URI_PUBLIC int URI_FUNC(RemoveBaseUriMm)(URI_TYPE(Uri) * dest, - const URI_TYPE(Uri) * absoluteSource, - const URI_TYPE(Uri) * absoluteBase, - UriBool domainRootMode, UriMemoryManager * memory); - - + const URI_TYPE(Uri) * absoluteSource, + const URI_TYPE(Uri) * absoluteBase, + UriBool domainRootMode, + UriMemoryManager * memory); /** * Checks two URIs for equivalence. Comparison is done @@ -582,10 +619,7 @@ URI_PUBLIC int URI_FUNC(RemoveBaseUriMm)(URI_TYPE(Uri) * dest, * * @since 0.4.0 */ -URI_PUBLIC UriBool URI_FUNC(EqualsUri)(const URI_TYPE(Uri) * a, - const URI_TYPE(Uri) * b); - - +URI_PUBLIC UriBool URI_FUNC(EqualsUri)(const URI_TYPE(Uri) * a, const URI_TYPE(Uri) * b); /** * Calculates the number of characters needed to store the @@ -593,34 +627,67 @@ URI_PUBLIC UriBool URI_FUNC(EqualsUri)(const URI_TYPE(Uri) * a, * terminator. * * @param uri IN: %URI to measure - * @param charsRequired OUT: Length of the string representation in characters excluding terminator + * @param charsRequired OUT: Length of the string representation in characters + * excluding terminator * @return Error code or 0 on success * * @see uriToStringA * @since 0.5.0 */ URI_PUBLIC int URI_FUNC(ToStringCharsRequired)(const URI_TYPE(Uri) * uri, - int * charsRequired); - - + int * charsRequired); /** * Converts a %URI structure back to text as described in - * section 5.3 of RFC 3986. + * section 5.3 of RFC + * 3986. + * + * NOTE: Scheme-based normalization + * (section 6.2.3 of + * RFC 3986) is not applied and is considered a responsibility of the application + * using uriparser. * * @param dest OUT: Output destination * @param uri IN: %URI to convert - * @param maxChars IN: Maximum number of characters to copy including terminator - * @param charsWritten OUT: Number of characters written, can be lower than maxChars even if the %URI is too long! + * @param maxChars IN: Maximum number of characters to copy including + * terminator + * @param charsWritten OUT: Number of characters written, can be lower than + * maxChars even if the %URI is too long! * @return Error code or 0 on success * * @see uriToStringCharsRequiredA * @since 0.4.0 */ URI_PUBLIC int URI_FUNC(ToString)(URI_CHAR * dest, const URI_TYPE(Uri) * uri, - int maxChars, int * charsWritten); + int maxChars, int * charsWritten); +/** + * Copies a %URI structure. + * + * @param destUri OUT: Output destination + * @param sourceUri IN: %URI to copy + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriCopyUriA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(CopyUriMm)(URI_TYPE(Uri) * destUri, + const URI_TYPE(Uri) * sourceUri, + UriMemoryManager * memory); +/** + * Copies a %URI structure. + * + * @param destUri OUT: Output destination + * @param sourceUri IN: %URI to copy + * @return Error code or 0 on success + * + * @see uriCopyUriMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(CopyUri)(URI_TYPE(Uri) * destUri, + const URI_TYPE(Uri) * sourceUri); /** * Determines the components of a %URI that are not normalized. @@ -633,12 +700,10 @@ URI_PUBLIC int URI_FUNC(ToString)(URI_CHAR * dest, const URI_TYPE(Uri) * uri, * @see uriNormalizeSyntaxExMmA * @see uriNormalizeSyntaxMaskRequiredExA * @since 0.5.0 - * @deprecated Deprecated since 0.9.0, please migrate to uriNormalizeSyntaxMaskRequiredExA (with "Ex"). + * @deprecated Deprecated since 0.9.0, please migrate to uriNormalizeSyntaxMaskRequiredExA + * (with "Ex"). */ -URI_PUBLIC unsigned int URI_FUNC(NormalizeSyntaxMaskRequired)( - const URI_TYPE(Uri) * uri); - - +URI_PUBLIC unsigned int URI_FUNC(NormalizeSyntaxMaskRequired)(const URI_TYPE(Uri) * uri); /** * Determines the components of a %URI that are not normalized. @@ -653,10 +718,8 @@ URI_PUBLIC unsigned int URI_FUNC(NormalizeSyntaxMaskRequired)( * @see uriNormalizeSyntaxMaskRequiredA * @since 0.9.0 */ -URI_PUBLIC int URI_FUNC(NormalizeSyntaxMaskRequiredEx)( - const URI_TYPE(Uri) * uri, unsigned int * outMask); - - +URI_PUBLIC int URI_FUNC(NormalizeSyntaxMaskRequiredEx)(const URI_TYPE(Uri) * uri, + unsigned int * outMask); /** * Normalizes a %URI using a normalization mask. @@ -675,10 +738,7 @@ URI_PUBLIC int URI_FUNC(NormalizeSyntaxMaskRequiredEx)( * @see uriNormalizeSyntaxMaskRequiredA * @since 0.5.0 */ -URI_PUBLIC int URI_FUNC(NormalizeSyntaxEx)(URI_TYPE(Uri) * uri, - unsigned int mask); - - +URI_PUBLIC int URI_FUNC(NormalizeSyntaxEx)(URI_TYPE(Uri) * uri, unsigned int mask); /** * Normalizes a %URI using a normalization mask. @@ -697,10 +757,8 @@ URI_PUBLIC int URI_FUNC(NormalizeSyntaxEx)(URI_TYPE(Uri) * uri, * @see uriNormalizeSyntaxMaskRequiredA * @since 0.9.0 */ -URI_PUBLIC int URI_FUNC(NormalizeSyntaxExMm)(URI_TYPE(Uri) * uri, - unsigned int mask, UriMemoryManager * memory); - - +URI_PUBLIC int URI_FUNC(NormalizeSyntaxExMm)(URI_TYPE(Uri) * uri, unsigned int mask, + UriMemoryManager * memory); /** * Normalizes all components of a %URI. @@ -719,8 +777,6 @@ URI_PUBLIC int URI_FUNC(NormalizeSyntaxExMm)(URI_TYPE(Uri) * uri, */ URI_PUBLIC int URI_FUNC(NormalizeSyntax)(URI_TYPE(Uri) * uri); - - /** * Converts a Unix filename to a %URI string. * The destination buffer must be large enough to hold 7 + 3 * len(filename) + 1 @@ -740,9 +796,7 @@ URI_PUBLIC int URI_FUNC(NormalizeSyntax)(URI_TYPE(Uri) * uri); * @since 0.5.2 */ URI_PUBLIC int URI_FUNC(UnixFilenameToUriString)(const URI_CHAR * filename, - URI_CHAR * uriString); - - + URI_CHAR * uriString); /** * Converts a Windows filename to a %URI string. @@ -763,13 +817,11 @@ URI_PUBLIC int URI_FUNC(UnixFilenameToUriString)(const URI_CHAR * filename, * @since 0.5.2 */ URI_PUBLIC int URI_FUNC(WindowsFilenameToUriString)(const URI_CHAR * filename, - URI_CHAR * uriString); - - + URI_CHAR * uriString); /** * Extracts a Unix filename from a %URI string. - * The destination buffer must be large enough to hold len(uriString) + 1 - 7 + * The destination buffer must be large enough to hold len(uriString) + 1 - 5 * characters in case of an absolute %URI or len(uriString) + 1 in case * of a relative %URI. * @@ -782,9 +834,7 @@ URI_PUBLIC int URI_FUNC(WindowsFilenameToUriString)(const URI_CHAR * filename, * @since 0.5.2 */ URI_PUBLIC int URI_FUNC(UriStringToUnixFilename)(const URI_CHAR * uriString, - URI_CHAR * filename); - - + URI_CHAR * filename); /** * Extracts a Windows filename from a %URI string. @@ -801,9 +851,7 @@ URI_PUBLIC int URI_FUNC(UriStringToUnixFilename)(const URI_CHAR * uriString, * @since 0.5.2 */ URI_PUBLIC int URI_FUNC(UriStringToWindowsFilename)(const URI_CHAR * uriString, - URI_CHAR * filename); - - + URI_CHAR * filename); /** * Calculates the number of characters needed to store the @@ -812,17 +860,16 @@ URI_PUBLIC int URI_FUNC(UriStringToWindowsFilename)(const URI_CHAR * uriString, * normalized to "%0D%0A". * * @param queryList IN: Query list to measure - * @param charsRequired OUT: Length of the string representation in characters excluding terminator + * @param charsRequired OUT: Length of the string representation in characters + * excluding terminator * @return Error code or 0 on success * * @see uriComposeQueryCharsRequiredExA * @see uriComposeQueryA * @since 0.7.0 */ -URI_PUBLIC int URI_FUNC(ComposeQueryCharsRequired)( - const URI_TYPE(QueryList) * queryList, int * charsRequired); - - +URI_PUBLIC int URI_FUNC(ComposeQueryCharsRequired)(const URI_TYPE(QueryList) * queryList, + int * charsRequired); /** * Calculates the number of characters needed to store the @@ -830,7 +877,8 @@ URI_PUBLIC int URI_FUNC(ComposeQueryCharsRequired)( * terminator. * * @param queryList IN: Query list to measure - * @param charsRequired OUT: Length of the string representation in characters excluding terminator + * @param charsRequired OUT: Length of the string representation in characters + * excluding terminator * @param spaceToPlus IN: Whether to convert ' ' to '+' or not * @param normalizeBreaks IN: Whether to convert CR and LF to CR-LF or not. * @return Error code or 0 on success @@ -839,11 +887,10 @@ URI_PUBLIC int URI_FUNC(ComposeQueryCharsRequired)( * @see uriComposeQueryExA * @since 0.7.0 */ -URI_PUBLIC int URI_FUNC(ComposeQueryCharsRequiredEx)( - const URI_TYPE(QueryList) * queryList, - int * charsRequired, UriBool spaceToPlus, UriBool normalizeBreaks); - - +URI_PUBLIC int + URI_FUNC(ComposeQueryCharsRequiredEx)(const URI_TYPE(QueryList) * queryList, + int * charsRequired, UriBool spaceToPlus, + UriBool normalizeBreaks); /** * Converts a query list structure back to a query string. @@ -853,8 +900,10 @@ URI_PUBLIC int URI_FUNC(ComposeQueryCharsRequiredEx)( * * @param dest OUT: Output destination * @param queryList IN: Query list to convert - * @param maxChars IN: Maximum number of characters to copy including terminator - * @param charsWritten OUT: Number of characters written, can be lower than maxChars even if the query list is too long! + * @param maxChars IN: Maximum number of characters to copy + * including terminator + * @param charsWritten OUT: Number of characters written, can be lower than + * maxChars even if the query list is too long! * @return Error code or 0 on success * * @see uriComposeQueryExA @@ -868,9 +917,8 @@ URI_PUBLIC int URI_FUNC(ComposeQueryCharsRequiredEx)( * @since 0.7.0 */ URI_PUBLIC int URI_FUNC(ComposeQuery)(URI_CHAR * dest, - const URI_TYPE(QueryList) * queryList, int maxChars, int * charsWritten); - - + const URI_TYPE(QueryList) * queryList, int maxChars, + int * charsWritten); /** * Converts a query list structure back to a query string. @@ -878,8 +926,10 @@ URI_PUBLIC int URI_FUNC(ComposeQuery)(URI_CHAR * dest, * * @param dest OUT: Output destination * @param queryList IN: Query list to convert - * @param maxChars IN: Maximum number of characters to copy including terminator - * @param charsWritten OUT: Number of characters written, can be lower than maxChars even if the query list is too long! + * @param maxChars IN: Maximum number of characters to copy + * including terminator + * @param charsWritten OUT: Number of characters written, can be lower than + * maxChars even if the query list is too long! * @param spaceToPlus IN: Whether to convert ' ' to '+' or not * @param normalizeBreaks IN: Whether to convert CR and LF to CR-LF or not. * @return Error code or 0 on success @@ -895,10 +945,9 @@ URI_PUBLIC int URI_FUNC(ComposeQuery)(URI_CHAR * dest, * @since 0.7.0 */ URI_PUBLIC int URI_FUNC(ComposeQueryEx)(URI_CHAR * dest, - const URI_TYPE(QueryList) * queryList, int maxChars, int * charsWritten, - UriBool spaceToPlus, UriBool normalizeBreaks); - - + const URI_TYPE(QueryList) * queryList, + int maxChars, int * charsWritten, + UriBool spaceToPlus, UriBool normalizeBreaks); /** * Converts a query list structure back to a query string. @@ -921,9 +970,7 @@ URI_PUBLIC int URI_FUNC(ComposeQueryEx)(URI_CHAR * dest, * @since 0.7.0 */ URI_PUBLIC int URI_FUNC(ComposeQueryMalloc)(URI_CHAR ** dest, - const URI_TYPE(QueryList) * queryList); - - + const URI_TYPE(QueryList) * queryList); /** * Converts a query list structure back to a query string. @@ -946,10 +993,9 @@ URI_PUBLIC int URI_FUNC(ComposeQueryMalloc)(URI_CHAR ** dest, * @since 0.7.0 */ URI_PUBLIC int URI_FUNC(ComposeQueryMallocEx)(URI_CHAR ** dest, - const URI_TYPE(QueryList) * queryList, - UriBool spaceToPlus, UriBool normalizeBreaks); - - + const URI_TYPE(QueryList) * queryList, + UriBool spaceToPlus, + UriBool normalizeBreaks); /** * Converts a query list structure back to a query string. @@ -972,11 +1018,10 @@ URI_PUBLIC int URI_FUNC(ComposeQueryMallocEx)(URI_CHAR ** dest, * @since 0.9.0 */ URI_PUBLIC int URI_FUNC(ComposeQueryMallocExMm)(URI_CHAR ** dest, - const URI_TYPE(QueryList) * queryList, - UriBool spaceToPlus, UriBool normalizeBreaks, - UriMemoryManager * memory); - - + const URI_TYPE(QueryList) * queryList, + UriBool spaceToPlus, + UriBool normalizeBreaks, + UriMemoryManager * memory); /** * Constructs a query list from the raw query string of a given URI. @@ -996,10 +1041,9 @@ URI_PUBLIC int URI_FUNC(ComposeQueryMallocExMm)(URI_CHAR ** dest, * @see uriFreeQueryListMmA * @since 0.7.0 */ -URI_PUBLIC int URI_FUNC(DissectQueryMalloc)(URI_TYPE(QueryList) ** dest, - int * itemCount, const URI_CHAR * first, const URI_CHAR * afterLast); - - +URI_PUBLIC int URI_FUNC(DissectQueryMalloc)(URI_TYPE(QueryList) * *dest, int * itemCount, + const URI_CHAR * first, + const URI_CHAR * afterLast); /** * Constructs a query list from the raw query string of a given URI. @@ -1019,11 +1063,11 @@ URI_PUBLIC int URI_FUNC(DissectQueryMalloc)(URI_TYPE(QueryList) ** dest, * @see uriFreeQueryListA * @since 0.7.0 */ -URI_PUBLIC int URI_FUNC(DissectQueryMallocEx)(URI_TYPE(QueryList) ** dest, - int * itemCount, const URI_CHAR * first, const URI_CHAR * afterLast, - UriBool plusToSpace, UriBreakConversion breakConversion); - - +URI_PUBLIC int URI_FUNC(DissectQueryMallocEx)(URI_TYPE(QueryList) * *dest, + int * itemCount, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriBool plusToSpace, + UriBreakConversion breakConversion); /** * Constructs a query list from the raw query string of a given URI. @@ -1044,12 +1088,12 @@ URI_PUBLIC int URI_FUNC(DissectQueryMallocEx)(URI_TYPE(QueryList) ** dest, * @see uriFreeQueryListMmA * @since 0.9.0 */ -URI_PUBLIC int URI_FUNC(DissectQueryMallocExMm)(URI_TYPE(QueryList) ** dest, - int * itemCount, const URI_CHAR * first, const URI_CHAR * afterLast, - UriBool plusToSpace, UriBreakConversion breakConversion, - UriMemoryManager * memory); - - +URI_PUBLIC int URI_FUNC(DissectQueryMallocExMm)(URI_TYPE(QueryList) * *dest, + int * itemCount, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriBool plusToSpace, + UriBreakConversion breakConversion, + UriMemoryManager * memory); /** * Frees all memory associated with the given query list. @@ -1062,8 +1106,6 @@ URI_PUBLIC int URI_FUNC(DissectQueryMallocExMm)(URI_TYPE(QueryList) ** dest, */ URI_PUBLIC void URI_FUNC(FreeQueryList)(URI_TYPE(QueryList) * queryList); - - /** * Frees all memory associated with the given query list. * The structure itself is freed as well. @@ -1076,15 +1118,1192 @@ URI_PUBLIC void URI_FUNC(FreeQueryList)(URI_TYPE(QueryList) * queryList); * @since 0.9.0 */ URI_PUBLIC int URI_FUNC(FreeQueryListMm)(URI_TYPE(QueryList) * queryList, - UriMemoryManager * memory); + UriMemoryManager * memory); + +/** + * Makes the %URI hold copies of strings so that it no longer depends + * on the original %URI string. If the %URI is already owner of copies, + * this function returns URI_TRUE and does not modify the %URI further. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to make independent + * @return Error code or 0 on success + * + * @see uriMakeOwnerMmA + * @since 0.9.4 + */ +URI_PUBLIC int URI_FUNC(MakeOwner)(URI_TYPE(Uri) * uri); +/** + * Makes the %URI hold copies of strings so that it no longer depends + * on the original %URI string. If the %URI is already owner of copies, + * this function returns URI_TRUE and does not modify the %URI further. + * + * @param uri INOUT: %URI to make independent + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriMakeOwnerA + * @since 0.9.4 + */ +URI_PUBLIC int URI_FUNC(MakeOwnerMm)(URI_TYPE(Uri) * uri, UriMemoryManager * memory); +/** + * Determines if the given text range contains a well-formed fragment + * according to RFC 3986 or not. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @return URI_TRUE if non-NULL and well-formed, else + * URI_FALSE + * + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @see uriSetFragmentA + * @see uriSetFragmentMmA + * @since 0.9.9 + */ +URI_PUBLIC UriBool URI_FUNC(IsWellFormedFragment)(const URI_CHAR * first, + const URI_CHAR * afterLast); -#ifdef __cplusplus -} -#endif +/** + * Determines if the given text range contains a well-formed IPv4 address + * according to RFC 3986 or not. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @return URI_TRUE if non-NULL and well-formed, else + * URI_FALSE + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @see uriSetHostIp4A + * @see uriSetHostIp4MmA + * @since 0.9.9 + */ +URI_PUBLIC UriBool URI_FUNC(IsWellFormedHostIp4)(const URI_CHAR * first, + const URI_CHAR * afterLast); +/** + * Determines if the given text range contains a well-formed IPv6 address + * according to RFC 3986 or not. + * + * Uses default libc-based memory manager. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @return URI_SUCCESS if non-NULL and well-formed, else an error + * code + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6MmA + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @see uriParseIpSixAddressA + * @see uriParseIpSixAddressMmA + * @since 0.9.9 + */ +int URI_FUNC(IsWellFormedHostIp6)(const URI_CHAR * first, const URI_CHAR * afterLast); +/** + * Determines if the given text range contains a well-formed IPv6 address + * according to RFC 3986 or not. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @param memory IN: Memory manager to use, NULL for default libc + * @return URI_SUCCESS if non-NULL and well-formed, else an error + * code + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureMmA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @since 0.9.9 + */ +int URI_FUNC(IsWellFormedHostIp6Mm)(const URI_CHAR * first, const URI_CHAR * afterLast, + UriMemoryManager * memory); -#endif +/** + * Determines if the given text range contains a well-formed IPvFuture address + * according to RFC 3986 or not. + * + * Uses default libc-based memory manager. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @return URI_SUCCESS if non-NULL and well-formed, else an error + * code + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureMmA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @see uriSetHostIpFutureA + * @see uriSetHostIpFutureMmA + * @since 0.9.9 + */ +int URI_FUNC(IsWellFormedHostIpFuture)(const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Determines if the given text range contains a well-formed IPvFuture address + * according to RFC 3986 or not. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @param memory IN: Memory manager to use, NULL for default libc + * @return URI_SUCCESS if non-NULL and well-formed, else an error + * code + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @see uriSetHostIpFutureA + * @see uriSetHostIpFutureMmA + * @since 0.9.9 + */ +int URI_FUNC(IsWellFormedHostIpFutureMm)(const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Determines if the given text range contains a well-formed registered host name + * according to RFC 3986 or not. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @return URI_TRUE if non-NULL and well-formed, else + * URI_FALSE + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @see uriSetHostRegNameA + * @see uriSetHostRegNameMmA + * @since 0.9.9 + */ +URI_PUBLIC UriBool URI_FUNC(IsWellFormedHostRegName)(const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Determines if the given text range contains a well-formed path + * according to RFC 3986 or not. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @param hasHost IN: Whether the target %URI has a non-NULL host set or + * not + * @return URI_TRUE if non-NULL and well-formed, else + * URI_FALSE + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @see uriSetPathA + * @see uriSetPathMmA + * @since 0.9.9 + */ +URI_PUBLIC UriBool URI_FUNC(IsWellFormedPath)(const URI_CHAR * first, + const URI_CHAR * afterLast, + UriBool hasHost); + +/** + * Determines if the given text range contains a well-formed port text + * according to RFC 3986 or not. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @return URI_TRUE if non-NULL and well-formed, else + * URI_FALSE + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @see uriSetPortTextA + * @see uriSetPortTextMmA + * @since 0.9.9 + */ +URI_PUBLIC UriBool URI_FUNC(IsWellFormedPort)(const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Determines if the given text range contains a well-formed query + * according to RFC 3986 or not. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @return URI_TRUE if non-NULL and well-formed, else + * URI_FALSE + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @see uriSetQueryA + * @see uriSetQueryMmA + * @since 0.9.9 + */ +URI_PUBLIC UriBool URI_FUNC(IsWellFormedQuery)(const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Determines if the given text range contains a well-formed scheme + * according to RFC 3986 or not. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @return URI_TRUE if non-NULL and well-formed, else + * URI_FALSE + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriIsWellFormedUserInfoA + * @see uriSetSchemeA + * @see uriSetSchemeMmA + * @since 0.9.9 + */ +URI_PUBLIC UriBool URI_FUNC(IsWellFormedScheme)(const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Determines if the given text range contains a well-formed user info + * according to RFC 3986 or not. + * + * @param first IN: Pointer to first character + * @param afterLast IN: Pointer to character after the last one still in + * @return URI_TRUE if non-NULL and well-formed, else + * URI_FALSE + * + * @see uriIsWellFormedFragmentA + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostRegNameA + * @see uriIsWellFormedPathA + * @see uriIsWellFormedPortA + * @see uriIsWellFormedQueryA + * @see uriIsWellFormedSchemeA + * @see uriSetUserInfoA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC UriBool URI_FUNC(IsWellFormedUserInfo)(const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the fragment of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedFragmentA + * @see uriSetFragmentMmA + * @see uriSetHostAutoA + * @see uriSetHostIp4A + * @see uriSetHostIp6A + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameA + * @see uriSetPortTextA + * @see uriSetQueryA + * @see uriSetSchemeA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetFragment)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the fragment of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedFragmentA + * @see uriSetFragmentA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameMmA + * @see uriSetPortTextMmA + * @see uriSetQueryMmA + * @see uriSetSchemeMmA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetFragmentMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Sets the host of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIp6MmA + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostIpFutureMmA + * @see uriIsWellFormedHostRegNameA + * @see uriSetFragmentA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4A + * @see uriSetHostIp6A + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameA + * @see uriSetPortTextA + * @see uriSetQueryA + * @see uriSetSchemeA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetHostAuto)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the host of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedHostIp4A + * @see uriIsWellFormedHostIp6A + * @see uriIsWellFormedHostIp6MmA + * @see uriIsWellFormedHostIpFutureA + * @see uriIsWellFormedHostIpFutureMmA + * @see uriIsWellFormedHostRegNameA + * @see uriSetFragmentMmA + * @see uriSetHostAutoA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameMmA + * @see uriSetPortTextMmA + * @see uriSetQueryMmA + * @see uriSetSchemeMmA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetHostAutoMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Sets the host of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedHostRegNameA + * @see uriSetFragmentA + * @see uriSetHostAutoA + * @see uriSetHostIp4A + * @see uriSetHostIp6A + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameMmA + * @see uriSetPortTextA + * @see uriSetQueryA + * @see uriSetSchemeA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetHostRegName)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the host of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedHostRegNameA + * @see uriSetFragmentMmA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameA + * @see uriSetPortTextMmA + * @see uriSetQueryMmA + * @see uriSetSchemeMmA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetHostRegNameMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Sets the host of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedHostIp4A + * @see uriSetFragmentA + * @see uriSetHostAutoA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6A + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameA + * @see uriSetPortTextA + * @see uriSetQueryA + * @see uriSetSchemeA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetHostIp4)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the host of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedHostIp4A + * @see uriSetFragmentMmA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4A + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameMmA + * @see uriSetPortTextMmA + * @see uriSetQueryMmA + * @see uriSetSchemeMmA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetHostIp4Mm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Sets the host of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedHostIp6A + * @see uriSetFragmentA + * @see uriSetHostAutoA + * @see uriSetHostIp4A + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameA + * @see uriSetPortTextA + * @see uriSetQueryA + * @see uriSetSchemeA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetHostIp6)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the host of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedHostIp6A + * @see uriSetFragmentMmA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6A + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameMmA + * @see uriSetPortTextMmA + * @see uriSetQueryMmA + * @see uriSetSchemeMmA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetHostIp6Mm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Sets the host of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedHostIp4A + * @see uriSetFragmentA + * @see uriSetHostAutoA + * @see uriSetHostIp4A + * @see uriSetHostIp6A + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameA + * @see uriSetPortTextA + * @see uriSetQueryA + * @see uriSetSchemeA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetHostIpFuture)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the host of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedHostIp4A + * @see uriSetFragmentMmA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameMmA + * @see uriSetPortTextMmA + * @see uriSetQueryMmA + * @see uriSetSchemeMmA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetHostIpFutureMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Sets the path of the given %URI to the given value. + * + * Non-NULL values must start with a leading slash for %URIs that have a host. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedPathA + * @see uriSetFragmentA + * @see uriSetHostAutoA + * @see uriSetHostIp4A + * @see uriSetHostIp6A + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameA + * @see uriSetPathMmA + * @see uriSetPortTextA + * @see uriSetQueryA + * @see uriSetSchemeA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetPath)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the path of the given %URI to the given value. + * + * Non-NULL values must start with a leading slash for %URIs that have a host. + * + * The %URI must have a non-NULL host set. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedPathA + * @see uriSetFragmentMmA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameMmA + * @see uriSetPathA + * @see uriSetPortTextMmA + * @see uriSetQueryMmA + * @see uriSetSchemeMmA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetPathMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory); + +/** + * Sets the port text of the given %URI to the given value. + * + * The %URI must have a non-NULL host set. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedPortA + * @see uriSetFragmentA + * @see uriSetHostAutoA + * @see uriSetHostIp4A + * @see uriSetHostIp6A + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameA + * @see uriSetPortTextMmA + * @see uriSetQueryA + * @see uriSetSchemeA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetPortText)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the port text of the given %URI to the given value. + * + * The %URI must have a non-NULL host set. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedPortA + * @see uriSetFragmentMmA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameMmA + * @see uriSetPortTextA + * @see uriSetQueryMmA + * @see uriSetSchemeMmA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetPortTextMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Sets the query of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedQueryA + * @see uriSetFragmentA + * @see uriSetHostAutoA + * @see uriSetHostIp4A + * @see uriSetHostIp6A + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameA + * @see uriSetPortTextA + * @see uriSetQueryMmA + * @see uriSetSchemeA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetQuery)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the query of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedQueryA + * @see uriSetFragmentMmA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameMmA + * @see uriSetPortTextMmA + * @see uriSetQueryA + * @see uriSetSchemeMmA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetQueryMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Sets the scheme of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedSchemeA + * @see uriSetFragmentA + * @see uriSetHostAutoA + * @see uriSetHostIp4A + * @see uriSetHostIp6A + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameA + * @see uriSetPortTextA + * @see uriSetQueryA + * @see uriSetSchemeMmA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetScheme)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the scheme of the given %URI to the given value. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedSchemeA + * @see uriSetFragmentMmA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameMmA + * @see uriSetPortTextMmA + * @see uriSetQueryMmA + * @see uriSetSchemeA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetSchemeMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Sets the user info of the given %URI to the given value. + * + * The %URI must have a non-NULL host set. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * Uses default libc-based memory manager. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @return Error code or 0 on success + * + * @see uriIsWellFormedUserInfoA + * @see uriSetFragmentA + * @see uriSetHostAutoA + * @see uriSetHostIp4A + * @see uriSetHostIp6A + * @see uriSetHostIpFutureA + * @see uriSetHostRegNameA + * @see uriSetPortTextA + * @see uriSetQueryA + * @see uriSetSchemeA + * @see uriSetUserInfoMmA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetUserInfo)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast); + +/** + * Sets the user info of the given %URI to the given value. + * + * The %URI must have a non-NULL host set. + * + * Parameters first and afterLast must both be NULL + * or non-NULL at the same time. + * + * The function may make the %URI own its memory if needed (if it is not already owned). + * + * For all return values but URI_ERROR_MALLOC, all-or-nothing behavior + * can be expected, e.g. trying to apply a malformed value will leave the + * %URI unchanged. + * + * @param uri INOUT: %URI to modify + * @param first IN: Pointer to first character, can be NULL + * @param afterLast IN: Pointer to character after the last one still in, can be + * NULL + * @param memory IN: Memory manager to use, NULL for default libc + * @return Error code or 0 on success + * + * @see uriIsWellFormedUserInfoA + * @see uriSetFragmentMmA + * @see uriSetHostAutoMmA + * @see uriSetHostIp4MmA + * @see uriSetHostIp6MmA + * @see uriSetHostIpFutureMmA + * @see uriSetHostRegNameMmA + * @see uriSetPortTextMmA + * @see uriSetQueryMmA + * @see uriSetSchemeMmA + * @see uriSetUserInfoA + * @since 0.9.9 + */ +URI_PUBLIC int URI_FUNC(SetUserInfoMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +/** + * Obtain the base runtime version of uriparser. + * + * The string returned is based on the compile time version + * of uriparser. + * + * @note Distributors may have applied backports of security + * fixes (potentially with adjusting packaging version but often + * without adjusting runtime version) + * or even fully custom patches. As a result, the version string + * returned serves as nothing more than "based on that version", + * it does not guarantee equivalence to vanilla upstream releases + * or absence of additional downstream patches. + * It is nothing more than "a hint" and MUST NEVER be used to + * make decisions on in application code at runtime. + * + * @return Pointer to a read-only zero terminated version string + * + * @since 0.9.9 + */ +URI_PUBLIC const URI_CHAR * URI_FUNC(BaseRuntimeVersion)(void); + +# ifdef __cplusplus +} +# endif + +# endif #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriBase.h b/cpp/src/arrow/vendored/uriparser/UriBase.h index 166915950c29..1cc91ceab772 100644 --- a/cpp/src/arrow/vendored/uriparser/UriBase.h +++ b/cpp/src/arrow/vendored/uriparser/UriBase.h @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -43,144 +43,136 @@ */ #ifndef URI_BASE_H -#define URI_BASE_H 1 - - +# define URI_BASE_H 1 /* Version helper macro */ -#define URI_ANSI_TO_UNICODE(x) L##x - - +# define URI_ANSI_TO_UNICODE_HELPER(x) L##x +# define URI_ANSI_TO_UNICODE(x) URI_ANSI_TO_UNICODE_HELPER(x) /* Version */ -#define URI_VER_MAJOR 0 -#define URI_VER_MINOR 9 -#define URI_VER_RELEASE 3 -#define URI_VER_SUFFIX_ANSI "" -#define URI_VER_SUFFIX_UNICODE URI_ANSI_TO_UNICODE(URI_VER_SUFFIX_ANSI) - - +# define URI_VER_MAJOR 1 +# define URI_VER_MINOR 0 +# define URI_VER_RELEASE 2 +# define URI_VER_SUFFIX_ANSI "" +# define URI_VER_SUFFIX_UNICODE URI_ANSI_TO_UNICODE(URI_VER_SUFFIX_ANSI) /* More version helper macros */ -#define URI_INT_TO_ANSI_HELPER(x) #x -#define URI_INT_TO_ANSI(x) URI_INT_TO_ANSI_HELPER(x) - -#define URI_INT_TO_UNICODE_HELPER(x) URI_ANSI_TO_UNICODE(#x) -#define URI_INT_TO_UNICODE(x) URI_INT_TO_UNICODE_HELPER(x) - -#define URI_VER_ANSI_HELPER(ma, mi, r, s) \ - URI_INT_TO_ANSI(ma) "." \ - URI_INT_TO_ANSI(mi) "." \ - URI_INT_TO_ANSI(r) \ - s +# define URI_INT_TO_ANSI_HELPER(x) #x +# define URI_INT_TO_ANSI(x) URI_INT_TO_ANSI_HELPER(x) -#define URI_VER_UNICODE_HELPER(ma, mi, r, s) \ - URI_INT_TO_UNICODE(ma) L"." \ - URI_INT_TO_UNICODE(mi) L"." \ - URI_INT_TO_UNICODE(r) \ - s +# define URI_INT_TO_UNICODE_HELPER(x) URI_ANSI_TO_UNICODE(#x) +# define URI_INT_TO_UNICODE(x) URI_INT_TO_UNICODE_HELPER(x) +# define URI_VER_ANSI_HELPER(ma, mi, r, s) \ + URI_INT_TO_ANSI(ma) "." URI_INT_TO_ANSI(mi) "." URI_INT_TO_ANSI(r) s +# define URI_VER_UNICODE_HELPER(ma, mi, r, s) \ + URI_INT_TO_UNICODE(ma) L"." URI_INT_TO_UNICODE(mi) L"." URI_INT_TO_UNICODE(r) s /* Full version strings */ -#define URI_VER_ANSI URI_VER_ANSI_HELPER(URI_VER_MAJOR, URI_VER_MINOR, URI_VER_RELEASE, URI_VER_SUFFIX_ANSI) -#define URI_VER_UNICODE URI_VER_UNICODE_HELPER(URI_VER_MAJOR, URI_VER_MINOR, URI_VER_RELEASE, URI_VER_SUFFIX_UNICODE) - - +# define URI_VER_ANSI \ + URI_VER_ANSI_HELPER(URI_VER_MAJOR, URI_VER_MINOR, URI_VER_RELEASE, \ + URI_VER_SUFFIX_ANSI) +# define URI_VER_UNICODE \ + URI_VER_UNICODE_HELPER(URI_VER_MAJOR, URI_VER_MINOR, URI_VER_RELEASE, \ + URI_VER_SUFFIX_UNICODE) /* Unused parameter macro */ -#ifdef __GNUC__ -# define URI_UNUSED(x) unused_##x __attribute__((unused)) -#else -# define URI_UNUSED(x) x -#endif - - - -/* Import/export decorator */ -#if defined(URI_STATIC_BUILD) -# define URI_PUBLIC -#else -# if defined(URI_LIBRARY_BUILD) -# if defined(_MSC_VER) -# define URI_PUBLIC __declspec(dllexport) -# elif defined(URI_VISIBILITY) -# define URI_PUBLIC __attribute__ ((visibility("default"))) +# ifdef __GNUC__ +# define URI_UNUSED(x) unused_##x __attribute__((unused)) # else -# define URI_PUBLIC +# define URI_UNUSED(x) x # endif -# else + +/* Import/export decorator */ # if defined(_MSC_VER) -# define URI_PUBLIC __declspec(dllimport) +# if defined(URI_STATIC_BUILD) +# define URI_PUBLIC +# elif defined(URI_LIBRARY_BUILD) +# define URI_PUBLIC __declspec(dllexport) +# else +# define URI_PUBLIC __declspec(dllimport) +# endif # else -# define URI_PUBLIC +# if !defined(URI_LIBRARY_BUILD) || !defined(URI_VISIBILITY) +# define URI_PUBLIC +# else +# define URI_PUBLIC __attribute__((visibility("default"))) +# endif # endif -# endif -#endif - - typedef int UriBool; /**< Boolean type */ -#define URI_TRUE 1 -#define URI_FALSE 0 - - +# define URI_TRUE 1 +# define URI_FALSE 0 /* Shared errors */ -#define URI_SUCCESS 0 -#define URI_ERROR_SYNTAX 1 /* Parsed text violates expected format */ -#define URI_ERROR_NULL 2 /* One of the params passed was NULL - although it mustn't be */ -#define URI_ERROR_MALLOC 3 /* Requested memory could not be allocated */ -#define URI_ERROR_OUTPUT_TOO_LARGE 4 /* Some output is to large for the receiving buffer */ -#define URI_ERROR_NOT_IMPLEMENTED 8 /* The called function is not implemented yet */ -#define URI_ERROR_RANGE_INVALID 9 /* The parameters passed contained invalid ranges */ -#define URI_ERROR_MEMORY_MANAGER_INCOMPLETE 10 /* [>=0.9.0] The UriMemoryManager passed does not implement all needed functions */ - +# define URI_SUCCESS 0 +# define URI_ERROR_SYNTAX 1 /* Parsed text violates expected format */ +# define URI_ERROR_NULL \ + 2 /* One of the params passed was NULL although it mustn't be \ + */ +# define URI_ERROR_MALLOC 3 /* Requested memory could not be allocated */ +# define URI_ERROR_OUTPUT_TOO_LARGE \ + 4 /* Some output is to large for the receiving buffer */ +# define URI_ERROR_NOT_IMPLEMENTED 8 /* The called function is not implemented yet */ +# define URI_ERROR_RANGE_INVALID 9 /* The parameters passed contained invalid ranges */ +# define URI_ERROR_MEMORY_MANAGER_INCOMPLETE \ + 10 /* [>=0.9.0] The UriMemoryManager passed does not implement all needed \ + functions */ /* Errors specific to ToString */ -#define URI_ERROR_TOSTRING_TOO_LONG URI_ERROR_OUTPUT_TOO_LARGE /* Deprecated, test for URI_ERROR_OUTPUT_TOO_LARGE instead */ +# define URI_ERROR_TOSTRING_TOO_LONG \ + URI_ERROR_OUTPUT_TOO_LARGE /* Deprecated, test for URI_ERROR_OUTPUT_TOO_LARGE \ + instead */ /* Errors specific to AddBaseUri */ -#define URI_ERROR_ADDBASE_REL_BASE 5 /* Given base is not absolute */ +# define URI_ERROR_ADDBASE_REL_BASE 5 /* Given base is not absolute */ /* Errors specific to RemoveBaseUri */ -#define URI_ERROR_REMOVEBASE_REL_BASE 6 /* Given base is not absolute */ -#define URI_ERROR_REMOVEBASE_REL_SOURCE 7 /* Given base is not absolute */ +# define URI_ERROR_REMOVEBASE_REL_BASE 6 /* Given base is not absolute */ +# define URI_ERROR_REMOVEBASE_REL_SOURCE 7 /* Given base is not absolute */ /* Error specific to uriTestMemoryManager */ -#define URI_ERROR_MEMORY_MANAGER_FAULTY 11 /* [>=0.9.0] The UriMemoryManager given did not pass the test suite */ - - -#ifndef URI_DOXYGEN -# include /* For NULL, snprintf */ -# include /* For wchar_t */ -# include /* For strlen, memset, memcpy */ -# include /* For malloc */ -#endif /* URI_DOXYGEN */ - - +# define URI_ERROR_MEMORY_MANAGER_FAULTY \ + 11 /* [>=0.9.0] The UriMemoryManager given did not pass the test suite */ + +/* Error specific to uriSetUserInfo */ +# define URI_ERROR_SETUSERINFO_HOST_NOT_SET \ + 12 /* [>=0.9.9] The %URI given does not have the host set */ + +/* Error specific to uriSetPort */ +# define URI_ERROR_SETPORT_HOST_NOT_SET \ + 13 /* [>=0.9.9] The %URI given does not have the host set */ + +/* Error specific to uriSetHost* */ +# define URI_ERROR_SETHOST_USERINFO_SET \ + 14 /* [>=0.9.9] The %URI given does have user info set */ +# define URI_ERROR_SETHOST_PORT_SET \ + 15 /* [>=0.9.9] The %URI given does have a port set */ + +# ifndef URI_DOXYGEN +# include /* For NULL, snprintf */ +# include /* For wchar_t */ +# include /* For strlen, memset, memcpy */ +# include /* For malloc */ +# endif /* URI_DOXYGEN */ /** * Holds an IPv4 address. */ typedef struct UriIp4Struct { - unsigned char data[4]; /**< Each octet in one byte */ + unsigned char data[4]; /**< Each octet in one byte */ } UriIp4; /**< @copydoc UriIp4Struct */ - - /** * Holds an IPv6 address. */ typedef struct UriIp6Struct { - unsigned char data[16]; /**< Each quad in two bytes */ + unsigned char data[16]; /**< Each quad in two bytes */ } UriIp6; /**< @copydoc UriIp6Struct */ - -struct UriMemoryManagerStruct; /* foward declaration to break loop */ - +struct UriMemoryManagerStruct; /* forward declaration to break loop */ /** * Function signature that custom malloc(3) functions must conform to @@ -208,7 +200,8 @@ typedef void * (*UriFuncRealloc)(struct UriMemoryManagerStruct *, void *, size_t * * @since 0.9.0 */ -typedef void * (*UriFuncReallocarray)(struct UriMemoryManagerStruct *, void *, size_t, size_t); +typedef void * (*UriFuncReallocarray)(struct UriMemoryManagerStruct *, void *, size_t, + size_t); /** * Function signature that custom free(3) functions must conform to @@ -217,7 +210,6 @@ typedef void * (*UriFuncReallocarray)(struct UriMemoryManagerStruct *, void *, s */ typedef void (*UriFuncFree)(struct UriMemoryManagerStruct *, void *); - /** * Class-like interface of custom memory managers * @@ -228,78 +220,87 @@ typedef void (*UriFuncFree)(struct UriMemoryManagerStruct *, void *); * @since 0.9.0 */ typedef struct UriMemoryManagerStruct { - UriFuncMalloc malloc; /**< Pointer to custom malloc(3) */ - UriFuncCalloc calloc; /**< Pointer to custom calloc(3); to emulate using malloc and memset see uriEmulateCalloc */ - UriFuncRealloc realloc; /**< Pointer to custom realloc(3) */ - UriFuncReallocarray reallocarray; /**< Pointer to custom reallocarray(3); to emulate using realloc see uriEmulateReallocarray */ - UriFuncFree free; /**< Pointer to custom free(3) */ - void * userData; /**< Pointer to data that the other function members need access to */ + UriFuncMalloc malloc; /**< Pointer to custom malloc(3) */ + UriFuncCalloc calloc; /**< Pointer to custom calloc(3); to emulate using malloc and + memset see uriEmulateCalloc */ + UriFuncRealloc realloc; /**< Pointer to custom realloc(3) */ + UriFuncReallocarray reallocarray; /**< Pointer to custom reallocarray(3); to emulate + using realloc see uriEmulateReallocarray */ + UriFuncFree free; /**< Pointer to custom free(3) */ + void * + userData; /**< Pointer to data that the other function members need access to */ } UriMemoryManager; /**< @copydoc UriMemoryManagerStruct */ - /** * Specifies a line break conversion mode. */ typedef enum UriBreakConversionEnum { - URI_BR_TO_LF, /**< Convert to Unix line breaks ("\\x0a") */ - URI_BR_TO_CRLF, /**< Convert to Windows line breaks ("\\x0d\\x0a") */ - URI_BR_TO_CR, /**< Convert to Macintosh line breaks ("\\x0d") */ - URI_BR_TO_UNIX = URI_BR_TO_LF, /**< @copydoc UriBreakConversionEnum::URI_BR_TO_LF */ - URI_BR_TO_WINDOWS = URI_BR_TO_CRLF, /**< @copydoc UriBreakConversionEnum::URI_BR_TO_CRLF */ - URI_BR_TO_MAC = URI_BR_TO_CR, /**< @copydoc UriBreakConversionEnum::URI_BR_TO_CR */ - URI_BR_DONT_TOUCH /**< Copy line breaks unmodified */ + URI_BR_TO_LF, /**< Convert to Unix line breaks ("\\x0a") */ + URI_BR_TO_CRLF, /**< Convert to Windows line breaks ("\\x0d\\x0a") */ + URI_BR_TO_CR, /**< Convert to Macintosh line breaks ("\\x0d") */ + URI_BR_TO_UNIX = URI_BR_TO_LF, /**< @copydoc UriBreakConversionEnum::URI_BR_TO_LF */ + URI_BR_TO_WINDOWS = + URI_BR_TO_CRLF, /**< @copydoc UriBreakConversionEnum::URI_BR_TO_CRLF */ + URI_BR_TO_MAC = URI_BR_TO_CR, /**< @copydoc UriBreakConversionEnum::URI_BR_TO_CR */ + URI_BR_DONT_TOUCH /**< Copy line breaks unmodified */ } UriBreakConversion; /**< @copydoc UriBreakConversionEnum */ - - /** * Specifies which component of a %URI has to be normalized. */ typedef enum UriNormalizationMaskEnum { - URI_NORMALIZED = 0, /**< Do not normalize anything */ - URI_NORMALIZE_SCHEME = 1 << 0, /**< Normalize scheme (fix uppercase letters) */ - URI_NORMALIZE_USER_INFO = 1 << 1, /**< Normalize user info (fix uppercase percent-encodings) */ - URI_NORMALIZE_HOST = 1 << 2, /**< Normalize host (fix uppercase letters) */ - URI_NORMALIZE_PATH = 1 << 3, /**< Normalize path (fix uppercase percent-encodings and redundant dot segments) */ - URI_NORMALIZE_QUERY = 1 << 4, /**< Normalize query (fix uppercase percent-encodings) */ - URI_NORMALIZE_FRAGMENT = 1 << 5 /**< Normalize fragment (fix uppercase percent-encodings) */ + URI_NORMALIZED = 0, /**< Do not normalize anything */ + URI_NORMALIZE_SCHEME = 1 << 0, /**< Normalize scheme (fix uppercase letters) */ + URI_NORMALIZE_USER_INFO = + 1 << 1, /**< Normalize user info (fix uppercase percent-encodings) */ + URI_NORMALIZE_HOST = 1 << 2, /**< Normalize host (fix uppercase letters) */ + URI_NORMALIZE_PATH = 1 << 3, /**< Normalize path (fix uppercase percent-encodings and + redundant dot segments) */ + URI_NORMALIZE_QUERY = + 1 << 4, /**< Normalize query (fix uppercase percent-encodings) */ + URI_NORMALIZE_FRAGMENT = + 1 << 5, /**< Normalize fragment (fix uppercase percent-encodings) */ + URI_NORMALIZE_PORT = 1 << 6 /**< Normalize port (drop leading zeros) @since 0.9.9 */ } UriNormalizationMask; /**< @copydoc UriNormalizationMaskEnum */ - - /** * Specifies how to resolve %URI references. */ typedef enum UriResolutionOptionsEnum { - URI_RESOLVE_STRICTLY = 0, /**< Full RFC conformance */ - URI_RESOLVE_IDENTICAL_SCHEME_COMPAT = 1 << 0 /**< Treat %URI to resolve with identical scheme as having no scheme */ + URI_RESOLVE_STRICTLY = 0, /**< Full RFC conformance */ + URI_RESOLVE_IDENTICAL_SCHEME_COMPAT = + 1 << 0 /**< Treat %URI to resolve with identical scheme as having no scheme */ } UriResolutionOptions; /**< @copydoc UriResolutionOptionsEnum */ - - /** - * Wraps a memory manager backend that only provides malloc and free - * to make a complete memory manager ready to be used. + * Wraps a memory manager backend that only provides malloc(3) and + * free(3) to make a complete memory manager ready to be used. * * The core feature of this wrapper is that you don't need to implement - * realloc if you don't want to. The wrapped memory manager uses - * backend->malloc, memcpy, and backend->free and soieof(size_t) extra - * bytes per allocation to emulate fallback realloc for you. + * realloc(3) if you don't want to. The wrapped memory manager uses + * backend->malloc, memcpy(3), and backend->free and + * (at least) sizeof(size_t) extra bytes per allocation to emulate + * fallback realloc(3) for you. * - * memory->calloc is uriEmulateCalloc. - * memory->free uses backend->free and handles the size header. - * memory->malloc uses backend->malloc and adds a size header. - * memory->realloc uses memory->malloc, memcpy, and memory->free and reads - * the size header. - * memory->reallocarray is uriEmulateReallocarray. + *
    + *
  • memory->calloc is uriEmulateCalloc.
    • + *
  • memory->free uses backend->free, + * and handles the size header.
    • + *
  • memory->malloc uses backend->malloc, + * and adds a size header.
    • + *
  • memory->realloc uses memory->malloc, + * memcpy(3) and memory->free, + * and reads the size header.
    • + *
  • memory->reallocarray is uriEmulateReallocarray.
    • + *
* - * The internal workings behind memory->free, memory->malloc, and - * memory->realloc may change so the functions exposed by these function - * pointer sshould be consided internal and not public API. + * The internal workings behind memory->free, memory->malloc, + * and memory->realloc may change, and the functions exposed by these + * function pointers should be considered internal and not public API. * * @param memory OUT: Where to write the wrapped memory manager to * @param backend IN: Memory manager to use as a backend - * @return Error code or 0 on success + * @return Error code or 0 on success * * @see uriEmulateCalloc * @see uriEmulateReallocarray @@ -307,12 +308,10 @@ typedef enum UriResolutionOptionsEnum { * @since 0.9.0 */ URI_PUBLIC int uriCompleteMemoryManager(UriMemoryManager * memory, - UriMemoryManager * backend); - - + UriMemoryManager * backend); /** - * Offers emulation of calloc(3) based on memory->malloc and memset. + * Offers emulation of calloc(3) based on memory->malloc and memset. * See "man 3 calloc" as well. * * @param memory IN: Memory manager to use, should not be NULL @@ -325,17 +324,14 @@ URI_PUBLIC int uriCompleteMemoryManager(UriMemoryManager * memory, * @see UriMemoryManager * @since 0.9.0 */ -URI_PUBLIC void * uriEmulateCalloc(UriMemoryManager * memory, - size_t nmemb, size_t size); - - +URI_PUBLIC void * uriEmulateCalloc(UriMemoryManager * memory, size_t nmemb, size_t size); /** - * Offers emulation of reallocarray(3) based on memory->realloc. + * Offers emulation of reallocarray(3) based on memory->realloc. * See "man 3 reallocarray" as well. * * @param memory IN: Memory manager to use, should not be NULL - * @param ptr IN: Pointer allocated using memory->malloc/... or NULL + * @param ptr IN: Pointer allocated using memory->malloc/... or NULL * @param nmemb IN: Number of elements to allocate * @param size IN: Size in bytes per element * @return Pointer to allocated memory or NULL @@ -345,10 +341,8 @@ URI_PUBLIC void * uriEmulateCalloc(UriMemoryManager * memory, * @see UriMemoryManager * @since 0.9.0 */ -URI_PUBLIC void * uriEmulateReallocarray(UriMemoryManager * memory, - void * ptr, size_t nmemb, size_t size); - - +URI_PUBLIC void * uriEmulateReallocarray(UriMemoryManager * memory, void * ptr, + size_t nmemb, size_t size); /** * Run multiple tests against a given memory manager. @@ -360,7 +354,11 @@ URI_PUBLIC void * uriEmulateReallocarray(UriMemoryManager * memory, * 5. and frees that memory. * * It is recommended to compile with AddressSanitizer enabled - * to take full advantage of uriTestMemoryManager. + * to take full advantage of uriTestMemoryManager. + * + * For backwards-compatibility, uriTestMemoryManager + * does not challenge pointer alignment; please see + * uriTestMemoryManagerEx for that feature. * * @param memory IN: Memory manager to use, should not be NULL * @return Error code or 0 on success @@ -368,10 +366,37 @@ URI_PUBLIC void * uriEmulateReallocarray(UriMemoryManager * memory, * @see uriEmulateCalloc * @see uriEmulateReallocarray * @see UriMemoryManager + * @see uriTestMemoryManagerEx * @since 0.9.0 */ URI_PUBLIC int uriTestMemoryManager(UriMemoryManager * memory); - +/** + * Run multiple tests against a given memory manager. + * For example, one test + * 1. allocates a small amount of memory, + * 2. writes some magic bytes to it, + * 3. reallocates it, + * 4. checks that previous values are still present, + * 5. and frees that memory. + * + * It is recommended to compile with both AddressSanitizer and + * UndefinedBehaviorSanitizer enabled to take full advantage of + * uriTestMemoryManagerEx. Note that environment variable + * UBSAN_OPTIONS may need adjustment to make UndefinedBehaviorSanitizer + * fatal (which by default it is not). + * + * @param memory IN: Memory manager to use, should not be NULL + * @param challengeAlignment IN: Whether to challenge pointer alignment + * @return Error code or 0 on success + * + * @see uriEmulateCalloc + * @see uriEmulateReallocarray + * @see UriMemoryManager + * @see uriTestMemoryManager + * @since 1.0.0 + */ +URI_PUBLIC int uriTestMemoryManagerEx(UriMemoryManager * memory, + UriBool challengeAlignment); #endif /* URI_BASE_H */ diff --git a/cpp/src/arrow/vendored/uriparser/UriCommon.c b/cpp/src/arrow/vendored/uriparser/UriCommon.c index 824893d7c2cb..7d99959afab2 100644 --- a/cpp/src/arrow/vendored/uriparser/UriCommon.c +++ b/cpp/src/arrow/vendored/uriparser/UriCommon.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -41,532 +41,725 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriCommon.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriCommon.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriCommon.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriCommon.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -# include "UriCommon.h" -#endif - - +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriSets.h" +# endif + +# include +# include +# include // SIZE_MAX /*extern*/ const URI_CHAR * const URI_FUNC(SafeToPointTo) = _UT("X"); /*extern*/ const URI_CHAR * const URI_FUNC(ConstPwd) = _UT("."); /*extern*/ const URI_CHAR * const URI_FUNC(ConstParent) = _UT(".."); - - void URI_FUNC(ResetUri)(URI_TYPE(Uri) * uri) { - if (uri == NULL) { - return; - } - memset(uri, 0, sizeof(URI_TYPE(Uri))); + if (uri == NULL) { + return; + } + memset(uri, 0, sizeof(URI_TYPE(Uri))); } - +int URI_FUNC(FreeUriPath)(URI_TYPE(Uri) * uri, UriMemoryManager * memory) { + assert(uri != NULL); + assert(memory != NULL); + + if (uri->pathHead != NULL) { + URI_TYPE(PathSegment) * segWalk = uri->pathHead; + while (segWalk != NULL) { + URI_TYPE(PathSegment) * const next = segWalk->next; + if ((uri->owner == URI_TRUE) + && (segWalk->text.first != segWalk->text.afterLast)) { + memory->free(memory, (URI_CHAR *)segWalk->text.first); + } + segWalk->text.first = NULL; + segWalk->text.afterLast = NULL; + segWalk->next = NULL; + memory->free(memory, segWalk); + segWalk = next; + } + uri->pathHead = NULL; + uri->pathTail = NULL; + } + + return URI_SUCCESS; +} /* Compares two text ranges for equal text content */ -int URI_FUNC(CompareRange)( - const URI_TYPE(TextRange) * a, - const URI_TYPE(TextRange) * b) { - int diff; - - /* NOTE: Both NULL means equal! */ - if ((a == NULL) || (b == NULL)) { - return ((a == NULL) ? 0 : 1) - ((b == NULL) ? 0 : 1); - } - - /* NOTE: Both NULL means equal! */ - if ((a->first == NULL) || (b->first == NULL)) { - return ((a->first == NULL) ? 0 : 1) - ((b->first == NULL) ? 0 : 1); - } - - diff = ((int)(a->afterLast - a->first) - (int)(b->afterLast - b->first)); - if (diff > 0) { - return 1; - } else if (diff < 0) { - return -1; - } - - diff = URI_STRNCMP(a->first, b->first, (a->afterLast - a->first)); - - if (diff > 0) { - return 1; - } else if (diff < 0) { - return -1; - } - - return diff; -} +bool URI_FUNC(RangeEquals)(const URI_TYPE(TextRange) * a, const URI_TYPE(TextRange) * b) { + /* NOTE: Both NULL means equal! */ + if ((a == NULL) || (b == NULL)) { + return a == b; + } + /* NOTE: Both NULL means equal! */ + if ((a->first == NULL) || (b->first == NULL)) { + return a->first == b->first; + } + const size_t lenA = a->afterLast - a->first; + const size_t lenB = b->afterLast - b->first; -/* Properly removes "." and ".." path segments */ -UriBool URI_FUNC(RemoveDotSegments)(URI_TYPE(Uri) * uri, - UriBool relative, UriMemoryManager * memory) { - if (uri == NULL) { - return URI_TRUE; - } - return URI_FUNC(RemoveDotSegmentsEx)(uri, relative, uri->owner, memory); -} - + if (lenA != lenB) { + return false; + } + return URI_STRNCMP(a->first, b->first, lenA) == 0; +} -UriBool URI_FUNC(RemoveDotSegmentsEx)(URI_TYPE(Uri) * uri, - UriBool relative, UriBool pathOwned, UriMemoryManager * memory) { - URI_TYPE(PathSegment) * walker; - if ((uri == NULL) || (uri->pathHead == NULL)) { - return URI_TRUE; - } - - walker = uri->pathHead; - walker->reserved = NULL; /* Prev pointer */ - do { - UriBool removeSegment = URI_FALSE; - int len = (int)(walker->text.afterLast - walker->text.first); - switch (len) { - case 1: - if ((walker->text.first)[0] == _UT('.')) { - /* "." segment -> remove if not essential */ - URI_TYPE(PathSegment) * const prev = walker->reserved; - URI_TYPE(PathSegment) * const nextBackup = walker->next; - - /* Is this dot segment essential? */ - removeSegment = URI_TRUE; - if (relative && (walker == uri->pathHead) && (walker->next != NULL)) { - const URI_CHAR * ch = walker->next->text.first; - for (; ch < walker->next->text.afterLast; ch++) { - if (*ch == _UT(':')) { - removeSegment = URI_FALSE; - break; - } - } - } - - if (removeSegment) { - /* Last segment? */ - if (walker->next != NULL) { - /* Not last segment */ - walker->next->reserved = prev; - - if (prev == NULL) { - /* First but not last segment */ - uri->pathHead = walker->next; - } else { - /* Middle segment */ - prev->next = walker->next; - } - - if (pathOwned && (walker->text.first != walker->text.afterLast)) { - memory->free(memory, (URI_CHAR *)walker->text.first); - } - memory->free(memory, walker); - } else { - /* Last segment */ - if (pathOwned && (walker->text.first != walker->text.afterLast)) { - memory->free(memory, (URI_CHAR *)walker->text.first); - } - - if (prev == NULL) { - /* Last and first */ - if (URI_FUNC(IsHostSet)(uri)) { - /* Replace "." with empty segment to represent trailing slash */ - walker->text.first = URI_FUNC(SafeToPointTo); - walker->text.afterLast = URI_FUNC(SafeToPointTo); - } else { - memory->free(memory, walker); - - uri->pathHead = NULL; - uri->pathTail = NULL; - } - } else { - /* Last but not first, replace "." with empty segment to represent trailing slash */ - walker->text.first = URI_FUNC(SafeToPointTo); - walker->text.afterLast = URI_FUNC(SafeToPointTo); - } - } - - walker = nextBackup; - } - } - break; - - case 2: - if (((walker->text.first)[0] == _UT('.')) - && ((walker->text.first)[1] == _UT('.'))) { - /* Path ".." -> remove this and the previous segment */ - URI_TYPE(PathSegment) * const prev = walker->reserved; - URI_TYPE(PathSegment) * prevPrev; - URI_TYPE(PathSegment) * const nextBackup = walker->next; - - removeSegment = URI_TRUE; - if (relative) { - if (prev == NULL) { - removeSegment = URI_FALSE; - } else if ((prev != NULL) - && ((prev->text.afterLast - prev->text.first) == 2) - && ((prev->text.first)[0] == _UT('.')) - && ((prev->text.first)[1] == _UT('.'))) { - removeSegment = URI_FALSE; - } - } - - if (removeSegment) { - if (prev != NULL) { - /* Not first segment */ - prevPrev = prev->reserved; - if (prevPrev != NULL) { - /* Not even prev is the first one */ - prevPrev->next = walker->next; - if (walker->next != NULL) { - walker->next->reserved = prevPrev; - } else { - /* Last segment -> insert "" segment to represent trailing slash, update tail */ - URI_TYPE(PathSegment) * const segment = memory->calloc(memory, 1, sizeof(URI_TYPE(PathSegment))); - if (segment == NULL) { - if (pathOwned && (walker->text.first != walker->text.afterLast)) { - memory->free(memory, (URI_CHAR *)walker->text.first); - } - memory->free(memory, walker); - - if (pathOwned && (prev->text.first != prev->text.afterLast)) { - memory->free(memory, (URI_CHAR *)prev->text.first); - } - memory->free(memory, prev); - - return URI_FALSE; /* Raises malloc error */ - } - segment->text.first = URI_FUNC(SafeToPointTo); - segment->text.afterLast = URI_FUNC(SafeToPointTo); - prevPrev->next = segment; - uri->pathTail = segment; - } - - if (pathOwned && (walker->text.first != walker->text.afterLast)) { - memory->free(memory, (URI_CHAR *)walker->text.first); - } - memory->free(memory, walker); - - if (pathOwned && (prev->text.first != prev->text.afterLast)) { - memory->free(memory, (URI_CHAR *)prev->text.first); - } - memory->free(memory, prev); - - walker = nextBackup; - } else { - /* Prev is the first segment */ - if (walker->next != NULL) { - uri->pathHead = walker->next; - walker->next->reserved = NULL; - - if (pathOwned && (walker->text.first != walker->text.afterLast)) { - memory->free(memory, (URI_CHAR *)walker->text.first); - } - memory->free(memory, walker); - } else { - /* Re-use segment for "" path segment to represent trailing slash, update tail */ - URI_TYPE(PathSegment) * const segment = walker; - if (pathOwned && (segment->text.first != segment->text.afterLast)) { - memory->free(memory, (URI_CHAR *)segment->text.first); - } - segment->text.first = URI_FUNC(SafeToPointTo); - segment->text.afterLast = URI_FUNC(SafeToPointTo); - uri->pathHead = segment; - uri->pathTail = segment; - } - - if (pathOwned && (prev->text.first != prev->text.afterLast)) { - memory->free(memory, (URI_CHAR *)prev->text.first); - } - memory->free(memory, prev); - - walker = nextBackup; - } - } else { - URI_TYPE(PathSegment) * const anotherNextBackup = walker->next; - /* First segment -> update head pointer */ - uri->pathHead = walker->next; - if (walker->next != NULL) { - walker->next->reserved = NULL; - } else { - /* Last segment -> update tail */ - uri->pathTail = NULL; - } - - if (pathOwned && (walker->text.first != walker->text.afterLast)) { - memory->free(memory, (URI_CHAR *)walker->text.first); - } - memory->free(memory, walker); - - walker = anotherNextBackup; - } - } - } - break; - - } - - if (!removeSegment) { - if (walker->next != NULL) { - walker->next->reserved = walker; - } else { - /* Last segment -> update tail */ - uri->pathTail = walker; - } - walker = walker->next; - } - } while (walker != NULL); - - return URI_TRUE; +UriBool URI_FUNC(CopyRange)(URI_TYPE(TextRange) * destRange, + const URI_TYPE(TextRange) * sourceRange, + UriMemoryManager * memory) { + const size_t lenInChars = sourceRange->afterLast - sourceRange->first; + if (lenInChars > SIZE_MAX / sizeof(URI_CHAR)) { // detect integer overflow + return URI_FALSE; + } + const size_t lenInBytes = lenInChars * sizeof(URI_CHAR); + URI_CHAR * dup = memory->malloc(memory, lenInBytes); + if (dup == NULL) { + return URI_FALSE; + } + memcpy(dup, sourceRange->first, lenInBytes); + destRange->first = dup; + destRange->afterLast = dup + lenInChars; + + return URI_TRUE; } +UriBool URI_FUNC(CopyRangeAsNeeded)(URI_TYPE(TextRange) * destRange, + const URI_TYPE(TextRange) * sourceRange, + UriMemoryManager * memory) { + if (sourceRange->first == NULL) { + destRange->first = NULL; + destRange->afterLast = NULL; + } else if (sourceRange->first == sourceRange->afterLast) { + destRange->first = URI_FUNC(SafeToPointTo); + destRange->afterLast = URI_FUNC(SafeToPointTo); + } else { + return URI_FUNC(CopyRange)(destRange, sourceRange, memory); + } + + return URI_TRUE; +} +UriBool URI_FUNC(RemoveDotSegmentsEx)(URI_TYPE(Uri) * uri, UriBool relative, + UriBool pathOwned, UriMemoryManager * memory) { + URI_TYPE(PathSegment) * walker; + if ((uri == NULL) || (uri->pathHead == NULL)) { + return URI_TRUE; + } + + walker = uri->pathHead; + walker->reserved = NULL; /* Prev pointer */ + do { + UriBool removeSegment = URI_FALSE; + const size_t len = walker->text.afterLast - walker->text.first; + switch (len) { + case 1: + if ((walker->text.first)[0] == _UT('.')) { + /* "." segment -> remove if not essential */ + URI_TYPE(PathSegment) * const prev = walker->reserved; + URI_TYPE(PathSegment) * const nextBackup = walker->next; + + /* + * Is this dot segment essential, + * i.e. is there a chance of changing semantics by dropping this dot + * segment? + * + * For example, changing "./http://foo" into "http://foo" would change + * semantics and hence the dot segment is essential to that case and + * cannot be removed. + * + * Other examples that would change semantics are: + * - cutting "/.//" down to "//" + * - cutting "scheme:/.//" down to "scheme://". + */ + removeSegment = URI_TRUE; + if ((walker == uri->pathHead) && (walker->next != NULL)) { + /* Detect case "/.//" (with or without scheme) */ + if ((walker->next->text.first == walker->next->text.afterLast) + && (URI_FUNC(HasHost)(uri) == URI_FALSE)) { + removeSegment = URI_FALSE; + /* Detect case "./withcolon:" */ + } else if (relative) { + const URI_CHAR * ch = walker->next->text.first; + for (; ch < walker->next->text.afterLast; ch++) { + if (*ch == _UT(':')) { + removeSegment = URI_FALSE; + break; + } + } + } + } + + if (removeSegment) { + /* .. then let's go remove that segment. */ + /* Last segment? */ + if (walker->next != NULL) { + /* Not last segment, i.e. first or middle segment + * OLD: (prev|NULL) <- walker <- next + * NEW: (prev|NULL) <----------- next */ + walker->next->reserved = prev; + + if (prev == NULL) { + /* First but not last segment + * OLD: head -> walker -> next + * NEW: head -----------> next */ + uri->pathHead = walker->next; + } else { + /* Middle segment + * OLD: prev -> walker -> next + * NEW: prev -----------> next */ + prev->next = walker->next; + } + + if (pathOwned && (walker->text.first != walker->text.afterLast)) { + memory->free(memory, (URI_CHAR *)walker->text.first); + } + memory->free(memory, walker); + } else { + /* Last segment */ + if (pathOwned && (walker->text.first != walker->text.afterLast)) { + memory->free(memory, (URI_CHAR *)walker->text.first); + } + + if (prev == NULL) { + /* Last and first */ + if (URI_FUNC(HasHost)(uri)) { + /* Replace "." with empty segment to represent trailing + * slash */ + walker->text.first = URI_FUNC(SafeToPointTo); + walker->text.afterLast = URI_FUNC(SafeToPointTo); + } else { + memory->free(memory, walker); + + uri->pathHead = NULL; + uri->pathTail = NULL; + } + } else { + /* Last but not first, replace "." with empty segment to + * represent trailing slash */ + walker->text.first = URI_FUNC(SafeToPointTo); + walker->text.afterLast = URI_FUNC(SafeToPointTo); + } + } + + walker = nextBackup; + } + } + break; + + case 2: + if (((walker->text.first)[0] == _UT('.')) + && ((walker->text.first)[1] == _UT('.'))) { + /* Path ".." -> remove this and the previous segment */ + URI_TYPE(PathSegment) * const prev = walker->reserved; + URI_TYPE(PathSegment) * prevPrev; + URI_TYPE(PathSegment) * const nextBackup = walker->next; + + removeSegment = URI_TRUE; + if (relative) { + if (prev == NULL) { + /* We cannot remove traversal beyond because the + * URI is relative and may be resolved later. + * So we can simplify "a/../b/d" to "b/d" but + * we cannot simplify "../b/d" (outside of reference resolution). + */ + removeSegment = URI_FALSE; + } else if ((prev != NULL) + && ((prev->text.afterLast - prev->text.first) == 2) + && ((prev->text.first)[0] == _UT('.')) + && ((prev->text.first)[1] == _UT('.'))) { + /* We need to protect against mis-simplifying "a/../../b" to + * "a/b". */ + removeSegment = URI_FALSE; + } + } + + if (removeSegment) { + if (prev != NULL) { + /* Not first segment */ + prevPrev = prev->reserved; + if (prevPrev != NULL) { + /* Not even prev is the first one + * OLD: prevPrev -> prev -> walker -> (next|NULL) + * NEW: prevPrev -------------------> (next|NULL) */ + prevPrev->next = walker->next; + if (walker->next != NULL) { + /* Update parent relationship as well + * OLD: prevPrev <- prev <- walker <- next + * NEW: prevPrev <------------------- next */ + walker->next->reserved = prevPrev; + } else { + /* Last segment -> insert "" segment to represent trailing + * slash, update tail */ + URI_TYPE(PathSegment) * const segment = memory->calloc( + memory, 1, sizeof(URI_TYPE(PathSegment))); + if (segment == NULL) { + if (pathOwned + && (walker->text.first + != walker->text.afterLast)) { + memory->free(memory, + (URI_CHAR *)walker->text.first); + } + memory->free(memory, walker); + + if (pathOwned + && (prev->text.first != prev->text.afterLast)) { + memory->free(memory, + (URI_CHAR *)prev->text.first); + } + memory->free(memory, prev); + + return URI_FALSE; /* Raises malloc error */ + } + segment->text.first = URI_FUNC(SafeToPointTo); + segment->text.afterLast = URI_FUNC(SafeToPointTo); + prevPrev->next = segment; + uri->pathTail = segment; + } + + if (pathOwned + && (walker->text.first != walker->text.afterLast)) { + memory->free(memory, (URI_CHAR *)walker->text.first); + } + memory->free(memory, walker); + + if (pathOwned && (prev->text.first != prev->text.afterLast)) { + memory->free(memory, (URI_CHAR *)prev->text.first); + } + memory->free(memory, prev); + + walker = nextBackup; + } else { + /* Prev is the first segment */ + if (walker->next != NULL) { + uri->pathHead = walker->next; + walker->next->reserved = NULL; + + if (pathOwned + && (walker->text.first != walker->text.afterLast)) { + memory->free(memory, (URI_CHAR *)walker->text.first); + } + memory->free(memory, walker); + } else { + /* Reuse segment for "" path segment to represent trailing + * slash, update tail */ + URI_TYPE(PathSegment) * const segment = walker; + if (pathOwned + && (segment->text.first != segment->text.afterLast)) { + memory->free(memory, (URI_CHAR *)segment->text.first); + } + segment->text.first = URI_FUNC(SafeToPointTo); + segment->text.afterLast = URI_FUNC(SafeToPointTo); + uri->pathHead = segment; + uri->pathTail = segment; + } + + if (pathOwned && (prev->text.first != prev->text.afterLast)) { + memory->free(memory, (URI_CHAR *)prev->text.first); + } + memory->free(memory, prev); + + walker = nextBackup; + } + } else { + URI_TYPE(PathSegment) * const anotherNextBackup = walker->next; + int freeWalker = URI_TRUE; + + /* First segment */ + if (walker->next != NULL) { + /* First segment of multiple -> update head + * OLD: head -> walker -> next + * NEW: head -----------> next */ + uri->pathHead = walker->next; + + /* Update parent link as well + * OLD: head <- walker <- next + * NEW: head <----------- next */ + walker->next->reserved = NULL; + } else { + if (uri->absolutePath) { + /* First and only segment -> update head + * OLD: head -> walker -> NULL + * NEW: head -----------> NULL */ + uri->pathHead = NULL; + + /* Last segment -> update tail + * OLD: tail -> walker + * NEW: tail -> NULL */ + uri->pathTail = NULL; + } else { + /* Reuse segment for "" path segment to represent trailing + * slash, then update head and tail */ + if (pathOwned + && (walker->text.first != walker->text.afterLast)) { + memory->free(memory, (URI_CHAR *)walker->text.first); + } + walker->text.first = URI_FUNC(SafeToPointTo); + walker->text.afterLast = URI_FUNC(SafeToPointTo); + freeWalker = URI_FALSE; + } + } + + if (freeWalker) { + if (pathOwned + && (walker->text.first != walker->text.afterLast)) { + memory->free(memory, (URI_CHAR *)walker->text.first); + } + memory->free(memory, walker); + } + + walker = anotherNextBackup; + } + } + } + break; + } /* end of switch */ + + if (!removeSegment) { + /* .. then let's move to the next element, and start again. */ + if (walker->next != NULL) { + walker->next->reserved = walker; + } else { + /* Last segment -> update tail */ + uri->pathTail = walker; + } + walker = walker->next; + } + } while (walker != NULL); + + return URI_TRUE; +} /* Properly removes "." and ".." path segments */ UriBool URI_FUNC(RemoveDotSegmentsAbsolute)(URI_TYPE(Uri) * uri, - UriMemoryManager * memory) { - const UriBool ABSOLUTE = URI_FALSE; - return URI_FUNC(RemoveDotSegments)(uri, ABSOLUTE, memory); + UriMemoryManager * memory) { + const UriBool ABSOLUTE = URI_FALSE; + if (uri == NULL) { + return URI_TRUE; + } + return URI_FUNC(RemoveDotSegmentsEx)(uri, ABSOLUTE, uri->owner, memory); } - - unsigned char URI_FUNC(HexdigToInt)(URI_CHAR hexdig) { - switch (hexdig) { - case _UT('0'): - case _UT('1'): - case _UT('2'): - case _UT('3'): - case _UT('4'): - case _UT('5'): - case _UT('6'): - case _UT('7'): - case _UT('8'): - case _UT('9'): - return (unsigned char)(9 + hexdig - _UT('9')); - - case _UT('a'): - case _UT('b'): - case _UT('c'): - case _UT('d'): - case _UT('e'): - case _UT('f'): - return (unsigned char)(15 + hexdig - _UT('f')); - - case _UT('A'): - case _UT('B'): - case _UT('C'): - case _UT('D'): - case _UT('E'): - case _UT('F'): - return (unsigned char)(15 + hexdig - _UT('F')); - - default: - return 0; - } -} - - - -URI_CHAR URI_FUNC(HexToLetter)(unsigned int value) { - /* Uppercase recommended in section 2.1. of RFC 3986 * - * http://tools.ietf.org/html/rfc3986#section-2.1 */ - return URI_FUNC(HexToLetterEx)(value, URI_TRUE); + switch (hexdig) { + case URI_SET_DIGIT(_UT): + return (unsigned char)(9 + hexdig - _UT('9')); + case URI_SET_HEX_LETTER_LOWER(_UT): + return (unsigned char)(15 + hexdig - _UT('f')); + case URI_SET_HEX_LETTER_UPPER(_UT): + return (unsigned char)(15 + hexdig - _UT('F')); + + default: + return 0; + } } - - URI_CHAR URI_FUNC(HexToLetterEx)(unsigned int value, UriBool uppercase) { - switch (value) { - case 0: return _UT('0'); - case 1: return _UT('1'); - case 2: return _UT('2'); - case 3: return _UT('3'); - case 4: return _UT('4'); - case 5: return _UT('5'); - case 6: return _UT('6'); - case 7: return _UT('7'); - case 8: return _UT('8'); - case 9: return _UT('9'); - - case 10: return (uppercase == URI_TRUE) ? _UT('A') : _UT('a'); - case 11: return (uppercase == URI_TRUE) ? _UT('B') : _UT('b'); - case 12: return (uppercase == URI_TRUE) ? _UT('C') : _UT('c'); - case 13: return (uppercase == URI_TRUE) ? _UT('D') : _UT('d'); - case 14: return (uppercase == URI_TRUE) ? _UT('E') : _UT('e'); - default: return (uppercase == URI_TRUE) ? _UT('F') : _UT('f'); - } + switch (value) { + case 0: + return _UT('0'); + case 1: + return _UT('1'); + case 2: + return _UT('2'); + case 3: + return _UT('3'); + case 4: + return _UT('4'); + case 5: + return _UT('5'); + case 6: + return _UT('6'); + case 7: + return _UT('7'); + case 8: + return _UT('8'); + case 9: + return _UT('9'); + + case 10: + return (uppercase == URI_TRUE) ? _UT('A') : _UT('a'); + case 11: + return (uppercase == URI_TRUE) ? _UT('B') : _UT('b'); + case 12: + return (uppercase == URI_TRUE) ? _UT('C') : _UT('c'); + case 13: + return (uppercase == URI_TRUE) ? _UT('D') : _UT('d'); + case 14: + return (uppercase == URI_TRUE) ? _UT('E') : _UT('e'); + default: + return (uppercase == URI_TRUE) ? _UT('F') : _UT('f'); + } } - - /* Checks if a URI has the host component set. */ -UriBool URI_FUNC(IsHostSet)(const URI_TYPE(Uri) * uri) { - return (uri != NULL) - && ((uri->hostText.first != NULL) - || (uri->hostData.ip4 != NULL) - || (uri->hostData.ip6 != NULL) - || (uri->hostData.ipFuture.first != NULL) - ); +UriBool URI_FUNC(HasHost)(const URI_TYPE(Uri) * uri) { + /* NOTE: .hostData.ipFuture.first is not being checked, * + * because we do check .hostText.first and * + * .hostData.ipFuture.first has to be identical to * + * .hostText.first if set, and hence there is * + * no more information to be gained. */ + return (uri != NULL) + && ((uri->hostText.first != NULL) || (uri->hostData.ip4 != NULL) + || (uri->hostData.ip6 != NULL)); } +/* Copies the path segment list from one URI to another. */ +UriBool URI_FUNC(CopyPath)(URI_TYPE(Uri) * dest, const URI_TYPE(Uri) * source, + UriMemoryManager * memory) { + if (source->pathHead == NULL) { + /* No path component */ + dest->pathHead = NULL; + dest->pathTail = NULL; + } else { + /* Copy list but not the text contained */ + URI_TYPE(PathSegment) * sourceWalker = source->pathHead; + URI_TYPE(PathSegment) * destPrev = NULL; + do { + URI_TYPE(PathSegment) * cur = + memory->malloc(memory, sizeof(URI_TYPE(PathSegment))); + if (cur == NULL) { + /* Fix broken list */ + if (destPrev != NULL) { + destPrev->next = NULL; + } + return URI_FALSE; /* Raises malloc error */ + } + + /* From this functions usage we know that * + * the dest URI cannot be uri->owner */ + cur->text = sourceWalker->text; + if (destPrev == NULL) { + /* First segment ever */ + dest->pathHead = cur; + } else { + destPrev->next = cur; + } + destPrev = cur; + sourceWalker = sourceWalker->next; + } while (sourceWalker != NULL); + dest->pathTail = destPrev; + dest->pathTail->next = NULL; + } + + dest->absolutePath = source->absolutePath; + return URI_TRUE; +} +/* Copies the authority part of an URI over to another. */ +UriBool URI_FUNC(CopyAuthority)(URI_TYPE(Uri) * dest, const URI_TYPE(Uri) * source, + UriMemoryManager * memory) { + /* From this functions usage we know that * + * the dest URI cannot be uri->owner */ + + /* Copy userInfo */ + dest->userInfo = source->userInfo; + + /* Copy hostText */ + dest->hostText = source->hostText; + + /* Copy hostData */ + if (source->hostData.ip4 != NULL) { + dest->hostData.ip4 = memory->malloc(memory, sizeof(UriIp4)); + if (dest->hostData.ip4 == NULL) { + return URI_FALSE; /* Raises malloc error */ + } + *(dest->hostData.ip4) = *(source->hostData.ip4); + dest->hostData.ip6 = NULL; + dest->hostData.ipFuture.first = NULL; + dest->hostData.ipFuture.afterLast = NULL; + } else if (source->hostData.ip6 != NULL) { + dest->hostData.ip4 = NULL; + dest->hostData.ip6 = memory->malloc(memory, sizeof(UriIp6)); + if (dest->hostData.ip6 == NULL) { + return URI_FALSE; /* Raises malloc error */ + } + *(dest->hostData.ip6) = *(source->hostData.ip6); + dest->hostData.ipFuture.first = NULL; + dest->hostData.ipFuture.afterLast = NULL; + } else { + dest->hostData.ip4 = NULL; + dest->hostData.ip6 = NULL; + dest->hostData.ipFuture = source->hostData.ipFuture; + } + + /* Copy portText */ + dest->portText = source->portText; + + return URI_TRUE; +} -/* Copies the path segment list from one URI to another. */ -UriBool URI_FUNC(CopyPath)(URI_TYPE(Uri) * dest, - const URI_TYPE(Uri) * source, UriMemoryManager * memory) { - if (source->pathHead == NULL) { - /* No path component */ - dest->pathHead = NULL; - dest->pathTail = NULL; - } else { - /* Copy list but not the text contained */ - URI_TYPE(PathSegment) * sourceWalker = source->pathHead; - URI_TYPE(PathSegment) * destPrev = NULL; - do { - URI_TYPE(PathSegment) * cur = memory->malloc(memory, sizeof(URI_TYPE(PathSegment))); - if (cur == NULL) { - /* Fix broken list */ - if (destPrev != NULL) { - destPrev->next = NULL; - } - return URI_FALSE; /* Raises malloc error */ - } - - /* From this functions usage we know that * - * the dest URI cannot be uri->owner */ - cur->text = sourceWalker->text; - if (destPrev == NULL) { - /* First segment ever */ - dest->pathHead = cur; - } else { - destPrev->next = cur; - } - destPrev = cur; - sourceWalker = sourceWalker->next; - } while (sourceWalker != NULL); - dest->pathTail = destPrev; - dest->pathTail->next = NULL; - } - - dest->absolutePath = source->absolutePath; - return URI_TRUE; +UriBool URI_FUNC(FixAmbiguity)(URI_TYPE(Uri) * uri, UriMemoryManager * memory) { + URI_TYPE(PathSegment) * segment; + + if (/* Case 1: absolute path, empty first segment */ + (uri->absolutePath && (uri->pathHead != NULL) + && (uri->pathHead->text.afterLast == uri->pathHead->text.first)) + + /* Case 2: relative path, empty first and second segment */ + || (!uri->absolutePath && (uri->pathHead != NULL) && (uri->pathHead->next != NULL) + && (uri->pathHead->text.afterLast == uri->pathHead->text.first) + && (uri->pathHead->next->text.afterLast + == uri->pathHead->next->text.first))) { + /* NOOP */ + } else { + return URI_TRUE; + } + + segment = memory->malloc(memory, 1 * sizeof(URI_TYPE(PathSegment))); + if (segment == NULL) { + return URI_FALSE; /* Raises malloc error */ + } + + /* Insert "." segment in front */ + segment->next = uri->pathHead; + segment->text.first = URI_FUNC(ConstPwd); + segment->text.afterLast = URI_FUNC(ConstPwd) + 1; + uri->pathHead = segment; + return URI_TRUE; } +static UriBool URI_FUNC(PrependNewDotSegment)(URI_TYPE(Uri) * uri, + UriMemoryManager * memory) { + assert(uri != NULL); + assert(memory != NULL); + URI_TYPE(PathSegment) * const segment = + memory->malloc(memory, 1 * sizeof(URI_TYPE(PathSegment))); -/* Copies the authority part of an URI over to another. */ -UriBool URI_FUNC(CopyAuthority)(URI_TYPE(Uri) * dest, - const URI_TYPE(Uri) * source, UriMemoryManager * memory) { - /* From this functions usage we know that * - * the dest URI cannot be uri->owner */ - - /* Copy userInfo */ - dest->userInfo = source->userInfo; - - /* Copy hostText */ - dest->hostText = source->hostText; - - /* Copy hostData */ - if (source->hostData.ip4 != NULL) { - dest->hostData.ip4 = memory->malloc(memory, sizeof(UriIp4)); - if (dest->hostData.ip4 == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - *(dest->hostData.ip4) = *(source->hostData.ip4); - dest->hostData.ip6 = NULL; - dest->hostData.ipFuture.first = NULL; - dest->hostData.ipFuture.afterLast = NULL; - } else if (source->hostData.ip6 != NULL) { - dest->hostData.ip4 = NULL; - dest->hostData.ip6 = memory->malloc(memory, sizeof(UriIp6)); - if (dest->hostData.ip6 == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - *(dest->hostData.ip6) = *(source->hostData.ip6); - dest->hostData.ipFuture.first = NULL; - dest->hostData.ipFuture.afterLast = NULL; - } else { - dest->hostData.ip4 = NULL; - dest->hostData.ip6 = NULL; - dest->hostData.ipFuture = source->hostData.ipFuture; - } - - /* Copy portText */ - dest->portText = source->portText; - - return URI_TRUE; -} + if (segment == NULL) { + return URI_FALSE; /* i.e. raise malloc error */ + } + segment->next = uri->pathHead; + URI_TYPE(TextRange) dotRange; + dotRange.first = URI_FUNC(ConstPwd); + dotRange.afterLast = URI_FUNC(ConstPwd) + 1; -UriBool URI_FUNC(FixAmbiguity)(URI_TYPE(Uri) * uri, - UriMemoryManager * memory) { - URI_TYPE(PathSegment) * segment; - - if ( /* Case 1: absolute path, empty first segment */ - (uri->absolutePath - && (uri->pathHead != NULL) - && (uri->pathHead->text.afterLast == uri->pathHead->text.first)) - - /* Case 2: relative path, empty first and second segment */ - || (!uri->absolutePath - && (uri->pathHead != NULL) - && (uri->pathHead->next != NULL) - && (uri->pathHead->text.afterLast == uri->pathHead->text.first) - && (uri->pathHead->next->text.afterLast == uri->pathHead->next->text.first))) { - /* NOOP */ - } else { - return URI_TRUE; - } - - segment = memory->malloc(memory, 1 * sizeof(URI_TYPE(PathSegment))); - if (segment == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - - /* Insert "." segment in front */ - segment->next = uri->pathHead; - segment->text.first = URI_FUNC(ConstPwd); - segment->text.afterLast = URI_FUNC(ConstPwd) + 1; - uri->pathHead = segment; - return URI_TRUE; -} + if (uri->owner == URI_TRUE) { + if (URI_FUNC(CopyRange)(&(segment->text), &dotRange, memory) == URI_FALSE) { + memory->free(memory, segment); + return URI_FALSE; /* i.e. raise malloc error */ + } + } else { + segment->text = dotRange; /* copies all members */ + } + uri->pathHead = segment; + return URI_TRUE; +} -void URI_FUNC(FixEmptyTrailSegment)(URI_TYPE(Uri) * uri, - UriMemoryManager * memory) { - /* Fix path if only one empty segment */ - if (!uri->absolutePath - && !URI_FUNC(IsHostSet)(uri) - && (uri->pathHead != NULL) - && (uri->pathHead->next == NULL) - && (uri->pathHead->text.first == uri->pathHead->text.afterLast)) { - memory->free(memory, uri->pathHead); - uri->pathHead = NULL; - uri->pathTail = NULL; - } +/* When dropping a scheme from a URI without a host and with a colon (":") + * in the first path segment, a consecutive reparse would rightfully + * mis-classify the first path segment as a scheme due to the colon. + * To protect against this case, we prepend an artificial "." segment + * to the path in here; the function is called after the scheme has + * just been dropped. + * + * 0. We start with parsed URI "scheme:path1:/path2/path3". + * 1. We drop the scheme naively and yield "path1:/path2/path3". + * 2. We prepend "." and yield unambiguous "./path1:/path2/path3". + * + * From the view of the RFC 3986 grammar, this is replacing rule path-rootless + * by path-noscheme content. + * + * Returns URI_TRUE for (a) nothing to do or (b) successful changes. + * Returns URI_FALSE to signal out-of-memory. + */ +UriBool URI_FUNC(FixPathNoScheme)(URI_TYPE(Uri) * uri, UriMemoryManager * memory) { + assert(uri != NULL); + assert(memory != NULL); + + if ((uri->absolutePath == URI_TRUE) || (uri->pathHead == NULL) + || (uri->scheme.first != NULL) || URI_FUNC(HasHost)(uri)) { + return URI_TRUE; /* i.e. nothing to do */ + } + + /* Check for troublesome first path segment containing a colon */ + UriBool colonFound = URI_FALSE; + const URI_CHAR * walker = uri->pathHead->text.first; + + while (walker < uri->pathHead->text.afterLast) { + if (walker[0] == _UT(':')) { + colonFound = URI_TRUE; + break; + } + walker++; + } + + assert((walker == uri->pathHead->text.afterLast) || (colonFound == URI_TRUE)); + + if (colonFound == URI_FALSE) { + return URI_TRUE; /* i.e. nothing to do */ + } + + /* Insert "." segment in front */ + return URI_FUNC(PrependNewDotSegment)(uri, memory); } +/* When dropping a host from a URI without a scheme, an absolute path + * and empty first path segment, a consecutive reparse would rightfully + * mis-classify the first path segment as a host marker due to the "//". + * To protect against this case, we prepend an artificial "." segment + * to the path in here; the function is called after the host has + * just been dropped. + * + * 0. We start with parsed URI "//host//path1/path2". + * 1. We drop the host naively and yield "//path1/path2". + * 2. We insert "./" and yield unambiguous "/.//path1/path2". + * + * Returns URI_TRUE for (a) nothing to do or (b) successful changes. + * Returns URI_FALSE to signal out-of-memory. + */ +UriBool URI_FUNC(EnsureThatPathIsNotMistakenForHost)(URI_TYPE(Uri) * uri, + UriMemoryManager * memory) { + assert(uri != NULL); + assert(memory != NULL); + + if ((URI_FUNC(HasHost)(uri) == URI_TRUE) || (uri->absolutePath == URI_FALSE) + || (uri->pathHead == NULL) + || (uri->pathHead == uri->pathTail) /* i.e. no second slash */ + || (uri->pathHead->text.first != uri->pathHead->text.afterLast)) { + return URI_TRUE; /* i.e. nothing to do */ + } + + /* Insert "." segment in front */ + return URI_FUNC(PrependNewDotSegment)(uri, memory); +} +void URI_FUNC(FixEmptyTrailSegment)(URI_TYPE(Uri) * uri, UriMemoryManager * memory) { + /* Fix path if only one empty segment */ + if (!uri->absolutePath && !URI_FUNC(HasHost)(uri) && (uri->pathHead != NULL) + && (uri->pathHead->next == NULL) + && (uri->pathHead->text.first == uri->pathHead->text.afterLast)) { + memory->free(memory, uri->pathHead); + uri->pathHead = NULL; + uri->pathTail = NULL; + } +} #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriCommon.h b/cpp/src/arrow/vendored/uriparser/UriCommon.h index 97b997a9afa3..34d9ddaecda5 100644 --- a/cpp/src/arrow/vendored/uriparser/UriCommon.h +++ b/cpp/src/arrow/vendored/uriparser/UriCommon.h @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -38,35 +38,36 @@ */ #if (defined(URI_PASS_ANSI) && !defined(URI_COMMON_H_ANSI)) \ - || (defined(URI_PASS_UNICODE) && !defined(URI_COMMON_H_UNICODE)) \ - || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) + || (defined(URI_PASS_UNICODE) && !defined(URI_COMMON_H_UNICODE)) \ + || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* What encodings are enabled? */ -#include "UriDefsConfig.h" -#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +# include "UriDefsConfig.h" +# if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriCommon.h" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriCommon.h" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriCommon.h" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriCommon.h" +# undef URI_PASS_UNICODE +# endif /* Only one pass for each encoding */ -#elif (defined(URI_PASS_ANSI) && !defined(URI_COMMON_H_ANSI) \ - && defined(URI_ENABLE_ANSI)) || (defined(URI_PASS_UNICODE) \ - && !defined(URI_COMMON_H_UNICODE) && defined(URI_ENABLE_UNICODE)) -# ifdef URI_PASS_ANSI -# define URI_COMMON_H_ANSI 1 -# include "UriDefsAnsi.h" -# else -# define URI_COMMON_H_UNICODE 1 -# include "UriDefsUnicode.h" -# endif - +# elif (defined(URI_PASS_ANSI) && !defined(URI_COMMON_H_ANSI) \ + && defined(URI_ENABLE_ANSI)) \ + || (defined(URI_PASS_UNICODE) && !defined(URI_COMMON_H_UNICODE) \ + && defined(URI_ENABLE_UNICODE)) +# ifdef URI_PASS_ANSI +# define URI_COMMON_H_ANSI 1 +# include "UriDefsAnsi.h" +# else +# define URI_COMMON_H_UNICODE 1 +# include "UriDefsUnicode.h" +# endif +# include /* Used to point to from empty path segments. * X.first and X.afterLast must be the same non-NULL value then. */ @@ -74,36 +75,37 @@ extern const URI_CHAR * const URI_FUNC(SafeToPointTo); extern const URI_CHAR * const URI_FUNC(ConstPwd); extern const URI_CHAR * const URI_FUNC(ConstParent); +void URI_FUNC(ResetUri)(URI_TYPE(Uri) * uri); +int URI_FUNC(FreeUriPath)(URI_TYPE(Uri) * uri, UriMemoryManager * memory); -void URI_FUNC(ResetUri)(URI_TYPE(Uri) * uri); +bool URI_FUNC(RangeEquals)(const URI_TYPE(TextRange) * a, const URI_TYPE(TextRange) * b); -int URI_FUNC(CompareRange)( - const URI_TYPE(TextRange) * a, - const URI_TYPE(TextRange) * b); +UriBool URI_FUNC(CopyRange)(URI_TYPE(TextRange) * destRange, + const URI_TYPE(TextRange) * sourceRange, + UriMemoryManager * memory); +UriBool URI_FUNC(CopyRangeAsNeeded)(URI_TYPE(TextRange) * destRange, + const URI_TYPE(TextRange) * sourceRange, + UriMemoryManager * memory); UriBool URI_FUNC(RemoveDotSegmentsAbsolute)(URI_TYPE(Uri) * uri, - UriMemoryManager * memory); -UriBool URI_FUNC(RemoveDotSegments)(URI_TYPE(Uri) * uri, UriBool relative, - UriMemoryManager * memory); -UriBool URI_FUNC(RemoveDotSegmentsEx)(URI_TYPE(Uri) * uri, - UriBool relative, UriBool pathOwned, UriMemoryManager * memory); + UriMemoryManager * memory); +UriBool URI_FUNC(RemoveDotSegmentsEx)(URI_TYPE(Uri) * uri, UriBool relative, + UriBool pathOwned, UriMemoryManager * memory); unsigned char URI_FUNC(HexdigToInt)(URI_CHAR hexdig); -URI_CHAR URI_FUNC(HexToLetter)(unsigned int value); URI_CHAR URI_FUNC(HexToLetterEx)(unsigned int value, UriBool uppercase); -UriBool URI_FUNC(IsHostSet)(const URI_TYPE(Uri) * uri); - UriBool URI_FUNC(CopyPath)(URI_TYPE(Uri) * dest, const URI_TYPE(Uri) * source, - UriMemoryManager * memory); -UriBool URI_FUNC(CopyAuthority)(URI_TYPE(Uri) * dest, - const URI_TYPE(Uri) * source, UriMemoryManager * memory); + UriMemoryManager * memory); +UriBool URI_FUNC(CopyAuthority)(URI_TYPE(Uri) * dest, const URI_TYPE(Uri) * source, + UriMemoryManager * memory); UriBool URI_FUNC(FixAmbiguity)(URI_TYPE(Uri) * uri, UriMemoryManager * memory); -void URI_FUNC(FixEmptyTrailSegment)(URI_TYPE(Uri) * uri, - UriMemoryManager * memory); - +UriBool URI_FUNC(FixPathNoScheme)(URI_TYPE(Uri) * uri, UriMemoryManager * memory); +UriBool URI_FUNC(EnsureThatPathIsNotMistakenForHost)(URI_TYPE(Uri) * uri, + UriMemoryManager * memory); +void URI_FUNC(FixEmptyTrailSegment)(URI_TYPE(Uri) * uri, UriMemoryManager * memory); -#endif +# endif #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriCompare.c b/cpp/src/arrow/vendored/uriparser/UriCompare.c index e63573a1f718..f331d3dda616 100644 --- a/cpp/src/arrow/vendored/uriparser/UriCompare.c +++ b/cpp/src/arrow/vendored/uriparser/UriCompare.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -41,128 +41,120 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriCompare.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriCompare.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriCompare.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriCompare.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -# include "UriIp4.h" -# include "UriCommon.h" -#endif - - - -UriBool URI_FUNC(EqualsUri)(const URI_TYPE(Uri) * a, - const URI_TYPE(Uri) * b) { - /* NOTE: Both NULL means equal! */ - if ((a == NULL) || (b == NULL)) { - return ((a == NULL) && (b == NULL)) ? URI_TRUE : URI_FALSE; - } - - /* scheme */ - if (URI_FUNC(CompareRange)(&(a->scheme), &(b->scheme))) { - return URI_FALSE; - } - - /* absolutePath */ - if ((a->scheme.first == NULL)&& (a->absolutePath != b->absolutePath)) { - return URI_FALSE; - } - - /* userInfo */ - if (URI_FUNC(CompareRange)(&(a->userInfo), &(b->userInfo))) { - return URI_FALSE; - } - - /* Host */ - if (((a->hostData.ip4 == NULL) != (b->hostData.ip4 == NULL)) - || ((a->hostData.ip6 == NULL) != (b->hostData.ip6 == NULL)) - || ((a->hostData.ipFuture.first == NULL) - != (b->hostData.ipFuture.first == NULL))) { - return URI_FALSE; - } - - if (a->hostData.ip4 != NULL) { - if (memcmp(a->hostData.ip4->data, b->hostData.ip4->data, 4)) { - return URI_FALSE; - } - } - - if (a->hostData.ip6 != NULL) { - if (memcmp(a->hostData.ip6->data, b->hostData.ip6->data, 16)) { - return URI_FALSE; - } - } - - if (a->hostData.ipFuture.first != NULL) { - if (URI_FUNC(CompareRange)(&(a->hostData.ipFuture), &(b->hostData.ipFuture))) { - return URI_FALSE; - } - } - - if ((a->hostData.ip4 == NULL) - && (a->hostData.ip6 == NULL) - && (a->hostData.ipFuture.first == NULL)) { - if (URI_FUNC(CompareRange)(&(a->hostText), &(b->hostText))) { - return URI_FALSE; - } - } - - /* portText */ - if (URI_FUNC(CompareRange)(&(a->portText), &(b->portText))) { - return URI_FALSE; - } - - /* Path */ - if ((a->pathHead == NULL) != (b->pathHead == NULL)) { - return URI_FALSE; - } - - if (a->pathHead != NULL) { - URI_TYPE(PathSegment) * walkA = a->pathHead; - URI_TYPE(PathSegment) * walkB = b->pathHead; - do { - if (URI_FUNC(CompareRange)(&(walkA->text), &(walkB->text))) { - return URI_FALSE; - } - if ((walkA->next == NULL) != (walkB->next == NULL)) { - return URI_FALSE; - } - walkA = walkA->next; - walkB = walkB->next; - } while (walkA != NULL); - } - - /* query */ - if (URI_FUNC(CompareRange)(&(a->query), &(b->query))) { - return URI_FALSE; - } - - /* fragment */ - if (URI_FUNC(CompareRange)(&(a->fragment), &(b->fragment))) { - return URI_FALSE; - } - - return URI_TRUE; /* Equal*/ +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriIp4.h" +# include "UriCommon.h" +# endif + +UriBool URI_FUNC(EqualsUri)(const URI_TYPE(Uri) * a, const URI_TYPE(Uri) * b) { + /* NOTE: Both NULL means equal! */ + if ((a == NULL) || (b == NULL)) { + return ((a == NULL) && (b == NULL)) ? URI_TRUE : URI_FALSE; + } + + /* scheme */ + if (!URI_FUNC(RangeEquals)(&(a->scheme), &(b->scheme))) { + return URI_FALSE; + } + + /* absolutePath -- not meaningful for URIs with a host set! */ + if (!URI_FUNC(HasHost)(a) && (a->absolutePath != b->absolutePath)) { + return URI_FALSE; + } + + /* userInfo */ + if (!URI_FUNC(RangeEquals)(&(a->userInfo), &(b->userInfo))) { + return URI_FALSE; + } + + /* Host */ + if (((a->hostData.ip4 == NULL) != (b->hostData.ip4 == NULL)) + || ((a->hostData.ip6 == NULL) != (b->hostData.ip6 == NULL)) + || ((a->hostData.ipFuture.first == NULL) + != (b->hostData.ipFuture.first == NULL))) { + return URI_FALSE; + } + + if (a->hostData.ip4 != NULL) { + if (memcmp(a->hostData.ip4->data, b->hostData.ip4->data, 4)) { + return URI_FALSE; + } + } + + if (a->hostData.ip6 != NULL) { + if (memcmp(a->hostData.ip6->data, b->hostData.ip6->data, 16)) { + return URI_FALSE; + } + } + + if (a->hostData.ipFuture.first != NULL) { + if (!URI_FUNC(RangeEquals)(&(a->hostData.ipFuture), &(b->hostData.ipFuture))) { + return URI_FALSE; + } + } + + if ((a->hostData.ip4 == NULL) && (a->hostData.ip6 == NULL) + && (a->hostData.ipFuture.first == NULL)) { + if (!URI_FUNC(RangeEquals)(&(a->hostText), &(b->hostText))) { + return URI_FALSE; + } + } + + /* portText */ + if (!URI_FUNC(RangeEquals)(&(a->portText), &(b->portText))) { + return URI_FALSE; + } + + /* Path */ + if ((a->pathHead == NULL) != (b->pathHead == NULL)) { + return URI_FALSE; + } + + if (a->pathHead != NULL) { + URI_TYPE(PathSegment) * walkA = a->pathHead; + URI_TYPE(PathSegment) * walkB = b->pathHead; + do { + if (!URI_FUNC(RangeEquals)(&(walkA->text), &(walkB->text))) { + return URI_FALSE; + } + if ((walkA->next == NULL) != (walkB->next == NULL)) { + return URI_FALSE; + } + walkA = walkA->next; + walkB = walkB->next; + } while (walkA != NULL); + } + + /* query */ + if (!URI_FUNC(RangeEquals)(&(a->query), &(b->query))) { + return URI_FALSE; + } + + /* fragment */ + if (!URI_FUNC(RangeEquals)(&(a->fragment), &(b->fragment))) { + return URI_FALSE; + } + + return URI_TRUE; /* Equal*/ } - - #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriConfig.h b/cpp/src/arrow/vendored/uriparser/UriConfig.h new file mode 100644 index 000000000000..7428e3fa817e --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriConfig.h @@ -0,0 +1,47 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2018, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#if !defined(URI_CONFIG_H) +# define URI_CONFIG_H 1 + +# define PACKAGE_VERSION "1.0.2" + +// #define HAVE_WPRINTF +// #define HAVE_REALLOCARRAY + +#endif /* !defined(URI_CONFIG_H) */ diff --git a/cpp/src/arrow/vendored/uriparser/UriCopy.c b/cpp/src/arrow/vendored/uriparser/UriCopy.c new file mode 100644 index 000000000000..1cb01f9236bd --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriCopy.c @@ -0,0 +1,248 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2007, Weijia Song + * Copyright (C) 2007, Sebastian Pipping + * Copyright (C) 2025, Máté Kocsis + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/** + * @file UriCopy.c + * Holds the RFC 3986 %URI normalization implementation. + * NOTE: This source file includes itself twice. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriCopy.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriCopy.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriMemory.h" +# include "UriNormalize.h" +# include "UriCopy.h" +# endif + +static void URI_FUNC(PreventLeakageAfterCopy)(URI_TYPE(Uri) * uri, + unsigned int revertMask, + UriMemoryManager * memory) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + + if (uri->hostData.ip4 != NULL) { + memory->free(memory, uri->hostData.ip4); + uri->hostData.ip4 = NULL; + } else if (uri->hostData.ip6 != NULL) { + memory->free(memory, uri->hostData.ip6); + uri->hostData.ip6 = NULL; + } + + if (revertMask & URI_NORMALIZE_PORT) { + if (uri->portText.first != uri->portText.afterLast) { + memory->free(memory, (URI_CHAR *)uri->portText.first); + } + uri->portText.first = NULL; + uri->portText.afterLast = NULL; + } +} + +int URI_FUNC(CopyUriMm)(URI_TYPE(Uri) * destUri, const URI_TYPE(Uri) * sourceUri, + UriMemoryManager * memory) { + unsigned int revertMask = URI_NORMALIZED; + + if (sourceUri == NULL || destUri == NULL) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + URI_FUNC(ResetUri)(destUri); + + if (URI_FUNC(CopyRangeAsNeeded)(&destUri->scheme, &sourceUri->scheme, memory) + == URI_FALSE) { + return URI_ERROR_MALLOC; + } + + revertMask |= URI_NORMALIZE_SCHEME; + + if (URI_FUNC(CopyRangeAsNeeded)(&destUri->userInfo, &sourceUri->userInfo, memory) + == URI_FALSE) { + URI_FUNC(PreventLeakageAfterCopy)(destUri, revertMask, memory); + return URI_ERROR_MALLOC; + } + + revertMask |= URI_NORMALIZE_USER_INFO; + + if (URI_FUNC(CopyRangeAsNeeded)(&destUri->hostText, &sourceUri->hostText, memory) + == URI_FALSE) { + URI_FUNC(PreventLeakageAfterCopy)(destUri, revertMask, memory); + return URI_ERROR_MALLOC; + } + + revertMask |= URI_NORMALIZE_HOST; + + if (sourceUri->hostData.ip4 == NULL) { + destUri->hostData.ip4 = NULL; + } else { + destUri->hostData.ip4 = memory->malloc(memory, sizeof(UriIp4)); + if (destUri->hostData.ip4 == NULL) { + URI_FUNC(PreventLeakageAfterCopy)(destUri, revertMask, memory); + return URI_ERROR_MALLOC; + } + *(destUri->hostData.ip4) = *(sourceUri->hostData.ip4); + } + + if (sourceUri->hostData.ip6 == NULL) { + destUri->hostData.ip6 = NULL; + } else { + destUri->hostData.ip6 = memory->malloc(memory, sizeof(UriIp6)); + if (destUri->hostData.ip6 == NULL) { + URI_FUNC(PreventLeakageAfterCopy)(destUri, revertMask, memory); + return URI_ERROR_MALLOC; + } + *(destUri->hostData.ip6) = *(sourceUri->hostData.ip6); + } + + if (sourceUri->hostData.ipFuture.first != NULL) { + destUri->hostData.ipFuture.first = destUri->hostText.first; + destUri->hostData.ipFuture.afterLast = destUri->hostText.afterLast; + } else if (URI_FUNC(CopyRangeAsNeeded)(&destUri->hostData.ipFuture, + &sourceUri->hostData.ipFuture, memory) + == URI_FALSE) { + URI_FUNC(PreventLeakageAfterCopy)(destUri, revertMask, memory); + return URI_ERROR_MALLOC; + } + + if (URI_FUNC(CopyRangeAsNeeded)(&destUri->portText, &sourceUri->portText, memory) + == URI_FALSE) { + URI_FUNC(PreventLeakageAfterCopy)(destUri, revertMask, memory); + return URI_ERROR_MALLOC; + } + + revertMask |= URI_NORMALIZE_PORT; + + destUri->pathHead = NULL; + destUri->pathTail = NULL; + + if (sourceUri->pathHead != NULL) { + URI_TYPE(PathSegment) * sourceWalker = sourceUri->pathHead; + URI_TYPE(PathSegment) * destPrev = NULL; + + while (sourceWalker != NULL) { + URI_TYPE(PathSegment) * destWalker = + memory->malloc(memory, sizeof(URI_TYPE(PathSegment))); + if (destWalker == NULL) { + URI_FUNC(PreventLeakageAfterCopy)(destUri, revertMask, memory); + return URI_ERROR_MALLOC; + } + + destWalker->text.first = NULL; + destWalker->text.afterLast = NULL; + destWalker->next = NULL; + destWalker->reserved = NULL; + + if (destUri->pathHead == NULL) { + destUri->pathHead = destWalker; + revertMask |= URI_NORMALIZE_PATH; + } + + if (URI_FUNC(CopyRangeAsNeeded)(&destWalker->text, &sourceWalker->text, + memory) + == URI_FALSE) { + // Unless wired to `destUri` above, `destWalker` may be hanging + // in the air now + if (destUri->pathHead != destWalker) { + memory->free(memory, destWalker); + } + + URI_FUNC(PreventLeakageAfterCopy)(destUri, revertMask, memory); + return URI_ERROR_MALLOC; + } + + if (destPrev != NULL) { + destPrev->next = destWalker; + } + + destPrev = destWalker; + sourceWalker = sourceWalker->next; + + destUri->pathTail = destWalker; + } + } + + if (URI_FUNC(CopyRangeAsNeeded)(&destUri->query, &sourceUri->query, memory) + == URI_FALSE) { + URI_FUNC(PreventLeakageAfterCopy)(destUri, revertMask, memory); + return URI_ERROR_MALLOC; + } + + revertMask |= URI_NORMALIZE_QUERY; + + if (URI_FUNC(CopyRangeAsNeeded)(&destUri->fragment, &sourceUri->fragment, memory) + == URI_FALSE) { + URI_FUNC(PreventLeakageAfterCopy)(destUri, revertMask, memory); + return URI_ERROR_MALLOC; + } + + destUri->absolutePath = sourceUri->absolutePath; + destUri->owner = URI_TRUE; + destUri->reserved = NULL; + + return URI_SUCCESS; +} + +int URI_FUNC(CopyUri)(URI_TYPE(Uri) * destUri, const URI_TYPE(Uri) * sourceUri) { + return URI_FUNC(CopyUriMm)(destUri, sourceUri, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriCopy.h b/cpp/src/arrow/vendored/uriparser/UriCopy.h new file mode 100644 index 000000000000..94bab76f6048 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriCopy.h @@ -0,0 +1,76 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2007, Weijia Song + * Copyright (C) 2007, Sebastian Pipping + * Copyright (C) 2025, Máté Kocsis + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#if (defined(URI_PASS_ANSI) && !defined(URI_COPY_H_ANSI)) \ + || (defined(URI_PASS_UNICODE) && !defined(URI_COPY_H_UNICODE)) \ + || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* What encodings are enabled? */ +# include "UriDefsConfig.h" +# if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriCopy.h" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriCopy.h" +# undef URI_PASS_UNICODE +# endif +/* Only one pass for each encoding */ +# elif (defined(URI_PASS_ANSI) && !defined(URI_COPY_H_ANSI) \ + && defined(URI_ENABLE_ANSI)) \ + || (defined(URI_PASS_UNICODE) && !defined(URI_COPY_H_UNICODE) \ + && defined(URI_ENABLE_UNICODE)) +# ifdef URI_PASS_ANSI +# define URI_COPY_H_ANSI 1 +# include "UriDefsAnsi.h" +# else +# define URI_COPY_H_UNICODE 1 +# include "UriDefsUnicode.h" +# endif + +int URI_FUNC(CopyUriMm)(URI_TYPE(Uri) * destUri, const URI_TYPE(Uri) * sourceUri, + UriMemoryManager * memory); +int URI_FUNC(CopyUri)(URI_TYPE(Uri) * destUri, const URI_TYPE(Uri) * sourceUri); + +# endif +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriDefsAnsi.h b/cpp/src/arrow/vendored/uriparser/UriDefsAnsi.h index d6dbcad1b2a7..17aed1b3b59e 100644 --- a/cpp/src/arrow/vendored/uriparser/UriDefsAnsi.h +++ b/cpp/src/arrow/vendored/uriparser/UriDefsAnsi.h @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -46,24 +46,18 @@ /* Allow multi inclusion */ #include "UriDefsConfig.h" - - #undef URI_CHAR #define URI_CHAR char #undef _UT #define _UT(x) x - - #undef URI_FUNC #define URI_FUNC(x) uri##x##A #undef URI_TYPE #define URI_TYPE(x) Uri##x##A - - #undef URI_STRLEN #define URI_STRLEN strlen #undef URI_STRCPY @@ -76,7 +70,7 @@ /* TODO Remove on next source-compatibility break */ #undef URI_SNPRINTF #if (defined(__WIN32__) || defined(_WIN32) || defined(WIN32)) -# define URI_SNPRINTF _snprintf +# define URI_SNPRINTF _snprintf #else -# define URI_SNPRINTF snprintf +# define URI_SNPRINTF snprintf #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriDefsConfig.h b/cpp/src/arrow/vendored/uriparser/UriDefsConfig.h index d604494b002c..79d485d90c38 100644 --- a/cpp/src/arrow/vendored/uriparser/UriDefsConfig.h +++ b/cpp/src/arrow/vendored/uriparser/UriDefsConfig.h @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -43,60 +43,51 @@ */ #ifndef URI_DEFS_CONFIG_H -#define URI_DEFS_CONFIG_H 1 - - +# define URI_DEFS_CONFIG_H 1 /* Deny external overriding */ -#undef URI_ENABLE_ANSI /* Internal for !URI_NO_ANSI */ -#undef URI_ENABLE_UNICODE /* Internal for !URI_NO_UNICODE */ - - +# undef URI_ENABLE_ANSI /* Internal for !URI_NO_ANSI */ +# undef URI_ENABLE_UNICODE /* Internal for !URI_NO_UNICODE */ /* Encoding */ -#ifdef URI_NO_ANSI -# ifdef URI_NO_UNICODE +# ifdef URI_NO_ANSI +# ifdef URI_NO_UNICODE /* No encoding at all */ -# error URI_NO_ANSI and URI_NO_UNICODE cannot go together. -# else -/* Unicode only */ -# define URI_ENABLE_UNICODE 1 -# endif -#else -# ifdef URI_NO_UNICODE -/* ANSI only */ -# define URI_ENABLE_ANSI 1 -# else -/* Both ANSI and Unicode */ -# define URI_ENABLE_ANSI 1 -# define URI_ENABLE_UNICODE 1 -# endif -#endif - - +# error URI_NO_ANSI and URI_NO_UNICODE cannot go together. +# else +/* Wide strings only */ +# define URI_ENABLE_UNICODE 1 +# endif +# else +# ifdef URI_NO_UNICODE +/* Narrow strings only */ +# define URI_ENABLE_ANSI 1 +# else +/* Both narrow and wide strings */ +# define URI_ENABLE_ANSI 1 +# define URI_ENABLE_UNICODE 1 +# endif +# endif /* Function inlining, not ANSI/ISO C! */ -#if defined(URI_DOXYGEN) -# define URI_INLINE -#elif defined(__INTEL_COMPILER) +# if defined(URI_DOXYGEN) +# define URI_INLINE +# elif defined(__INTEL_COMPILER) /* Intel C/C++ */ /* http://predef.sourceforge.net/precomp.html#sec20 */ /* http://www.intel.com/support/performancetools/c/windows/sb/CS-007751.htm#2 */ -/* EDIT 11/5/20. Intel changed __force_inline to __forceinline */ -# define URI_INLINE __forceinline -#elif defined(_MSC_VER) +# define URI_INLINE __forceinline +# elif defined(_MSC_VER) /* Microsoft Visual C++ */ /* http://predef.sourceforge.net/precomp.html#sec32 */ /* http://msdn2.microsoft.com/en-us/library/ms882281.aspx */ -# define URI_INLINE __forceinline -#elif (defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)) +# define URI_INLINE __forceinline +# elif (defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)) /* C99, "inline" is a keyword */ -# define URI_INLINE inline -#else +# define URI_INLINE inline +# else /* No inlining */ -# define URI_INLINE -#endif - - +# define URI_INLINE +# endif #endif /* URI_DEFS_CONFIG_H */ diff --git a/cpp/src/arrow/vendored/uriparser/UriDefsUnicode.h b/cpp/src/arrow/vendored/uriparser/UriDefsUnicode.h index 8bb8bc2bec31..6fce6725e69b 100644 --- a/cpp/src/arrow/vendored/uriparser/UriDefsUnicode.h +++ b/cpp/src/arrow/vendored/uriparser/UriDefsUnicode.h @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -39,31 +39,25 @@ /** * @file UriDefsUnicode.h - * Holds definitions for the Unicode pass. + * Holds definitions for the wide string pass. * NOTE: This header is included N times, not once. */ /* Allow multi inclusion */ #include "UriDefsConfig.h" - - #undef URI_CHAR #define URI_CHAR wchar_t #undef _UT #define _UT(x) L##x - - #undef URI_FUNC #define URI_FUNC(x) uri##x##W #undef URI_TYPE #define URI_TYPE(x) Uri##x##W - - #undef URI_STRLEN #define URI_STRLEN wcslen #undef URI_STRCPY @@ -76,7 +70,7 @@ /* TODO Remove on next source-compatibility break */ #undef URI_SNPRINTF #if (defined(__WIN32__) || defined(_WIN32) || defined(WIN32)) -# define URI_SNPRINTF _snwprintf +# define URI_SNPRINTF _snwprintf #else -# define URI_SNPRINTF swprintf +# define URI_SNPRINTF swprintf #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriEscape.c b/cpp/src/arrow/vendored/uriparser/UriEscape.c index 46f3fd4ace56..8fc434e3a8b1 100644 --- a/cpp/src/arrow/vendored/uriparser/UriEscape.c +++ b/cpp/src/arrow/vendored/uriparser/UriEscape.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -41,413 +41,293 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriEscape.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriEscape.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriEscape.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriEscape.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -# include "UriCommon.h" -#endif - - - -URI_CHAR * URI_FUNC(Escape)(const URI_CHAR * in, URI_CHAR * out, - UriBool spaceToPlus, UriBool normalizeBreaks) { - return URI_FUNC(EscapeEx)(in, NULL, out, spaceToPlus, normalizeBreaks); +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriSets.h" +# endif + +URI_CHAR * URI_FUNC(Escape)(const URI_CHAR * in, URI_CHAR * out, UriBool spaceToPlus, + UriBool normalizeBreaks) { + return URI_FUNC(EscapeEx)(in, NULL, out, spaceToPlus, normalizeBreaks); } - - -URI_CHAR * URI_FUNC(EscapeEx)(const URI_CHAR * inFirst, - const URI_CHAR * inAfterLast, URI_CHAR * out, - UriBool spaceToPlus, UriBool normalizeBreaks) { - const URI_CHAR * read = inFirst; - URI_CHAR * write = out; - UriBool prevWasCr = URI_FALSE; - if ((out == NULL) || (inFirst == out)) { - return NULL; - } else if (inFirst == NULL) { - if (out != NULL) { - out[0] = _UT('\0'); - } - return out; - } - - for (;;) { - if ((inAfterLast != NULL) && (read >= inAfterLast)) { - write[0] = _UT('\0'); - return write; - } - - switch (read[0]) { - case _UT('\0'): - write[0] = _UT('\0'); - return write; - - case _UT(' '): - if (spaceToPlus) { - write[0] = _UT('+'); - write++; - } else { - write[0] = _UT('%'); - write[1] = _UT('2'); - write[2] = _UT('0'); - write += 3; - } - prevWasCr = URI_FALSE; - break; - - case _UT('a'): /* ALPHA */ - case _UT('A'): - case _UT('b'): - case _UT('B'): - case _UT('c'): - case _UT('C'): - case _UT('d'): - case _UT('D'): - case _UT('e'): - case _UT('E'): - case _UT('f'): - case _UT('F'): - case _UT('g'): - case _UT('G'): - case _UT('h'): - case _UT('H'): - case _UT('i'): - case _UT('I'): - case _UT('j'): - case _UT('J'): - case _UT('k'): - case _UT('K'): - case _UT('l'): - case _UT('L'): - case _UT('m'): - case _UT('M'): - case _UT('n'): - case _UT('N'): - case _UT('o'): - case _UT('O'): - case _UT('p'): - case _UT('P'): - case _UT('q'): - case _UT('Q'): - case _UT('r'): - case _UT('R'): - case _UT('s'): - case _UT('S'): - case _UT('t'): - case _UT('T'): - case _UT('u'): - case _UT('U'): - case _UT('v'): - case _UT('V'): - case _UT('w'): - case _UT('W'): - case _UT('x'): - case _UT('X'): - case _UT('y'): - case _UT('Y'): - case _UT('z'): - case _UT('Z'): - case _UT('0'): /* DIGIT */ - case _UT('1'): - case _UT('2'): - case _UT('3'): - case _UT('4'): - case _UT('5'): - case _UT('6'): - case _UT('7'): - case _UT('8'): - case _UT('9'): - case _UT('-'): /* "-" / "." / "_" / "~" */ - case _UT('.'): - case _UT('_'): - case _UT('~'): - /* Copy unmodified */ - write[0] = read[0]; - write++; - - prevWasCr = URI_FALSE; - break; - - case _UT('\x0a'): - if (normalizeBreaks) { - if (!prevWasCr) { - write[0] = _UT('%'); - write[1] = _UT('0'); - write[2] = _UT('D'); - write[3] = _UT('%'); - write[4] = _UT('0'); - write[5] = _UT('A'); - write += 6; - } - } else { - write[0] = _UT('%'); - write[1] = _UT('0'); - write[2] = _UT('A'); - write += 3; - } - prevWasCr = URI_FALSE; - break; - - case _UT('\x0d'): - if (normalizeBreaks) { - write[0] = _UT('%'); - write[1] = _UT('0'); - write[2] = _UT('D'); - write[3] = _UT('%'); - write[4] = _UT('0'); - write[5] = _UT('A'); - write += 6; - } else { - write[0] = _UT('%'); - write[1] = _UT('0'); - write[2] = _UT('D'); - write += 3; - } - prevWasCr = URI_TRUE; - break; - - default: - /* Percent encode */ - { - const unsigned char code = (unsigned char)read[0]; - write[0] = _UT('%'); - write[1] = URI_FUNC(HexToLetter)(code >> 4); - write[2] = URI_FUNC(HexToLetter)(code & 0x0f); - write += 3; - } - prevWasCr = URI_FALSE; - break; - } - - read++; - } +URI_CHAR * URI_FUNC(EscapeEx)(const URI_CHAR * inFirst, const URI_CHAR * inAfterLast, + URI_CHAR * out, UriBool spaceToPlus, + UriBool normalizeBreaks) { + const URI_CHAR * read = inFirst; + URI_CHAR * write = out; + UriBool prevWasCr = URI_FALSE; + if ((out == NULL) || (inFirst == out)) { + return NULL; + } else if (inFirst == NULL) { + if (out != NULL) { + out[0] = _UT('\0'); + } + return out; + } + + for (;;) { + if ((inAfterLast != NULL) && (read >= inAfterLast)) { + write[0] = _UT('\0'); + return write; + } + + switch (read[0]) { + case _UT('\0'): + write[0] = _UT('\0'); + return write; + + case _UT(' '): + if (spaceToPlus) { + write[0] = _UT('+'); + write++; + } else { + write[0] = _UT('%'); + write[1] = _UT('2'); + write[2] = _UT('0'); + write += 3; + } + prevWasCr = URI_FALSE; + break; + + case URI_SET_UNRESERVED(_UT): + /* Copy unmodified */ + write[0] = read[0]; + write++; + + prevWasCr = URI_FALSE; + break; + + case _UT('\x0a'): + if (normalizeBreaks) { + if (!prevWasCr) { + write[0] = _UT('%'); + write[1] = _UT('0'); + write[2] = _UT('D'); + write[3] = _UT('%'); + write[4] = _UT('0'); + write[5] = _UT('A'); + write += 6; + } + } else { + write[0] = _UT('%'); + write[1] = _UT('0'); + write[2] = _UT('A'); + write += 3; + } + prevWasCr = URI_FALSE; + break; + + case _UT('\x0d'): + if (normalizeBreaks) { + write[0] = _UT('%'); + write[1] = _UT('0'); + write[2] = _UT('D'); + write[3] = _UT('%'); + write[4] = _UT('0'); + write[5] = _UT('A'); + write += 6; + } else { + write[0] = _UT('%'); + write[1] = _UT('0'); + write[2] = _UT('D'); + write += 3; + } + prevWasCr = URI_TRUE; + break; + + default: + /* Percent encode */ + { + const unsigned char code = (unsigned char)read[0]; + /* Uppercase recommended in (last sentence of) section 2.1 * + * of RFC 3986: * + * https://datatracker.ietf.org/doc/html/rfc3986#section-2.1 */ + write[0] = _UT('%'); + write[1] = URI_FUNC(HexToLetterEx)(code >> 4, URI_TRUE); + write[2] = URI_FUNC(HexToLetterEx)(code & 0x0f, URI_TRUE); + write += 3; + } + prevWasCr = URI_FALSE; + break; + } + + read++; + } } - - const URI_CHAR * URI_FUNC(UnescapeInPlace)(URI_CHAR * inout) { - return URI_FUNC(UnescapeInPlaceEx)(inout, URI_FALSE, URI_BR_DONT_TOUCH); + return URI_FUNC(UnescapeInPlaceEx)(inout, URI_FALSE, URI_BR_DONT_TOUCH); } - - -const URI_CHAR * URI_FUNC(UnescapeInPlaceEx)(URI_CHAR * inout, - UriBool plusToSpace, UriBreakConversion breakConversion) { - URI_CHAR * read = inout; - URI_CHAR * write = inout; - UriBool prevWasCr = URI_FALSE; - - if (inout == NULL) { - return NULL; - } - - for (;;) { - switch (read[0]) { - case _UT('\0'): - if (read > write) { - write[0] = _UT('\0'); - } - return write; - - case _UT('%'): - switch (read[1]) { - case _UT('0'): - case _UT('1'): - case _UT('2'): - case _UT('3'): - case _UT('4'): - case _UT('5'): - case _UT('6'): - case _UT('7'): - case _UT('8'): - case _UT('9'): - case _UT('a'): - case _UT('b'): - case _UT('c'): - case _UT('d'): - case _UT('e'): - case _UT('f'): - case _UT('A'): - case _UT('B'): - case _UT('C'): - case _UT('D'): - case _UT('E'): - case _UT('F'): - switch (read[2]) { - case _UT('0'): - case _UT('1'): - case _UT('2'): - case _UT('3'): - case _UT('4'): - case _UT('5'): - case _UT('6'): - case _UT('7'): - case _UT('8'): - case _UT('9'): - case _UT('a'): - case _UT('b'): - case _UT('c'): - case _UT('d'): - case _UT('e'): - case _UT('f'): - case _UT('A'): - case _UT('B'): - case _UT('C'): - case _UT('D'): - case _UT('E'): - case _UT('F'): - { - /* Percent group found */ - const unsigned char left = URI_FUNC(HexdigToInt)(read[1]); - const unsigned char right = URI_FUNC(HexdigToInt)(read[2]); - const int code = 16 * left + right; - switch (code) { - case 10: - switch (breakConversion) { - case URI_BR_TO_LF: - if (!prevWasCr) { - write[0] = (URI_CHAR)10; - write++; - } - break; - - case URI_BR_TO_CRLF: - if (!prevWasCr) { - write[0] = (URI_CHAR)13; - write[1] = (URI_CHAR)10; - write += 2; - } - break; - - case URI_BR_TO_CR: - if (!prevWasCr) { - write[0] = (URI_CHAR)13; - write++; - } - break; - - case URI_BR_DONT_TOUCH: - default: - write[0] = (URI_CHAR)10; - write++; - - } - prevWasCr = URI_FALSE; - break; - - case 13: - switch (breakConversion) { - case URI_BR_TO_LF: - write[0] = (URI_CHAR)10; - write++; - break; - - case URI_BR_TO_CRLF: - write[0] = (URI_CHAR)13; - write[1] = (URI_CHAR)10; - write += 2; - break; - - case URI_BR_TO_CR: - write[0] = (URI_CHAR)13; - write++; - break; - - case URI_BR_DONT_TOUCH: - default: - write[0] = (URI_CHAR)13; - write++; - - } - prevWasCr = URI_TRUE; - break; - - default: - write[0] = (URI_CHAR)(code); - write++; - - prevWasCr = URI_FALSE; - - } - read += 3; - } - break; - - default: - /* Copy two chars unmodified and */ - /* look at this char again */ - if (read > write) { - write[0] = read[0]; - write[1] = read[1]; - } - read += 2; - write += 2; - - prevWasCr = URI_FALSE; - } - break; - - default: - /* Copy one char unmodified and */ - /* look at this char again */ - if (read > write) { - write[0] = read[0]; - } - read++; - write++; - - prevWasCr = URI_FALSE; - } - break; - - case _UT('+'): - if (plusToSpace) { - /* Convert '+' to ' ' */ - write[0] = _UT(' '); - } else { - /* Copy one char unmodified */ - if (read > write) { - write[0] = read[0]; - } - } - read++; - write++; - - prevWasCr = URI_FALSE; - break; - - default: - /* Copy one char unmodified */ - if (read > write) { - write[0] = read[0]; - } - read++; - write++; - - prevWasCr = URI_FALSE; - } - } +const URI_CHAR * URI_FUNC(UnescapeInPlaceEx)(URI_CHAR * inout, UriBool plusToSpace, + UriBreakConversion breakConversion) { + URI_CHAR * read = inout; + URI_CHAR * write = inout; + UriBool prevWasCr = URI_FALSE; + + if (inout == NULL) { + return NULL; + } + + for (;;) { + switch (read[0]) { + case _UT('\0'): + if (read > write) { + write[0] = _UT('\0'); + } + return write; + + case _UT('%'): + switch (read[1]) { + case URI_SET_HEXDIG(_UT): + switch (read[2]) { + case URI_SET_HEXDIG(_UT): { + /* Percent group found */ + const unsigned char left = URI_FUNC(HexdigToInt)(read[1]); + const unsigned char right = URI_FUNC(HexdigToInt)(read[2]); + const int code = 16 * left + right; + switch (code) { + case 10: + switch (breakConversion) { + case URI_BR_TO_LF: + if (!prevWasCr) { + write[0] = (URI_CHAR)10; + write++; + } + break; + + case URI_BR_TO_CRLF: + if (!prevWasCr) { + write[0] = (URI_CHAR)13; + write[1] = (URI_CHAR)10; + write += 2; + } + break; + + case URI_BR_TO_CR: + if (!prevWasCr) { + write[0] = (URI_CHAR)13; + write++; + } + break; + + case URI_BR_DONT_TOUCH: + default: + write[0] = (URI_CHAR)10; + write++; + } + prevWasCr = URI_FALSE; + break; + + case 13: + switch (breakConversion) { + case URI_BR_TO_LF: + write[0] = (URI_CHAR)10; + write++; + break; + + case URI_BR_TO_CRLF: + write[0] = (URI_CHAR)13; + write[1] = (URI_CHAR)10; + write += 2; + break; + + case URI_BR_TO_CR: + write[0] = (URI_CHAR)13; + write++; + break; + + case URI_BR_DONT_TOUCH: + default: + write[0] = (URI_CHAR)13; + write++; + } + prevWasCr = URI_TRUE; + break; + + default: + write[0] = (URI_CHAR)(code); + write++; + + prevWasCr = URI_FALSE; + } + read += 3; + } break; + + default: + /* Copy two chars unmodified and */ + /* look at this char again */ + if (read > write) { + write[0] = read[0]; + write[1] = read[1]; + } + read += 2; + write += 2; + + prevWasCr = URI_FALSE; + } + break; + + default: + /* Copy one char unmodified and */ + /* look at this char again */ + if (read > write) { + write[0] = read[0]; + } + read++; + write++; + + prevWasCr = URI_FALSE; + } + break; + + case _UT('+'): + if (plusToSpace) { + /* Convert '+' to ' ' */ + write[0] = _UT(' '); + } else { + /* Copy one char unmodified */ + if (read > write) { + write[0] = read[0]; + } + } + read++; + write++; + + prevWasCr = URI_FALSE; + break; + + default: + /* Copy one char unmodified */ + if (read > write) { + write[0] = read[0]; + } + read++; + write++; + + prevWasCr = URI_FALSE; + } + } } - - #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriFile.c b/cpp/src/arrow/vendored/uriparser/UriFile.c index abaa84f9adf8..8373fc4c818b 100644 --- a/cpp/src/arrow/vendored/uriparser/UriFile.c +++ b/cpp/src/arrow/vendored/uriparser/UriFile.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -41,202 +41,193 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriFile.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriFile.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriFile.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriFile.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -#endif - - - -#include /* for size_t, avoiding stddef.h for older MSVCs */ +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif +# ifndef URI_DOXYGEN +# include "Uri.h" +# endif +# include // size_t +# include // SIZE_MAX static URI_INLINE int URI_FUNC(FilenameToUriString)(const URI_CHAR * filename, - URI_CHAR * uriString, UriBool fromUnix) { - const URI_CHAR * input = filename; - const URI_CHAR * lastSep = input - 1; - UriBool firstSegment = URI_TRUE; - URI_CHAR * output = uriString; - UriBool absolute; - UriBool is_windows_network; - - if ((filename == NULL) || (uriString == NULL)) { - return URI_ERROR_NULL; - } - - is_windows_network = (filename[0] == _UT('\\')) && (filename[1] == _UT('\\')); - absolute = fromUnix - ? (filename[0] == _UT('/')) - : (((filename[0] != _UT('\0')) && (filename[1] == _UT(':'))) - || is_windows_network); - - if (absolute) { - const URI_CHAR * const prefix = fromUnix - ? _UT("file://") - : is_windows_network - ? _UT("file:") - : _UT("file:///"); - const size_t prefixLen = URI_STRLEN(prefix); - - /* Copy prefix */ - memcpy(uriString, prefix, prefixLen * sizeof(URI_CHAR)); - output += prefixLen; - } - - /* Copy and escape on the fly */ - for (;;) { - if ((input[0] == _UT('\0')) - || (fromUnix && input[0] == _UT('/')) - || (!fromUnix && input[0] == _UT('\\'))) { - /* Copy text after last separator */ - if (lastSep + 1 < input) { - if (!fromUnix && absolute && (firstSegment == URI_TRUE)) { - /* Quick hack to not convert "C:" to "C%3A" */ - const int charsToCopy = (int)(input - (lastSep + 1)); - memcpy(output, lastSep + 1, charsToCopy * sizeof(URI_CHAR)); - output += charsToCopy; - } else { - output = URI_FUNC(EscapeEx)(lastSep + 1, input, output, - URI_FALSE, URI_FALSE); - } - } - firstSegment = URI_FALSE; - } - - if (input[0] == _UT('\0')) { - output[0] = _UT('\0'); - break; - } else if (fromUnix && (input[0] == _UT('/'))) { - /* Copy separators unmodified */ - output[0] = _UT('/'); - output++; - lastSep = input; - } else if (!fromUnix && (input[0] == _UT('\\'))) { - /* Convert backslashes to forward slashes */ - output[0] = _UT('/'); - output++; - lastSep = input; - } - input++; - } - - return URI_SUCCESS; + URI_CHAR * uriString, + UriBool fromUnix) { + const URI_CHAR * input = filename; + const URI_CHAR * lastSep = input - 1; + UriBool firstSegment = URI_TRUE; + URI_CHAR * output = uriString; + UriBool absolute; + UriBool is_windows_network; + + if ((filename == NULL) || (uriString == NULL)) { + return URI_ERROR_NULL; + } + + is_windows_network = (filename[0] == _UT('\\')) && (filename[1] == _UT('\\')); + absolute = fromUnix ? (filename[0] == _UT('/')) + : (((filename[0] != _UT('\0')) && (filename[1] == _UT(':'))) + || is_windows_network); + + if (absolute) { + const URI_CHAR * const prefix = fromUnix ? _UT("file://") + : is_windows_network ? _UT("file:") + : _UT("file:///"); + const size_t prefixLen = URI_STRLEN(prefix); + + // Detect and avoid integer overflow + if (prefixLen > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_ERROR_OUTPUT_TOO_LARGE; + } + + /* Copy prefix */ + memcpy(uriString, prefix, prefixLen * sizeof(URI_CHAR)); + output += prefixLen; + } + + /* Copy and escape on the fly */ + for (;;) { + if ((input[0] == _UT('\0')) || (fromUnix && input[0] == _UT('/')) + || (!fromUnix && input[0] == _UT('\\'))) { + /* Copy text after last separator */ + if (lastSep + 1 < input) { + if (!fromUnix && absolute && (firstSegment == URI_TRUE)) { + /* Quick hack to not convert "C:" to "C%3A" */ + const size_t charsToCopy = input - (lastSep + 1); + + // Detect and avoid integer overflow + if (charsToCopy > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_ERROR_OUTPUT_TOO_LARGE; + } + + memcpy(output, lastSep + 1, charsToCopy * sizeof(URI_CHAR)); + output += charsToCopy; + } else { + output = URI_FUNC(EscapeEx)(lastSep + 1, input, output, URI_FALSE, + URI_FALSE); + } + } + firstSegment = URI_FALSE; + } + + if (input[0] == _UT('\0')) { + output[0] = _UT('\0'); + break; + } else if (fromUnix && (input[0] == _UT('/'))) { + /* Copy separators unmodified */ + output[0] = _UT('/'); + output++; + lastSep = input; + } else if (!fromUnix && (input[0] == _UT('\\'))) { + /* Convert backslashes to forward slashes */ + output[0] = _UT('/'); + output++; + lastSep = input; + } + input++; + } + + return URI_SUCCESS; } - - static URI_INLINE int URI_FUNC(UriStringToFilename)(const URI_CHAR * uriString, - URI_CHAR * filename, UriBool toUnix) { - if ((uriString == NULL) || (filename == NULL)) { - return URI_ERROR_NULL; - } - - { - const UriBool file_unknown_slashes = - URI_STRNCMP(uriString, _UT("file:"), URI_STRLEN(_UT("file:"))) == 0; - const UriBool file_one_or_more_slashes = file_unknown_slashes - && (URI_STRNCMP(uriString, _UT("file:/"), URI_STRLEN(_UT("file:/"))) == 0); - const UriBool file_two_or_more_slashes = file_one_or_more_slashes - && (URI_STRNCMP(uriString, _UT("file://"), URI_STRLEN(_UT("file://"))) == 0); - const UriBool file_three_or_more_slashes = file_two_or_more_slashes - && (URI_STRNCMP(uriString, _UT("file:///"), URI_STRLEN(_UT("file:///"))) == 0); - - const size_t charsToSkip = file_two_or_more_slashes - ? file_three_or_more_slashes - ? toUnix - /* file:///bin/bash */ - ? URI_STRLEN(_UT("file://")) - /* file:///E:/Documents%20and%20Settings */ - : URI_STRLEN(_UT("file:///")) - /* file://Server01/Letter.txt */ - : URI_STRLEN(_UT("file://")) - : ((file_one_or_more_slashes && toUnix) - /* file:/bin/bash */ - /* https://tools.ietf.org/html/rfc8089#appendix-B */ - ? URI_STRLEN(_UT("file:")) - : ((! toUnix && file_unknown_slashes && ! file_one_or_more_slashes) - /* file:c:/path/to/file */ - /* https://tools.ietf.org/html/rfc8089#appendix-E.2 */ - ? URI_STRLEN(_UT("file:")) - : 0)); - const size_t charsToCopy = URI_STRLEN(uriString + charsToSkip) + 1; - - const UriBool is_windows_network_with_authority = - (toUnix == URI_FALSE) - && file_two_or_more_slashes - && ! file_three_or_more_slashes; - - URI_CHAR * const unescape_target = is_windows_network_with_authority - ? (filename + 2) - : filename; - - if (is_windows_network_with_authority) { - filename[0] = '\\'; - filename[1] = '\\'; - } - - memcpy(unescape_target, uriString + charsToSkip, charsToCopy * sizeof(URI_CHAR)); - URI_FUNC(UnescapeInPlaceEx)(filename, URI_FALSE, URI_BR_DONT_TOUCH); - } - - /* Convert forward slashes to backslashes */ - if (!toUnix) { - URI_CHAR * walker = filename; - while (walker[0] != _UT('\0')) { - if (walker[0] == _UT('/')) { - walker[0] = _UT('\\'); - } - walker++; - } - } - - return URI_SUCCESS; + URI_CHAR * filename, UriBool toUnix) { + if ((uriString == NULL) || (filename == NULL)) { + return URI_ERROR_NULL; + } + + const UriBool file_unknown_slashes = + URI_STRNCMP(uriString, _UT("file:"), URI_STRLEN(_UT("file:"))) == 0; + const UriBool file_one_or_more_slashes = + file_unknown_slashes + && (URI_STRNCMP(uriString, _UT("file:/"), URI_STRLEN(_UT("file:/"))) == 0); + const UriBool file_two_or_more_slashes = + file_one_or_more_slashes + && (URI_STRNCMP(uriString, _UT("file://"), URI_STRLEN(_UT("file://"))) == 0); + const UriBool file_three_or_more_slashes = + file_two_or_more_slashes + && (URI_STRNCMP(uriString, _UT("file:///"), URI_STRLEN(_UT("file:///"))) == 0); + + const size_t charsToSkip = + file_two_or_more_slashes + ? file_three_or_more_slashes ? toUnix + /* file:///bin/bash */ + ? URI_STRLEN(_UT("file://")) + /* file:///E:/Documents%20and%20Settings */ + : URI_STRLEN(_UT("file:///")) + /* file://Server01/Letter.txt */ + : URI_STRLEN(_UT("file://")) + : ((file_one_or_more_slashes && toUnix) + /* file:/bin/bash */ + /* https://tools.ietf.org/html/rfc8089#appendix-B */ + ? URI_STRLEN(_UT("file:")) + : ((!toUnix && file_unknown_slashes && !file_one_or_more_slashes) + /* file:c:/path/to/file */ + /* https://tools.ietf.org/html/rfc8089#appendix-E.2 */ + ? URI_STRLEN(_UT("file:")) + : 0)); + const size_t charsToCopy = URI_STRLEN(uriString + charsToSkip) + 1; + + const UriBool is_windows_network_with_authority = + (toUnix == URI_FALSE) && file_two_or_more_slashes && !file_three_or_more_slashes; + + URI_CHAR * const unescape_target = + is_windows_network_with_authority ? (filename + 2) : filename; + + if (is_windows_network_with_authority) { + filename[0] = '\\'; + filename[1] = '\\'; + } + + memcpy(unescape_target, uriString + charsToSkip, charsToCopy * sizeof(URI_CHAR)); + URI_FUNC(UnescapeInPlaceEx)(filename, URI_FALSE, URI_BR_DONT_TOUCH); + + /* Convert forward slashes to backslashes */ + if (!toUnix) { + URI_CHAR * walker = filename; + while (walker[0] != _UT('\0')) { + if (walker[0] == _UT('/')) { + walker[0] = _UT('\\'); + } + walker++; + } + } + + return URI_SUCCESS; } - - int URI_FUNC(UnixFilenameToUriString)(const URI_CHAR * filename, URI_CHAR * uriString) { - return URI_FUNC(FilenameToUriString)(filename, uriString, URI_TRUE); + return URI_FUNC(FilenameToUriString)(filename, uriString, URI_TRUE); } - - -int URI_FUNC(WindowsFilenameToUriString)(const URI_CHAR * filename, URI_CHAR * uriString) { - return URI_FUNC(FilenameToUriString)(filename, uriString, URI_FALSE); +int URI_FUNC(WindowsFilenameToUriString)(const URI_CHAR * filename, + URI_CHAR * uriString) { + return URI_FUNC(FilenameToUriString)(filename, uriString, URI_FALSE); } - - int URI_FUNC(UriStringToUnixFilename)(const URI_CHAR * uriString, URI_CHAR * filename) { - return URI_FUNC(UriStringToFilename)(uriString, filename, URI_TRUE); + return URI_FUNC(UriStringToFilename)(uriString, filename, URI_TRUE); } - - -int URI_FUNC(UriStringToWindowsFilename)(const URI_CHAR * uriString, URI_CHAR * filename) { - return URI_FUNC(UriStringToFilename)(uriString, filename, URI_FALSE); +int URI_FUNC(UriStringToWindowsFilename)(const URI_CHAR * uriString, + URI_CHAR * filename) { + return URI_FUNC(UriStringToFilename)(uriString, filename, URI_FALSE); } - - #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriIp4.c b/cpp/src/arrow/vendored/uriparser/UriIp4.c index 1d4ab39bc67a..d791c9b5fee6 100644 --- a/cpp/src/arrow/vendored/uriparser/UriIp4.c +++ b/cpp/src/arrow/vendored/uriparser/UriIp4.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -47,97 +47,94 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriIp4.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriIp4.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriIp4.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriIp4.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# endif - - - -#ifndef URI_DOXYGEN -# include "UriIp4.h" -# include "UriIp4Base.h" -# include "UriBase.h" -#endif - - +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# endif + +# ifndef URI_DOXYGEN +# include "UriIp4.h" +# include "UriIp4Base.h" +# include "UriBase.h" +# include "UriSets.h" +# endif /* Prototypes */ static const URI_CHAR * URI_FUNC(ParseDecOctet)(UriIp4Parser * parser, - const URI_CHAR * first, const URI_CHAR * afterLast); + const URI_CHAR * first, + const URI_CHAR * afterLast); static const URI_CHAR * URI_FUNC(ParseDecOctetOne)(UriIp4Parser * parser, - const URI_CHAR * first, const URI_CHAR * afterLast); + const URI_CHAR * first, + const URI_CHAR * afterLast); static const URI_CHAR * URI_FUNC(ParseDecOctetTwo)(UriIp4Parser * parser, - const URI_CHAR * first, const URI_CHAR * afterLast); + const URI_CHAR * first, + const URI_CHAR * afterLast); static const URI_CHAR * URI_FUNC(ParseDecOctetThree)(UriIp4Parser * parser, - const URI_CHAR * first, const URI_CHAR * afterLast); + const URI_CHAR * first, + const URI_CHAR * afterLast); static const URI_CHAR * URI_FUNC(ParseDecOctetFour)(UriIp4Parser * parser, - const URI_CHAR * first, const URI_CHAR * afterLast); - - + const URI_CHAR * first, + const URI_CHAR * afterLast); /* * [ipFourAddress]->[decOctet]<.>[decOctet]<.>[decOctet]<.>[decOctet] */ -int URI_FUNC(ParseIpFourAddress)(unsigned char * octetOutput, - const URI_CHAR * first, const URI_CHAR * afterLast) { - const URI_CHAR * after; - UriIp4Parser parser; - - /* Essential checks */ - if ((octetOutput == NULL) || (first == NULL) - || (afterLast <= first)) { - return URI_ERROR_SYNTAX; - } - - /* Reset parser */ - parser.stackCount = 0; - - /* Octet #1 */ - after = URI_FUNC(ParseDecOctet)(&parser, first, afterLast); - if ((after == NULL) || (after >= afterLast) || (*after != _UT('.'))) { - return URI_ERROR_SYNTAX; - } - uriStackToOctet(&parser, octetOutput); - - /* Octet #2 */ - after = URI_FUNC(ParseDecOctet)(&parser, after + 1, afterLast); - if ((after == NULL) || (after >= afterLast) || (*after != _UT('.'))) { - return URI_ERROR_SYNTAX; - } - uriStackToOctet(&parser, octetOutput + 1); - - /* Octet #3 */ - after = URI_FUNC(ParseDecOctet)(&parser, after + 1, afterLast); - if ((after == NULL) || (after >= afterLast) || (*after != _UT('.'))) { - return URI_ERROR_SYNTAX; - } - uriStackToOctet(&parser, octetOutput + 2); - - /* Octet #4 */ - after = URI_FUNC(ParseDecOctet)(&parser, after + 1, afterLast); - if (after != afterLast) { - return URI_ERROR_SYNTAX; - } - uriStackToOctet(&parser, octetOutput + 3); - - return URI_SUCCESS; +int URI_FUNC(ParseIpFourAddress)(unsigned char * octetOutput, const URI_CHAR * first, + const URI_CHAR * afterLast) { + const URI_CHAR * after; + UriIp4Parser parser; + + /* Essential checks */ + if ((octetOutput == NULL) || (first == NULL) || (afterLast <= first)) { + return URI_ERROR_SYNTAX; + } + + /* Reset parser */ + parser.stackCount = 0; + + /* Octet #1 */ + after = URI_FUNC(ParseDecOctet)(&parser, first, afterLast); + if ((after == NULL) || (after >= afterLast) || (*after != _UT('.'))) { + return URI_ERROR_SYNTAX; + } + uriStackToOctet(&parser, octetOutput); + + /* Octet #2 */ + after = URI_FUNC(ParseDecOctet)(&parser, after + 1, afterLast); + if ((after == NULL) || (after >= afterLast) || (*after != _UT('.'))) { + return URI_ERROR_SYNTAX; + } + uriStackToOctet(&parser, octetOutput + 1); + + /* Octet #3 */ + after = URI_FUNC(ParseDecOctet)(&parser, after + 1, afterLast); + if ((after == NULL) || (after >= afterLast) || (*after != _UT('.'))) { + return URI_ERROR_SYNTAX; + } + uriStackToOctet(&parser, octetOutput + 2); + + /* Octet #4 */ + after = URI_FUNC(ParseDecOctet)(&parser, after + 1, afterLast); + if (after != afterLast) { + return URI_ERROR_SYNTAX; + } + uriStackToOctet(&parser, octetOutput + 3); + + return URI_SUCCESS; } - - /* * [decOctet]-><0> * [decOctet]-><1>[decOctetOne] @@ -151,72 +148,63 @@ int URI_FUNC(ParseIpFourAddress)(unsigned char * octetOutput, * [decOctet]-><9>[decOctetThree] */ static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctet)(UriIp4Parser * parser, - const URI_CHAR * first, const URI_CHAR * afterLast) { - if (first >= afterLast) { - return NULL; - } - - switch (*first) { - case _UT('0'): - uriPushToStack(parser, 0); - return first + 1; - - case _UT('1'): - uriPushToStack(parser, 1); - return (const URI_CHAR *)URI_FUNC(ParseDecOctetOne)(parser, first + 1, afterLast); - - case _UT('2'): - uriPushToStack(parser, 2); - return (const URI_CHAR *)URI_FUNC(ParseDecOctetTwo)(parser, first + 1, afterLast); - - case _UT('3'): - case _UT('4'): - case _UT('5'): - case _UT('6'): - case _UT('7'): - case _UT('8'): - case _UT('9'): - uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); - return (const URI_CHAR *)URI_FUNC(ParseDecOctetThree)(parser, first + 1, afterLast); - - default: - return NULL; - } + const URI_CHAR * first, + const URI_CHAR * afterLast) { + if (first >= afterLast) { + return NULL; + } + + switch (*first) { + case _UT('0'): + uriPushToStack(parser, 0); + return first + 1; + + case _UT('1'): + uriPushToStack(parser, 1); + return (const URI_CHAR *)URI_FUNC(ParseDecOctetOne)(parser, first + 1, afterLast); + + case _UT('2'): + uriPushToStack(parser, 2); + return (const URI_CHAR *)URI_FUNC(ParseDecOctetTwo)(parser, first + 1, afterLast); + + case _UT('3'): + case _UT('4'): + case _UT('5'): + case _UT('6'): + case _UT('7'): + case _UT('8'): + case _UT('9'): + uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); + return (const URI_CHAR *)URI_FUNC(ParseDecOctetThree)(parser, first + 1, + afterLast); + + default: + return NULL; + } } - - /* * [decOctetOne]-> * [decOctetOne]->[DIGIT][decOctetThree] */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctetOne)(UriIp4Parser * parser, - const URI_CHAR * first, const URI_CHAR * afterLast) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('0'): - case _UT('1'): - case _UT('2'): - case _UT('3'): - case _UT('4'): - case _UT('5'): - case _UT('6'): - case _UT('7'): - case _UT('8'): - case _UT('9'): - uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); - return (const URI_CHAR *)URI_FUNC(ParseDecOctetThree)(parser, first + 1, afterLast); - - default: - return first; - } +static URI_INLINE const URI_CHAR * +URI_FUNC(ParseDecOctetOne)(UriIp4Parser * parser, const URI_CHAR * first, + const URI_CHAR * afterLast) { + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case URI_SET_DIGIT(_UT): + uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); + return (const URI_CHAR *)URI_FUNC(ParseDecOctetThree)(parser, first + 1, + afterLast); + + default: + return first; + } } - - /* * [decOctetTwo]-> * [decOctetTwo]-><0>[decOctetThree] @@ -229,71 +217,62 @@ static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctetOne)(UriIp4Parser * par * [decOctetTwo]-><7> * [decOctetTwo]-><8> * [decOctetTwo]-><9> -*/ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctetTwo)(UriIp4Parser * parser, - const URI_CHAR * first, const URI_CHAR * afterLast) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('0'): - case _UT('1'): - case _UT('2'): - case _UT('3'): - case _UT('4'): - uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); - return (const URI_CHAR *)URI_FUNC(ParseDecOctetThree)(parser, first + 1, afterLast); - - case _UT('5'): - uriPushToStack(parser, 5); - return (const URI_CHAR *)URI_FUNC(ParseDecOctetFour)(parser, first + 1, afterLast); - - case _UT('6'): - case _UT('7'): - case _UT('8'): - case _UT('9'): - uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); - return first + 1; - - default: - return first; - } + */ +static URI_INLINE const URI_CHAR * +URI_FUNC(ParseDecOctetTwo)(UriIp4Parser * parser, const URI_CHAR * first, + const URI_CHAR * afterLast) { + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case _UT('0'): + case _UT('1'): + case _UT('2'): + case _UT('3'): + case _UT('4'): + uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); + return (const URI_CHAR *)URI_FUNC(ParseDecOctetThree)(parser, first + 1, + afterLast); + + case _UT('5'): + uriPushToStack(parser, 5); + return (const URI_CHAR *)URI_FUNC(ParseDecOctetFour)(parser, first + 1, + afterLast); + + case _UT('6'): + case _UT('7'): + case _UT('8'): + case _UT('9'): + uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); + return first + 1; + + default: + return first; + } } - - /* * [decOctetThree]-> * [decOctetThree]->[DIGIT] */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctetThree)(UriIp4Parser * parser, - const URI_CHAR * first, const URI_CHAR * afterLast) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('0'): - case _UT('1'): - case _UT('2'): - case _UT('3'): - case _UT('4'): - case _UT('5'): - case _UT('6'): - case _UT('7'): - case _UT('8'): - case _UT('9'): - uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); - return first + 1; - - default: - return first; - } +static URI_INLINE const URI_CHAR * +URI_FUNC(ParseDecOctetThree)(UriIp4Parser * parser, const URI_CHAR * first, + const URI_CHAR * afterLast) { + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case URI_SET_DIGIT(_UT): + uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); + return first + 1; + + default: + return first; + } } - - /* * [decOctetFour]-> * [decOctetFour]-><0> @@ -303,27 +282,26 @@ static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctetThree)(UriIp4Parser * p * [decOctetFour]-><4> * [decOctetFour]-><5> */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseDecOctetFour)(UriIp4Parser * parser, - const URI_CHAR * first, const URI_CHAR * afterLast) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('0'): - case _UT('1'): - case _UT('2'): - case _UT('3'): - case _UT('4'): - case _UT('5'): - uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); - return first + 1; - - default: - return first; - } +static URI_INLINE const URI_CHAR * +URI_FUNC(ParseDecOctetFour)(UriIp4Parser * parser, const URI_CHAR * first, + const URI_CHAR * afterLast) { + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case _UT('0'): + case _UT('1'): + case _UT('2'): + case _UT('3'): + case _UT('4'): + case _UT('5'): + uriPushToStack(parser, (unsigned char)(9 + *first - _UT('9'))); + return first + 1; + + default: + return first; + } } - - #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriIp4.h b/cpp/src/arrow/vendored/uriparser/UriIp4.h index e1d7f1e4ad16..5f84b62dd2fc 100644 --- a/cpp/src/arrow/vendored/uriparser/UriIp4.h +++ b/cpp/src/arrow/vendored/uriparser/UriIp4.h @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -44,67 +44,62 @@ */ #if (defined(URI_PASS_ANSI) && !defined(URI_IP4_TWICE_H_ANSI)) \ - || (defined(URI_PASS_UNICODE) && !defined(URI_IP4_TWICE_H_UNICODE)) \ - || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) + || (defined(URI_PASS_UNICODE) && !defined(URI_IP4_TWICE_H_UNICODE)) \ + || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* What encodings are enabled? */ -#include "UriDefsConfig.h" -#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +# include "UriDefsConfig.h" +# if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriIp4.h" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriIp4.h" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriIp4.h" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriIp4.h" +# undef URI_PASS_UNICODE +# endif /* Only one pass for each encoding */ -#elif (defined(URI_PASS_ANSI) && !defined(URI_IP4_TWICE_H_ANSI) \ - && defined(URI_ENABLE_ANSI)) || (defined(URI_PASS_UNICODE) \ - && !defined(URI_IP4_TWICE_H_UNICODE) && defined(URI_ENABLE_UNICODE)) -# ifdef URI_PASS_ANSI -# define URI_IP4_TWICE_H_ANSI 1 -# include "UriDefsAnsi.h" -# else -# define URI_IP4_TWICE_H_UNICODE 1 -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifdef __cplusplus +# elif (defined(URI_PASS_ANSI) && !defined(URI_IP4_TWICE_H_ANSI) \ + && defined(URI_ENABLE_ANSI)) \ + || (defined(URI_PASS_UNICODE) && !defined(URI_IP4_TWICE_H_UNICODE) \ + && defined(URI_ENABLE_UNICODE)) +# ifdef URI_PASS_ANSI +# define URI_IP4_TWICE_H_ANSI 1 +# include "UriDefsAnsi.h" +# else +# define URI_IP4_TWICE_H_UNICODE 1 +# include "UriDefsUnicode.h" +# include +# endif + +# ifdef __cplusplus extern "C" { -#endif - - - -#ifndef URI_DOXYGEN -# include "UriBase.h" -#endif - +# endif +# ifndef URI_DOXYGEN +# include "UriBase.h" +# endif /** - * Converts a IPv4 text representation into four bytes. + * Converts an IPv4 text representation into four bytes. * * @param octetOutput Output destination * @param first First character of IPv4 text to parse * @param afterLast Position to stop parsing at * @return Error code or 0 on success + * + * @see uriParseIpSixAddressA + * @see uriParseIpSixAddressMmA */ URI_PUBLIC int URI_FUNC(ParseIpFourAddress)(unsigned char * octetOutput, - const URI_CHAR * first, const URI_CHAR * afterLast); - - + const URI_CHAR * first, + const URI_CHAR * afterLast); -#ifdef __cplusplus +# ifdef __cplusplus } -#endif - +# endif - -#endif +# endif #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriIp4Base.c b/cpp/src/arrow/vendored/uriparser/UriIp4Base.c index f0662465d7ea..900234cd782d 100644 --- a/cpp/src/arrow/vendored/uriparser/UriIp4Base.c +++ b/cpp/src/arrow/vendored/uriparser/UriIp4Base.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -43,54 +43,45 @@ */ #ifndef URI_DOXYGEN -# include "UriIp4Base.h" +# include "UriIp4Base.h" #endif - - void uriStackToOctet(UriIp4Parser * parser, unsigned char * octet) { - switch (parser->stackCount) { - case 1: - *octet = parser->stackOne; - break; + switch (parser->stackCount) { + case 1: + *octet = parser->stackOne; + break; - case 2: - *octet = parser->stackOne * 10 - + parser->stackTwo; - break; + case 2: + *octet = parser->stackOne * 10 + parser->stackTwo; + break; - case 3: - *octet = parser->stackOne * 100 - + parser->stackTwo * 10 - + parser->stackThree; - break; + case 3: + *octet = parser->stackOne * 100 + parser->stackTwo * 10 + parser->stackThree; + break; - default: - ; - } - parser->stackCount = 0; + default:; + } + parser->stackCount = 0; } - - void uriPushToStack(UriIp4Parser * parser, unsigned char digit) { - switch (parser->stackCount) { - case 0: - parser->stackOne = digit; - parser->stackCount = 1; - break; + switch (parser->stackCount) { + case 0: + parser->stackOne = digit; + parser->stackCount = 1; + break; - case 1: - parser->stackTwo = digit; - parser->stackCount = 2; - break; + case 1: + parser->stackTwo = digit; + parser->stackCount = 2; + break; - case 2: - parser->stackThree = digit; - parser->stackCount = 3; - break; + case 2: + parser->stackThree = digit; + parser->stackCount = 3; + break; - default: - ; - } + default:; + } } diff --git a/cpp/src/arrow/vendored/uriparser/UriIp4Base.h b/cpp/src/arrow/vendored/uriparser/UriIp4Base.h index bef028f7c943..1a2ff148b628 100644 --- a/cpp/src/arrow/vendored/uriparser/UriIp4Base.h +++ b/cpp/src/arrow/vendored/uriparser/UriIp4Base.h @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -38,22 +38,16 @@ */ #ifndef URI_IP4_BASE_H -#define URI_IP4_BASE_H 1 - - +# define URI_IP4_BASE_H 1 typedef struct UriIp4ParserStruct { - unsigned char stackCount; - unsigned char stackOne; - unsigned char stackTwo; - unsigned char stackThree; + unsigned char stackCount; + unsigned char stackOne; + unsigned char stackTwo; + unsigned char stackThree; } UriIp4Parser; - - void uriPushToStack(UriIp4Parser * parser, unsigned char digit); void uriStackToOctet(UriIp4Parser * parser, unsigned char * octet); - - #endif /* URI_IP4_BASE_H */ diff --git a/cpp/src/arrow/vendored/uriparser/UriMemory.c b/cpp/src/arrow/vendored/uriparser/UriMemory.c index 2d23f1b6f6e8..669f48dd6827 100644 --- a/cpp/src/arrow/vendored/uriparser/UriMemory.c +++ b/cpp/src/arrow/vendored/uriparser/UriMemory.c @@ -5,32 +5,32 @@ * Copyright (C) 2018, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -42,427 +42,436 @@ * Holds memory manager implementation. */ -#include "config.h" +#include "UriConfig.h" /* for HAVE_REALLOCARRAY */ #ifdef HAVE_REALLOCARRAY -# ifndef _GNU_SOURCE -# define _GNU_SOURCE 1 -# endif +// For glibc >=2.29 of 2019-02-01 +# if !defined(_DEFAULT_SOURCE) +# define _DEFAULT_SOURCE 1 +# endif + +// For NetBSD (stdlib.h revision 1.122 of 2020-05-26) +# if defined(__NetBSD__) && !defined(_OPENBSD_SOURCE) +# define _OPENBSD_SOURCE 1 +# endif + +// POSIX 2024 (XPG8) for e.g. Illumos/SmartOS +# if !defined(_XOPEN_SOURCE) || (_XOPEN_SOURCE - 0 < 800) +# undef _XOPEN_SOURCE +# define _XOPEN_SOURCE 800 +# endif +# if !defined(_POSIX_C_SOURCE) || (_POSIX_C_SOURCE - 0 < 202405L) +# undef _POSIX_C_SOURCE +# define _POSIX_C_SOURCE 202405L +# endif #endif #include #include - - #ifndef URI_DOXYGEN -# include "UriMemory.h" +# include "UriMemory.h" #endif +#define URI_MAX(a, b) (((a) > (b)) ? (a) : (b)) +/* NOTE: This intends to mimic MALLOC_ALIGNMENT of glibc */ +#define URI_MALLOC_ALIGNMENT URI_MAX(2 * sizeof(size_t), sizeof(long double)) -#define URI_CHECK_ALLOC_OVERFLOW(total_size, nmemb, size) \ - do { \ - /* check for unsigned overflow */ \ - if ((nmemb != 0) && (total_size / nmemb != size)) { \ - errno = ENOMEM; \ - return NULL; \ - } \ - } while (0) - - +#define URI_MALLOC_PADDING (URI_MALLOC_ALIGNMENT - sizeof(size_t)) -static void * uriDefaultMalloc(UriMemoryManager * URI_UNUSED(memory), - size_t size) { - return malloc(size); +#define URI_CHECK_ALLOC_OVERFLOW(total_size, nmemb, size) \ + do { \ + /* check for unsigned overflow */ \ + if ((nmemb != 0) && (total_size / nmemb != size)) { \ + errno = ENOMEM; \ + return NULL; \ + } \ + } while (0) + +static void * uriDefaultMalloc(UriMemoryManager * URI_UNUSED(memory), size_t size) { + return malloc(size); } - - -static void * uriDefaultCalloc(UriMemoryManager * URI_UNUSED(memory), - size_t nmemb, size_t size) { - return calloc(nmemb, size); +static void * uriDefaultCalloc(UriMemoryManager * URI_UNUSED(memory), size_t nmemb, + size_t size) { + return calloc(nmemb, size); } - - -static void * uriDefaultRealloc(UriMemoryManager * URI_UNUSED(memory), - void * ptr, size_t size) { - return realloc(ptr, size); +static void * uriDefaultRealloc(UriMemoryManager * URI_UNUSED(memory), void * ptr, + size_t size) { + return realloc(ptr, size); } - - -static void * uriDefaultReallocarray(UriMemoryManager * URI_UNUSED(memory), - void * ptr, size_t nmemb, size_t size) { +static void * uriDefaultReallocarray(UriMemoryManager * URI_UNUSED(memory), void * ptr, + size_t nmemb, size_t size) { #ifdef HAVE_REALLOCARRAY - return reallocarray(ptr, nmemb, size); + return reallocarray(ptr, nmemb, size); #else - const size_t total_size = nmemb * size; + const size_t total_size = nmemb * size; - URI_CHECK_ALLOC_OVERFLOW(total_size, nmemb, size); /* may return */ + URI_CHECK_ALLOC_OVERFLOW(total_size, nmemb, size); /* may return */ - return realloc(ptr, total_size); + return realloc(ptr, total_size); #endif } - - -static void uriDefaultFree(UriMemoryManager * URI_UNUSED(memory), - void * ptr) { - free(ptr); +static void uriDefaultFree(UriMemoryManager * URI_UNUSED(memory), void * ptr) { + free(ptr); } - - UriBool uriMemoryManagerIsComplete(const UriMemoryManager * memory) { - return (memory - && memory->malloc - && memory->calloc - && memory->realloc - && memory->reallocarray - && memory->free) ? URI_TRUE : URI_FALSE; + return (memory && memory->malloc && memory->calloc && memory->realloc + && memory->reallocarray && memory->free) + ? URI_TRUE + : URI_FALSE; } - - void * uriEmulateCalloc(UriMemoryManager * memory, size_t nmemb, size_t size) { - void * buffer; - const size_t total_size = nmemb * size; - - if (memory == NULL) { - errno = EINVAL; - return NULL; - } - - URI_CHECK_ALLOC_OVERFLOW(total_size, nmemb, size); /* may return */ - - buffer = memory->malloc(memory, total_size); - if (buffer == NULL) { - /* errno set by malloc */ - return NULL; - } - memset(buffer, 0, total_size); - return buffer; + void * buffer; + const size_t total_size = nmemb * size; + + if (memory == NULL) { + errno = EINVAL; + return NULL; + } + + URI_CHECK_ALLOC_OVERFLOW(total_size, nmemb, size); /* may return */ + + buffer = memory->malloc(memory, total_size); + if (buffer == NULL) { + /* errno set by malloc */ + return NULL; + } + memset(buffer, 0, total_size); + return buffer; } +void * uriEmulateReallocarray(UriMemoryManager * memory, void * ptr, size_t nmemb, + size_t size) { + const size_t total_size = nmemb * size; + if (memory == NULL) { + errno = EINVAL; + return NULL; + } -void * uriEmulateReallocarray(UriMemoryManager * memory, - void * ptr, size_t nmemb, size_t size) { - const size_t total_size = nmemb * size; - - if (memory == NULL) { - errno = EINVAL; - return NULL; - } - - URI_CHECK_ALLOC_OVERFLOW(total_size, nmemb, size); /* may return */ + URI_CHECK_ALLOC_OVERFLOW(total_size, nmemb, size); /* may return */ - return memory->realloc(memory, ptr, total_size); + return memory->realloc(memory, ptr, total_size); } +static void * uriDecorateMalloc(UriMemoryManager * memory, size_t size) { + UriMemoryManager * backend; + const size_t extraBytes = sizeof(size_t) + URI_MALLOC_PADDING; + void * buffer; + if (memory == NULL) { + errno = EINVAL; + return NULL; + } -static void * uriDecorateMalloc(UriMemoryManager * memory, - size_t size) { - UriMemoryManager * backend; - const size_t extraBytes = sizeof(size_t); - void * buffer; + /* check for unsigned overflow */ + if (size > ((size_t)-1) - extraBytes) { + errno = ENOMEM; + return NULL; + } - if (memory == NULL) { - errno = EINVAL; - return NULL; - } + backend = (UriMemoryManager *)memory->userData; + if (backend == NULL) { + errno = EINVAL; + return NULL; + } - /* check for unsigned overflow */ - if (size > ((size_t)-1) - extraBytes) { - errno = ENOMEM; - return NULL; - } + buffer = backend->malloc(backend, extraBytes + size); + if (buffer == NULL) { + return NULL; + } - backend = (UriMemoryManager *)memory->userData; - if (backend == NULL) { - errno = EINVAL; - return NULL; - } + *(size_t *)buffer = size; - buffer = backend->malloc(backend, extraBytes + size); - if (buffer == NULL) { - return NULL; - } - - *(size_t *)buffer = size; - - return (char *)buffer + extraBytes; + return (char *)buffer + extraBytes; } +static void * uriDecorateRealloc(UriMemoryManager * memory, void * ptr, size_t size) { + void * newBuffer; + size_t prevSize; + if (memory == NULL) { + errno = EINVAL; + return NULL; + } -static void * uriDecorateRealloc(UriMemoryManager * memory, - void * ptr, size_t size) { - void * newBuffer; - size_t prevSize; - - if (memory == NULL) { - errno = EINVAL; - return NULL; - } - - /* man realloc: "If ptr is NULL, then the call is equivalent to - * malloc(size), for *all* values of size" */ - if (ptr == NULL) { - return memory->malloc(memory, size); - } + /* man realloc: "If ptr is NULL, then the call is equivalent to + * malloc(size), for *all* values of size" */ + if (ptr == NULL) { + return memory->malloc(memory, size); + } - /* man realloc: "If size is equal to zero, and ptr is *not* NULL, - * then the call is equivalent to free(ptr)." */ - if (size == 0) { - memory->free(memory, ptr); - return NULL; - } + /* man realloc: "If size is equal to zero, and ptr is *not* NULL, + * then the call is equivalent to free(ptr)." */ + if (size == 0) { + memory->free(memory, ptr); + return NULL; + } - prevSize = *((size_t *)((char *)ptr - sizeof(size_t))); + prevSize = *((size_t *)((char *)ptr - sizeof(size_t) - URI_MALLOC_PADDING)); - /* Anything to do? */ - if (size <= prevSize) { - return ptr; - } + /* Anything to do? */ + /* mull-ignore-next cxx_le_to_lt */ + if (size <= prevSize) { + return ptr; + } - newBuffer = memory->malloc(memory, size); - if (newBuffer == NULL) { - /* errno set by malloc */ - return NULL; - } + newBuffer = memory->malloc(memory, size); + if (newBuffer == NULL) { + /* errno set by malloc */ + return NULL; + } - memcpy(newBuffer, ptr, prevSize); + memcpy(newBuffer, ptr, prevSize); - memory->free(memory, ptr); + memory->free(memory, ptr); - return newBuffer; + return newBuffer; } - - static void uriDecorateFree(UriMemoryManager * memory, void * ptr) { - UriMemoryManager * backend; + UriMemoryManager * backend; - if ((ptr == NULL) || (memory == NULL)) { - return; - } + if ((ptr == NULL) || (memory == NULL)) { + return; + } - backend = (UriMemoryManager *)memory->userData; - if (backend == NULL) { - return; - } + backend = (UriMemoryManager *)memory->userData; + if (backend == NULL) { + return; + } - backend->free(backend, (char *)ptr - sizeof(size_t)); + backend->free(backend, (char *)ptr - sizeof(size_t) - URI_MALLOC_PADDING); } +int uriCompleteMemoryManager(UriMemoryManager * memory, UriMemoryManager * backend) { + if ((memory == NULL) || (backend == NULL)) { + return URI_ERROR_NULL; + } + if ((backend->malloc == NULL) || (backend->free == NULL)) { + return URI_ERROR_MEMORY_MANAGER_INCOMPLETE; + } -int uriCompleteMemoryManager(UriMemoryManager * memory, - UriMemoryManager * backend) { - if ((memory == NULL) || (backend == NULL)) { - return URI_ERROR_NULL; - } - - if ((backend->malloc == NULL) || (backend->free == NULL)) { - return URI_ERROR_MEMORY_MANAGER_INCOMPLETE; - } - - memory->calloc = uriEmulateCalloc; - memory->reallocarray = uriEmulateReallocarray; + memory->calloc = uriEmulateCalloc; + memory->reallocarray = uriEmulateReallocarray; - memory->malloc = uriDecorateMalloc; - memory->realloc = uriDecorateRealloc; - memory->free = uriDecorateFree; + memory->malloc = uriDecorateMalloc; + memory->realloc = uriDecorateRealloc; + memory->free = uriDecorateFree; - memory->userData = backend; + memory->userData = backend; - return URI_SUCCESS; + return URI_SUCCESS; } - +/* mull-off */ +int uriTestMemoryManagerEx(UriMemoryManager * memory, UriBool challengeAlignment) { + const size_t mallocSize = 7; + const size_t callocNmemb = 3; + const size_t callocSize = 5; + const size_t callocTotalSize = callocNmemb * callocSize; + const size_t reallocSize = 11; + const size_t reallocarrayNmemb = 5; + const size_t reallocarraySize = 7; + const size_t reallocarrayTotal = reallocarrayNmemb * reallocarraySize; + size_t index; + char * buffer; + + if (memory == NULL) { + return URI_ERROR_NULL; + } + + if (uriMemoryManagerIsComplete(memory) != URI_TRUE) { + return URI_ERROR_MEMORY_MANAGER_INCOMPLETE; + } + + /* malloc + free*/ + buffer = memory->malloc(memory, mallocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + buffer[mallocSize - 1] = '\xF1'; + memory->free(memory, buffer); + buffer = NULL; + + /* calloc + free */ + buffer = memory->calloc(memory, callocNmemb, callocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + for (index = 0; index < callocTotalSize; index++) { /* all zeros? */ + if (buffer[index] != '\0') { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + } + buffer[callocTotalSize - 1] = '\xF2'; + memory->free(memory, buffer); + buffer = NULL; + + /* malloc + realloc + free */ + buffer = memory->malloc(memory, mallocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + for (index = 0; index < mallocSize; index++) { + buffer[index] = '\xF3'; + } + buffer = memory->realloc(memory, buffer, reallocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + for (index = 0; index < mallocSize; index++) { /* previous content? */ + if (buffer[index] != '\xF3') { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + } + buffer[reallocSize - 1] = '\xF4'; + memory->free(memory, buffer); + buffer = NULL; + + /* malloc + realloc ptr!=NULL size==0 (equals free) */ + buffer = memory->malloc(memory, mallocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + buffer[mallocSize - 1] = '\xF5'; + memory->realloc(memory, buffer, 0); + buffer = NULL; + + /* realloc ptr==NULL size!=0 (equals malloc) + free */ + buffer = memory->realloc(memory, NULL, mallocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + buffer[mallocSize - 1] = '\xF6'; + memory->free(memory, buffer); + buffer = NULL; + + /* realloc ptr==NULL size==0 (equals malloc) + free */ + buffer = memory->realloc(memory, NULL, 0); + if (buffer != NULL) { + memory->free(memory, buffer); + buffer = NULL; + } + + /* malloc + reallocarray + free */ + buffer = memory->malloc(memory, mallocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + for (index = 0; index < mallocSize; index++) { + buffer[index] = '\xF7'; + } + buffer = memory->reallocarray(memory, buffer, reallocarrayNmemb, reallocarraySize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + for (index = 0; index < mallocSize; index++) { /* previous content? */ + if (buffer[index] != '\xF7') { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + } + buffer[reallocarrayTotal - 1] = '\xF8'; + memory->free(memory, buffer); + buffer = NULL; + + /* malloc + reallocarray ptr!=NULL nmemb==0 size!=0 (equals free) */ + buffer = memory->malloc(memory, mallocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + buffer[mallocSize - 1] = '\xF9'; + memory->reallocarray(memory, buffer, 0, reallocarraySize); + buffer = NULL; + + /* malloc + reallocarray ptr!=NULL nmemb!=0 size==0 (equals free) */ + buffer = memory->malloc(memory, mallocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + buffer[mallocSize - 1] = '\xFA'; + memory->reallocarray(memory, buffer, reallocarrayNmemb, 0); + buffer = NULL; + + /* malloc + reallocarray ptr!=NULL nmemb==0 size==0 (equals free) */ + buffer = memory->malloc(memory, mallocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + buffer[mallocSize - 1] = '\xFB'; + memory->reallocarray(memory, buffer, 0, 0); + buffer = NULL; + + /* reallocarray ptr==NULL nmemb!=0 size!=0 (equals malloc) + free */ + buffer = memory->reallocarray(memory, NULL, callocNmemb, callocSize); + if (buffer == NULL) { + return URI_ERROR_MEMORY_MANAGER_FAULTY; + } + buffer[callocTotalSize - 1] = '\xFC'; + memory->free(memory, buffer); + buffer = NULL; + + /* reallocarray ptr==NULL nmemb==0 size!=0 (equals malloc) + free */ + buffer = memory->reallocarray(memory, NULL, 0, callocSize); + if (buffer != NULL) { + memory->free(memory, buffer); + buffer = NULL; + } + + /* reallocarray ptr==NULL nmemb!=0 size==0 (equals malloc) + free */ + buffer = memory->reallocarray(memory, NULL, callocNmemb, 0); + if (buffer != NULL) { + memory->free(memory, buffer); + buffer = NULL; + } + + /* reallocarray ptr==NULL nmemb==0 size==0 (equals malloc) + free */ + buffer = memory->reallocarray(memory, NULL, 0, 0); + if (buffer != NULL) { + memory->free(memory, buffer); + buffer = NULL; + } + + /* challenge pointer alignment */ + if (challengeAlignment == URI_TRUE) { + long double * ptr = memory->malloc(memory, 4 * sizeof(long double)); + if (ptr != NULL) { + ptr[0] = 0.0L; + ptr[1] = 1.1L; + ptr[2] = 2.2L; + ptr[3] = 3.3L; + + long double * const ptrNew = + memory->realloc(memory, ptr, 8 * sizeof(long double)); + if (ptrNew != NULL) { + ptr = ptrNew; + ptr[4] = 4.4L; + ptr[5] = 5.5L; + ptr[6] = 6.6L; + ptr[7] = 7.7L; + } + + memory->free(memory, ptr); + } + } + + return URI_SUCCESS; +} +/* mull-on */ int uriTestMemoryManager(UriMemoryManager * memory) { - const size_t mallocSize = 7; - const size_t callocNmemb = 3; - const size_t callocSize = 5; - const size_t callocTotalSize = callocNmemb * callocSize; - const size_t reallocSize = 11; - const size_t reallocarrayNmemb = 5; - const size_t reallocarraySize = 7; - const size_t reallocarrayTotal = reallocarrayNmemb * reallocarraySize; - size_t index; - char * buffer; - - if (memory == NULL) { - return URI_ERROR_NULL; - } - - if (uriMemoryManagerIsComplete(memory) != URI_TRUE) { - return URI_ERROR_MEMORY_MANAGER_INCOMPLETE; - } - - /* malloc + free*/ - buffer = memory->malloc(memory, mallocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - buffer[mallocSize - 1] = '\xF1'; - memory->free(memory, buffer); - buffer = NULL; - - /* calloc + free */ - buffer = memory->calloc(memory, callocNmemb, callocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - for (index = 0; index < callocTotalSize; index++) { /* all zeros? */ - if (buffer[index] != '\0') { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - } - buffer[callocTotalSize - 1] = '\xF2'; - memory->free(memory, buffer); - buffer = NULL; - - /* malloc + realloc + free */ - buffer = memory->malloc(memory, mallocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - for (index = 0; index < mallocSize; index++) { - buffer[index] = '\xF3'; - } - buffer = memory->realloc(memory, buffer, reallocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - for (index = 0; index < mallocSize; index++) { /* previous content? */ - if (buffer[index] != '\xF3') { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - } - buffer[reallocSize - 1] = '\xF4'; - memory->free(memory, buffer); - buffer = NULL; - - /* malloc + realloc ptr!=NULL size==0 (equals free) */ - buffer = memory->malloc(memory, mallocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - buffer[mallocSize - 1] = '\xF5'; - memory->realloc(memory, buffer, 0); - buffer = NULL; - - /* realloc ptr==NULL size!=0 (equals malloc) + free */ - buffer = memory->realloc(memory, NULL, mallocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - buffer[mallocSize - 1] = '\xF6'; - memory->free(memory, buffer); - buffer = NULL; - - /* realloc ptr==NULL size==0 (equals malloc) + free */ - buffer = memory->realloc(memory, NULL, 0); - if (buffer != NULL) { - memory->free(memory, buffer); - buffer = NULL; - } - - /* malloc + reallocarray + free */ - buffer = memory->malloc(memory, mallocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - for (index = 0; index < mallocSize; index++) { - buffer[index] = '\xF7'; - } - buffer = memory->reallocarray(memory, buffer, reallocarrayNmemb, - reallocarraySize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - for (index = 0; index < mallocSize; index++) { /* previous content? */ - if (buffer[index] != '\xF7') { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - } - buffer[reallocarrayTotal - 1] = '\xF8'; - memory->free(memory, buffer); - buffer = NULL; - - /* malloc + reallocarray ptr!=NULL nmemb==0 size!=0 (equals free) */ - buffer = memory->malloc(memory, mallocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - buffer[mallocSize - 1] = '\xF9'; - memory->reallocarray(memory, buffer, 0, reallocarraySize); - buffer = NULL; - - /* malloc + reallocarray ptr!=NULL nmemb!=0 size==0 (equals free) */ - buffer = memory->malloc(memory, mallocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - buffer[mallocSize - 1] = '\xFA'; - memory->reallocarray(memory, buffer, reallocarrayNmemb, 0); - buffer = NULL; - - /* malloc + reallocarray ptr!=NULL nmemb==0 size==0 (equals free) */ - buffer = memory->malloc(memory, mallocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - buffer[mallocSize - 1] = '\xFB'; - memory->reallocarray(memory, buffer, 0, 0); - buffer = NULL; - - /* reallocarray ptr==NULL nmemb!=0 size!=0 (equals malloc) + free */ - buffer = memory->reallocarray(memory, NULL, callocNmemb, callocSize); - if (buffer == NULL) { - return URI_ERROR_MEMORY_MANAGER_FAULTY; - } - buffer[callocTotalSize - 1] = '\xFC'; - memory->free(memory, buffer); - buffer = NULL; - - /* reallocarray ptr==NULL nmemb==0 size!=0 (equals malloc) + free */ - buffer = memory->reallocarray(memory, NULL, 0, callocSize); - if (buffer != NULL) { - memory->free(memory, buffer); - buffer = NULL; - } - - /* reallocarray ptr==NULL nmemb!=0 size==0 (equals malloc) + free */ - buffer = memory->reallocarray(memory, NULL, callocNmemb, 0); - if (buffer != NULL) { - memory->free(memory, buffer); - buffer = NULL; - } - - /* reallocarray ptr==NULL nmemb==0 size==0 (equals malloc) + free */ - buffer = memory->reallocarray(memory, NULL, 0, 0); - if (buffer != NULL) { - memory->free(memory, buffer); - buffer = NULL; - } - - return URI_SUCCESS; + return uriTestMemoryManagerEx(memory, /*challengeAlignment=*/URI_FALSE); } - - /*extern*/ UriMemoryManager defaultMemoryManager = { - uriDefaultMalloc, - uriDefaultCalloc, - uriDefaultRealloc, - uriDefaultReallocarray, - uriDefaultFree, - NULL /* userData */ + uriDefaultMalloc, uriDefaultCalloc, uriDefaultRealloc, + uriDefaultReallocarray, uriDefaultFree, NULL /* userData */ }; diff --git a/cpp/src/arrow/vendored/uriparser/UriMemory.h b/cpp/src/arrow/vendored/uriparser/UriMemory.h index 5d6bf678415f..9c3c241356d9 100644 --- a/cpp/src/arrow/vendored/uriparser/UriMemory.h +++ b/cpp/src/arrow/vendored/uriparser/UriMemory.h @@ -5,32 +5,32 @@ * Copyright (C) 2018, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -38,41 +38,31 @@ */ #ifndef URI_MEMORY_H -#define URI_MEMORY_H 1 - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -#endif - - - -#define URI_CHECK_MEMORY_MANAGER(memory) \ - do { \ - if (memory == NULL) { \ - memory = &defaultMemoryManager; \ - } else if (uriMemoryManagerIsComplete(memory) != URI_TRUE) { \ - return URI_ERROR_MEMORY_MANAGER_INCOMPLETE; \ - } \ - } while (0) - - - -#ifdef __cplusplus -# define URIPARSER_EXTERN extern "C" -#else -# define URIPARSER_EXTERN extern -#endif +# define URI_MEMORY_H 1 + +# ifndef URI_DOXYGEN +# include "Uri.h" +# endif + +# define URI_CHECK_MEMORY_MANAGER(memory) \ + do { \ + if (memory == NULL) { \ + memory = &defaultMemoryManager; \ + } else if (uriMemoryManagerIsComplete(memory) != URI_TRUE) { \ + return URI_ERROR_MEMORY_MANAGER_INCOMPLETE; \ + } \ + } while (0) + +# ifdef __cplusplus +# define URIPARSER_EXTERN extern "C" +# else +# define URIPARSER_EXTERN extern +# endif URIPARSER_EXTERN UriMemoryManager defaultMemoryManager; -#undef URIPARSER_EXTERN - - +# undef URIPARSER_EXTERN UriBool uriMemoryManagerIsComplete(const UriMemoryManager * memory); - - #endif /* URI_MEMORY_H */ diff --git a/cpp/src/arrow/vendored/uriparser/UriNormalize.c b/cpp/src/arrow/vendored/uriparser/UriNormalize.c index 3fc74920399a..fe049f695aff 100644 --- a/cpp/src/arrow/vendored/uriparser/UriNormalize.c +++ b/cpp/src/arrow/vendored/uriparser/UriNormalize.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -47,725 +47,823 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriNormalize.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriNormalize.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriNormalize.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriNormalize.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -# include "UriNormalizeBase.h" -# include "UriCommon.h" -# include "UriMemory.h" -#endif - - - -#include - - +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriNormalizeBase.h" +# include "UriCommon.h" +# include "UriMemory.h" +# endif + +# include +# include // SIZE_MAX static int URI_FUNC(NormalizeSyntaxEngine)(URI_TYPE(Uri) * uri, unsigned int inMask, - unsigned int * outMask, UriMemoryManager * memory); + unsigned int * outMask, + UriMemoryManager * memory); -static UriBool URI_FUNC(MakeRangeOwner)(unsigned int * doneMask, - unsigned int maskTest, URI_TYPE(TextRange) * range, - UriMemoryManager * memory); -static UriBool URI_FUNC(MakeOwner)(URI_TYPE(Uri) * uri, - unsigned int * doneMask, UriMemoryManager * memory); +static UriBool URI_FUNC(MakeRangeOwner)(unsigned int * revertMask, unsigned int maskTest, + URI_TYPE(TextRange) * range, + UriMemoryManager * memory); +static UriBool URI_FUNC(MakeOwnerEngine)(URI_TYPE(Uri) * uri, unsigned int * revertMask, + UriMemoryManager * memory); static void URI_FUNC(FixPercentEncodingInplace)(const URI_CHAR * first, - const URI_CHAR ** afterLast); + const URI_CHAR ** afterLast); static UriBool URI_FUNC(FixPercentEncodingMalloc)(const URI_CHAR ** first, - const URI_CHAR ** afterLast, UriMemoryManager * memory); -static void URI_FUNC(FixPercentEncodingEngine)( - const URI_CHAR * inFirst, const URI_CHAR * inAfterLast, - const URI_CHAR * outFirst, const URI_CHAR ** outAfterLast); + const URI_CHAR ** afterLast, + UriMemoryManager * memory); +static void URI_FUNC(FixPercentEncodingEngine)(const URI_CHAR * inFirst, + const URI_CHAR * inAfterLast, + const URI_CHAR * outFirst, + const URI_CHAR ** outAfterLast); static UriBool URI_FUNC(ContainsUppercaseLetters)(const URI_CHAR * first, - const URI_CHAR * afterLast); + const URI_CHAR * afterLast); static UriBool URI_FUNC(ContainsUglyPercentEncoding)(const URI_CHAR * first, - const URI_CHAR * afterLast); + const URI_CHAR * afterLast); static void URI_FUNC(LowercaseInplace)(const URI_CHAR * first, - const URI_CHAR * afterLast); + const URI_CHAR * afterLast); +static void URI_FUNC(LowercaseInplaceExceptPercentEncoding)(const URI_CHAR * first, + const URI_CHAR * afterLast); static UriBool URI_FUNC(LowercaseMalloc)(const URI_CHAR ** first, - const URI_CHAR ** afterLast, UriMemoryManager * memory); - -static void URI_FUNC(PreventLeakage)(URI_TYPE(Uri) * uri, - unsigned int revertMask, UriMemoryManager * memory); - - - -static URI_INLINE void URI_FUNC(PreventLeakage)(URI_TYPE(Uri) * uri, - unsigned int revertMask, UriMemoryManager * memory) { - if (revertMask & URI_NORMALIZE_SCHEME) { - memory->free(memory, (URI_CHAR *)uri->scheme.first); - uri->scheme.first = NULL; - uri->scheme.afterLast = NULL; - } - - if (revertMask & URI_NORMALIZE_USER_INFO) { - memory->free(memory, (URI_CHAR *)uri->userInfo.first); - uri->userInfo.first = NULL; - uri->userInfo.afterLast = NULL; - } - - if (revertMask & URI_NORMALIZE_HOST) { - if (uri->hostData.ipFuture.first != NULL) { - /* IPvFuture */ - memory->free(memory, (URI_CHAR *)uri->hostData.ipFuture.first); - uri->hostData.ipFuture.first = NULL; - uri->hostData.ipFuture.afterLast = NULL; - uri->hostText.first = NULL; - uri->hostText.afterLast = NULL; - } else if ((uri->hostText.first != NULL) - && (uri->hostData.ip4 == NULL) - && (uri->hostData.ip6 == NULL)) { - /* Regname */ - memory->free(memory, (URI_CHAR *)uri->hostText.first); - uri->hostText.first = NULL; - uri->hostText.afterLast = NULL; - } - } - - /* NOTE: Port cannot happen! */ - - if (revertMask & URI_NORMALIZE_PATH) { - URI_TYPE(PathSegment) * walker = uri->pathHead; - while (walker != NULL) { - URI_TYPE(PathSegment) * const next = walker->next; - if (walker->text.afterLast > walker->text.first) { - memory->free(memory, (URI_CHAR *)walker->text.first); - } - memory->free(memory, walker); - walker = next; - } - uri->pathHead = NULL; - uri->pathTail = NULL; - } - - if (revertMask & URI_NORMALIZE_QUERY) { - memory->free(memory, (URI_CHAR *)uri->query.first); - uri->query.first = NULL; - uri->query.afterLast = NULL; - } - - if (revertMask & URI_NORMALIZE_FRAGMENT) { - memory->free(memory, (URI_CHAR *)uri->fragment.first); - uri->fragment.first = NULL; - uri->fragment.afterLast = NULL; - } + const URI_CHAR ** afterLast, + UriMemoryManager * memory); + +void URI_FUNC(PreventLeakage)(URI_TYPE(Uri) * uri, unsigned int revertMask, + UriMemoryManager * memory) { + if (revertMask & URI_NORMALIZE_SCHEME) { + /* NOTE: A scheme cannot be the empty string + * so no need to compare .first with .afterLast, here. */ + memory->free(memory, (URI_CHAR *)uri->scheme.first); + uri->scheme.first = NULL; + uri->scheme.afterLast = NULL; + } + + if (revertMask & URI_NORMALIZE_USER_INFO) { + if (uri->userInfo.first != uri->userInfo.afterLast) { + memory->free(memory, (URI_CHAR *)uri->userInfo.first); + } + uri->userInfo.first = NULL; + uri->userInfo.afterLast = NULL; + } + + if (revertMask & URI_NORMALIZE_HOST) { + if (uri->hostData.ipFuture.first != NULL) { + /* IPvFuture */ + /* NOTE: An IPvFuture address cannot be the empty string + * so no need to compare .first with .afterLast, here. */ + memory->free(memory, (URI_CHAR *)uri->hostData.ipFuture.first); + uri->hostData.ipFuture.first = NULL; + uri->hostData.ipFuture.afterLast = NULL; + uri->hostText.first = NULL; + uri->hostText.afterLast = NULL; + } else if (uri->hostText.first != NULL) { + /* Regname */ + if (uri->hostText.first != uri->hostText.afterLast) { + memory->free(memory, (URI_CHAR *)uri->hostText.first); + } + uri->hostText.first = NULL; + uri->hostText.afterLast = NULL; + } + } + + /* NOTE: Port cannot happen! */ + + if (revertMask & URI_NORMALIZE_PATH) { + URI_TYPE(PathSegment) * walker = uri->pathHead; + while (walker != NULL) { + URI_TYPE(PathSegment) * const next = walker->next; + if (walker->text.afterLast > walker->text.first) { + memory->free(memory, (URI_CHAR *)walker->text.first); + } + memory->free(memory, walker); + walker = next; + } + uri->pathHead = NULL; + uri->pathTail = NULL; + } + + if (revertMask & URI_NORMALIZE_QUERY) { + if (uri->query.first != uri->query.afterLast) { + memory->free(memory, (URI_CHAR *)uri->query.first); + } + uri->query.first = NULL; + uri->query.afterLast = NULL; + } + + if (revertMask & URI_NORMALIZE_FRAGMENT) { + if (uri->fragment.first != uri->fragment.afterLast) { + memory->free(memory, (URI_CHAR *)uri->fragment.first); + } + uri->fragment.first = NULL; + uri->fragment.afterLast = NULL; + } } - - static URI_INLINE UriBool URI_FUNC(ContainsUppercaseLetters)(const URI_CHAR * first, - const URI_CHAR * afterLast) { - if ((first != NULL) && (afterLast != NULL) && (afterLast > first)) { - const URI_CHAR * i = first; - for (; i < afterLast; i++) { - /* 6.2.2.1 Case Normalization: uppercase letters in scheme or host */ - if ((*i >= _UT('A')) && (*i <= _UT('Z'))) { - return URI_TRUE; - } - } - } - return URI_FALSE; + const URI_CHAR * afterLast) { + if ((first != NULL) && (afterLast != NULL) && (afterLast > first)) { + const URI_CHAR * i = first; + for (; i < afterLast; i++) { + /* 6.2.2.1 Case Normalization: uppercase letters in scheme or host */ + if ((*i >= _UT('A')) && (*i <= _UT('Z'))) { + return URI_TRUE; + } + } + } + return URI_FALSE; } - - -static URI_INLINE UriBool URI_FUNC(ContainsUglyPercentEncoding)(const URI_CHAR * first, - const URI_CHAR * afterLast) { - if ((first != NULL) && (afterLast != NULL) && (afterLast > first)) { - const URI_CHAR * i = first; - for (; i + 2 < afterLast; i++) { - if (i[0] == _UT('%')) { - /* 6.2.2.1 Case Normalization: * - * lowercase percent-encodings */ - if (((i[1] >= _UT('a')) && (i[1] <= _UT('f'))) - || ((i[2] >= _UT('a')) && (i[2] <= _UT('f')))) { - return URI_TRUE; - } else { - /* 6.2.2.2 Percent-Encoding Normalization: * - * percent-encoded unreserved characters */ - const unsigned char left = URI_FUNC(HexdigToInt)(i[1]); - const unsigned char right = URI_FUNC(HexdigToInt)(i[2]); - const int code = 16 * left + right; - if (uriIsUnreserved(code)) { - return URI_TRUE; - } - } - } - } - } - return URI_FALSE; +static URI_INLINE UriBool URI_FUNC(ContainsUglyPercentEncoding)( + const URI_CHAR * first, const URI_CHAR * afterLast) { + if ((first != NULL) && (afterLast != NULL) && (afterLast > first)) { + const URI_CHAR * i = first; + for (; i + 2 < afterLast; i++) { + if (i[0] == _UT('%')) { + /* 6.2.2.1 Case Normalization: * + * lowercase percent-encodings */ + if (((i[1] >= _UT('a')) && (i[1] <= _UT('f'))) + || ((i[2] >= _UT('a')) && (i[2] <= _UT('f')))) { + return URI_TRUE; + } else { + /* 6.2.2.2 Percent-Encoding Normalization: * + * percent-encoded unreserved characters */ + const unsigned char left = URI_FUNC(HexdigToInt)(i[1]); + const unsigned char right = URI_FUNC(HexdigToInt)(i[2]); + const int code = 16 * left + right; + if (uriIsUnreserved(code)) { + return URI_TRUE; + } + } + } + } + } + return URI_FALSE; } - - static URI_INLINE void URI_FUNC(LowercaseInplace)(const URI_CHAR * first, - const URI_CHAR * afterLast) { - if ((first != NULL) && (afterLast != NULL) && (afterLast > first)) { - URI_CHAR * i = (URI_CHAR *)first; - const int lowerUpperDiff = (_UT('a') - _UT('A')); - for (; i < afterLast; i++) { - if ((*i >= _UT('A')) && (*i <=_UT('Z'))) { - *i = (URI_CHAR)(*i + lowerUpperDiff); - } - } - } + const URI_CHAR * afterLast) { + if ((first != NULL) && (afterLast != NULL) && (afterLast > first)) { + URI_CHAR * i = (URI_CHAR *)first; + const int lowerUpperDiff = (_UT('a') - _UT('A')); + for (; i < afterLast; i++) { + if ((*i >= _UT('A')) && (*i <= _UT('Z'))) { + *i = (URI_CHAR)(*i + lowerUpperDiff); + } + } + } } - +static URI_INLINE void +URI_FUNC(LowercaseInplaceExceptPercentEncoding)(const URI_CHAR * first, + const URI_CHAR * afterLast) { + if ((first != NULL) && (afterLast != NULL) && (afterLast > first)) { + URI_CHAR * i = (URI_CHAR *)first; + const int lowerUpperDiff = (_UT('a') - _UT('A')); + for (; i < afterLast; i++) { + if ((*i >= _UT('A')) && (*i <= _UT('Z'))) { + *i = (URI_CHAR)(*i + lowerUpperDiff); + } else if (*i == _UT('%')) { + if (i + 3 >= afterLast) { + return; + } + i += 2; + } + } + } +} static URI_INLINE UriBool URI_FUNC(LowercaseMalloc)(const URI_CHAR ** first, - const URI_CHAR ** afterLast, UriMemoryManager * memory) { - int lenInChars; - const int lowerUpperDiff = (_UT('a') - _UT('A')); - URI_CHAR * buffer; - int i = 0; - - if ((first == NULL) || (afterLast == NULL) || (*first == NULL) - || (*afterLast == NULL)) { - return URI_FALSE; - } - - lenInChars = (int)(*afterLast - *first); - if (lenInChars == 0) { - return URI_TRUE; - } else if (lenInChars < 0) { - return URI_FALSE; - } - - buffer = memory->malloc(memory, lenInChars * sizeof(URI_CHAR)); - if (buffer == NULL) { - return URI_FALSE; - } - - for (; i < lenInChars; i++) { - if (((*first)[i] >= _UT('A')) && ((*first)[i] <=_UT('Z'))) { - buffer[i] = (URI_CHAR)((*first)[i] + lowerUpperDiff); - } else { - buffer[i] = (*first)[i]; - } - } - - *first = buffer; - *afterLast = buffer + lenInChars; - return URI_TRUE; + const URI_CHAR ** afterLast, + UriMemoryManager * memory) { + const int lowerUpperDiff = (_UT('a') - _UT('A')); + URI_CHAR * buffer; + size_t i = 0; + + if ((first == NULL) || (afterLast == NULL) || (*first == NULL) + || (*afterLast == NULL)) { + return URI_FALSE; + } + + const size_t lenInChars = *afterLast - *first; + if (lenInChars == 0) { + return URI_TRUE; + } + + // Detect and avoid integer overflow + if (lenInChars > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_FALSE; + } + + buffer = memory->malloc(memory, lenInChars * sizeof(URI_CHAR)); + if (buffer == NULL) { + return URI_FALSE; + } + + for (; i < lenInChars; i++) { + if (((*first)[i] >= _UT('A')) && ((*first)[i] <= _UT('Z'))) { + buffer[i] = (URI_CHAR)((*first)[i] + lowerUpperDiff); + } else { + buffer[i] = (*first)[i]; + } + } + + *first = buffer; + *afterLast = buffer + lenInChars; + return URI_TRUE; } - - /* NOTE: Implementation must stay inplace-compatible */ -static URI_INLINE void URI_FUNC(FixPercentEncodingEngine)( - const URI_CHAR * inFirst, const URI_CHAR * inAfterLast, - const URI_CHAR * outFirst, const URI_CHAR ** outAfterLast) { - URI_CHAR * write = (URI_CHAR *)outFirst; - const int lenInChars = (int)(inAfterLast - inFirst); - int i = 0; - - /* All but last two */ - for (; i + 2 < lenInChars; i++) { - if (inFirst[i] != _UT('%')) { - write[0] = inFirst[i]; - write++; - } else { - /* 6.2.2.2 Percent-Encoding Normalization: * - * percent-encoded unreserved characters */ - const URI_CHAR one = inFirst[i + 1]; - const URI_CHAR two = inFirst[i + 2]; - const unsigned char left = URI_FUNC(HexdigToInt)(one); - const unsigned char right = URI_FUNC(HexdigToInt)(two); - const int code = 16 * left + right; - if (uriIsUnreserved(code)) { - write[0] = (URI_CHAR)(code); - write++; - } else { - /* 6.2.2.1 Case Normalization: * - * lowercase percent-encodings */ - write[0] = _UT('%'); - write[1] = URI_FUNC(HexToLetter)(left); - write[2] = URI_FUNC(HexToLetter)(right); - write += 3; - } - - i += 2; /* For the two chars of the percent group we just ate */ - } - } - - /* Last two */ - for (; i < lenInChars; i++) { - write[0] = inFirst[i]; - write++; - } - - *outAfterLast = write; +static URI_INLINE void +URI_FUNC(FixPercentEncodingEngine)(const URI_CHAR * inFirst, const URI_CHAR * inAfterLast, + const URI_CHAR * outFirst, + const URI_CHAR ** outAfterLast) { + URI_CHAR * write = (URI_CHAR *)outFirst; + const size_t lenInChars = inAfterLast - inFirst; + size_t i = 0; + + /* All but last two */ + for (; i + 2 < lenInChars; i++) { + if (inFirst[i] != _UT('%')) { + write[0] = inFirst[i]; + write++; + } else { + /* 6.2.2.2 Percent-Encoding Normalization: * + * percent-encoded unreserved characters */ + const URI_CHAR one = inFirst[i + 1]; + const URI_CHAR two = inFirst[i + 2]; + const unsigned char left = URI_FUNC(HexdigToInt)(one); + const unsigned char right = URI_FUNC(HexdigToInt)(two); + const int code = 16 * left + right; + if (uriIsUnreserved(code)) { + write[0] = (URI_CHAR)(code); + write++; + } else { + /* 6.2.2.1 Case Normalization: * + * uppercase percent-encodings */ + write[0] = _UT('%'); + write[1] = URI_FUNC(HexToLetterEx)(left, URI_TRUE); + write[2] = URI_FUNC(HexToLetterEx)(right, URI_TRUE); + write += 3; + } + + i += 2; /* For the two chars of the percent group we just ate */ + } + } + + /* Last two */ + for (; i < lenInChars; i++) { + write[0] = inFirst[i]; + write++; + } + + *outAfterLast = write; } - - static URI_INLINE void URI_FUNC(FixPercentEncodingInplace)(const URI_CHAR * first, - const URI_CHAR ** afterLast) { - /* Death checks */ - if ((first == NULL) || (afterLast == NULL) || (*afterLast == NULL)) { - return; - } - - /* Fix inplace */ - URI_FUNC(FixPercentEncodingEngine)(first, *afterLast, first, afterLast); + const URI_CHAR ** afterLast) { + /* Death checks */ + if ((first == NULL) || (afterLast == NULL) || (*afterLast == NULL)) { + return; + } + + /* Fix inplace */ + URI_FUNC(FixPercentEncodingEngine)(first, *afterLast, first, afterLast); } - - static URI_INLINE UriBool URI_FUNC(FixPercentEncodingMalloc)(const URI_CHAR ** first, - const URI_CHAR ** afterLast, UriMemoryManager * memory) { - int lenInChars; - URI_CHAR * buffer; - - /* Death checks */ - if ((first == NULL) || (afterLast == NULL) - || (*first == NULL) || (*afterLast == NULL)) { - return URI_FALSE; - } - - /* Old text length */ - lenInChars = (int)(*afterLast - *first); - if (lenInChars == 0) { - return URI_TRUE; - } else if (lenInChars < 0) { - return URI_FALSE; - } - - /* New buffer */ - buffer = memory->malloc(memory, lenInChars * sizeof(URI_CHAR)); - if (buffer == NULL) { - return URI_FALSE; - } - - /* Fix on copy */ - URI_FUNC(FixPercentEncodingEngine)(*first, *afterLast, buffer, afterLast); - *first = buffer; - return URI_TRUE; + const URI_CHAR ** afterLast, + UriMemoryManager * memory) { + URI_CHAR * buffer; + + /* Death checks */ + if ((first == NULL) || (afterLast == NULL) || (*first == NULL) + || (*afterLast == NULL)) { + return URI_FALSE; + } + + /* Old text length */ + const size_t lenInChars = *afterLast - *first; + if (lenInChars == 0) { + return URI_TRUE; + } + + // Detect and avoid integer overflow + if (lenInChars > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_FALSE; + } + + /* New buffer */ + buffer = memory->malloc(memory, lenInChars * sizeof(URI_CHAR)); + if (buffer == NULL) { + return URI_FALSE; + } + + /* Fix on copy */ + URI_FUNC(FixPercentEncodingEngine)(*first, *afterLast, buffer, afterLast); + *first = buffer; + return URI_TRUE; } - - -static URI_INLINE UriBool URI_FUNC(MakeRangeOwner)(unsigned int * doneMask, - unsigned int maskTest, URI_TYPE(TextRange) * range, - UriMemoryManager * memory) { - if (((*doneMask & maskTest) == 0) - && (range->first != NULL) - && (range->afterLast != NULL) - && (range->afterLast > range->first)) { - const int lenInChars = (int)(range->afterLast - range->first); - const int lenInBytes = lenInChars * sizeof(URI_CHAR); - URI_CHAR * dup = memory->malloc(memory, lenInBytes); - if (dup == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - memcpy(dup, range->first, lenInBytes); - range->first = dup; - range->afterLast = dup + lenInChars; - *doneMask |= maskTest; - } - return URI_TRUE; +static URI_INLINE UriBool URI_FUNC(MakeRangeOwner)(unsigned int * revertMask, + unsigned int maskTest, + URI_TYPE(TextRange) * range, + UriMemoryManager * memory) { + if (((*revertMask & maskTest) == 0) && (range->first != NULL) + && (range->afterLast != NULL) && (range->afterLast > range->first)) { + if (URI_FUNC(CopyRange)(range, range, memory) == URI_FALSE) { + return URI_FALSE; + } + *revertMask |= maskTest; + } + return URI_TRUE; } +static URI_INLINE UriBool URI_FUNC(MakeOwnerEngine)(URI_TYPE(Uri) * uri, + unsigned int * revertMask, + UriMemoryManager * memory) { + URI_TYPE(PathSegment) * walker = uri->pathHead; + if (!URI_FUNC(MakeRangeOwner)(revertMask, URI_NORMALIZE_SCHEME, &(uri->scheme), + memory) + || !URI_FUNC(MakeRangeOwner)(revertMask, URI_NORMALIZE_USER_INFO, + &(uri->userInfo), memory) + || !URI_FUNC(MakeRangeOwner)(revertMask, URI_NORMALIZE_QUERY, &(uri->query), + memory) + || !URI_FUNC(MakeRangeOwner)(revertMask, URI_NORMALIZE_FRAGMENT, &(uri->fragment), + memory)) { + return URI_FALSE; /* Raises malloc error */ + } + + /* Host */ + if ((*revertMask & URI_NORMALIZE_HOST) == 0) { + if (uri->hostData.ipFuture.first != NULL) { + /* IPvFuture */ + if (!URI_FUNC(MakeRangeOwner)(revertMask, URI_NORMALIZE_HOST, + &(uri->hostData.ipFuture), memory)) { + return URI_FALSE; /* Raises malloc error */ + } + uri->hostText.first = uri->hostData.ipFuture.first; + uri->hostText.afterLast = uri->hostData.ipFuture.afterLast; + } else if (uri->hostText.first != NULL) { + /* Regname */ + if (!URI_FUNC(MakeRangeOwner)(revertMask, URI_NORMALIZE_HOST, + &(uri->hostText), memory)) { + return URI_FALSE; /* Raises malloc error */ + } + } + } + + /* Path */ + if ((*revertMask & URI_NORMALIZE_PATH) == 0) { + while (walker != NULL) { + if (!URI_FUNC(MakeRangeOwner)(revertMask, 0, &(walker->text), memory)) { + /* Free allocations done so far and kill path */ + + /* Kill path to one before walker (if any) */ + URI_TYPE(PathSegment) * ranger = uri->pathHead; + while (ranger != walker) { + URI_TYPE(PathSegment) * const next = ranger->next; + if ((ranger->text.first != NULL) && (ranger->text.afterLast != NULL) + && (ranger->text.afterLast > ranger->text.first)) { + memory->free(memory, (URI_CHAR *)ranger->text.first); + } + memory->free(memory, ranger); + ranger = next; + } + + /* Kill path from walker */ + while (walker != NULL) { + URI_TYPE(PathSegment) * const next = walker->next; + memory->free(memory, walker); + walker = next; + } + + uri->pathHead = NULL; + uri->pathTail = NULL; + return URI_FALSE; /* Raises malloc error */ + } + walker = walker->next; + } + *revertMask |= URI_NORMALIZE_PATH; + } + + /* Port text, must come last so we don't have to undo that one if it fails. * + * Otherwise we would need and extra enum flag for it although the port * + * cannot go unnormalized... */ + if (!URI_FUNC(MakeRangeOwner)(revertMask, 0, &(uri->portText), memory)) { + return URI_FALSE; /* Raises malloc error */ + } + + return URI_TRUE; +} - -static URI_INLINE UriBool URI_FUNC(MakeOwner)(URI_TYPE(Uri) * uri, - unsigned int * doneMask, UriMemoryManager * memory) { - URI_TYPE(PathSegment) * walker = uri->pathHead; - if (!URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_SCHEME, - &(uri->scheme), memory) - || !URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_USER_INFO, - &(uri->userInfo), memory) - || !URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_QUERY, - &(uri->query), memory) - || !URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_FRAGMENT, - &(uri->fragment), memory)) { - return URI_FALSE; /* Raises malloc error */ - } - - /* Host */ - if ((*doneMask & URI_NORMALIZE_HOST) == 0) { - if ((uri->hostData.ip4 == NULL) - && (uri->hostData.ip6 == NULL)) { - if (uri->hostData.ipFuture.first != NULL) { - /* IPvFuture */ - if (!URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_HOST, - &(uri->hostData.ipFuture), memory)) { - return URI_FALSE; /* Raises malloc error */ - } - uri->hostText.first = uri->hostData.ipFuture.first; - uri->hostText.afterLast = uri->hostData.ipFuture.afterLast; - } else if (uri->hostText.first != NULL) { - /* Regname */ - if (!URI_FUNC(MakeRangeOwner)(doneMask, URI_NORMALIZE_HOST, - &(uri->hostText), memory)) { - return URI_FALSE; /* Raises malloc error */ - } - } - } - } - - /* Path */ - if ((*doneMask & URI_NORMALIZE_PATH) == 0) { - while (walker != NULL) { - if (!URI_FUNC(MakeRangeOwner)(doneMask, 0, &(walker->text), memory)) { - /* Free allocations done so far and kill path */ - - /* Kill path to one before walker (if any) */ - URI_TYPE(PathSegment) * ranger = uri->pathHead; - while (ranger != walker) { - URI_TYPE(PathSegment) * const next = ranger->next; - if ((ranger->text.first != NULL) - && (ranger->text.afterLast != NULL) - && (ranger->text.afterLast > ranger->text.first)) { - memory->free(memory, (URI_CHAR *)ranger->text.first); - } - memory->free(memory, ranger); - ranger = next; - } - - /* Kill path from walker */ - while (walker != NULL) { - URI_TYPE(PathSegment) * const next = walker->next; - memory->free(memory, walker); - walker = next; - } - - uri->pathHead = NULL; - uri->pathTail = NULL; - return URI_FALSE; /* Raises malloc error */ - } - walker = walker->next; - } - *doneMask |= URI_NORMALIZE_PATH; - } - - /* Port text, must come last so we don't have to undo that one if it fails. * - * Otherwise we would need and extra enum flag for it although the port * - * cannot go unnormalized... */ - if (!URI_FUNC(MakeRangeOwner)(doneMask, 0, &(uri->portText), memory)) { - return URI_FALSE; /* Raises malloc error */ - } - - return URI_TRUE; +unsigned int URI_FUNC(NormalizeSyntaxMaskRequired)(const URI_TYPE(Uri) * uri) { + unsigned int outMask = URI_NORMALIZED; /* for NULL uri */ + URI_FUNC(NormalizeSyntaxMaskRequiredEx)(uri, &outMask); + return outMask; } +int URI_FUNC(NormalizeSyntaxMaskRequiredEx)(const URI_TYPE(Uri) * uri, + unsigned int * outMask) { + UriMemoryManager * const memory = NULL; /* no use of memory manager */ + +# if defined(__GNUC__) \ + && ((__GNUC__ > 4) \ + || ((__GNUC__ == 4) && defined(__GNUC_MINOR__) && (__GNUC_MINOR__ >= 2))) + /* Slower code that fixes a warning, not sure if this is a smart idea */ + URI_TYPE(Uri) writeableClone; +# endif + + if ((uri == NULL) || (outMask == NULL)) { + return URI_ERROR_NULL; + } + +# if defined(__GNUC__) \ + && ((__GNUC__ > 4) \ + || ((__GNUC__ == 4) && defined(__GNUC_MINOR__) && (__GNUC_MINOR__ >= 2))) + /* Slower code that fixes a warning, not sure if this is a smart idea */ + memcpy(&writeableClone, uri, 1 * sizeof(URI_TYPE(Uri))); + URI_FUNC(NormalizeSyntaxEngine)(&writeableClone, 0, outMask, memory); +# else + URI_FUNC(NormalizeSyntaxEngine)((URI_TYPE(Uri) *)uri, 0, outMask, memory); +# endif + return URI_SUCCESS; +} +int URI_FUNC(NormalizeSyntaxEx)(URI_TYPE(Uri) * uri, unsigned int mask) { + return URI_FUNC(NormalizeSyntaxExMm)(uri, mask, NULL); +} -unsigned int URI_FUNC(NormalizeSyntaxMaskRequired)(const URI_TYPE(Uri) * uri) { - unsigned int outMask = URI_NORMALIZED; /* for NULL uri */ - URI_FUNC(NormalizeSyntaxMaskRequiredEx)(uri, &outMask); - return outMask; +int URI_FUNC(NormalizeSyntaxExMm)(URI_TYPE(Uri) * uri, unsigned int mask, + UriMemoryManager * memory) { + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + return URI_FUNC(NormalizeSyntaxEngine)(uri, mask, NULL, memory); } +int URI_FUNC(NormalizeSyntax)(URI_TYPE(Uri) * uri) { + return URI_FUNC(NormalizeSyntaxEx)(uri, (unsigned int)-1); +} +static const URI_CHAR * URI_FUNC(PastLeadingZeros)(const URI_CHAR * first, + const URI_CHAR * afterLast) { + assert(first != NULL); + assert(afterLast != NULL); + assert(first != afterLast); + + /* Find the first non-zero character */ + const URI_CHAR * remainderFirst = first; + while ((remainderFirst < afterLast) && (remainderFirst[0] == _UT('0'))) { + remainderFirst++; + } + + /* Is the string /all/ zeros? */ + if (remainderFirst == afterLast) { + /* Yes, and length is >=1 because we ruled out the empty string earlier; + * pull back onto rightmost zero */ + assert(remainderFirst > first); + remainderFirst--; + assert(remainderFirst[0] == _UT('0')); + } + + return remainderFirst; +} -int URI_FUNC(NormalizeSyntaxMaskRequiredEx)(const URI_TYPE(Uri) * uri, - unsigned int * outMask) { - UriMemoryManager * const memory = NULL; /* no use of memory manager */ +static void URI_FUNC(DropLeadingZerosInplace)(URI_CHAR * first, + const URI_CHAR ** afterLast) { + assert(first != NULL); + assert(afterLast != NULL); + assert(*afterLast != NULL); -#if defined(__GNUC__) && ((__GNUC__ > 4) \ - || ((__GNUC__ == 4) && defined(__GNUC_MINOR__) && (__GNUC_MINOR__ >= 2))) - /* Slower code that fixes a warning, not sure if this is a smart idea */ - URI_TYPE(Uri) writeableClone; -#endif + if (first == *afterLast) { + return; + } - if ((uri == NULL) || (outMask == NULL)) { - return URI_ERROR_NULL; - } + const URI_CHAR * const remainderFirst = URI_FUNC(PastLeadingZeros)(first, *afterLast); -#if defined(__GNUC__) && ((__GNUC__ > 4) \ - || ((__GNUC__ == 4) && defined(__GNUC_MINOR__) && (__GNUC_MINOR__ >= 2))) - /* Slower code that fixes a warning, not sure if this is a smart idea */ - memcpy(&writeableClone, uri, 1 * sizeof(URI_TYPE(Uri))); - URI_FUNC(NormalizeSyntaxEngine)(&writeableClone, 0, outMask, memory); -#else - URI_FUNC(NormalizeSyntaxEngine)((URI_TYPE(Uri) *)uri, 0, outMask, memory); -#endif - return URI_SUCCESS; + if (remainderFirst > first) { + const size_t remainderLen = *afterLast - remainderFirst; + memmove(first, remainderFirst, remainderLen * sizeof(URI_CHAR)); + first[remainderLen] = _UT('\0'); + *afterLast = first + remainderLen; + } } +static void URI_FUNC(AdvancePastLeadingZeros)(const URI_CHAR ** first, + const URI_CHAR * afterLast) { + assert(first != NULL); + assert(*first != NULL); + assert(afterLast != NULL); + if (*first == afterLast) { + return; + } -int URI_FUNC(NormalizeSyntaxEx)(URI_TYPE(Uri) * uri, unsigned int mask) { - return URI_FUNC(NormalizeSyntaxExMm)(uri, mask, NULL); -} - + const URI_CHAR * const remainderFirst = URI_FUNC(PastLeadingZeros)(*first, afterLast); + /* Cut off leading zeros */ + *first = remainderFirst; +} -int URI_FUNC(NormalizeSyntaxExMm)(URI_TYPE(Uri) * uri, unsigned int mask, - UriMemoryManager * memory) { - URI_CHECK_MEMORY_MANAGER(memory); /* may return */ - return URI_FUNC(NormalizeSyntaxEngine)(uri, mask, NULL, memory); +static URI_INLINE int URI_FUNC(NormalizeSyntaxEngine)(URI_TYPE(Uri) * uri, + unsigned int inMask, + unsigned int * outMask, + UriMemoryManager * memory) { + unsigned int revertMask = URI_NORMALIZED; + + /* Not just doing inspection? -> memory manager required! */ + if (outMask == NULL) { + assert(memory != NULL); + } + + if (uri == NULL) { + if (outMask != NULL) { + *outMask = URI_NORMALIZED; + return URI_SUCCESS; + } else { + return URI_ERROR_NULL; + } + } + + if (outMask != NULL) { + /* Reset mask */ + *outMask = URI_NORMALIZED; + } else if (inMask == URI_NORMALIZED) { + /* Nothing to do */ + return URI_SUCCESS; + } + + /* Scheme, host */ + if (outMask != NULL) { + const UriBool normalizeScheme = + URI_FUNC(ContainsUppercaseLetters)(uri->scheme.first, uri->scheme.afterLast); + const UriBool normalizeHostCase = URI_FUNC(ContainsUppercaseLetters)( + uri->hostText.first, uri->hostText.afterLast); + if (normalizeScheme) { + *outMask |= URI_NORMALIZE_SCHEME; + } + + if (normalizeHostCase) { + *outMask |= URI_NORMALIZE_HOST; + } else { + const UriBool normalizeHostPrecent = URI_FUNC(ContainsUglyPercentEncoding)( + uri->hostText.first, uri->hostText.afterLast); + if (normalizeHostPrecent) { + *outMask |= URI_NORMALIZE_HOST; + } + } + } else { + /* Scheme */ + if ((inMask & URI_NORMALIZE_SCHEME) && (uri->scheme.first != NULL)) { + if (uri->owner) { + URI_FUNC(LowercaseInplace)(uri->scheme.first, uri->scheme.afterLast); + } else { + if (!URI_FUNC(LowercaseMalloc)(&(uri->scheme.first), + &(uri->scheme.afterLast), memory)) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + return URI_ERROR_MALLOC; + } + revertMask |= URI_NORMALIZE_SCHEME; + } + } + + /* Host */ + if (inMask & URI_NORMALIZE_HOST) { + if (uri->hostData.ipFuture.first != NULL) { + /* IPvFuture */ + if (uri->owner) { + URI_FUNC(LowercaseInplace)(uri->hostData.ipFuture.first, + uri->hostData.ipFuture.afterLast); + } else { + if (!URI_FUNC(LowercaseMalloc)(&(uri->hostData.ipFuture.first), + &(uri->hostData.ipFuture.afterLast), + memory)) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + return URI_ERROR_MALLOC; + } + revertMask |= URI_NORMALIZE_HOST; + } + uri->hostText.first = uri->hostData.ipFuture.first; + uri->hostText.afterLast = uri->hostData.ipFuture.afterLast; + } else if ((uri->hostText.first != NULL) && (uri->hostData.ip4 == NULL)) { + /* Regname or IPv6 */ + if (uri->owner) { + URI_FUNC(FixPercentEncodingInplace)(uri->hostText.first, + &(uri->hostText.afterLast)); + } else { + if (!URI_FUNC(FixPercentEncodingMalloc)( + &(uri->hostText.first), &(uri->hostText.afterLast), memory)) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + return URI_ERROR_MALLOC; + } + revertMask |= URI_NORMALIZE_HOST; + } + + URI_FUNC(LowercaseInplaceExceptPercentEncoding)(uri->hostText.first, + uri->hostText.afterLast); + } + } + } + + /* Port */ + if (outMask != NULL) { + /* Is there a port even? */ + if (uri->portText.first != NULL) { + /* Determine whether the port is already normalized, i.e. either "", "0" or no + * leading zeros */ + const size_t portLen = uri->portText.afterLast - uri->portText.first; + if ((portLen > 1) && (uri->portText.first[0] == _UT('0'))) { + *outMask |= URI_NORMALIZE_PORT; + } + } + } else { + /* Normalize the port, i.e. drop leading zeros (except for string "0") */ + if ((inMask & URI_NORMALIZE_PORT) && (uri->portText.first != NULL)) { + if (uri->owner) { + URI_FUNC(DropLeadingZerosInplace)((URI_CHAR *)uri->portText.first, + &(uri->portText.afterLast)); + } else { + URI_FUNC(AdvancePastLeadingZeros)(&(uri->portText.first), + uri->portText.afterLast); + } + } + } + + /* User info */ + if (outMask != NULL) { + const UriBool normalizeUserInfo = URI_FUNC(ContainsUglyPercentEncoding)( + uri->userInfo.first, uri->userInfo.afterLast); + if (normalizeUserInfo) { + *outMask |= URI_NORMALIZE_USER_INFO; + } + } else { + if ((inMask & URI_NORMALIZE_USER_INFO) && (uri->userInfo.first != NULL)) { + if (uri->owner) { + URI_FUNC(FixPercentEncodingInplace)(uri->userInfo.first, + &(uri->userInfo.afterLast)); + } else { + if (!URI_FUNC(FixPercentEncodingMalloc)( + &(uri->userInfo.first), &(uri->userInfo.afterLast), memory)) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + return URI_ERROR_MALLOC; + } + revertMask |= URI_NORMALIZE_USER_INFO; + } + } + } + + /* Path */ + if (outMask != NULL) { + const URI_TYPE(PathSegment) * walker = uri->pathHead; + while (walker != NULL) { + const URI_CHAR * const first = walker->text.first; + const URI_CHAR * const afterLast = walker->text.afterLast; + if ((first != NULL) && (afterLast != NULL) && (afterLast > first) + && ((((afterLast - first) == 1) && (first[0] == _UT('.'))) + || (((afterLast - first) == 2) && (first[0] == _UT('.')) + && (first[1] == _UT('.'))) + || URI_FUNC(ContainsUglyPercentEncoding)(first, afterLast))) { + *outMask |= URI_NORMALIZE_PATH; + break; + } + walker = walker->next; + } + } else if (inMask & URI_NORMALIZE_PATH) { + URI_TYPE(PathSegment) * walker; + const UriBool relative = + ((uri->scheme.first == NULL) && !uri->absolutePath) ? URI_TRUE : URI_FALSE; + + /* Fix percent-encoding for each segment */ + walker = uri->pathHead; + if (uri->owner) { + while (walker != NULL) { + URI_FUNC(FixPercentEncodingInplace)(walker->text.first, + &(walker->text.afterLast)); + walker = walker->next; + } + } else { + while (walker != NULL) { + if (!URI_FUNC(FixPercentEncodingMalloc)( + &(walker->text.first), &(walker->text.afterLast), memory)) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + return URI_ERROR_MALLOC; + } + walker = walker->next; + } + revertMask |= URI_NORMALIZE_PATH; + } + + /* 6.2.2.3 Path Segment Normalization */ + if (!URI_FUNC(RemoveDotSegmentsEx)( + uri, relative, + (uri->owner == URI_TRUE) || ((revertMask & URI_NORMALIZE_PATH) != 0), + memory)) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + return URI_ERROR_MALLOC; + } + URI_FUNC(FixEmptyTrailSegment)(uri, memory); + } + + /* Query, fragment */ + if (outMask != NULL) { + const UriBool normalizeQuery = + URI_FUNC(ContainsUglyPercentEncoding)(uri->query.first, uri->query.afterLast); + const UriBool normalizeFragment = URI_FUNC(ContainsUglyPercentEncoding)( + uri->fragment.first, uri->fragment.afterLast); + if (normalizeQuery) { + *outMask |= URI_NORMALIZE_QUERY; + } + + if (normalizeFragment) { + *outMask |= URI_NORMALIZE_FRAGMENT; + } + } else { + /* Query */ + if ((inMask & URI_NORMALIZE_QUERY) && (uri->query.first != NULL)) { + if (uri->owner) { + URI_FUNC(FixPercentEncodingInplace)(uri->query.first, + &(uri->query.afterLast)); + } else { + if (!URI_FUNC(FixPercentEncodingMalloc)( + &(uri->query.first), &(uri->query.afterLast), memory)) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + return URI_ERROR_MALLOC; + } + revertMask |= URI_NORMALIZE_QUERY; + } + } + + /* Fragment */ + if ((inMask & URI_NORMALIZE_FRAGMENT) && (uri->fragment.first != NULL)) { + if (uri->owner) { + URI_FUNC(FixPercentEncodingInplace)(uri->fragment.first, + &(uri->fragment.afterLast)); + } else { + if (!URI_FUNC(FixPercentEncodingMalloc)( + &(uri->fragment.first), &(uri->fragment.afterLast), memory)) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + return URI_ERROR_MALLOC; + } + revertMask |= URI_NORMALIZE_FRAGMENT; + } + } + } + + /* Dup all not duped yet */ + if ((outMask == NULL) && !uri->owner) { + if (!URI_FUNC(MakeOwnerEngine)(uri, &revertMask, memory)) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + return URI_ERROR_MALLOC; + } + uri->owner = URI_TRUE; + } + + return URI_SUCCESS; } +int URI_FUNC(MakeOwnerMm)(URI_TYPE(Uri) * uri, UriMemoryManager * memory) { + unsigned int revertMask = URI_NORMALIZED; + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ -int URI_FUNC(NormalizeSyntax)(URI_TYPE(Uri) * uri) { - return URI_FUNC(NormalizeSyntaxEx)(uri, (unsigned int)-1); -} + if (uri == NULL) { + return URI_ERROR_NULL; + } + if (uri->owner == URI_TRUE) { + return URI_SUCCESS; + } + if (!URI_FUNC(MakeOwnerEngine)(uri, &revertMask, memory)) { + URI_FUNC(PreventLeakage)(uri, revertMask, memory); + return URI_ERROR_MALLOC; + } -static URI_INLINE int URI_FUNC(NormalizeSyntaxEngine)(URI_TYPE(Uri) * uri, - unsigned int inMask, unsigned int * outMask, - UriMemoryManager * memory) { - unsigned int doneMask = URI_NORMALIZED; - - /* Not just doing inspection? -> memory manager required! */ - if (outMask == NULL) { - assert(memory != NULL); - } - - if (uri == NULL) { - if (outMask != NULL) { - *outMask = URI_NORMALIZED; - return URI_SUCCESS; - } else { - return URI_ERROR_NULL; - } - } - - if (outMask != NULL) { - /* Reset mask */ - *outMask = URI_NORMALIZED; - } else if (inMask == URI_NORMALIZED) { - /* Nothing to do */ - return URI_SUCCESS; - } - - /* Scheme, host */ - if (outMask != NULL) { - const UriBool normalizeScheme = URI_FUNC(ContainsUppercaseLetters)( - uri->scheme.first, uri->scheme.afterLast); - const UriBool normalizeHostCase = URI_FUNC(ContainsUppercaseLetters)( - uri->hostText.first, uri->hostText.afterLast); - if (normalizeScheme) { - *outMask |= URI_NORMALIZE_SCHEME; - } - - if (normalizeHostCase) { - *outMask |= URI_NORMALIZE_HOST; - } else { - const UriBool normalizeHostPrecent = URI_FUNC(ContainsUglyPercentEncoding)( - uri->hostText.first, uri->hostText.afterLast); - if (normalizeHostPrecent) { - *outMask |= URI_NORMALIZE_HOST; - } - } - } else { - /* Scheme */ - if ((inMask & URI_NORMALIZE_SCHEME) && (uri->scheme.first != NULL)) { - if (uri->owner) { - URI_FUNC(LowercaseInplace)(uri->scheme.first, uri->scheme.afterLast); - } else { - if (!URI_FUNC(LowercaseMalloc)(&(uri->scheme.first), &(uri->scheme.afterLast), memory)) { - URI_FUNC(PreventLeakage)(uri, doneMask, memory); - return URI_ERROR_MALLOC; - } - doneMask |= URI_NORMALIZE_SCHEME; - } - } - - /* Host */ - if (inMask & URI_NORMALIZE_HOST) { - if (uri->hostData.ipFuture.first != NULL) { - /* IPvFuture */ - if (uri->owner) { - URI_FUNC(LowercaseInplace)(uri->hostData.ipFuture.first, - uri->hostData.ipFuture.afterLast); - } else { - if (!URI_FUNC(LowercaseMalloc)(&(uri->hostData.ipFuture.first), - &(uri->hostData.ipFuture.afterLast), memory)) { - URI_FUNC(PreventLeakage)(uri, doneMask, memory); - return URI_ERROR_MALLOC; - } - doneMask |= URI_NORMALIZE_HOST; - } - uri->hostText.first = uri->hostData.ipFuture.first; - uri->hostText.afterLast = uri->hostData.ipFuture.afterLast; - } else if ((uri->hostText.first != NULL) - && (uri->hostData.ip4 == NULL) - && (uri->hostData.ip6 == NULL)) { - /* Regname */ - if (uri->owner) { - URI_FUNC(FixPercentEncodingInplace)(uri->hostText.first, - &(uri->hostText.afterLast)); - } else { - if (!URI_FUNC(FixPercentEncodingMalloc)( - &(uri->hostText.first), - &(uri->hostText.afterLast), - memory)) { - URI_FUNC(PreventLeakage)(uri, doneMask, memory); - return URI_ERROR_MALLOC; - } - doneMask |= URI_NORMALIZE_HOST; - } - - URI_FUNC(LowercaseInplace)(uri->hostText.first, - uri->hostText.afterLast); - } - } - } - - /* User info */ - if (outMask != NULL) { - const UriBool normalizeUserInfo = URI_FUNC(ContainsUglyPercentEncoding)( - uri->userInfo.first, uri->userInfo.afterLast); - if (normalizeUserInfo) { - *outMask |= URI_NORMALIZE_USER_INFO; - } - } else { - if ((inMask & URI_NORMALIZE_USER_INFO) && (uri->userInfo.first != NULL)) { - if (uri->owner) { - URI_FUNC(FixPercentEncodingInplace)(uri->userInfo.first, &(uri->userInfo.afterLast)); - } else { - if (!URI_FUNC(FixPercentEncodingMalloc)(&(uri->userInfo.first), - &(uri->userInfo.afterLast), memory)) { - URI_FUNC(PreventLeakage)(uri, doneMask, memory); - return URI_ERROR_MALLOC; - } - doneMask |= URI_NORMALIZE_USER_INFO; - } - } - } - - /* Path */ - if (outMask != NULL) { - const URI_TYPE(PathSegment) * walker = uri->pathHead; - while (walker != NULL) { - const URI_CHAR * const first = walker->text.first; - const URI_CHAR * const afterLast = walker->text.afterLast; - if ((first != NULL) - && (afterLast != NULL) - && (afterLast > first) - && ( - (((afterLast - first) == 1) - && (first[0] == _UT('.'))) - || - (((afterLast - first) == 2) - && (first[0] == _UT('.')) - && (first[1] == _UT('.'))) - || - URI_FUNC(ContainsUglyPercentEncoding)(first, afterLast) - )) { - *outMask |= URI_NORMALIZE_PATH; - break; - } - walker = walker->next; - } - } else if (inMask & URI_NORMALIZE_PATH) { - URI_TYPE(PathSegment) * walker; - const UriBool relative = ((uri->scheme.first == NULL) - && !uri->absolutePath) ? URI_TRUE : URI_FALSE; - - /* Fix percent-encoding for each segment */ - walker = uri->pathHead; - if (uri->owner) { - while (walker != NULL) { - URI_FUNC(FixPercentEncodingInplace)(walker->text.first, &(walker->text.afterLast)); - walker = walker->next; - } - } else { - while (walker != NULL) { - if (!URI_FUNC(FixPercentEncodingMalloc)(&(walker->text.first), - &(walker->text.afterLast), memory)) { - URI_FUNC(PreventLeakage)(uri, doneMask, memory); - return URI_ERROR_MALLOC; - } - walker = walker->next; - } - doneMask |= URI_NORMALIZE_PATH; - } - - /* 6.2.2.3 Path Segment Normalization */ - if (!URI_FUNC(RemoveDotSegmentsEx)(uri, relative, - (uri->owner == URI_TRUE) - || ((doneMask & URI_NORMALIZE_PATH) != 0), - memory)) { - URI_FUNC(PreventLeakage)(uri, doneMask, memory); - return URI_ERROR_MALLOC; - } - URI_FUNC(FixEmptyTrailSegment)(uri, memory); - } - - /* Query, fragment */ - if (outMask != NULL) { - const UriBool normalizeQuery = URI_FUNC(ContainsUglyPercentEncoding)( - uri->query.first, uri->query.afterLast); - const UriBool normalizeFragment = URI_FUNC(ContainsUglyPercentEncoding)( - uri->fragment.first, uri->fragment.afterLast); - if (normalizeQuery) { - *outMask |= URI_NORMALIZE_QUERY; - } - - if (normalizeFragment) { - *outMask |= URI_NORMALIZE_FRAGMENT; - } - } else { - /* Query */ - if ((inMask & URI_NORMALIZE_QUERY) && (uri->query.first != NULL)) { - if (uri->owner) { - URI_FUNC(FixPercentEncodingInplace)(uri->query.first, &(uri->query.afterLast)); - } else { - if (!URI_FUNC(FixPercentEncodingMalloc)(&(uri->query.first), - &(uri->query.afterLast), memory)) { - URI_FUNC(PreventLeakage)(uri, doneMask, memory); - return URI_ERROR_MALLOC; - } - doneMask |= URI_NORMALIZE_QUERY; - } - } - - /* Fragment */ - if ((inMask & URI_NORMALIZE_FRAGMENT) && (uri->fragment.first != NULL)) { - if (uri->owner) { - URI_FUNC(FixPercentEncodingInplace)(uri->fragment.first, &(uri->fragment.afterLast)); - } else { - if (!URI_FUNC(FixPercentEncodingMalloc)(&(uri->fragment.first), - &(uri->fragment.afterLast), memory)) { - URI_FUNC(PreventLeakage)(uri, doneMask, memory); - return URI_ERROR_MALLOC; - } - doneMask |= URI_NORMALIZE_FRAGMENT; - } - } - } - - /* Dup all not duped yet */ - if ((outMask == NULL) && !uri->owner) { - if (!URI_FUNC(MakeOwner)(uri, &doneMask, memory)) { - URI_FUNC(PreventLeakage)(uri, doneMask, memory); - return URI_ERROR_MALLOC; - } - uri->owner = URI_TRUE; - } - - return URI_SUCCESS; -} + uri->owner = URI_TRUE; + return URI_SUCCESS; +} +int URI_FUNC(MakeOwner)(URI_TYPE(Uri) * uri) { + return URI_FUNC(MakeOwnerMm)(uri, NULL); +} #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriNormalize.h b/cpp/src/arrow/vendored/uriparser/UriNormalize.h new file mode 100644 index 000000000000..376e7539be70 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriNormalize.h @@ -0,0 +1,75 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2018, Weijia Song + * Copyright (C) 2018, Sebastian Pipping + * Copyright (C) 2025, Máté Kocsis + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#if (defined(URI_PASS_ANSI) && !defined(URI_NORMALIZE_H_ANSI)) \ + || (defined(URI_PASS_UNICODE) && !defined(URI_NORMALIZE_H_UNICODE)) \ + || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* What encodings are enabled? */ +# include "UriDefsConfig.h" +# if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriNormalize.h" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriNormalize.h" +# undef URI_PASS_UNICODE +# endif +/* Only one pass for each encoding */ +# elif (defined(URI_PASS_ANSI) && !defined(URI_NORMALIZE_H_ANSI) \ + && defined(URI_ENABLE_ANSI)) \ + || (defined(URI_PASS_UNICODE) && !defined(URI_NORMALIZE_H_UNICODE) \ + && defined(URI_ENABLE_UNICODE)) +# ifdef URI_PASS_ANSI +# define URI_NORMALIZE_H_ANSI 1 +# include "UriDefsAnsi.h" +# else +# define URI_NORMALIZE_H_UNICODE 1 +# include "UriDefsUnicode.h" +# endif + +void URI_FUNC(PreventLeakage)(URI_TYPE(Uri) * uri, unsigned int revertMask, + UriMemoryManager * memory); + +# endif +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriNormalizeBase.c b/cpp/src/arrow/vendored/uriparser/UriNormalizeBase.c index 8dbe7fd19379..9c2acee1ee75 100644 --- a/cpp/src/arrow/vendored/uriparser/UriNormalizeBase.c +++ b/cpp/src/arrow/vendored/uriparser/UriNormalizeBase.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -38,82 +38,80 @@ */ #ifndef URI_DOXYGEN -# include "UriNormalizeBase.h" +# include "UriNormalizeBase.h" #endif - - UriBool uriIsUnreserved(int code) { - switch (code) { - case L'a': /* ALPHA */ - case L'A': - case L'b': - case L'B': - case L'c': - case L'C': - case L'd': - case L'D': - case L'e': - case L'E': - case L'f': - case L'F': - case L'g': - case L'G': - case L'h': - case L'H': - case L'i': - case L'I': - case L'j': - case L'J': - case L'k': - case L'K': - case L'l': - case L'L': - case L'm': - case L'M': - case L'n': - case L'N': - case L'o': - case L'O': - case L'p': - case L'P': - case L'q': - case L'Q': - case L'r': - case L'R': - case L's': - case L'S': - case L't': - case L'T': - case L'u': - case L'U': - case L'v': - case L'V': - case L'w': - case L'W': - case L'x': - case L'X': - case L'y': - case L'Y': - case L'z': - case L'Z': - case L'0': /* DIGIT */ - case L'1': - case L'2': - case L'3': - case L'4': - case L'5': - case L'6': - case L'7': - case L'8': - case L'9': - case L'-': /* "-" / "." / "_" / "~" */ - case L'.': - case L'_': - case L'~': - return URI_TRUE; + switch (code) { + case L'a': /* ALPHA */ + case L'A': + case L'b': + case L'B': + case L'c': + case L'C': + case L'd': + case L'D': + case L'e': + case L'E': + case L'f': + case L'F': + case L'g': + case L'G': + case L'h': + case L'H': + case L'i': + case L'I': + case L'j': + case L'J': + case L'k': + case L'K': + case L'l': + case L'L': + case L'm': + case L'M': + case L'n': + case L'N': + case L'o': + case L'O': + case L'p': + case L'P': + case L'q': + case L'Q': + case L'r': + case L'R': + case L's': + case L'S': + case L't': + case L'T': + case L'u': + case L'U': + case L'v': + case L'V': + case L'w': + case L'W': + case L'x': + case L'X': + case L'y': + case L'Y': + case L'z': + case L'Z': + case L'0': /* DIGIT */ + case L'1': + case L'2': + case L'3': + case L'4': + case L'5': + case L'6': + case L'7': + case L'8': + case L'9': + case L'-': /* "-" / "." / "_" / "~" */ + case L'.': + case L'_': + case L'~': + return URI_TRUE; - default: - return URI_FALSE; - } + default: + return URI_FALSE; + } } diff --git a/cpp/src/arrow/vendored/uriparser/UriNormalizeBase.h b/cpp/src/arrow/vendored/uriparser/UriNormalizeBase.h index a1290756409b..2ff1a01ad732 100644 --- a/cpp/src/arrow/vendored/uriparser/UriNormalizeBase.h +++ b/cpp/src/arrow/vendored/uriparser/UriNormalizeBase.h @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -38,16 +38,10 @@ */ #ifndef URI_NORMALIZE_BASE_H -#define URI_NORMALIZE_BASE_H 1 - - - -#include "UriBase.h" - +# define URI_NORMALIZE_BASE_H 1 +# include "UriBase.h" UriBool uriIsUnreserved(int code); - - #endif /* URI_NORMALIZE_BASE_H */ diff --git a/cpp/src/arrow/vendored/uriparser/UriParse.c b/cpp/src/arrow/vendored/uriparser/UriParse.c index f5972d8f4d34..9265465efd25 100644 --- a/cpp/src/arrow/vendored/uriparser/UriParse.c +++ b/cpp/src/arrow/vendored/uriparser/UriParse.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -47,842 +47,757 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriParse.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriParse.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriParse.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriParse.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -# include "UriIp4.h" -# include "UriCommon.h" -# include "UriMemory.h" -# include "UriParseBase.h" -#endif - - - -#define URI_SET_DIGIT \ - _UT('0'): \ - case _UT('1'): \ - case _UT('2'): \ - case _UT('3'): \ - case _UT('4'): \ - case _UT('5'): \ - case _UT('6'): \ - case _UT('7'): \ - case _UT('8'): \ - case _UT('9') - -#define URI_SET_HEX_LETTER_UPPER \ - _UT('A'): \ - case _UT('B'): \ - case _UT('C'): \ - case _UT('D'): \ - case _UT('E'): \ - case _UT('F') - -#define URI_SET_HEX_LETTER_LOWER \ - _UT('a'): \ - case _UT('b'): \ - case _UT('c'): \ - case _UT('d'): \ - case _UT('e'): \ - case _UT('f') - -#define URI_SET_HEXDIG \ - URI_SET_DIGIT: \ - case URI_SET_HEX_LETTER_UPPER: \ - case URI_SET_HEX_LETTER_LOWER - -#define URI_SET_ALPHA \ - URI_SET_HEX_LETTER_UPPER: \ - case URI_SET_HEX_LETTER_LOWER: \ - case _UT('g'): \ - case _UT('G'): \ - case _UT('h'): \ - case _UT('H'): \ - case _UT('i'): \ - case _UT('I'): \ - case _UT('j'): \ - case _UT('J'): \ - case _UT('k'): \ - case _UT('K'): \ - case _UT('l'): \ - case _UT('L'): \ - case _UT('m'): \ - case _UT('M'): \ - case _UT('n'): \ - case _UT('N'): \ - case _UT('o'): \ - case _UT('O'): \ - case _UT('p'): \ - case _UT('P'): \ - case _UT('q'): \ - case _UT('Q'): \ - case _UT('r'): \ - case _UT('R'): \ - case _UT('s'): \ - case _UT('S'): \ - case _UT('t'): \ - case _UT('T'): \ - case _UT('u'): \ - case _UT('U'): \ - case _UT('v'): \ - case _UT('V'): \ - case _UT('w'): \ - case _UT('W'): \ - case _UT('x'): \ - case _UT('X'): \ - case _UT('y'): \ - case _UT('Y'): \ - case _UT('z'): \ - case _UT('Z') - - - -static const URI_CHAR * URI_FUNC(ParseAuthority)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseAuthorityTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast); -static const URI_CHAR * URI_FUNC(ParseHexZero)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast); -static const URI_CHAR * URI_FUNC(ParseHierPart)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseIpFutLoop)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseIpFutStopGo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseIpLit2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseIPv6address2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseMustBeSegmentNzNc)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseOwnHost)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseOwnHost2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseOwnHostUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseOwnHostUserInfoNz)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseOwnPortUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseOwnUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParsePartHelperTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParsePathAbsEmpty)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParsePathAbsNoLeadSlash)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParsePathRootless)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParsePchar)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParsePctEncoded)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParsePctSubUnres)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParsePort)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast); -static const URI_CHAR * URI_FUNC(ParseQueryFrag)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseSegment)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseSegmentNz)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseSegmentNzNcOrScheme2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseUriReference)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseUriTail)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseUriTailTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); -static const URI_CHAR * URI_FUNC(ParseZeroMoreSlashSegs)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast, UriMemoryManager * memory); - -static UriBool URI_FUNC(OnExitOwnHost2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, UriMemoryManager * memory); -static UriBool URI_FUNC(OnExitOwnHostUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, UriMemoryManager * memory); -static UriBool URI_FUNC(OnExitOwnPortUserInfo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, UriMemoryManager * memory); -static UriBool URI_FUNC(OnExitSegmentNzNcOrScheme2)(URI_TYPE(ParserState) * state, const URI_CHAR * first, UriMemoryManager * memory); +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriIp4.h" +# include "UriCommon.h" +# include "UriMemory.h" +# include "UriParseBase.h" +# include "UriSets.h" +# endif + +static const URI_CHAR * URI_FUNC(ParseAuthority)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseAuthorityTwo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast); +static const URI_CHAR * URI_FUNC(ParseHexZero)(const URI_CHAR * first, + const URI_CHAR * afterLast); +static const URI_CHAR * URI_FUNC(ParseHierPart)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseIpFutLoop)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseIpLit2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseIPv6address2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseMustBeSegmentNzNc)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseOwnHost)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseOwnHost2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseOwnHostUserInfoNz)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseOwnPortUserInfo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseOwnUserInfo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParsePartHelperTwo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParsePathAbsEmpty)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParsePathAbsNoLeadSlash)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParsePathRootless)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParsePchar)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParsePctEncoded)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParsePctSubUnres)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParsePort)(const URI_CHAR * first, + const URI_CHAR * afterLast); +static const URI_CHAR * URI_FUNC(ParseQueryFrag)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseSegment)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseSegmentNz)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseSegmentNzNcOrScheme2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseUriReference)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseUriTail)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseUriTailTwo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static const URI_CHAR * URI_FUNC(ParseZeroMoreSlashSegs)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); + +static UriBool URI_FUNC(OnExitOwnHost2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + UriMemoryManager * memory); +static UriBool URI_FUNC(OnExitOwnHostUserInfo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + UriMemoryManager * memory); +static UriBool URI_FUNC(OnExitOwnPortUserInfo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + UriMemoryManager * memory); +static UriBool URI_FUNC(OnExitSegmentNzNcOrScheme2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + UriMemoryManager * memory); static void URI_FUNC(OnExitPartHelperTwo)(URI_TYPE(ParserState) * state); static void URI_FUNC(ResetParserStateExceptUri)(URI_TYPE(ParserState) * state); static UriBool URI_FUNC(PushPathSegment)(URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory); - -static void URI_FUNC(StopSyntax)(URI_TYPE(ParserState) * state, const URI_CHAR * errorPos, UriMemoryManager * memory); -static void URI_FUNC(StopMalloc)(URI_TYPE(ParserState) * state, UriMemoryManager * memory); - -static int URI_FUNC(ParseUriExMm)(URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory); + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory); +static void URI_FUNC(StopSyntax)(URI_TYPE(ParserState) * state, const URI_CHAR * errorPos, + UriMemoryManager * memory); +static void URI_FUNC(StopMalloc)(URI_TYPE(ParserState) * state, + UriMemoryManager * memory); +static int URI_FUNC(ParseUriExMm)(URI_TYPE(ParserState) * state, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory); static URI_INLINE void URI_FUNC(StopSyntax)(URI_TYPE(ParserState) * state, - const URI_CHAR * errorPos, UriMemoryManager * memory) { - URI_FUNC(FreeUriMembersMm)(state->uri, memory); - state->errorPos = errorPos; - state->errorCode = URI_ERROR_SYNTAX; + const URI_CHAR * errorPos, + UriMemoryManager * memory) { + URI_FUNC(FreeUriMembersMm)(state->uri, memory); + state->errorPos = errorPos; + state->errorCode = URI_ERROR_SYNTAX; } - - -static URI_INLINE void URI_FUNC(StopMalloc)(URI_TYPE(ParserState) * state, UriMemoryManager * memory) { - URI_FUNC(FreeUriMembersMm)(state->uri, memory); - state->errorPos = NULL; - state->errorCode = URI_ERROR_MALLOC; +static URI_INLINE void URI_FUNC(StopMalloc)(URI_TYPE(ParserState) * state, + UriMemoryManager * memory) { + URI_FUNC(FreeUriMembersMm)(state->uri, memory); + state->errorPos = NULL; + state->errorCode = URI_ERROR_MALLOC; } - - /* * [authority]-><[>[ipLit2][authorityTwo] * [authority]->[ownHostUserInfoNz] * [authority]-> */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseAuthority)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - /* "" regname host */ - state->uri->hostText.first = URI_FUNC(SafeToPointTo); - state->uri->hostText.afterLast = URI_FUNC(SafeToPointTo); - return afterLast; - } - - switch (*first) { - case _UT('['): - { - const URI_CHAR * const afterIpLit2 - = URI_FUNC(ParseIpLit2)(state, first + 1, afterLast, memory); - if (afterIpLit2 == NULL) { - return NULL; - } - state->uri->hostText.first = first + 1; /* HOST BEGIN */ - return URI_FUNC(ParseAuthorityTwo)(state, afterIpLit2, afterLast); - } - - case _UT('!'): - case _UT('$'): - case _UT('%'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(':'): - case _UT(';'): - case _UT('@'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - state->uri->userInfo.first = first; /* USERINFO BEGIN */ - return URI_FUNC(ParseOwnHostUserInfoNz)(state, first, afterLast, memory); - - default: - /* "" regname host */ - state->uri->hostText.first = URI_FUNC(SafeToPointTo); - state->uri->hostText.afterLast = URI_FUNC(SafeToPointTo); - return first; - } +static URI_INLINE const URI_CHAR * URI_FUNC(ParseAuthority)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if (first >= afterLast) { + /* "" regname host */ + state->uri->hostText.first = URI_FUNC(SafeToPointTo); + state->uri->hostText.afterLast = URI_FUNC(SafeToPointTo); + return afterLast; + } + + switch (*first) { + case _UT('['): { + const URI_CHAR * const afterIpLit2 = + URI_FUNC(ParseIpLit2)(state, first + 1, afterLast, memory); + if (afterIpLit2 == NULL) { + return NULL; + } + state->uri->hostText.first = first + 1; /* HOST BEGIN */ + return URI_FUNC(ParseAuthorityTwo)(state, afterIpLit2, afterLast); + } + + case URI_SET_PCHAR(_UT): + state->uri->userInfo.first = first; /* USERINFO BEGIN */ + return URI_FUNC(ParseOwnHostUserInfoNz)(state, first, afterLast, memory); + + default: + /* "" regname host */ + state->uri->hostText.first = URI_FUNC(SafeToPointTo); + state->uri->hostText.afterLast = URI_FUNC(SafeToPointTo); + return first; + } } - - /* * [authorityTwo]-><:>[port] * [authorityTwo]-> */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseAuthorityTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT(':'): - { - const URI_CHAR * const afterPort = URI_FUNC(ParsePort)(state, first + 1, afterLast); - if (afterPort == NULL) { - return NULL; - } - state->uri->portText.first = first + 1; /* PORT BEGIN */ - state->uri->portText.afterLast = afterPort; /* PORT END */ - return afterPort; - } - - default: - return first; - } +static URI_INLINE const URI_CHAR * +URI_FUNC(ParseAuthorityTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, + const URI_CHAR * afterLast) { + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case _UT(':'): { + const URI_CHAR * const afterPort = URI_FUNC(ParsePort)(first + 1, afterLast); + if (afterPort == NULL) { + return NULL; + } + state->uri->portText.first = first + 1; /* PORT BEGIN */ + state->uri->portText.afterLast = afterPort; /* PORT END */ + return afterPort; + } + + default: + return first; + } } - - /* * [hexZero]->[HEXDIG][hexZero] * [hexZero]-> */ -static const URI_CHAR * URI_FUNC(ParseHexZero)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case URI_SET_HEXDIG: - return URI_FUNC(ParseHexZero)(state, first + 1, afterLast); - - default: - return first; - } +static const URI_CHAR * URI_FUNC(ParseHexZero)(const URI_CHAR * first, + const URI_CHAR * afterLast) { +tail_call: + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case URI_SET_HEXDIG(_UT): + first += 1; + goto tail_call; + + default: + return first; + } } - - /* * [hierPart]->[pathRootless] * [hierPart]->[partHelperTwo] * [hierPart]-> */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseHierPart)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('!'): - case _UT('$'): - case _UT('%'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(':'): - case _UT(';'): - case _UT('@'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - return URI_FUNC(ParsePathRootless)(state, first, afterLast, memory); - - case _UT('/'): - return URI_FUNC(ParsePartHelperTwo)(state, first + 1, afterLast, memory); - - default: - return first; - } +static URI_INLINE const URI_CHAR * URI_FUNC(ParseHierPart)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case URI_SET_PCHAR(_UT): + return URI_FUNC(ParsePathRootless)(state, first, afterLast, memory); + + case _UT('/'): + return URI_FUNC(ParsePartHelperTwo)(state, first + 1, afterLast, memory); + + default: + return first; + } } - - /* * [ipFutLoop]->[subDelims][ipFutStopGo] * [ipFutLoop]->[unreserved][ipFutStopGo] * [ipFutLoop]-><:>[ipFutStopGo] - */ -static const URI_CHAR * URI_FUNC(ParseIpFutLoop)(URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - switch (*first) { - case _UT('!'): - case _UT('$'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(':'): - case _UT(';'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - return URI_FUNC(ParseIpFutStopGo)(state, first + 1, afterLast, memory); - - default: - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } -} - - - -/* + * * [ipFutStopGo]->[ipFutLoop] * [ipFutStopGo]-> */ -static const URI_CHAR * URI_FUNC(ParseIpFutStopGo)( - URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('!'): - case _UT('$'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(':'): - case _UT(';'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - return URI_FUNC(ParseIpFutLoop)(state, first, afterLast, memory); - - default: - return first; - } +static const URI_CHAR * URI_FUNC(ParseIpFutLoop)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + const URI_CHAR * const originalFirst = first; + + while (first < afterLast) { + switch (*first) { + case _UT(':'): + case URI_SET_SUB_DELIMS(_UT): + case URI_SET_UNRESERVED(_UT): + first += 1; + break; + + default: + goto done_looping; + break; + } + } + +done_looping: + if (first == originalFirst) { + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + + return first; } - - /* * [ipFuture]->[HEXDIG][hexZero]<.>[ipFutLoop] */ static const URI_CHAR * URI_FUNC(ParseIpFuture)(URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - /* - First character has already been - checked before entering this rule. - - switch (*first) { - case _UT('v'): - */ - if (first + 1 >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - switch (first[1]) { - case URI_SET_HEXDIG: - { - const URI_CHAR * afterIpFutLoop; - const URI_CHAR * const afterHexZero - = URI_FUNC(ParseHexZero)(state, first + 2, afterLast); - if (afterHexZero == NULL) { - return NULL; - } - if (afterHexZero >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - if (*afterHexZero != _UT('.')) { - URI_FUNC(StopSyntax)(state, afterHexZero, memory); - return NULL; - } - state->uri->hostText.first = first; /* HOST BEGIN */ - state->uri->hostData.ipFuture.first = first; /* IPFUTURE BEGIN */ - afterIpFutLoop = URI_FUNC(ParseIpFutLoop)(state, afterHexZero + 1, afterLast, memory); - if (afterIpFutLoop == NULL) { - return NULL; - } - state->uri->hostText.afterLast = afterIpFutLoop; /* HOST END */ - state->uri->hostData.ipFuture.afterLast = afterIpFutLoop; /* IPFUTURE END */ - return afterIpFutLoop; - } - - default: - URI_FUNC(StopSyntax)(state, first + 1, memory); - return NULL; - } - - /* - default: - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } - */ + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if (first >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + /* + First character has already been + checked before entering this rule. + + switch (*first) { + case _UT('v'): + case _UT('V'): + */ + if (afterLast - first < 2) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + switch (first[1]) { + case URI_SET_HEXDIG(_UT): { + const URI_CHAR * afterIpFutLoop; + const URI_CHAR * const afterHexZero = + URI_FUNC(ParseHexZero)(first + 2, afterLast); + if (afterHexZero == NULL) { + return NULL; + } + if (afterHexZero >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + if (*afterHexZero != _UT('.')) { + URI_FUNC(StopSyntax)(state, afterHexZero, memory); + return NULL; + } + state->uri->hostText.first = first; /* HOST BEGIN */ + state->uri->hostData.ipFuture.first = first; /* IPFUTURE BEGIN */ + afterIpFutLoop = + URI_FUNC(ParseIpFutLoop)(state, afterHexZero + 1, afterLast, memory); + if (afterIpFutLoop == NULL) { + return NULL; + } + state->uri->hostText.afterLast = afterIpFutLoop; /* HOST END */ + state->uri->hostData.ipFuture.afterLast = afterIpFutLoop; /* IPFUTURE END */ + return afterIpFutLoop; + } + + default: + URI_FUNC(StopSyntax)(state, first + 1, memory); + return NULL; + } + + /* + default: + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + */ } - - /* * [ipLit2]->[ipFuture]<]> * [ipLit2]->[IPv6address2] */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseIpLit2)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - switch (*first) { - case _UT('v'): - { - const URI_CHAR * const afterIpFuture - = URI_FUNC(ParseIpFuture)(state, first, afterLast, memory); - if (afterIpFuture == NULL) { - return NULL; - } - if (afterIpFuture >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - if (*afterIpFuture != _UT(']')) { - URI_FUNC(StopSyntax)(state, afterIpFuture, memory); - return NULL; - } - return afterIpFuture + 1; - } - - case _UT(':'): - case _UT(']'): - case URI_SET_HEXDIG: - state->uri->hostData.ip6 = memory->malloc(memory, 1 * sizeof(UriIp6)); /* Freed when stopping on parse error */ - if (state->uri->hostData.ip6 == NULL) { - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return URI_FUNC(ParseIPv6address2)(state, first, afterLast, memory); - - default: - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } +static URI_INLINE const URI_CHAR * URI_FUNC(ParseIpLit2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if (first >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + switch (*first) { + /* The leading "v" of IPvFuture is case-insensitive. */ + case _UT('v'): + case _UT('V'): { + const URI_CHAR * const afterIpFuture = + URI_FUNC(ParseIpFuture)(state, first, afterLast, memory); + if (afterIpFuture == NULL) { + return NULL; + } + if (afterIpFuture >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + if (*afterIpFuture != _UT(']')) { + URI_FUNC(StopSyntax)(state, afterIpFuture, memory); + return NULL; + } + return afterIpFuture + 1; + } + + case _UT(':'): + case _UT(']'): + case URI_SET_HEXDIG(_UT): + state->uri->hostData.ip6 = memory->malloc( + memory, 1 * sizeof(UriIp6)); /* Freed when stopping on parse error */ + if (state->uri->hostData.ip6 == NULL) { + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + return URI_FUNC(ParseIPv6address2)(state, first, afterLast, memory); + + default: + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } } - - /* * [IPv6address2]->..<]> */ -static const URI_CHAR * URI_FUNC(ParseIPv6address2)( - URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - int zipperEver = 0; - int quadsDone = 0; - int digitCount = 0; - unsigned char digitHistory[4]; - int ip4OctetsDone = 0; - - unsigned char quadsAfterZipper[14]; - int quadsAfterZipperCount = 0; - - - for (;;) { - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - /* Inside IPv4 part? */ - if (ip4OctetsDone > 0) { - /* Eat rest of IPv4 address */ - for (;;) { - switch (*first) { - case URI_SET_DIGIT: - if (digitCount == 4) { - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } - digitHistory[digitCount++] = (unsigned char)(9 + *first - _UT('9')); - break; - - case _UT('.'): - if ((ip4OctetsDone == 4) /* NOTE! */ - || (digitCount == 0) - || (digitCount == 4)) { - /* Invalid digit or octet count */ - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } else if ((digitCount > 1) - && (digitHistory[0] == 0)) { - /* Leading zero */ - URI_FUNC(StopSyntax)(state, first - digitCount, memory); - return NULL; - } else if ((digitCount > 2) - && (digitHistory[1] == 0)) { - /* Leading zero */ - URI_FUNC(StopSyntax)(state, first - digitCount + 1, memory); - return NULL; - } else if ((digitCount == 3) - && (100 * digitHistory[0] - + 10 * digitHistory[1] - + digitHistory[2] > 255)) { - /* Octet value too large */ - if (digitHistory[0] > 2) { - URI_FUNC(StopSyntax)(state, first - 3, memory); - } else if (digitHistory[1] > 5) { - URI_FUNC(StopSyntax)(state, first - 2, memory); - } else { - URI_FUNC(StopSyntax)(state, first - 1, memory); - } - return NULL; - } - - /* Copy IPv4 octet */ - state->uri->hostData.ip6->data[16 - 4 + ip4OctetsDone] = uriGetOctetValue(digitHistory, digitCount); - digitCount = 0; - ip4OctetsDone++; - break; - - case _UT(']'): - if ((ip4OctetsDone != 3) /* NOTE! */ - || (digitCount == 0) - || (digitCount == 4)) { - /* Invalid digit or octet count */ - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } else if ((digitCount > 1) - && (digitHistory[0] == 0)) { - /* Leading zero */ - URI_FUNC(StopSyntax)(state, first - digitCount, memory); - return NULL; - } else if ((digitCount > 2) - && (digitHistory[1] == 0)) { - /* Leading zero */ - URI_FUNC(StopSyntax)(state, first - digitCount + 1, memory); - return NULL; - } else if ((digitCount == 3) - && (100 * digitHistory[0] - + 10 * digitHistory[1] - + digitHistory[2] > 255)) { - /* Octet value too large */ - if (digitHistory[0] > 2) { - URI_FUNC(StopSyntax)(state, first - 3, memory); - } else if (digitHistory[1] > 5) { - URI_FUNC(StopSyntax)(state, first - 2, memory); - } else { - URI_FUNC(StopSyntax)(state, first - 1, memory); - } - return NULL; - } - - state->uri->hostText.afterLast = first; /* HOST END */ - - /* Copy missing quads right before IPv4 */ - memcpy(state->uri->hostData.ip6->data + 16 - 4 - 2 * quadsAfterZipperCount, - quadsAfterZipper, 2 * quadsAfterZipperCount); - - /* Copy last IPv4 octet */ - state->uri->hostData.ip6->data[16 - 4 + 3] = uriGetOctetValue(digitHistory, digitCount); - - return first + 1; - - default: - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } - first++; - - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - } - } else { - /* Eat while no dot in sight */ - int letterAmong = 0; - int walking = 1; - do { - switch (*first) { - case URI_SET_HEX_LETTER_LOWER: - letterAmong = 1; - if (digitCount == 4) { - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } - digitHistory[digitCount] = (unsigned char)(15 + *first - _UT('f')); - digitCount++; - break; - - case URI_SET_HEX_LETTER_UPPER: - letterAmong = 1; - if (digitCount == 4) { - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } - digitHistory[digitCount] = (unsigned char)(15 + *first - _UT('F')); - digitCount++; - break; - - case URI_SET_DIGIT: - if (digitCount == 4) { - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } - digitHistory[digitCount] = (unsigned char)(9 + *first - _UT('9')); - digitCount++; - break; - - case _UT(':'): - { - int setZipper = 0; - - if (digitCount > 0) { - if (zipperEver) { - uriWriteQuadToDoubleByte(digitHistory, digitCount, quadsAfterZipper + 2 * quadsAfterZipperCount); - quadsAfterZipperCount++; - } else { - uriWriteQuadToDoubleByte(digitHistory, digitCount, state->uri->hostData.ip6->data + 2 * quadsDone); - } - quadsDone++; - digitCount = 0; - } - letterAmong = 0; - - /* Too many quads? */ - if (quadsDone >= 8 - zipperEver) { - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } - - /* "::"? */ - if (first + 1 >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - if (first[1] == _UT(':')) { - const int resetOffset = 2 * (quadsDone + (digitCount > 0)); - - first++; - if (zipperEver) { - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; /* "::.+::" */ - } - - /* Zero everything after zipper */ - memset(state->uri->hostData.ip6->data + resetOffset, 0, 16 - resetOffset); - setZipper = 1; - - /* ":::+"? */ - if (first + 1 >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; /* No ']' yet */ - } - if (first[1] == _UT(':')) { - URI_FUNC(StopSyntax)(state, first + 1, memory); - return NULL; /* ":::+ "*/ - } - } - - if (setZipper) { - zipperEver = 1; - } - } - break; - - case _UT('.'): - if ((quadsDone > 6) /* NOTE */ - || (!zipperEver && (quadsDone < 6)) - || letterAmong - || (digitCount == 0) - || (digitCount == 4)) { - /* Invalid octet before */ - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } else if ((digitCount > 1) - && (digitHistory[0] == 0)) { - /* Leading zero */ - URI_FUNC(StopSyntax)(state, first - digitCount, memory); - return NULL; - } else if ((digitCount > 2) - && (digitHistory[1] == 0)) { - /* Leading zero */ - URI_FUNC(StopSyntax)(state, first - digitCount + 1, memory); - return NULL; - } else if ((digitCount == 3) - && (100 * digitHistory[0] - + 10 * digitHistory[1] - + digitHistory[2] > 255)) { - /* Octet value too large */ - if (digitHistory[0] > 2) { - URI_FUNC(StopSyntax)(state, first - 3, memory); - } else if (digitHistory[1] > 5) { - URI_FUNC(StopSyntax)(state, first - 2, memory); - } else { - URI_FUNC(StopSyntax)(state, first - 1, memory); - } - return NULL; - } - - /* Copy first IPv4 octet */ - state->uri->hostData.ip6->data[16 - 4] = uriGetOctetValue(digitHistory, digitCount); - digitCount = 0; - - /* Switch over to IPv4 loop */ - ip4OctetsDone = 1; - walking = 0; - break; - - case _UT(']'): - /* Too little quads? */ - if (!zipperEver && !((quadsDone == 7) && (digitCount > 0))) { - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } - - if (digitCount > 0) { - if (zipperEver) { - uriWriteQuadToDoubleByte(digitHistory, digitCount, quadsAfterZipper + 2 * quadsAfterZipperCount); - quadsAfterZipperCount++; - } else { - uriWriteQuadToDoubleByte(digitHistory, digitCount, state->uri->hostData.ip6->data + 2 * quadsDone); - } - /* - quadsDone++; - digitCount = 0; - */ - } - - /* Copy missing quads to the end */ - memcpy(state->uri->hostData.ip6->data + 16 - 2 * quadsAfterZipperCount, - quadsAfterZipper, 2 * quadsAfterZipperCount); - - state->uri->hostText.afterLast = first; /* HOST END */ - return first + 1; /* Fine */ - - default: - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } - first++; - - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; /* No ']' yet */ - } - } while (walking); - } - } +static const URI_CHAR * URI_FUNC(ParseIPv6address2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + int zipperEver = 0; + int quadsDone = 0; + int digitCount = 0; + unsigned char digitHistory[4]; + int ip4OctetsDone = 0; + + unsigned char quadsAfterZipper[14]; + int quadsAfterZipperCount = 0; + + for (;;) { + if (first >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + /* Inside IPv4 part? */ + if (ip4OctetsDone > 0) { + /* Eat rest of IPv4 address */ + for (;;) { + switch (*first) { + case URI_SET_DIGIT(_UT): + if (digitCount == 4) { + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + digitHistory[digitCount++] = (unsigned char)(9 + *first - _UT('9')); + break; + + case _UT('.'): + if ((ip4OctetsDone == 4) /* NOTE! */ + || (digitCount == 0) || (digitCount == 4)) { + /* Invalid digit or octet count */ + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } else if ((digitCount > 1) && (digitHistory[0] == 0)) { + /* Leading zero */ + URI_FUNC(StopSyntax)(state, first - digitCount, memory); + return NULL; + } else if ((digitCount == 3) + && (100 * digitHistory[0] + 10 * digitHistory[1] + + digitHistory[2] + > 255)) { + /* Octet value too large */ + if (digitHistory[0] > 2) { + URI_FUNC(StopSyntax)(state, first - 3, memory); + } else if (digitHistory[1] > 5) { + URI_FUNC(StopSyntax)(state, first - 2, memory); + } else { + URI_FUNC(StopSyntax)(state, first - 1, memory); + } + return NULL; + } + + /* Copy IPv4 octet */ + state->uri->hostData.ip6->data[16 - 4 + ip4OctetsDone] = + uriGetOctetValue(digitHistory, digitCount); + digitCount = 0; + ip4OctetsDone++; + break; + + case _UT(']'): + if ((ip4OctetsDone != 3) /* NOTE! */ + || (digitCount == 0) || (digitCount == 4)) { + /* Invalid digit or octet count */ + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } else if ((digitCount > 1) && (digitHistory[0] == 0)) { + /* Leading zero */ + URI_FUNC(StopSyntax)(state, first - digitCount, memory); + return NULL; + } else if ((digitCount == 3) + && (100 * digitHistory[0] + 10 * digitHistory[1] + + digitHistory[2] + > 255)) { + /* Octet value too large */ + if (digitHistory[0] > 2) { + URI_FUNC(StopSyntax)(state, first - 3, memory); + } else if (digitHistory[1] > 5) { + URI_FUNC(StopSyntax)(state, first - 2, memory); + } else { + URI_FUNC(StopSyntax)(state, first - 1, memory); + } + return NULL; + } + + state->uri->hostText.afterLast = first; /* HOST END */ + + /* Copy missing quads right before IPv4 */ + memcpy(state->uri->hostData.ip6->data + 16 - 4 + - 2 * quadsAfterZipperCount, + quadsAfterZipper, 2 * quadsAfterZipperCount); + + /* Copy last IPv4 octet */ + state->uri->hostData.ip6->data[16 - 4 + 3] = + uriGetOctetValue(digitHistory, digitCount); + + return first + 1; + + default: + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + first++; + + if (first >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + } + } else { + /* Eat while no dot in sight */ + int letterAmong = 0; + int walking = 1; + do { + switch (*first) { + case URI_SET_HEX_LETTER_LOWER(_UT): + letterAmong = 1; + if (digitCount == 4) { + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + digitHistory[digitCount] = (unsigned char)(15 + *first - _UT('f')); + digitCount++; + break; + + case URI_SET_HEX_LETTER_UPPER(_UT): + letterAmong = 1; + if (digitCount == 4) { + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + digitHistory[digitCount] = (unsigned char)(15 + *first - _UT('F')); + digitCount++; + break; + + case URI_SET_DIGIT(_UT): + if (digitCount == 4) { + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + digitHistory[digitCount] = (unsigned char)(9 + *first - _UT('9')); + digitCount++; + break; + + case _UT(':'): { + int setZipper = 0; + + if (digitCount > 0) { + if (zipperEver) { + uriWriteQuadToDoubleByte(digitHistory, digitCount, + quadsAfterZipper + + 2 * quadsAfterZipperCount); + quadsAfterZipperCount++; + } else { + uriWriteQuadToDoubleByte(digitHistory, digitCount, + state->uri->hostData.ip6->data + + 2 * quadsDone); + } + quadsDone++; + digitCount = 0; + } + letterAmong = 0; + + /* Too many quads? */ + if (quadsDone >= 8 - zipperEver) { + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + + /* "::"? */ + if (afterLast - first < 2) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + if (first[1] == _UT(':')) { + const int resetOffset = 2 * (quadsDone + (digitCount > 0)); + + first++; + if (zipperEver) { + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; /* "::.+::" */ + } + + /* Zero everything after zipper */ + memset(state->uri->hostData.ip6->data + resetOffset, 0, + 16 - resetOffset); + setZipper = 1; + + /* ":::+"? */ + if (afterLast - first < 2) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; /* No ']' yet */ + } + if (first[1] == _UT(':')) { + URI_FUNC(StopSyntax)(state, first + 1, memory); + return NULL; /* ":::+ "*/ + } + } else if (quadsDone == 0 || first[1] == _UT(']')) { + /* Single leading or trailing ":" */ + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + + if (setZipper) { + zipperEver = 1; + } + } break; + + case _UT('.'): + if ((quadsDone + zipperEver > 6) /* NOTE */ + || (!zipperEver && (quadsDone < 6)) || letterAmong + || (digitCount == 0) || (digitCount == 4)) { + /* Invalid octet before */ + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } else if ((digitCount > 1) && (digitHistory[0] == 0)) { + /* Leading zero */ + URI_FUNC(StopSyntax)(state, first - digitCount, memory); + return NULL; + } else if ((digitCount == 3) + && (100 * digitHistory[0] + 10 * digitHistory[1] + + digitHistory[2] + > 255)) { + /* Octet value too large */ + if (digitHistory[0] > 2) { + URI_FUNC(StopSyntax)(state, first - 3, memory); + } else if (digitHistory[1] > 5) { + URI_FUNC(StopSyntax)(state, first - 2, memory); + } else { + URI_FUNC(StopSyntax)(state, first - 1, memory); + } + return NULL; + } + + /* Copy first IPv4 octet */ + state->uri->hostData.ip6->data[16 - 4] = + uriGetOctetValue(digitHistory, digitCount); + digitCount = 0; + + /* Switch over to IPv4 loop */ + ip4OctetsDone = 1; + walking = 0; + break; + + case _UT(']'): + /* Too little quads? */ + if (!zipperEver && !((quadsDone == 7) && (digitCount > 0))) { + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + + if (digitCount > 0) { + if (zipperEver) { + /* Too many quads? */ + if (quadsDone >= 7) { + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + uriWriteQuadToDoubleByte(digitHistory, digitCount, + quadsAfterZipper + + 2 * quadsAfterZipperCount); + quadsAfterZipperCount++; + } else { + uriWriteQuadToDoubleByte(digitHistory, digitCount, + state->uri->hostData.ip6->data + + 2 * quadsDone); + } + /* + quadsDone++; + digitCount = 0; + */ + } + + /* Copy missing quads to the end */ + memcpy(state->uri->hostData.ip6->data + 16 + - 2 * quadsAfterZipperCount, + quadsAfterZipper, 2 * quadsAfterZipperCount); + + state->uri->hostText.afterLast = first; /* HOST END */ + return first + 1; /* Fine */ + + default: + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + first++; + + if (first >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; /* No ']' yet */ + } + } while (walking); + } + } } - - /* * [mustBeSegmentNzNc]->[pctEncoded][mustBeSegmentNzNc] * [mustBeSegmentNzNc]->[subDelims][mustBeSegmentNzNc] @@ -891,346 +806,277 @@ static const URI_CHAR * URI_FUNC(ParseIPv6address2)( * [mustBeSegmentNzNc]->[segment][zeroMoreSlashSegs][uriTail] * [mustBeSegmentNzNc]-><@>[mustBeSegmentNzNc] */ -static const URI_CHAR * URI_FUNC(ParseMustBeSegmentNzNc)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first, memory)) { /* SEGMENT BOTH */ - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - state->uri->scheme.first = NULL; /* Not a scheme, reset */ - return afterLast; - } - - switch (*first) { - case _UT('%'): - { - const URI_CHAR * const afterPctEncoded - = URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); - if (afterPctEncoded == NULL) { - return NULL; - } - return URI_FUNC(ParseMustBeSegmentNzNc)(state, afterPctEncoded, afterLast, memory); - } - - case _UT('@'): - case _UT('!'): - case _UT('$'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('*'): - case _UT(','): - case _UT(';'): - case _UT('\''): - case _UT('+'): - case _UT('='): - case _UT('-'): - case _UT('.'): - case _UT('_'): - case _UT('~'): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - return URI_FUNC(ParseMustBeSegmentNzNc)(state, first + 1, afterLast, memory); - - case _UT('/'): - { - const URI_CHAR * afterZeroMoreSlashSegs; - const URI_CHAR * afterSegment; - if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first, memory)) { /* SEGMENT BOTH */ - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - state->uri->scheme.first = NULL; /* Not a scheme, reset */ - afterSegment = URI_FUNC(ParseSegment)(state, first + 1, afterLast, memory); - if (afterSegment == NULL) { - return NULL; - } - if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment, memory)) { /* SEGMENT BOTH */ - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - afterZeroMoreSlashSegs - = URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegment, afterLast, memory); - if (afterZeroMoreSlashSegs == NULL) { - return NULL; - } - return URI_FUNC(ParseUriTail)(state, afterZeroMoreSlashSegs, afterLast, memory); - } - - default: - if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first, memory)) { /* SEGMENT BOTH */ - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - state->uri->scheme.first = NULL; /* Not a scheme, reset */ - return URI_FUNC(ParseUriTail)(state, first, afterLast, memory); - } +static const URI_CHAR * URI_FUNC(ParseMustBeSegmentNzNc)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { +tail_call: + if (first >= afterLast) { + if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first, + memory)) { /* SEGMENT BOTH */ + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + state->uri->scheme.first = NULL; /* Not a scheme, reset */ + return afterLast; + } + + switch (*first) { + case _UT('%'): { + const URI_CHAR * const afterPctEncoded = + URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); + if (afterPctEncoded == NULL) { + return NULL; + } + first = afterPctEncoded; + goto tail_call; + } + + case _UT('@'): + case URI_SET_SUB_DELIMS(_UT): + case URI_SET_UNRESERVED(_UT): + first += 1; + goto tail_call; + + case _UT('/'): { + const URI_CHAR * afterZeroMoreSlashSegs; + const URI_CHAR * afterSegment; + if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first, + memory)) { /* SEGMENT BOTH */ + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + state->uri->scheme.first = NULL; /* Not a scheme, reset */ + afterSegment = URI_FUNC(ParseSegment)(state, first + 1, afterLast, memory); + if (afterSegment == NULL) { + return NULL; + } + if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment, + memory)) { /* SEGMENT BOTH */ + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + afterZeroMoreSlashSegs = + URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegment, afterLast, memory); + if (afterZeroMoreSlashSegs == NULL) { + return NULL; + } + return URI_FUNC(ParseUriTail)(state, afterZeroMoreSlashSegs, afterLast, memory); + } + + default: + if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first, + memory)) { /* SEGMENT BOTH */ + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + state->uri->scheme.first = NULL; /* Not a scheme, reset */ + return URI_FUNC(ParseUriTail)(state, first, afterLast, memory); + } } - - /* * [ownHost]-><[>[ipLit2][authorityTwo] * [ownHost]->[ownHost2] // can take */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseOwnHost)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - state->uri->hostText.afterLast = afterLast; /* HOST END */ - return afterLast; - } - - switch (*first) { - case _UT('['): - { - const URI_CHAR * const afterIpLit2 - = URI_FUNC(ParseIpLit2)(state, first + 1, afterLast, memory); - if (afterIpLit2 == NULL) { - return NULL; - } - state->uri->hostText.first = first + 1; /* HOST BEGIN */ - return URI_FUNC(ParseAuthorityTwo)(state, afterIpLit2, afterLast); - } - - default: - return URI_FUNC(ParseOwnHost2)(state, first, afterLast, memory); - } +static URI_INLINE const URI_CHAR * URI_FUNC(ParseOwnHost)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if (first >= afterLast) { + state->uri->hostText.afterLast = afterLast; /* HOST END */ + return afterLast; + } + + switch (*first) { + case _UT('['): { + const URI_CHAR * const afterIpLit2 = + URI_FUNC(ParseIpLit2)(state, first + 1, afterLast, memory); + if (afterIpLit2 == NULL) { + return NULL; + } + state->uri->hostText.first = first + 1; /* HOST BEGIN */ + return URI_FUNC(ParseAuthorityTwo)(state, afterIpLit2, afterLast); + } + + default: + return URI_FUNC(ParseOwnHost2)(state, first, afterLast, memory); + } } - - -static URI_INLINE UriBool URI_FUNC(OnExitOwnHost2)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - UriMemoryManager * memory) { - state->uri->hostText.afterLast = first; /* HOST END */ - - /* Valid IPv4 or just a regname? */ - state->uri->hostData.ip4 = memory->malloc(memory, 1 * sizeof(UriIp4)); /* Freed when stopping on parse error */ - if (state->uri->hostData.ip4 == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - if (URI_FUNC(ParseIpFourAddress)(state->uri->hostData.ip4->data, - state->uri->hostText.first, state->uri->hostText.afterLast)) { - /* Not IPv4 */ - memory->free(memory, state->uri->hostData.ip4); - state->uri->hostData.ip4 = NULL; - } - return URI_TRUE; /* Success */ +static URI_INLINE UriBool URI_FUNC(OnExitOwnHost2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + UriMemoryManager * memory) { + state->uri->hostText.afterLast = first; /* HOST END */ + + /* Valid IPv4 or just a regname? */ + state->uri->hostData.ip4 = memory->malloc( + memory, 1 * sizeof(UriIp4)); /* Freed when stopping on parse error */ + if (state->uri->hostData.ip4 == NULL) { + return URI_FALSE; /* Raises malloc error */ + } + if (URI_FUNC(ParseIpFourAddress)(state->uri->hostData.ip4->data, + state->uri->hostText.first, + state->uri->hostText.afterLast)) { + /* Not IPv4 */ + memory->free(memory, state->uri->hostData.ip4); + state->uri->hostData.ip4 = NULL; + } + return URI_TRUE; /* Success */ } - - /* * [ownHost2]->[authorityTwo] // can take * [ownHost2]->[pctSubUnres][ownHost2] */ -static const URI_CHAR * URI_FUNC(ParseOwnHost2)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - if (!URI_FUNC(OnExitOwnHost2)(state, first, memory)) { - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return afterLast; - } - - switch (*first) { - case _UT('!'): - case _UT('$'): - case _UT('%'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(';'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - { - const URI_CHAR * const afterPctSubUnres - = URI_FUNC(ParsePctSubUnres)(state, first, afterLast, memory); - if (afterPctSubUnres == NULL) { - return NULL; - } - return URI_FUNC(ParseOwnHost2)(state, afterPctSubUnres, afterLast, memory); - } - - default: - if (!URI_FUNC(OnExitOwnHost2)(state, first, memory)) { - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return URI_FUNC(ParseAuthorityTwo)(state, first, afterLast); - } +static const URI_CHAR * URI_FUNC(ParseOwnHost2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { +tail_call: + if (first >= afterLast) { + if (!URI_FUNC(OnExitOwnHost2)(state, first, memory)) { + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + return afterLast; + } + + switch (*first) { + case _UT('%'): + case URI_SET_SUB_DELIMS(_UT): + case URI_SET_UNRESERVED(_UT): { + const URI_CHAR * const afterPctSubUnres = + URI_FUNC(ParsePctSubUnres)(state, first, afterLast, memory); + if (afterPctSubUnres == NULL) { + return NULL; + } + first = afterPctSubUnres; + goto tail_call; + } + + default: + if (!URI_FUNC(OnExitOwnHost2)(state, first, memory)) { + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + return URI_FUNC(ParseAuthorityTwo)(state, first, afterLast); + } } - - -static URI_INLINE UriBool URI_FUNC(OnExitOwnHostUserInfo)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - UriMemoryManager * memory) { - state->uri->hostText.first = state->uri->userInfo.first; /* Host instead of userInfo, update */ - state->uri->userInfo.first = NULL; /* Not a userInfo, reset */ - state->uri->hostText.afterLast = first; /* HOST END */ - - /* Valid IPv4 or just a regname? */ - state->uri->hostData.ip4 = memory->malloc(memory, 1 * sizeof(UriIp4)); /* Freed when stopping on parse error */ - if (state->uri->hostData.ip4 == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - if (URI_FUNC(ParseIpFourAddress)(state->uri->hostData.ip4->data, - state->uri->hostText.first, state->uri->hostText.afterLast)) { - /* Not IPv4 */ - memory->free(memory, state->uri->hostData.ip4); - state->uri->hostData.ip4 = NULL; - } - return URI_TRUE; /* Success */ +static URI_INLINE UriBool URI_FUNC(OnExitOwnHostUserInfo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + UriMemoryManager * memory) { + state->uri->hostText.first = + state->uri->userInfo.first; /* Host instead of userInfo, update */ + state->uri->userInfo.first = NULL; /* Not a userInfo, reset */ + state->uri->hostText.afterLast = first; /* HOST END */ + + /* Valid IPv4 or just a regname? */ + state->uri->hostData.ip4 = memory->malloc( + memory, 1 * sizeof(UriIp4)); /* Freed when stopping on parse error */ + if (state->uri->hostData.ip4 == NULL) { + return URI_FALSE; /* Raises malloc error */ + } + if (URI_FUNC(ParseIpFourAddress)(state->uri->hostData.ip4->data, + state->uri->hostText.first, + state->uri->hostText.afterLast)) { + /* Not IPv4 */ + memory->free(memory, state->uri->hostData.ip4); + state->uri->hostData.ip4 = NULL; + } + return URI_TRUE; /* Success */ } - - -/* - * [ownHostUserInfo]->[ownHostUserInfoNz] - * [ownHostUserInfo]-> - */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseOwnHostUserInfo)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - if (!URI_FUNC(OnExitOwnHostUserInfo)(state, first, memory)) { - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return afterLast; - } - - switch (*first) { - case _UT('!'): - case _UT('$'): - case _UT('%'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(':'): - case _UT(';'): - case _UT('@'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - return URI_FUNC(ParseOwnHostUserInfoNz)(state, first, afterLast, memory); - - default: - if (!URI_FUNC(OnExitOwnHostUserInfo)(state, first, memory)) { - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return first; - } -} - - - /* * [ownHostUserInfoNz]->[pctSubUnres][ownHostUserInfo] * [ownHostUserInfoNz]-><:>[ownPortUserInfo] * [ownHostUserInfoNz]-><@>[ownHost] + * + * [ownHostUserInfo]->[ownHostUserInfoNz] + * [ownHostUserInfo]-> */ -static const URI_CHAR * URI_FUNC(ParseOwnHostUserInfoNz)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - switch (*first) { - case _UT('!'): - case _UT('$'): - case _UT('%'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(';'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - { - const URI_CHAR * const afterPctSubUnres - = URI_FUNC(ParsePctSubUnres)(state, first, afterLast, memory); - if (afterPctSubUnres == NULL) { - return NULL; - } - return URI_FUNC(ParseOwnHostUserInfo)(state, afterPctSubUnres, afterLast, memory); - } - - case _UT(':'): - state->uri->hostText.afterLast = first; /* HOST END */ - state->uri->portText.first = first + 1; /* PORT BEGIN */ - return URI_FUNC(ParseOwnPortUserInfo)(state, first + 1, afterLast, memory); - - case _UT('@'): - state->uri->userInfo.afterLast = first; /* USERINFO END */ - state->uri->hostText.first = first + 1; /* HOST BEGIN */ - return URI_FUNC(ParseOwnHost)(state, first + 1, afterLast, memory); - - default: - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } +static const URI_CHAR * URI_FUNC(ParseOwnHostUserInfoNz)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + const URI_CHAR * const originalFirst = first; + + while (first < afterLast) { + switch (*first) { + case _UT('%'): + case URI_SET_SUB_DELIMS(_UT): + case URI_SET_UNRESERVED(_UT): { + const URI_CHAR * const afterPctSubUnres = + URI_FUNC(ParsePctSubUnres)(state, first, afterLast, memory); + if (afterPctSubUnres == NULL) { + return NULL; + } + first = afterPctSubUnres; + break; + } + + default: + goto done_looping; + break; + } + } + +done_looping: + if (first < afterLast) { + switch (*first) { + case _UT(':'): + state->uri->hostText.afterLast = first; /* HOST END */ + state->uri->portText.first = first + 1; /* PORT BEGIN */ + return URI_FUNC(ParseOwnPortUserInfo)(state, first + 1, afterLast, memory); + + case _UT('@'): + state->uri->userInfo.afterLast = first; /* USERINFO END */ + state->uri->hostText.first = first + 1; /* HOST BEGIN */ + return URI_FUNC(ParseOwnHost)(state, first + 1, afterLast, memory); + + default: + break; + } + } + + if (first == originalFirst) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + if (!URI_FUNC(OnExitOwnHostUserInfo)(state, first, memory)) { + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + + return first; } - - -static URI_INLINE UriBool URI_FUNC(OnExitOwnPortUserInfo)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - UriMemoryManager * memory) { - state->uri->hostText.first = state->uri->userInfo.first; /* Host instead of userInfo, update */ - state->uri->userInfo.first = NULL; /* Not a userInfo, reset */ - state->uri->portText.afterLast = first; /* PORT END */ - - /* Valid IPv4 or just a regname? */ - state->uri->hostData.ip4 = memory->malloc(memory, 1 * sizeof(UriIp4)); /* Freed when stopping on parse error */ - if (state->uri->hostData.ip4 == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - if (URI_FUNC(ParseIpFourAddress)(state->uri->hostData.ip4->data, - state->uri->hostText.first, state->uri->hostText.afterLast)) { - /* Not IPv4 */ - memory->free(memory, state->uri->hostData.ip4); - state->uri->hostData.ip4 = NULL; - } - return URI_TRUE; /* Success */ +static URI_INLINE UriBool URI_FUNC(OnExitOwnPortUserInfo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + UriMemoryManager * memory) { + state->uri->hostText.first = + state->uri->userInfo.first; /* Host instead of userInfo, update */ + state->uri->userInfo.first = NULL; /* Not a userInfo, reset */ + state->uri->portText.afterLast = first; /* PORT END */ + + /* Valid IPv4 or just a regname? */ + state->uri->hostData.ip4 = memory->malloc( + memory, 1 * sizeof(UriIp4)); /* Freed when stopping on parse error */ + if (state->uri->hostData.ip4 == NULL) { + return URI_FALSE; /* Raises malloc error */ + } + if (URI_FUNC(ParseIpFourAddress)(state->uri->hostData.ip4->data, + state->uri->hostText.first, + state->uri->hostText.afterLast)) { + /* Not IPv4 */ + memory->free(memory, state->uri->hostData.ip4); + state->uri->hostData.ip4 = NULL; + } + return URI_TRUE; /* Success */ } - - /* * [ownPortUserInfo]->[ALPHA][ownUserInfo] * [ownPortUserInfo]->[DIGIT][ownPortUserInfo] @@ -1244,281 +1090,229 @@ static URI_INLINE UriBool URI_FUNC(OnExitOwnPortUserInfo)( * [ownPortUserInfo]-><@>[ownHost] * [ownPortUserInfo]-> */ -static const URI_CHAR * URI_FUNC(ParseOwnPortUserInfo)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - if (!URI_FUNC(OnExitOwnPortUserInfo)(state, first, memory)) { - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return afterLast; - } - - switch (*first) { - /* begin sub-delims */ - case _UT('!'): - case _UT('$'): - case _UT('&'): - case _UT('\''): - case _UT('('): - case _UT(')'): - case _UT('*'): - case _UT('+'): - case _UT(','): - case _UT(';'): - case _UT('='): - /* end sub-delims */ - /* begin unreserved (except alpha and digit) */ - case _UT('-'): - case _UT('.'): - case _UT('_'): - case _UT('~'): - /* end unreserved (except alpha and digit) */ - case _UT(':'): - case URI_SET_ALPHA: - state->uri->hostText.afterLast = NULL; /* Not a host, reset */ - state->uri->portText.first = NULL; /* Not a port, reset */ - return URI_FUNC(ParseOwnUserInfo)(state, first + 1, afterLast, memory); - - case URI_SET_DIGIT: - return URI_FUNC(ParseOwnPortUserInfo)(state, first + 1, afterLast, memory); - - case _UT('%'): - state->uri->portText.first = NULL; /* Not a port, reset */ - { - const URI_CHAR * const afterPct - = URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); - if (afterPct == NULL) { - return NULL; - } - return URI_FUNC(ParseOwnUserInfo)(state, afterPct, afterLast, memory); - } - - case _UT('@'): - state->uri->hostText.afterLast = NULL; /* Not a host, reset */ - state->uri->portText.first = NULL; /* Not a port, reset */ - state->uri->userInfo.afterLast = first; /* USERINFO END */ - state->uri->hostText.first = first + 1; /* HOST BEGIN */ - return URI_FUNC(ParseOwnHost)(state, first + 1, afterLast, memory); - - default: - if (!URI_FUNC(OnExitOwnPortUserInfo)(state, first, memory)) { - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return first; - } +static const URI_CHAR * URI_FUNC(ParseOwnPortUserInfo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { +tail_call: + if (first >= afterLast) { + if (!URI_FUNC(OnExitOwnPortUserInfo)(state, first, memory)) { + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + return afterLast; + } + + switch (*first) { + case URI_SET_SUB_DELIMS(_UT): + /* begin unreserved (except alpha and digit) */ + case _UT('-'): + case _UT('.'): + case _UT('_'): + case _UT('~'): + /* end unreserved (except alpha and digit) */ + case _UT(':'): + case URI_SET_ALPHA(_UT): + state->uri->hostText.afterLast = NULL; /* Not a host, reset */ + state->uri->portText.first = NULL; /* Not a port, reset */ + return URI_FUNC(ParseOwnUserInfo)(state, first + 1, afterLast, memory); + + case URI_SET_DIGIT(_UT): + first += 1; + goto tail_call; + + case _UT('%'): + state->uri->portText.first = NULL; /* Not a port, reset */ + const URI_CHAR * const afterPct = + URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); + if (afterPct == NULL) { + return NULL; + } + return URI_FUNC(ParseOwnUserInfo)(state, afterPct, afterLast, memory); + + case _UT('@'): + state->uri->hostText.afterLast = NULL; /* Not a host, reset */ + state->uri->portText.first = NULL; /* Not a port, reset */ + state->uri->userInfo.afterLast = first; /* USERINFO END */ + state->uri->hostText.first = first + 1; /* HOST BEGIN */ + return URI_FUNC(ParseOwnHost)(state, first + 1, afterLast, memory); + + default: + if (!URI_FUNC(OnExitOwnPortUserInfo)(state, first, memory)) { + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + return first; + } } - - /* * [ownUserInfo]->[pctSubUnres][ownUserInfo] * [ownUserInfo]-><:>[ownUserInfo] * [ownUserInfo]-><@>[ownHost] */ -static const URI_CHAR * URI_FUNC(ParseOwnUserInfo)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - switch (*first) { - case _UT('!'): - case _UT('$'): - case _UT('%'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(';'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - { - const URI_CHAR * const afterPctSubUnres - = URI_FUNC(ParsePctSubUnres)(state, first, afterLast, memory); - if (afterPctSubUnres == NULL) { - return NULL; - } - return URI_FUNC(ParseOwnUserInfo)(state, afterPctSubUnres, afterLast, memory); - } - - case _UT(':'): - return URI_FUNC(ParseOwnUserInfo)(state, first + 1, afterLast, memory); - - case _UT('@'): - /* SURE */ - state->uri->userInfo.afterLast = first; /* USERINFO END */ - state->uri->hostText.first = first + 1; /* HOST BEGIN */ - return URI_FUNC(ParseOwnHost)(state, first + 1, afterLast, memory); - - default: - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } +static const URI_CHAR * URI_FUNC(ParseOwnUserInfo)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { +tail_call: + if (first >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + switch (*first) { + case _UT('%'): + case URI_SET_SUB_DELIMS(_UT): + case URI_SET_UNRESERVED(_UT): { + const URI_CHAR * const afterPctSubUnres = + URI_FUNC(ParsePctSubUnres)(state, first, afterLast, memory); + if (afterPctSubUnres == NULL) { + return NULL; + } + first = afterPctSubUnres; + goto tail_call; + } + + case _UT(':'): + first += 1; + goto tail_call; + + case _UT('@'): + /* SURE */ + state->uri->userInfo.afterLast = first; /* USERINFO END */ + state->uri->hostText.first = first + 1; /* HOST BEGIN */ + return URI_FUNC(ParseOwnHost)(state, first + 1, afterLast, memory); + + default: + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } } - - static URI_INLINE void URI_FUNC(OnExitPartHelperTwo)(URI_TYPE(ParserState) * state) { - state->uri->absolutePath = URI_TRUE; + state->uri->absolutePath = URI_TRUE; } - - /* * [partHelperTwo]->[pathAbsNoLeadSlash] // can take * [partHelperTwo]->[authority][pathAbsEmpty] */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParsePartHelperTwo)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - URI_FUNC(OnExitPartHelperTwo)(state); - return afterLast; - } - - switch (*first) { - case _UT('/'): - { - const URI_CHAR * const afterAuthority - = URI_FUNC(ParseAuthority)(state, first + 1, afterLast, memory); - const URI_CHAR * afterPathAbsEmpty; - if (afterAuthority == NULL) { - return NULL; - } - afterPathAbsEmpty = URI_FUNC(ParsePathAbsEmpty)(state, afterAuthority, afterLast, memory); - - URI_FUNC(FixEmptyTrailSegment)(state->uri, memory); - - return afterPathAbsEmpty; - } - - default: - URI_FUNC(OnExitPartHelperTwo)(state); - return URI_FUNC(ParsePathAbsNoLeadSlash)(state, first, afterLast, memory); - } +static URI_INLINE const URI_CHAR * +URI_FUNC(ParsePartHelperTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + if (first >= afterLast) { + URI_FUNC(OnExitPartHelperTwo)(state); + return afterLast; + } + + switch (*first) { + case _UT('/'): { + const URI_CHAR * const afterAuthority = + URI_FUNC(ParseAuthority)(state, first + 1, afterLast, memory); + const URI_CHAR * afterPathAbsEmpty; + if (afterAuthority == NULL) { + return NULL; + } + afterPathAbsEmpty = + URI_FUNC(ParsePathAbsEmpty)(state, afterAuthority, afterLast, memory); + + URI_FUNC(FixEmptyTrailSegment)(state->uri, memory); + + return afterPathAbsEmpty; + } + + default: + URI_FUNC(OnExitPartHelperTwo)(state); + return URI_FUNC(ParsePathAbsNoLeadSlash)(state, first, afterLast, memory); + } } - - /* * [pathAbsEmpty]->[segment][pathAbsEmpty] * [pathAbsEmpty]-> */ -static const URI_CHAR * URI_FUNC(ParsePathAbsEmpty)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('/'): - { - const URI_CHAR * const afterSegment - = URI_FUNC(ParseSegment)(state, first + 1, afterLast, memory); - if (afterSegment == NULL) { - return NULL; - } - if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment, memory)) { /* SEGMENT BOTH */ - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return URI_FUNC(ParsePathAbsEmpty)(state, afterSegment, afterLast, memory); - } - - default: - return first; - } +static const URI_CHAR * URI_FUNC(ParsePathAbsEmpty)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { +tail_call: + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case _UT('/'): { + const URI_CHAR * const afterSegment = + URI_FUNC(ParseSegment)(state, first + 1, afterLast, memory); + if (afterSegment == NULL) { + return NULL; + } + if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment, + memory)) { /* SEGMENT BOTH */ + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + first = afterSegment; + goto tail_call; + } + + default: + return first; + } } - - /* * [pathAbsNoLeadSlash]->[segmentNz][zeroMoreSlashSegs] * [pathAbsNoLeadSlash]-> */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParsePathAbsNoLeadSlash)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('!'): - case _UT('$'): - case _UT('%'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(':'): - case _UT(';'): - case _UT('@'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - { - const URI_CHAR * const afterSegmentNz - = URI_FUNC(ParseSegmentNz)(state, first, afterLast, memory); - if (afterSegmentNz == NULL) { - return NULL; - } - if (!URI_FUNC(PushPathSegment)(state, first, afterSegmentNz, memory)) { /* SEGMENT BOTH */ - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegmentNz, afterLast, memory); - } - - default: - return first; - } +static URI_INLINE const URI_CHAR * +URI_FUNC(ParsePathAbsNoLeadSlash)(URI_TYPE(ParserState) * state, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case URI_SET_PCHAR(_UT): { + const URI_CHAR * const afterSegmentNz = + URI_FUNC(ParseSegmentNz)(state, first, afterLast, memory); + if (afterSegmentNz == NULL) { + return NULL; + } + if (!URI_FUNC(PushPathSegment)(state, first, afterSegmentNz, + memory)) { /* SEGMENT BOTH */ + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + return URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegmentNz, afterLast, memory); + } + + default: + return first; + } } - - /* * [pathRootless]->[segmentNz][zeroMoreSlashSegs] */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParsePathRootless)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - const URI_CHAR * const afterSegmentNz - = URI_FUNC(ParseSegmentNz)(state, first, afterLast, memory); - if (afterSegmentNz == NULL) { - return NULL; - } else { - if (!URI_FUNC(PushPathSegment)(state, first, afterSegmentNz, memory)) { /* SEGMENT BOTH */ - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - } - return URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegmentNz, afterLast, memory); +static URI_INLINE const URI_CHAR * +URI_FUNC(ParsePathRootless)(URI_TYPE(ParserState) * state, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + const URI_CHAR * const afterSegmentNz = + URI_FUNC(ParseSegmentNz)(state, first, afterLast, memory); + if (afterSegmentNz == NULL) { + return NULL; + } else { + if (!URI_FUNC(PushPathSegment)(state, first, afterSegmentNz, + memory)) { /* SEGMENT BOTH */ + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + } + return URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegmentNz, afterLast, memory); } - - /* * [pchar]->[pctEncoded] * [pchar]->[subDelims] @@ -1527,166 +1321,129 @@ static URI_INLINE const URI_CHAR * URI_FUNC(ParsePathRootless)( * [pchar]-><@> */ static const URI_CHAR * URI_FUNC(ParsePchar)(URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - switch (*first) { - case _UT('%'): - return URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); - - case _UT(':'): - case _UT('@'): - case _UT('!'): - case _UT('$'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('*'): - case _UT(','): - case _UT(';'): - case _UT('\''): - case _UT('+'): - case _UT('='): - case _UT('-'): - case _UT('.'): - case _UT('_'): - case _UT('~'): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - return first + 1; - - default: - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if (first >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + switch (*first) { + case _UT('%'): + return URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); + + case URI_SET_PCHAR_WITHOUT_PERCENT(_UT): + return first + 1; + + default: + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } } - - /* * [pctEncoded]-><%>[HEXDIG][HEXDIG] */ -static const URI_CHAR * URI_FUNC(ParsePctEncoded)( - URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - /* - First character has already been - checked before entering this rule. - - switch (*first) { - case _UT('%'): - */ - if (first + 1 >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - switch (first[1]) { - case URI_SET_HEXDIG: - if (first + 2 >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - switch (first[2]) { - case URI_SET_HEXDIG: - return first + 3; - - default: - URI_FUNC(StopSyntax)(state, first + 2, memory); - return NULL; - } - - default: - URI_FUNC(StopSyntax)(state, first + 1, memory); - return NULL; - } - - /* - default: - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } - */ +static const URI_CHAR * URI_FUNC(ParsePctEncoded)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if (first >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + /* + First character has already been + checked before entering this rule. + + switch (*first) { + case _UT('%'): + */ + if (afterLast - first < 2) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + switch (first[1]) { + case URI_SET_HEXDIG(_UT): + if (afterLast - first < 3) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + switch (first[2]) { + case URI_SET_HEXDIG(_UT): + return first + 3; + + default: + URI_FUNC(StopSyntax)(state, first + 2, memory); + return NULL; + } + + default: + URI_FUNC(StopSyntax)(state, first + 1, memory); + return NULL; + } + + /* + default: + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } + */ } - - /* * [pctSubUnres]->[pctEncoded] * [pctSubUnres]->[subDelims] * [pctSubUnres]->[unreserved] */ -static const URI_CHAR * URI_FUNC(ParsePctSubUnres)( - URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - if (first >= afterLast) { - URI_FUNC(StopSyntax)(state, afterLast, memory); - return NULL; - } - - switch (*first) { - case _UT('%'): - return URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); - - case _UT('!'): - case _UT('$'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('*'): - case _UT(','): - case _UT(';'): - case _UT('\''): - case _UT('+'): - case _UT('='): - case _UT('-'): - case _UT('.'): - case _UT('_'): - case _UT('~'): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - return first + 1; - - default: - URI_FUNC(StopSyntax)(state, first, memory); - return NULL; - } +static const URI_CHAR * URI_FUNC(ParsePctSubUnres)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if (first >= afterLast) { + URI_FUNC(StopSyntax)(state, afterLast, memory); + return NULL; + } + + switch (*first) { + case _UT('%'): + return URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); + + case URI_SET_SUB_DELIMS(_UT): + case URI_SET_UNRESERVED(_UT): + return first + 1; + + default: + URI_FUNC(StopSyntax)(state, first, memory); + return NULL; + } } - - /* * [port]->[DIGIT][port] * [port]-> */ -static const URI_CHAR * URI_FUNC(ParsePort)(URI_TYPE(ParserState) * state, const URI_CHAR * first, const URI_CHAR * afterLast) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case URI_SET_DIGIT: - return URI_FUNC(ParsePort)(state, first + 1, afterLast); - - default: - return first; - } +static const URI_CHAR * URI_FUNC(ParsePort)(const URI_CHAR * first, + const URI_CHAR * afterLast) { +tail_call: + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case URI_SET_DIGIT(_UT): + first += 1; + goto tail_call; + + default: + return first; + } } - - /* * [queryFrag]->[pchar][queryFrag] * [queryFrag]->[queryFrag] @@ -1694,130 +1451,89 @@ static const URI_CHAR * URI_FUNC(ParsePort)(URI_TYPE(ParserState) * state, const * [queryFrag]-> */ static const URI_CHAR * URI_FUNC(ParseQueryFrag)(URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('!'): - case _UT('$'): - case _UT('%'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(':'): - case _UT(';'): - case _UT('@'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - { - const URI_CHAR * const afterPchar - = URI_FUNC(ParsePchar)(state, first, afterLast, memory); - if (afterPchar == NULL) { - return NULL; - } - return URI_FUNC(ParseQueryFrag)(state, afterPchar, afterLast, memory); - } - - case _UT('/'): - case _UT('?'): - return URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast, memory); - - default: - return first; - } + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { +tail_call: + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case URI_SET_PCHAR(_UT): { + const URI_CHAR * const afterPchar = + URI_FUNC(ParsePchar)(state, first, afterLast, memory); + if (afterPchar == NULL) { + return NULL; + } + first = afterPchar; + goto tail_call; + } + + case _UT('/'): + case _UT('?'): + first += 1; + goto tail_call; + + default: + return first; + } } - - /* * [segment]->[pchar][segment] * [segment]-> */ static const URI_CHAR * URI_FUNC(ParseSegment)(URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('!'): - case _UT('$'): - case _UT('%'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('-'): - case _UT('*'): - case _UT(','): - case _UT('.'): - case _UT(':'): - case _UT(';'): - case _UT('@'): - case _UT('\''): - case _UT('_'): - case _UT('~'): - case _UT('+'): - case _UT('='): - case URI_SET_DIGIT: - case URI_SET_ALPHA: - { - const URI_CHAR * const afterPchar - = URI_FUNC(ParsePchar)(state, first, afterLast, memory); - if (afterPchar == NULL) { - return NULL; - } - return URI_FUNC(ParseSegment)(state, afterPchar, afterLast, memory); - } - - default: - return first; - } + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { +tail_call: + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case URI_SET_PCHAR(_UT): { + const URI_CHAR * const afterPchar = + URI_FUNC(ParsePchar)(state, first, afterLast, memory); + if (afterPchar == NULL) { + return NULL; + } + first = afterPchar; + goto tail_call; + } + + default: + return first; + } } - - /* * [segmentNz]->[pchar][segment] */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseSegmentNz)( - URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - const URI_CHAR * const afterPchar - = URI_FUNC(ParsePchar)(state, first, afterLast, memory); - if (afterPchar == NULL) { - return NULL; - } - return URI_FUNC(ParseSegment)(state, afterPchar, afterLast, memory); +static URI_INLINE const URI_CHAR * URI_FUNC(ParseSegmentNz)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + const URI_CHAR * const afterPchar = + URI_FUNC(ParsePchar)(state, first, afterLast, memory); + if (afterPchar == NULL) { + return NULL; + } + return URI_FUNC(ParseSegment)(state, afterPchar, afterLast, memory); } - - static URI_INLINE UriBool URI_FUNC(OnExitSegmentNzNcOrScheme2)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - UriMemoryManager * memory) { - if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first, memory)) { /* SEGMENT BOTH */ - return URI_FALSE; /* Raises malloc error*/ - } - state->uri->scheme.first = NULL; /* Not a scheme, reset */ - return URI_TRUE; /* Success */ + URI_TYPE(ParserState) * state, const URI_CHAR * first, UriMemoryManager * memory) { + if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first, + memory)) { /* SEGMENT BOTH */ + return URI_FALSE; /* Raises malloc error*/ + } + state->uri->scheme.first = NULL; /* Not a scheme, reset */ + return URI_TRUE; /* Success */ } - - /* * [segmentNzNcOrScheme2]->[ALPHA][segmentNzNcOrScheme2] * [segmentNzNcOrScheme2]->[DIGIT][segmentNzNcOrScheme2] @@ -1842,97 +1558,98 @@ static URI_INLINE UriBool URI_FUNC(OnExitSegmentNzNcOrScheme2)( * [segmentNzNcOrScheme2]-><'>[mustBeSegmentNzNc] * [segmentNzNcOrScheme2]-><->[segmentNzNcOrScheme2] */ -static const URI_CHAR * URI_FUNC(ParseSegmentNzNcOrScheme2)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - if (!URI_FUNC(OnExitSegmentNzNcOrScheme2)(state, first, memory)) { - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return afterLast; - } - - switch (*first) { - case _UT('.'): - case _UT('+'): - case _UT('-'): - case URI_SET_ALPHA: - case URI_SET_DIGIT: - return URI_FUNC(ParseSegmentNzNcOrScheme2)(state, first + 1, afterLast, memory); - - case _UT('%'): - { - const URI_CHAR * const afterPctEncoded - = URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); - if (afterPctEncoded == NULL) { - return NULL; - } - return URI_FUNC(ParseMustBeSegmentNzNc)(state, afterPctEncoded, afterLast, memory); - } - - case _UT('!'): - case _UT('$'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('*'): - case _UT(','): - case _UT(';'): - case _UT('@'): - case _UT('_'): - case _UT('~'): - case _UT('='): - case _UT('\''): - return URI_FUNC(ParseMustBeSegmentNzNc)(state, first + 1, afterLast, memory); - - case _UT('/'): - { - const URI_CHAR * afterZeroMoreSlashSegs; - const URI_CHAR * const afterSegment - = URI_FUNC(ParseSegment)(state, first + 1, afterLast, memory); - if (afterSegment == NULL) { - return NULL; - } - if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first, memory)) { /* SEGMENT BOTH */ - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - state->uri->scheme.first = NULL; /* Not a scheme, reset */ - if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment, memory)) { /* SEGMENT BOTH */ - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - afterZeroMoreSlashSegs - = URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegment, afterLast, memory); - if (afterZeroMoreSlashSegs == NULL) { - return NULL; - } - return URI_FUNC(ParseUriTail)(state, afterZeroMoreSlashSegs, afterLast, memory); - } - - case _UT(':'): - { - const URI_CHAR * const afterHierPart - = URI_FUNC(ParseHierPart)(state, first + 1, afterLast, memory); - state->uri->scheme.afterLast = first; /* SCHEME END */ - if (afterHierPart == NULL) { - return NULL; - } - return URI_FUNC(ParseUriTail)(state, afterHierPart, afterLast, memory); - } - - default: - if (!URI_FUNC(OnExitSegmentNzNcOrScheme2)(state, first, memory)) { - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return URI_FUNC(ParseUriTail)(state, first, afterLast, memory); - } +static const URI_CHAR * URI_FUNC(ParseSegmentNzNcOrScheme2)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { +tail_call: + if (first >= afterLast) { + if (!URI_FUNC(OnExitSegmentNzNcOrScheme2)(state, first, memory)) { + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + return afterLast; + } + + switch (*first) { + case _UT('.'): + case _UT('+'): + case _UT('-'): + case URI_SET_ALPHA(_UT): + case URI_SET_DIGIT(_UT): + first += 1; + goto tail_call; + + case _UT('%'): { + const URI_CHAR * const afterPctEncoded = + URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); + if (afterPctEncoded == NULL) { + return NULL; + } + return URI_FUNC(ParseMustBeSegmentNzNc)(state, afterPctEncoded, afterLast, + memory); + } + + case _UT('!'): + case _UT('$'): + case _UT('&'): + case _UT('('): + case _UT(')'): + case _UT('*'): + case _UT(','): + case _UT(';'): + case _UT('@'): + case _UT('_'): + case _UT('~'): + case _UT('='): + case _UT('\''): + return URI_FUNC(ParseMustBeSegmentNzNc)(state, first + 1, afterLast, memory); + + case _UT('/'): { + const URI_CHAR * afterZeroMoreSlashSegs; + const URI_CHAR * const afterSegment = + URI_FUNC(ParseSegment)(state, first + 1, afterLast, memory); + if (afterSegment == NULL) { + return NULL; + } + if (!URI_FUNC(PushPathSegment)(state, state->uri->scheme.first, first, + memory)) { /* SEGMENT BOTH */ + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + state->uri->scheme.first = NULL; /* Not a scheme, reset */ + if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment, + memory)) { /* SEGMENT BOTH */ + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + afterZeroMoreSlashSegs = + URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegment, afterLast, memory); + if (afterZeroMoreSlashSegs == NULL) { + return NULL; + } + return URI_FUNC(ParseUriTail)(state, afterZeroMoreSlashSegs, afterLast, memory); + } + + case _UT(':'): { + const URI_CHAR * const afterHierPart = + URI_FUNC(ParseHierPart)(state, first + 1, afterLast, memory); + state->uri->scheme.afterLast = first; /* SCHEME END */ + if (afterHierPart == NULL) { + return NULL; + } + return URI_FUNC(ParseUriTail)(state, afterHierPart, afterLast, memory); + } + + default: + if (!URI_FUNC(OnExitSegmentNzNcOrScheme2)(state, first, memory)) { + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + return URI_FUNC(ParseUriTail)(state, first, afterLast, memory); + } } - - /* * [uriReference]->[ALPHA][segmentNzNcOrScheme2] * [uriReference]->[DIGIT][mustBeSegmentNzNc] @@ -1946,465 +1663,410 @@ static const URI_CHAR * URI_FUNC(ParseSegmentNzNcOrScheme2)( * [uriReference]-><~>[mustBeSegmentNzNc] * [uriReference]-><->[mustBeSegmentNzNc] */ -static const URI_CHAR * URI_FUNC(ParseUriReference)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case URI_SET_ALPHA: - state->uri->scheme.first = first; /* SCHEME BEGIN */ - return URI_FUNC(ParseSegmentNzNcOrScheme2)(state, first + 1, afterLast, memory); - - case URI_SET_DIGIT: - case _UT('!'): - case _UT('$'): - case _UT('&'): - case _UT('('): - case _UT(')'): - case _UT('*'): - case _UT(','): - case _UT(';'): - case _UT('\''): - case _UT('+'): - case _UT('='): - case _UT('.'): - case _UT('_'): - case _UT('~'): - case _UT('-'): - case _UT('@'): - state->uri->scheme.first = first; /* SEGMENT BEGIN, ABUSE SCHEME POINTER */ - return URI_FUNC(ParseMustBeSegmentNzNc)(state, first + 1, afterLast, memory); - - case _UT('%'): - { - const URI_CHAR * const afterPctEncoded - = URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); - if (afterPctEncoded == NULL) { - return NULL; - } - state->uri->scheme.first = first; /* SEGMENT BEGIN, ABUSE SCHEME POINTER */ - return URI_FUNC(ParseMustBeSegmentNzNc)(state, afterPctEncoded, afterLast, memory); - } - - case _UT('/'): - { - const URI_CHAR * const afterPartHelperTwo - = URI_FUNC(ParsePartHelperTwo)(state, first + 1, afterLast, memory); - if (afterPartHelperTwo == NULL) { - return NULL; - } - return URI_FUNC(ParseUriTail)(state, afterPartHelperTwo, afterLast, memory); - } - - default: - return URI_FUNC(ParseUriTail)(state, first, afterLast, memory); - } +static const URI_CHAR * URI_FUNC(ParseUriReference)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case URI_SET_ALPHA(_UT): + state->uri->scheme.first = first; /* SCHEME BEGIN */ + return URI_FUNC(ParseSegmentNzNcOrScheme2)(state, first + 1, afterLast, memory); + + case URI_SET_DIGIT(_UT): + case URI_SET_SUB_DELIMS(_UT): + case _UT('.'): + case _UT('_'): + case _UT('~'): + case _UT('-'): + case _UT('@'): + state->uri->scheme.first = first; /* SEGMENT BEGIN, ABUSE SCHEME POINTER */ + return URI_FUNC(ParseMustBeSegmentNzNc)(state, first + 1, afterLast, memory); + + case _UT('%'): { + const URI_CHAR * const afterPctEncoded = + URI_FUNC(ParsePctEncoded)(state, first, afterLast, memory); + if (afterPctEncoded == NULL) { + return NULL; + } + state->uri->scheme.first = first; /* SEGMENT BEGIN, ABUSE SCHEME POINTER */ + return URI_FUNC(ParseMustBeSegmentNzNc)(state, afterPctEncoded, afterLast, + memory); + } + + case _UT('/'): { + const URI_CHAR * const afterPartHelperTwo = + URI_FUNC(ParsePartHelperTwo)(state, first + 1, afterLast, memory); + if (afterPartHelperTwo == NULL) { + return NULL; + } + return URI_FUNC(ParseUriTail)(state, afterPartHelperTwo, afterLast, memory); + } + + default: + return URI_FUNC(ParseUriTail)(state, first, afterLast, memory); + } } - - /* * [uriTail]-><#>[queryFrag] * [uriTail]->[queryFrag][uriTailTwo] * [uriTail]-> */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseUriTail)( - URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('#'): - { - const URI_CHAR * const afterQueryFrag = URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast, memory); - if (afterQueryFrag == NULL) { - return NULL; - } - state->uri->fragment.first = first + 1; /* FRAGMENT BEGIN */ - state->uri->fragment.afterLast = afterQueryFrag; /* FRAGMENT END */ - return afterQueryFrag; - } - - case _UT('?'): - { - const URI_CHAR * const afterQueryFrag - = URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast, memory); - if (afterQueryFrag == NULL) { - return NULL; - } - state->uri->query.first = first + 1; /* QUERY BEGIN */ - state->uri->query.afterLast = afterQueryFrag; /* QUERY END */ - return URI_FUNC(ParseUriTailTwo)(state, afterQueryFrag, afterLast, memory); - } - - default: - return first; - } +static URI_INLINE const URI_CHAR * URI_FUNC(ParseUriTail)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case _UT('#'): { + const URI_CHAR * const afterQueryFrag = + URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast, memory); + if (afterQueryFrag == NULL) { + return NULL; + } + state->uri->fragment.first = first + 1; /* FRAGMENT BEGIN */ + state->uri->fragment.afterLast = afterQueryFrag; /* FRAGMENT END */ + return afterQueryFrag; + } + + case _UT('?'): { + const URI_CHAR * const afterQueryFrag = + URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast, memory); + if (afterQueryFrag == NULL) { + return NULL; + } + state->uri->query.first = first + 1; /* QUERY BEGIN */ + state->uri->query.afterLast = afterQueryFrag; /* QUERY END */ + return URI_FUNC(ParseUriTailTwo)(state, afterQueryFrag, afterLast, memory); + } + + default: + return first; + } } - - /* * [uriTailTwo]-><#>[queryFrag] * [uriTailTwo]-> */ -static URI_INLINE const URI_CHAR * URI_FUNC(ParseUriTailTwo)( - URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('#'): - { - const URI_CHAR * const afterQueryFrag = URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast, memory); - if (afterQueryFrag == NULL) { - return NULL; - } - state->uri->fragment.first = first + 1; /* FRAGMENT BEGIN */ - state->uri->fragment.afterLast = afterQueryFrag; /* FRAGMENT END */ - return afterQueryFrag; - } - - default: - return first; - } +static URI_INLINE const URI_CHAR * +URI_FUNC(ParseUriTailTwo)(URI_TYPE(ParserState) * state, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case _UT('#'): { + const URI_CHAR * const afterQueryFrag = + URI_FUNC(ParseQueryFrag)(state, first + 1, afterLast, memory); + if (afterQueryFrag == NULL) { + return NULL; + } + state->uri->fragment.first = first + 1; /* FRAGMENT BEGIN */ + state->uri->fragment.afterLast = afterQueryFrag; /* FRAGMENT END */ + return afterQueryFrag; + } + + default: + return first; + } } - - /* * [zeroMoreSlashSegs]->[segment][zeroMoreSlashSegs] * [zeroMoreSlashSegs]-> */ -static const URI_CHAR * URI_FUNC(ParseZeroMoreSlashSegs)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - if (first >= afterLast) { - return afterLast; - } - - switch (*first) { - case _UT('/'): - { - const URI_CHAR * const afterSegment - = URI_FUNC(ParseSegment)(state, first + 1, afterLast, memory); - if (afterSegment == NULL) { - return NULL; - } - if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment, memory)) { /* SEGMENT BOTH */ - URI_FUNC(StopMalloc)(state, memory); - return NULL; - } - return URI_FUNC(ParseZeroMoreSlashSegs)(state, afterSegment, afterLast, memory); - } - - default: - return first; - } +static const URI_CHAR * URI_FUNC(ParseZeroMoreSlashSegs)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { +tail_call: + if (first >= afterLast) { + return afterLast; + } + + switch (*first) { + case _UT('/'): { + const URI_CHAR * const afterSegment = + URI_FUNC(ParseSegment)(state, first + 1, afterLast, memory); + if (afterSegment == NULL) { + return NULL; + } + if (!URI_FUNC(PushPathSegment)(state, first + 1, afterSegment, + memory)) { /* SEGMENT BOTH */ + URI_FUNC(StopMalloc)(state, memory); + return NULL; + } + first = afterSegment; + goto tail_call; + } + + default: + return first; + } } - - -static URI_INLINE void URI_FUNC(ResetParserStateExceptUri)(URI_TYPE(ParserState) * state) { - URI_TYPE(Uri) * const uriBackup = state->uri; - memset(state, 0, sizeof(URI_TYPE(ParserState))); - state->uri = uriBackup; +static URI_INLINE void URI_FUNC(ResetParserStateExceptUri)(URI_TYPE(ParserState) + * state) { + URI_TYPE(Uri) * const uriBackup = state->uri; + memset(state, 0, sizeof(URI_TYPE(ParserState))); + state->uri = uriBackup; } - - -static URI_INLINE UriBool URI_FUNC(PushPathSegment)( - URI_TYPE(ParserState) * state, const URI_CHAR * first, - const URI_CHAR * afterLast, UriMemoryManager * memory) { - URI_TYPE(PathSegment) * segment = memory->calloc(memory, 1, sizeof(URI_TYPE(PathSegment))); - if (segment == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - if (first == afterLast) { - segment->text.first = URI_FUNC(SafeToPointTo); - segment->text.afterLast = URI_FUNC(SafeToPointTo); - } else { - segment->text.first = first; - segment->text.afterLast = afterLast; - } - - /* First segment ever? */ - if (state->uri->pathHead == NULL) { - /* First segment ever, set head and tail */ - state->uri->pathHead = segment; - state->uri->pathTail = segment; - } else { - /* Append, update tail */ - state->uri->pathTail->next = segment; - state->uri->pathTail = segment; - } - - return URI_TRUE; /* Success */ +static URI_INLINE UriBool URI_FUNC(PushPathSegment)(URI_TYPE(ParserState) * state, + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + URI_TYPE(PathSegment) * segment = + memory->calloc(memory, 1, sizeof(URI_TYPE(PathSegment))); + if (segment == NULL) { + return URI_FALSE; /* Raises malloc error */ + } + if (first == afterLast) { + segment->text.first = URI_FUNC(SafeToPointTo); + segment->text.afterLast = URI_FUNC(SafeToPointTo); + } else { + segment->text.first = first; + segment->text.afterLast = afterLast; + } + + /* First segment ever? */ + if (state->uri->pathHead == NULL) { + /* First segment ever, set head and tail */ + state->uri->pathHead = segment; + state->uri->pathTail = segment; + } else { + /* Append, update tail */ + state->uri->pathTail->next = segment; + state->uri->pathTail = segment; + } + + return URI_TRUE; /* Success */ } - - -int URI_FUNC(ParseUriEx)(URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast) { - return URI_FUNC(ParseUriExMm)(state, first, afterLast, NULL); +int URI_FUNC(ParseUriEx)(URI_TYPE(ParserState) * state, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(ParseUriExMm)(state, first, afterLast, NULL); } - - -static int URI_FUNC(ParseUriExMm)(URI_TYPE(ParserState) * state, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - const URI_CHAR * afterUriReference; - URI_TYPE(Uri) * uri; - - /* Check params */ - if ((state == NULL) || (first == NULL) || (afterLast == NULL)) { - return URI_ERROR_NULL; - } - URI_CHECK_MEMORY_MANAGER(memory); /* may return */ - - uri = state->uri; - - /* Init parser */ - URI_FUNC(ResetParserStateExceptUri)(state); - URI_FUNC(ResetUri)(uri); - - /* Parse */ - afterUriReference = URI_FUNC(ParseUriReference)(state, first, afterLast, memory); - if (afterUriReference == NULL) { - /* Waterproof errorPos <= afterLast */ - if (state->errorPos && (state->errorPos > afterLast)) { - state->errorPos = afterLast; - } - return state->errorCode; - } - if (afterUriReference != afterLast) { - if (afterUriReference < afterLast) { - URI_FUNC(StopSyntax)(state, afterUriReference, memory); - } else { - URI_FUNC(StopSyntax)(state, afterLast, memory); - } - return state->errorCode; - } - return URI_SUCCESS; +static int URI_FUNC(ParseUriExMm)(URI_TYPE(ParserState) * state, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + const URI_CHAR * afterUriReference; + URI_TYPE(Uri) * uri; + + /* Check params */ + if ((state == NULL) || (first == NULL) || (afterLast == NULL)) { + return URI_ERROR_NULL; + } + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + uri = state->uri; + + /* Init parser */ + URI_FUNC(ResetParserStateExceptUri)(state); + URI_FUNC(ResetUri)(uri); + + /* Parse */ + afterUriReference = URI_FUNC(ParseUriReference)(state, first, afterLast, memory); + if (afterUriReference == NULL) { + /* Waterproof errorPos <= afterLast */ + if (state->errorPos && (state->errorPos > afterLast)) { + state->errorPos = afterLast; + } + return state->errorCode; + } + if (afterUriReference != afterLast) { + if (afterUriReference < afterLast) { + URI_FUNC(StopSyntax)(state, afterUriReference, memory); + } else { + URI_FUNC(StopSyntax)(state, afterLast, memory); + } + return state->errorCode; + } + return URI_SUCCESS; } - - int URI_FUNC(ParseUri)(URI_TYPE(ParserState) * state, const URI_CHAR * text) { - if ((state == NULL) || (text == NULL)) { - return URI_ERROR_NULL; - } - return URI_FUNC(ParseUriEx)(state, text, text + URI_STRLEN(text)); + if ((state == NULL) || (text == NULL)) { + return URI_ERROR_NULL; + } + return URI_FUNC(ParseUriEx)(state, text, text + URI_STRLEN(text)); } - - int URI_FUNC(ParseSingleUri)(URI_TYPE(Uri) * uri, const URI_CHAR * text, - const URI_CHAR ** errorPos) { - return URI_FUNC(ParseSingleUriEx)(uri, text, NULL, errorPos); + const URI_CHAR ** errorPos) { + return URI_FUNC(ParseSingleUriEx)(uri, text, NULL, errorPos); } - - -int URI_FUNC(ParseSingleUriEx)(URI_TYPE(Uri) * uri, - const URI_CHAR * first, const URI_CHAR * afterLast, - const URI_CHAR ** errorPos) { +int URI_FUNC(ParseSingleUriEx)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, const URI_CHAR ** errorPos) { if ((afterLast == NULL) && (first != NULL)) { - afterLast = first + URI_STRLEN(first); - } - return URI_FUNC(ParseSingleUriExMm)(uri, first, afterLast, errorPos, NULL); + afterLast = first + URI_STRLEN(first); + } + return URI_FUNC(ParseSingleUriExMm)(uri, first, afterLast, errorPos, NULL); } +int URI_FUNC(ParseSingleUriExMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, const URI_CHAR ** errorPos, + UriMemoryManager * memory) { + URI_TYPE(ParserState) state; + int res; + /* Check params */ + if ((uri == NULL) || (first == NULL) || (afterLast == NULL)) { + return URI_ERROR_NULL; + } + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ -int URI_FUNC(ParseSingleUriExMm)(URI_TYPE(Uri) * uri, - const URI_CHAR * first, const URI_CHAR * afterLast, - const URI_CHAR ** errorPos, UriMemoryManager * memory) { - URI_TYPE(ParserState) state; - int res; - - /* Check params */ - if ((uri == NULL) || (first == NULL) || (afterLast == NULL)) { - return URI_ERROR_NULL; - } - URI_CHECK_MEMORY_MANAGER(memory); /* may return */ - - state.uri = uri; + state.uri = uri; - res = URI_FUNC(ParseUriExMm)(&state, first, afterLast, memory); + res = URI_FUNC(ParseUriExMm)(&state, first, afterLast, memory); - if (res != URI_SUCCESS) { - if (errorPos != NULL) { - *errorPos = state.errorPos; - } - URI_FUNC(FreeUriMembersMm)(uri, memory); - } + if (res != URI_SUCCESS) { + if (errorPos != NULL) { + *errorPos = state.errorPos; + } + URI_FUNC(FreeUriMembersMm)(uri, memory); + } - return res; + return res; } - - void URI_FUNC(FreeUriMembers)(URI_TYPE(Uri) * uri) { - URI_FUNC(FreeUriMembersMm)(uri, NULL); + URI_FUNC(FreeUriMembersMm)(uri, NULL); } - - int URI_FUNC(FreeUriMembersMm)(URI_TYPE(Uri) * uri, UriMemoryManager * memory) { - if (uri == NULL) { - return URI_ERROR_NULL; - } - - URI_CHECK_MEMORY_MANAGER(memory); /* may return */ - - if (uri->owner) { - /* Scheme */ - if (uri->scheme.first != NULL) { - if (uri->scheme.first != uri->scheme.afterLast) { - memory->free(memory, (URI_CHAR *)uri->scheme.first); - } - uri->scheme.first = NULL; - uri->scheme.afterLast = NULL; - } - - /* User info */ - if (uri->userInfo.first != NULL) { - if (uri->userInfo.first != uri->userInfo.afterLast) { - memory->free(memory, (URI_CHAR *)uri->userInfo.first); - } - uri->userInfo.first = NULL; - uri->userInfo.afterLast = NULL; - } - - /* Host data - IPvFuture */ - if (uri->hostData.ipFuture.first != NULL) { - if (uri->hostData.ipFuture.first != uri->hostData.ipFuture.afterLast) { - memory->free(memory, (URI_CHAR *)uri->hostData.ipFuture.first); - } - uri->hostData.ipFuture.first = NULL; - uri->hostData.ipFuture.afterLast = NULL; - uri->hostText.first = NULL; - uri->hostText.afterLast = NULL; - } - - /* Host text (if regname, after IPvFuture!) */ - if ((uri->hostText.first != NULL) - && (uri->hostData.ip4 == NULL) - && (uri->hostData.ip6 == NULL)) { - /* Real regname */ - if (uri->hostText.first != uri->hostText.afterLast) { - memory->free(memory, (URI_CHAR *)uri->hostText.first); - } - uri->hostText.first = NULL; - uri->hostText.afterLast = NULL; - } - } - - /* Host data - IPv4 */ - if (uri->hostData.ip4 != NULL) { - memory->free(memory, uri->hostData.ip4); - uri->hostData.ip4 = NULL; - } - - /* Host data - IPv6 */ - if (uri->hostData.ip6 != NULL) { - memory->free(memory, uri->hostData.ip6); - uri->hostData.ip6 = NULL; - } - - /* Port text */ - if (uri->owner && (uri->portText.first != NULL)) { - if (uri->portText.first != uri->portText.afterLast) { - memory->free(memory, (URI_CHAR *)uri->portText.first); - } - uri->portText.first = NULL; - uri->portText.afterLast = NULL; - } - - /* Path */ - if (uri->pathHead != NULL) { - URI_TYPE(PathSegment) * segWalk = uri->pathHead; - while (segWalk != NULL) { - URI_TYPE(PathSegment) * const next = segWalk->next; - if (uri->owner && (segWalk->text.first != NULL) - && (segWalk->text.first < segWalk->text.afterLast)) { - memory->free(memory, (URI_CHAR *)segWalk->text.first); - } - memory->free(memory, segWalk); - segWalk = next; - } - uri->pathHead = NULL; - uri->pathTail = NULL; - } - - if (uri->owner) { - /* Query */ - if (uri->query.first != NULL) { - if (uri->query.first != uri->query.afterLast) { - memory->free(memory, (URI_CHAR *)uri->query.first); - } - uri->query.first = NULL; - uri->query.afterLast = NULL; - } - - /* Fragment */ - if (uri->fragment.first != NULL) { - if (uri->fragment.first != uri->fragment.afterLast) { - memory->free(memory, (URI_CHAR *)uri->fragment.first); - } - uri->fragment.first = NULL; - uri->fragment.afterLast = NULL; - } - } - - return URI_SUCCESS; + if (uri == NULL) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + if (uri->owner) { + /* Scheme */ + if (uri->scheme.first != NULL) { + if (uri->scheme.first != uri->scheme.afterLast) { + memory->free(memory, (URI_CHAR *)uri->scheme.first); + } + uri->scheme.first = NULL; + uri->scheme.afterLast = NULL; + } + + /* User info */ + if (uri->userInfo.first != NULL) { + if (uri->userInfo.first != uri->userInfo.afterLast) { + memory->free(memory, (URI_CHAR *)uri->userInfo.first); + } + uri->userInfo.first = NULL; + uri->userInfo.afterLast = NULL; + } + + /* Host data - IPvFuture (may affect host text) */ + if (uri->hostData.ipFuture.first != NULL) { + /* NOTE: .hostData.ipFuture holds the very same range pointers + * as .hostText; we must not free memory twice. */ + uri->hostText.first = NULL; + uri->hostText.afterLast = NULL; + + if (uri->hostData.ipFuture.first != uri->hostData.ipFuture.afterLast) { + memory->free(memory, (URI_CHAR *)uri->hostData.ipFuture.first); + } + uri->hostData.ipFuture.first = NULL; + uri->hostData.ipFuture.afterLast = NULL; + } + + /* Host text (after IPvFuture, see above) */ + if (uri->hostText.first != NULL) { + if (uri->hostText.first != uri->hostText.afterLast) { + memory->free(memory, (URI_CHAR *)uri->hostText.first); + } + uri->hostText.first = NULL; + uri->hostText.afterLast = NULL; + } + } + + /* Host data - IPv4 */ + if (uri->hostData.ip4 != NULL) { + memory->free(memory, uri->hostData.ip4); + uri->hostData.ip4 = NULL; + } + + /* Host data - IPv6 */ + if (uri->hostData.ip6 != NULL) { + memory->free(memory, uri->hostData.ip6); + uri->hostData.ip6 = NULL; + } + + /* Port text */ + if (uri->owner && (uri->portText.first != NULL)) { + if (uri->portText.first != uri->portText.afterLast) { + memory->free(memory, (URI_CHAR *)uri->portText.first); + } + uri->portText.first = NULL; + uri->portText.afterLast = NULL; + } + + /* Path */ + URI_FUNC(FreeUriPath)(uri, memory); + + if (uri->owner) { + /* Query */ + if (uri->query.first != NULL) { + if (uri->query.first != uri->query.afterLast) { + memory->free(memory, (URI_CHAR *)uri->query.first); + } + uri->query.first = NULL; + uri->query.afterLast = NULL; + } + + /* Fragment */ + if (uri->fragment.first != NULL) { + if (uri->fragment.first != uri->fragment.afterLast) { + memory->free(memory, (URI_CHAR *)uri->fragment.first); + } + uri->fragment.first = NULL; + uri->fragment.afterLast = NULL; + } + } + + return URI_SUCCESS; } - - UriBool URI_FUNC(_TESTING_ONLY_ParseIpSix)(const URI_CHAR * text) { - UriMemoryManager * const memory = &defaultMemoryManager; - URI_TYPE(Uri) uri; - URI_TYPE(ParserState) parser; - const URI_CHAR * const afterIpSix = text + URI_STRLEN(text); - const URI_CHAR * res; - - URI_FUNC(ResetUri)(&uri); - parser.uri = &uri; - URI_FUNC(ResetParserStateExceptUri)(&parser); - parser.uri->hostData.ip6 = memory->malloc(memory, 1 * sizeof(UriIp6)); - res = URI_FUNC(ParseIPv6address2)(&parser, text, afterIpSix, memory); - URI_FUNC(FreeUriMembersMm)(&uri, memory); - return res == afterIpSix ? URI_TRUE : URI_FALSE; + UriMemoryManager * const memory = &defaultMemoryManager; + URI_TYPE(Uri) uri; + URI_TYPE(ParserState) parser; + const URI_CHAR * const afterIpSix = text + URI_STRLEN(text); + const URI_CHAR * res; + + URI_FUNC(ResetUri)(&uri); + parser.uri = &uri; + URI_FUNC(ResetParserStateExceptUri)(&parser); + parser.uri->hostData.ip6 = memory->malloc(memory, 1 * sizeof(UriIp6)); + res = URI_FUNC(ParseIPv6address2)(&parser, text, afterIpSix, memory); + URI_FUNC(FreeUriMembersMm)(&uri, memory); + return res == afterIpSix ? URI_TRUE : URI_FALSE; } - - UriBool URI_FUNC(_TESTING_ONLY_ParseIpFour)(const URI_CHAR * text) { - unsigned char octets[4]; - int res = URI_FUNC(ParseIpFourAddress)(octets, text, text + URI_STRLEN(text)); - return (res == URI_SUCCESS) ? URI_TRUE : URI_FALSE; + unsigned char octets[4]; + int res = URI_FUNC(ParseIpFourAddress)(octets, text, text + URI_STRLEN(text)); + return (res == URI_SUCCESS) ? URI_TRUE : URI_FALSE; } - - -#undef URI_SET_DIGIT -#undef URI_SET_HEX_LETTER_UPPER -#undef URI_SET_HEX_LETTER_LOWER -#undef URI_SET_HEXDIG -#undef URI_SET_ALPHA - - +# undef URI_SET_DIGIT +# undef URI_SET_HEX_LETTER_UPPER +# undef URI_SET_HEX_LETTER_LOWER +# undef URI_SET_HEXDIG +# undef URI_SET_ALPHA #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriParseBase.c b/cpp/src/arrow/vendored/uriparser/UriParseBase.c index 1d4ef6e2bca1..0ee66d4427e3 100644 --- a/cpp/src/arrow/vendored/uriparser/UriParseBase.c +++ b/cpp/src/arrow/vendored/uriparser/UriParseBase.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -38,53 +38,48 @@ */ #ifndef URI_DOXYGEN -# include "UriParseBase.h" +# include "UriParseBase.h" #endif +void uriWriteQuadToDoubleByte(const unsigned char * hexDigits, int digitCount, + unsigned char * output) { + switch (digitCount) { + case 1: + /* 0x___? -> \x00 \x0? */ + output[0] = 0; + output[1] = hexDigits[0]; + break; + case 2: + /* 0x__?? -> \0xx \x?? */ + output[0] = 0; + output[1] = 16 * hexDigits[0] + hexDigits[1]; + break; -void uriWriteQuadToDoubleByte(const unsigned char * hexDigits, int digitCount, unsigned char * output) { - switch (digitCount) { - case 1: - /* 0x___? -> \x00 \x0? */ - output[0] = 0; - output[1] = hexDigits[0]; - break; + case 3: + /* 0x_??? -> \0x? \x?? */ + output[0] = hexDigits[0]; + output[1] = 16 * hexDigits[1] + hexDigits[2]; + break; - case 2: - /* 0x__?? -> \0xx \x?? */ - output[0] = 0; - output[1] = 16 * hexDigits[0] + hexDigits[1]; - break; - - case 3: - /* 0x_??? -> \0x? \x?? */ - output[0] = hexDigits[0]; - output[1] = 16 * hexDigits[1] + hexDigits[2]; - break; - - case 4: - /* 0x???? -> \0?? \x?? */ - output[0] = 16 * hexDigits[0] + hexDigits[1]; - output[1] = 16 * hexDigits[2] + hexDigits[3]; - break; - - } + case 4: + /* 0x???? -> \0?? \x?? */ + output[0] = 16 * hexDigits[0] + hexDigits[1]; + output[1] = 16 * hexDigits[2] + hexDigits[3]; + break; + } } - - unsigned char uriGetOctetValue(const unsigned char * digits, int digitCount) { - switch (digitCount) { - case 1: - return digits[0]; - - case 2: - return 10 * digits[0] + digits[1]; + switch (digitCount) { + case 1: + return digits[0]; - case 3: - default: - return 100 * digits[0] + 10 * digits[1] + digits[2]; + case 2: + return 10 * digits[0] + digits[1]; - } + case 3: + default: + return 100 * digits[0] + 10 * digits[1] + digits[2]; + } } diff --git a/cpp/src/arrow/vendored/uriparser/UriParseBase.h b/cpp/src/arrow/vendored/uriparser/UriParseBase.h index eea9ffa29d78..e7e3695ece35 100644 --- a/cpp/src/arrow/vendored/uriparser/UriParseBase.h +++ b/cpp/src/arrow/vendored/uriparser/UriParseBase.h @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -38,18 +38,12 @@ */ #ifndef URI_PARSE_BASE_H -#define URI_PARSE_BASE_H 1 - - - -#include "UriBase.h" - +# define URI_PARSE_BASE_H 1 +# include "UriBase.h" void uriWriteQuadToDoubleByte(const unsigned char * hexDigits, int digitCount, - unsigned char * output); + unsigned char * output); unsigned char uriGetOctetValue(const unsigned char * digits, int digitCount); - - #endif /* URI_PARSE_BASE_H */ diff --git a/cpp/src/arrow/vendored/uriparser/UriQuery.c b/cpp/src/arrow/vendored/uriparser/UriQuery.c index 42a74aeec474..9565ba70da68 100644 --- a/cpp/src/arrow/vendored/uriparser/UriQuery.c +++ b/cpp/src/arrow/vendored/uriparser/UriQuery.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -41,461 +41,450 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriQuery.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriQuery.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriQuery.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriQuery.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -# include "UriCommon.h" -# include "UriMemory.h" -#endif - - - -#include - - +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriMemory.h" +# endif + +# include +# include /* size_t */ +# include // SIZE_MAX static int URI_FUNC(ComposeQueryEngine)(URI_CHAR * dest, - const URI_TYPE(QueryList) * queryList, - int maxChars, int * charsWritten, int * charsRequired, - UriBool spaceToPlus, UriBool normalizeBreaks); - -static UriBool URI_FUNC(AppendQueryItem)(URI_TYPE(QueryList) ** prevNext, - int * itemCount, const URI_CHAR * keyFirst, const URI_CHAR * keyAfter, - const URI_CHAR * valueFirst, const URI_CHAR * valueAfter, - UriBool plusToSpace, UriBreakConversion breakConversion, - UriMemoryManager * memory); - + const URI_TYPE(QueryList) * queryList, + int maxChars, int * charsWritten, + int * charsRequired, UriBool spaceToPlus, + UriBool normalizeBreaks); +static UriBool URI_FUNC(AppendQueryItem)( + URI_TYPE(QueryList) * *prevNext, int * itemCount, const URI_CHAR * keyFirst, + const URI_CHAR * keyAfter, const URI_CHAR * valueFirst, const URI_CHAR * valueAfter, + UriBool plusToSpace, UriBreakConversion breakConversion, UriMemoryManager * memory); int URI_FUNC(ComposeQueryCharsRequired)(const URI_TYPE(QueryList) * queryList, - int * charsRequired) { - const UriBool spaceToPlus = URI_TRUE; - const UriBool normalizeBreaks = URI_TRUE; + int * charsRequired) { + const UriBool spaceToPlus = URI_TRUE; + const UriBool normalizeBreaks = URI_TRUE; - return URI_FUNC(ComposeQueryCharsRequiredEx)(queryList, charsRequired, - spaceToPlus, normalizeBreaks); + return URI_FUNC(ComposeQueryCharsRequiredEx)(queryList, charsRequired, spaceToPlus, + normalizeBreaks); } - - int URI_FUNC(ComposeQueryCharsRequiredEx)(const URI_TYPE(QueryList) * queryList, - int * charsRequired, UriBool spaceToPlus, UriBool normalizeBreaks) { - if ((queryList == NULL) || (charsRequired == NULL)) { - return URI_ERROR_NULL; - } - - return URI_FUNC(ComposeQueryEngine)(NULL, queryList, 0, NULL, - charsRequired, spaceToPlus, normalizeBreaks); + int * charsRequired, UriBool spaceToPlus, + UriBool normalizeBreaks) { + if ((queryList == NULL) || (charsRequired == NULL)) { + return URI_ERROR_NULL; + } + + return URI_FUNC(ComposeQueryEngine)(NULL, queryList, 0, NULL, charsRequired, + spaceToPlus, normalizeBreaks); } +int URI_FUNC(ComposeQuery)(URI_CHAR * dest, const URI_TYPE(QueryList) * queryList, + int maxChars, int * charsWritten) { + const UriBool spaceToPlus = URI_TRUE; + const UriBool normalizeBreaks = URI_TRUE; - -int URI_FUNC(ComposeQuery)(URI_CHAR * dest, - const URI_TYPE(QueryList) * queryList, int maxChars, int * charsWritten) { - const UriBool spaceToPlus = URI_TRUE; - const UriBool normalizeBreaks = URI_TRUE; - - return URI_FUNC(ComposeQueryEx)(dest, queryList, maxChars, charsWritten, - spaceToPlus, normalizeBreaks); + return URI_FUNC(ComposeQueryEx)(dest, queryList, maxChars, charsWritten, spaceToPlus, + normalizeBreaks); } +int URI_FUNC(ComposeQueryEx)(URI_CHAR * dest, const URI_TYPE(QueryList) * queryList, + int maxChars, int * charsWritten, UriBool spaceToPlus, + UriBool normalizeBreaks) { + if ((dest == NULL) || (queryList == NULL)) { + return URI_ERROR_NULL; + } + if (maxChars < 1) { + return URI_ERROR_OUTPUT_TOO_LARGE; + } -int URI_FUNC(ComposeQueryEx)(URI_CHAR * dest, - const URI_TYPE(QueryList) * queryList, int maxChars, int * charsWritten, - UriBool spaceToPlus, UriBool normalizeBreaks) { - if ((dest == NULL) || (queryList == NULL)) { - return URI_ERROR_NULL; - } - - if (maxChars < 1) { - return URI_ERROR_OUTPUT_TOO_LARGE; - } - - return URI_FUNC(ComposeQueryEngine)(dest, queryList, maxChars, - charsWritten, NULL, spaceToPlus, normalizeBreaks); + return URI_FUNC(ComposeQueryEngine)(dest, queryList, maxChars, charsWritten, NULL, + spaceToPlus, normalizeBreaks); } - - int URI_FUNC(ComposeQueryMalloc)(URI_CHAR ** dest, - const URI_TYPE(QueryList) * queryList) { - const UriBool spaceToPlus = URI_TRUE; - const UriBool normalizeBreaks = URI_TRUE; + const URI_TYPE(QueryList) * queryList) { + const UriBool spaceToPlus = URI_TRUE; + const UriBool normalizeBreaks = URI_TRUE; - return URI_FUNC(ComposeQueryMallocEx)(dest, queryList, - spaceToPlus, normalizeBreaks); + return URI_FUNC(ComposeQueryMallocEx)(dest, queryList, spaceToPlus, normalizeBreaks); } - - int URI_FUNC(ComposeQueryMallocEx)(URI_CHAR ** dest, - const URI_TYPE(QueryList) * queryList, - UriBool spaceToPlus, UriBool normalizeBreaks) { - return URI_FUNC(ComposeQueryMallocExMm)(dest, queryList, spaceToPlus, - normalizeBreaks, NULL); + const URI_TYPE(QueryList) * queryList, + UriBool spaceToPlus, UriBool normalizeBreaks) { + return URI_FUNC(ComposeQueryMallocExMm)(dest, queryList, spaceToPlus, normalizeBreaks, + NULL); } - - int URI_FUNC(ComposeQueryMallocExMm)(URI_CHAR ** dest, - const URI_TYPE(QueryList) * queryList, - UriBool spaceToPlus, UriBool normalizeBreaks, - UriMemoryManager * memory) { - int charsRequired; - int res; - URI_CHAR * queryString; - - if (dest == NULL) { - return URI_ERROR_NULL; - } - - URI_CHECK_MEMORY_MANAGER(memory); /* may return */ - - /* Calculate space */ - res = URI_FUNC(ComposeQueryCharsRequiredEx)(queryList, &charsRequired, - spaceToPlus, normalizeBreaks); - if (res != URI_SUCCESS) { - return res; - } - charsRequired++; - - /* Allocate space */ - queryString = memory->malloc(memory, charsRequired * sizeof(URI_CHAR)); - if (queryString == NULL) { - return URI_ERROR_MALLOC; - } - - /* Put query in */ - res = URI_FUNC(ComposeQueryEx)(queryString, queryList, charsRequired, - NULL, spaceToPlus, normalizeBreaks); - if (res != URI_SUCCESS) { - memory->free(memory, queryString); - return res; - } - - *dest = queryString; - return URI_SUCCESS; + const URI_TYPE(QueryList) * queryList, + UriBool spaceToPlus, UriBool normalizeBreaks, + UriMemoryManager * memory) { + int charsRequired; + int res; + URI_CHAR * queryString; + + if (dest == NULL) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + /* Calculate space */ + res = URI_FUNC(ComposeQueryCharsRequiredEx)(queryList, &charsRequired, spaceToPlus, + normalizeBreaks); + if (res != URI_SUCCESS) { + return res; + } + if (charsRequired == INT_MAX) { + return URI_ERROR_MALLOC; + } + charsRequired++; + + /* Allocate space */ + queryString = memory->calloc(memory, charsRequired, sizeof(URI_CHAR)); + if (queryString == NULL) { + return URI_ERROR_MALLOC; + } + + /* Put query in */ + res = URI_FUNC(ComposeQueryEx)(queryString, queryList, charsRequired, NULL, + spaceToPlus, normalizeBreaks); + if (res != URI_SUCCESS) { + memory->free(memory, queryString); + return res; + } + + *dest = queryString; + return URI_SUCCESS; } - - -int URI_FUNC(ComposeQueryEngine)(URI_CHAR * dest, - const URI_TYPE(QueryList) * queryList, - int maxChars, int * charsWritten, int * charsRequired, - UriBool spaceToPlus, UriBool normalizeBreaks) { - UriBool firstItem = URI_TRUE; - int ampersandLen = 0; /* increased to 1 from second item on */ - URI_CHAR * write = dest; - - /* Subtract terminator */ - if (dest == NULL) { - *charsRequired = 0; - } else { - maxChars--; - } - - while (queryList != NULL) { - const URI_CHAR * const key = queryList->key; - const URI_CHAR * const value = queryList->value; - const int worstCase = (normalizeBreaks == URI_TRUE ? 6 : 3); - const int keyLen = (key == NULL) ? 0 : (int)URI_STRLEN(key); - int keyRequiredChars; - const int valueLen = (value == NULL) ? 0 : (int)URI_STRLEN(value); - int valueRequiredChars; - - if ((keyLen >= INT_MAX / worstCase) || (valueLen >= INT_MAX / worstCase)) { - return URI_ERROR_OUTPUT_TOO_LARGE; - } - keyRequiredChars = worstCase * keyLen; - valueRequiredChars = worstCase * valueLen; - - if (dest == NULL) { - (*charsRequired) += ampersandLen + keyRequiredChars + ((value == NULL) - ? 0 - : 1 + valueRequiredChars); - - if (firstItem == URI_TRUE) { - ampersandLen = 1; - firstItem = URI_FALSE; - } - } else { - if ((write - dest) + ampersandLen + keyRequiredChars > maxChars) { - return URI_ERROR_OUTPUT_TOO_LARGE; - } - - /* Copy key */ - if (firstItem == URI_TRUE) { - ampersandLen = 1; - firstItem = URI_FALSE; - } else { - write[0] = _UT('&'); - write++; - } - write = URI_FUNC(EscapeEx)(key, key + keyLen, - write, spaceToPlus, normalizeBreaks); - - if (value != NULL) { - if ((write - dest) + 1 + valueRequiredChars > maxChars) { - return URI_ERROR_OUTPUT_TOO_LARGE; - } - - /* Copy value */ - write[0] = _UT('='); - write++; - write = URI_FUNC(EscapeEx)(value, value + valueLen, - write, spaceToPlus, normalizeBreaks); - } - } - - queryList = queryList->next; - } - - if (dest != NULL) { - write[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = (int)(write - dest) + 1; /* .. for terminator */ - } - } - - return URI_SUCCESS; +int URI_FUNC(ComposeQueryEngine)(URI_CHAR * dest, const URI_TYPE(QueryList) * queryList, + int maxChars, int * charsWritten, int * charsRequired, + UriBool spaceToPlus, UriBool normalizeBreaks) { + UriBool firstItem = URI_TRUE; + int ampersandLen = 0; /* increased to 1 from second item on */ + URI_CHAR * write = dest; + + /* Subtract terminator */ + if (dest == NULL) { + *charsRequired = 0; + } else { + maxChars--; + } + + while (queryList != NULL) { + const URI_CHAR * const key = queryList->key; + const URI_CHAR * const value = queryList->value; + const int worstCase = (normalizeBreaks == URI_TRUE ? 6 : 3); + const size_t keyLen = (key == NULL) ? 0 : URI_STRLEN(key); + int keyRequiredChars; + const size_t valueLen = (value == NULL) ? 0 : URI_STRLEN(value); + int valueRequiredChars; + + if ((keyLen >= (size_t)INT_MAX / worstCase) + || (valueLen >= (size_t)INT_MAX / worstCase)) { + return URI_ERROR_OUTPUT_TOO_LARGE; + } + keyRequiredChars = worstCase * (int)keyLen; + valueRequiredChars = worstCase * (int)valueLen; + + if (dest == NULL) { + (*charsRequired) += ampersandLen + keyRequiredChars + + ((value == NULL) ? 0 : 1 + valueRequiredChars); + + if (firstItem == URI_TRUE) { + ampersandLen = 1; + firstItem = URI_FALSE; + } + } else { + if ((write - dest) + ampersandLen + keyRequiredChars > maxChars) { + return URI_ERROR_OUTPUT_TOO_LARGE; + } + + /* Copy key */ + if (firstItem == URI_TRUE) { + ampersandLen = 1; + firstItem = URI_FALSE; + } else { + write[0] = _UT('&'); + write++; + } + write = URI_FUNC(EscapeEx)(key, key + keyLen, write, spaceToPlus, + normalizeBreaks); + + if (value != NULL) { + if ((write - dest) + 1 + valueRequiredChars > maxChars) { + return URI_ERROR_OUTPUT_TOO_LARGE; + } + + /* Copy value */ + write[0] = _UT('='); + write++; + write = URI_FUNC(EscapeEx)(value, value + valueLen, write, spaceToPlus, + normalizeBreaks); + } + } + + queryList = queryList->next; + } + + if (dest != NULL) { + write[0] = _UT('\0'); + if (charsWritten != NULL) { + const size_t lenInChars = write - dest; + + // Detect and avoid integer overflow + if (lenInChars > INT_MAX - 1) { + return URI_ERROR_OUTPUT_TOO_LARGE; + } + + *charsWritten = (int)(lenInChars + 1); /* .. for terminator */ + } + } + + return URI_SUCCESS; } - - -UriBool URI_FUNC(AppendQueryItem)(URI_TYPE(QueryList) ** prevNext, - int * itemCount, const URI_CHAR * keyFirst, const URI_CHAR * keyAfter, - const URI_CHAR * valueFirst, const URI_CHAR * valueAfter, - UriBool plusToSpace, UriBreakConversion breakConversion, - UriMemoryManager * memory) { - const int keyLen = (int)(keyAfter - keyFirst); - const int valueLen = (int)(valueAfter - valueFirst); - URI_CHAR * key; - URI_CHAR * value; - - if ((prevNext == NULL) || (itemCount == NULL) - || (keyFirst == NULL) || (keyAfter == NULL) - || (keyFirst > keyAfter) || (valueFirst > valueAfter) - || ((keyFirst == keyAfter) - && (valueFirst == NULL) && (valueAfter == NULL))) { - return URI_TRUE; - } - - /* Append new empty item */ - *prevNext = memory->malloc(memory, 1 * sizeof(URI_TYPE(QueryList))); - if (*prevNext == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - (*prevNext)->next = NULL; - - - /* Fill key */ - key = memory->malloc(memory, (keyLen + 1) * sizeof(URI_CHAR)); - if (key == NULL) { - memory->free(memory, *prevNext); - *prevNext = NULL; - return URI_FALSE; /* Raises malloc error */ - } - - key[keyLen] = _UT('\0'); - if (keyLen > 0) { - /* Copy 1:1 */ - memcpy(key, keyFirst, keyLen * sizeof(URI_CHAR)); - - /* Unescape */ - URI_FUNC(UnescapeInPlaceEx)(key, plusToSpace, breakConversion); - } - (*prevNext)->key = key; - - - /* Fill value */ - if (valueFirst != NULL) { - value = memory->malloc(memory, (valueLen + 1) * sizeof(URI_CHAR)); - if (value == NULL) { - memory->free(memory, key); - memory->free(memory, *prevNext); - *prevNext = NULL; - return URI_FALSE; /* Raises malloc error */ - } - - value[valueLen] = _UT('\0'); - if (valueLen > 0) { - /* Copy 1:1 */ - memcpy(value, valueFirst, valueLen * sizeof(URI_CHAR)); - - /* Unescape */ - URI_FUNC(UnescapeInPlaceEx)(value, plusToSpace, breakConversion); - } - (*prevNext)->value = value; - } else { - value = NULL; - } - (*prevNext)->value = value; - - (*itemCount)++; - return URI_TRUE; +UriBool URI_FUNC(AppendQueryItem)(URI_TYPE(QueryList) * *prevNext, int * itemCount, + const URI_CHAR * keyFirst, const URI_CHAR * keyAfter, + const URI_CHAR * valueFirst, + const URI_CHAR * valueAfter, UriBool plusToSpace, + UriBreakConversion breakConversion, + UriMemoryManager * memory) { + const size_t keyLen = keyAfter - keyFirst; + const size_t valueLen = valueAfter - valueFirst; + URI_CHAR * key; + URI_CHAR * value; + + if ((prevNext == NULL) || (itemCount == NULL) || (keyFirst == NULL) + || (keyAfter == NULL) || (keyFirst > keyAfter) || (valueFirst > valueAfter) + || ((keyFirst == keyAfter) && (valueFirst == NULL) && (valueAfter == NULL))) { + return URI_TRUE; + } + + /* Append new empty item */ + *prevNext = memory->malloc(memory, 1 * sizeof(URI_TYPE(QueryList))); + if (*prevNext == NULL) { + return URI_FALSE; /* Raises malloc error */ + } + (*prevNext)->next = NULL; + + // Detect integer overflow + if ((keyLen > SIZE_MAX - 1) || (keyLen + 1 > SIZE_MAX / sizeof(URI_CHAR))) { + memory->free(memory, *prevNext); + *prevNext = NULL; + return URI_FALSE; // Raises malloc error + } + + /* Fill key */ + key = memory->malloc(memory, (keyLen + 1) * sizeof(URI_CHAR)); + if (key == NULL) { + memory->free(memory, *prevNext); + *prevNext = NULL; + return URI_FALSE; /* Raises malloc error */ + } + + key[keyLen] = _UT('\0'); + if (keyLen > 0) { + /* Copy 1:1 */ + memcpy(key, keyFirst, keyLen * sizeof(URI_CHAR)); + + /* Unescape */ + URI_FUNC(UnescapeInPlaceEx)(key, plusToSpace, breakConversion); + } + (*prevNext)->key = key; + + /* Fill value */ + if (valueFirst != NULL) { + // Detect integer overflow + if ((valueLen > SIZE_MAX - 1) || (valueLen + 1 > SIZE_MAX / sizeof(URI_CHAR))) { + memory->free(memory, key); + memory->free(memory, *prevNext); + *prevNext = NULL; + return URI_FALSE; // Raises malloc error + } + + value = memory->malloc(memory, (valueLen + 1) * sizeof(URI_CHAR)); + if (value == NULL) { + memory->free(memory, key); + memory->free(memory, *prevNext); + *prevNext = NULL; + return URI_FALSE; /* Raises malloc error */ + } + + value[valueLen] = _UT('\0'); + if (valueLen > 0) { + /* Copy 1:1 */ + memcpy(value, valueFirst, valueLen * sizeof(URI_CHAR)); + + /* Unescape */ + URI_FUNC(UnescapeInPlaceEx)(value, plusToSpace, breakConversion); + } + (*prevNext)->value = value; + } else { + value = NULL; + } + (*prevNext)->value = value; + + (*itemCount)++; + return URI_TRUE; } - - void URI_FUNC(FreeQueryList)(URI_TYPE(QueryList) * queryList) { - URI_FUNC(FreeQueryListMm)(queryList, NULL); + URI_FUNC(FreeQueryListMm)(queryList, NULL); } - - int URI_FUNC(FreeQueryListMm)(URI_TYPE(QueryList) * queryList, - UriMemoryManager * memory) { - URI_CHECK_MEMORY_MANAGER(memory); /* may return */ - while (queryList != NULL) { - URI_TYPE(QueryList) * nextBackup = queryList->next; - memory->free(memory, (URI_CHAR *)queryList->key); /* const cast */ - memory->free(memory, (URI_CHAR *)queryList->value); /* const cast */ - memory->free(memory, queryList); - queryList = nextBackup; - } - return URI_SUCCESS; + UriMemoryManager * memory) { + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + while (queryList != NULL) { + URI_TYPE(QueryList) * nextBackup = queryList->next; + memory->free(memory, (URI_CHAR *)queryList->key); /* const cast */ + memory->free(memory, (URI_CHAR *)queryList->value); /* const cast */ + memory->free(memory, queryList); + queryList = nextBackup; + } + return URI_SUCCESS; } +int URI_FUNC(DissectQueryMalloc)(URI_TYPE(QueryList) * *dest, int * itemCount, + const URI_CHAR * first, const URI_CHAR * afterLast) { + const UriBool plusToSpace = URI_TRUE; + const UriBreakConversion breakConversion = URI_BR_DONT_TOUCH; - -int URI_FUNC(DissectQueryMalloc)(URI_TYPE(QueryList) ** dest, int * itemCount, - const URI_CHAR * first, const URI_CHAR * afterLast) { - const UriBool plusToSpace = URI_TRUE; - const UriBreakConversion breakConversion = URI_BR_DONT_TOUCH; - - return URI_FUNC(DissectQueryMallocEx)(dest, itemCount, first, afterLast, - plusToSpace, breakConversion); + return URI_FUNC(DissectQueryMallocEx)(dest, itemCount, first, afterLast, plusToSpace, + breakConversion); } - - -int URI_FUNC(DissectQueryMallocEx)(URI_TYPE(QueryList) ** dest, int * itemCount, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriBool plusToSpace, UriBreakConversion breakConversion) { - return URI_FUNC(DissectQueryMallocExMm)(dest, itemCount, first, afterLast, - plusToSpace, breakConversion, NULL); +int URI_FUNC(DissectQueryMallocEx)(URI_TYPE(QueryList) * *dest, int * itemCount, + const URI_CHAR * first, const URI_CHAR * afterLast, + UriBool plusToSpace, + UriBreakConversion breakConversion) { + return URI_FUNC(DissectQueryMallocExMm)(dest, itemCount, first, afterLast, + plusToSpace, breakConversion, NULL); } - - -int URI_FUNC(DissectQueryMallocExMm)(URI_TYPE(QueryList) ** dest, int * itemCount, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriBool plusToSpace, UriBreakConversion breakConversion, - UriMemoryManager * memory) { - const URI_CHAR * walk = first; - const URI_CHAR * keyFirst = first; - const URI_CHAR * keyAfter = NULL; - const URI_CHAR * valueFirst = NULL; - const URI_CHAR * valueAfter = NULL; - URI_TYPE(QueryList) ** prevNext = dest; - int nullCounter; - int * itemsAppended = (itemCount == NULL) ? &nullCounter : itemCount; - - if ((dest == NULL) || (first == NULL) || (afterLast == NULL)) { - return URI_ERROR_NULL; - } - - if (first > afterLast) { - return URI_ERROR_RANGE_INVALID; - } - - URI_CHECK_MEMORY_MANAGER(memory); /* may return */ - - *dest = NULL; - *itemsAppended = 0; - - /* Parse query string */ - for (; walk < afterLast; walk++) { - switch (*walk) { - case _UT('&'): - if (valueFirst != NULL) { - valueAfter = walk; - } else { - keyAfter = walk; - } - - if (URI_FUNC(AppendQueryItem)(prevNext, itemsAppended, - keyFirst, keyAfter, valueFirst, valueAfter, - plusToSpace, breakConversion, memory) - == URI_FALSE) { - /* Free list we built */ - *itemsAppended = 0; - URI_FUNC(FreeQueryListMm)(*dest, memory); - return URI_ERROR_MALLOC; - } - - /* Make future items children of the current */ - if ((prevNext != NULL) && (*prevNext != NULL)) { - prevNext = &((*prevNext)->next); - } - - if (walk + 1 < afterLast) { - keyFirst = walk + 1; - } else { - keyFirst = NULL; - } - keyAfter = NULL; - valueFirst = NULL; - valueAfter = NULL; - break; - - case _UT('='): - /* NOTE: WE treat the first '=' as a separator, */ - /* all following go into the value part */ - if (keyAfter == NULL) { - keyAfter = walk; - if (walk + 1 <= afterLast) { - valueFirst = walk + 1; - valueAfter = walk + 1; - } - } - break; - - default: - break; - } - } - - if (valueFirst != NULL) { - /* Must be key/value pair */ - valueAfter = walk; - } else { - /* Must be key only */ - keyAfter = walk; - } - - if (URI_FUNC(AppendQueryItem)(prevNext, itemsAppended, keyFirst, keyAfter, - valueFirst, valueAfter, plusToSpace, breakConversion, memory) - == URI_FALSE) { - /* Free list we built */ - *itemsAppended = 0; - URI_FUNC(FreeQueryListMm)(*dest, memory); - return URI_ERROR_MALLOC; - } - - return URI_SUCCESS; +int URI_FUNC(DissectQueryMallocExMm)(URI_TYPE(QueryList) * *dest, int * itemCount, + const URI_CHAR * first, const URI_CHAR * afterLast, + UriBool plusToSpace, + UriBreakConversion breakConversion, + UriMemoryManager * memory) { + const URI_CHAR * walk = first; + const URI_CHAR * keyFirst = first; + const URI_CHAR * keyAfter = NULL; + const URI_CHAR * valueFirst = NULL; + const URI_CHAR * valueAfter = NULL; + URI_TYPE(QueryList) ** prevNext = dest; + int nullCounter; + int * itemsAppended = (itemCount == NULL) ? &nullCounter : itemCount; + + if ((dest == NULL) || (first == NULL) || (afterLast == NULL)) { + return URI_ERROR_NULL; + } + + if (first > afterLast) { + return URI_ERROR_RANGE_INVALID; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + *dest = NULL; + *itemsAppended = 0; + + /* Parse query string */ + for (; walk < afterLast; walk++) { + switch (*walk) { + case _UT('&'): + if (valueFirst != NULL) { + valueAfter = walk; + } else { + keyAfter = walk; + } + + if (URI_FUNC(AppendQueryItem)(prevNext, itemsAppended, keyFirst, keyAfter, + valueFirst, valueAfter, plusToSpace, + breakConversion, memory) + == URI_FALSE) { + /* Free list we built */ + *itemsAppended = 0; + URI_FUNC(FreeQueryListMm)(*dest, memory); + return URI_ERROR_MALLOC; + } + + /* Make future items children of the current */ + if ((prevNext != NULL) && (*prevNext != NULL)) { + prevNext = &((*prevNext)->next); + } + + if (walk + 1 < afterLast) { + keyFirst = walk + 1; + } else { + keyFirst = NULL; + } + keyAfter = NULL; + valueFirst = NULL; + valueAfter = NULL; + break; + + case _UT('='): + /* NOTE: WE treat the first '=' as a separator, */ + /* all following go into the value part */ + if (keyAfter == NULL) { + keyAfter = walk; + if (walk + 1 <= afterLast) { + valueFirst = walk + 1; + valueAfter = walk + 1; + } + } + break; + + default: + break; + } + } + + if (valueFirst != NULL) { + /* Must be key/value pair */ + valueAfter = walk; + } else { + /* Must be key only */ + keyAfter = walk; + } + + if (URI_FUNC(AppendQueryItem)(prevNext, itemsAppended, keyFirst, keyAfter, valueFirst, + valueAfter, plusToSpace, breakConversion, memory) + == URI_FALSE) { + /* Free list we built */ + *itemsAppended = 0; + URI_FUNC(FreeQueryListMm)(*dest, memory); + return URI_ERROR_MALLOC; + } + + return URI_SUCCESS; } - - #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriRecompose.c b/cpp/src/arrow/vendored/uriparser/UriRecompose.c index 2f987215fa18..6633c28e8f09 100644 --- a/cpp/src/arrow/vendored/uriparser/UriRecompose.c +++ b/cpp/src/arrow/vendored/uriparser/UriRecompose.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -41,537 +41,691 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriRecompose.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriRecompose.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriRecompose.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriRecompose.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -# include "UriCommon.h" -#endif +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# endif +# include // INT_MAX +# include // SIZE_MAX static int URI_FUNC(ToStringEngine)(URI_CHAR * dest, const URI_TYPE(Uri) * uri, - int maxChars, int * charsWritten, int * charsRequired); - - + int maxChars, int * charsWritten, + int * charsRequired); -int URI_FUNC(ToStringCharsRequired)(const URI_TYPE(Uri) * uri, - int * charsRequired) { - const int MAX_CHARS = ((unsigned int)-1) >> 1; - return URI_FUNC(ToStringEngine)(NULL, uri, MAX_CHARS, NULL, charsRequired); +int URI_FUNC(ToStringCharsRequired)(const URI_TYPE(Uri) * uri, int * charsRequired) { + const int MAX_CHARS = ((unsigned int)-1) >> 1; + return URI_FUNC(ToStringEngine)(NULL, uri, MAX_CHARS, NULL, charsRequired); } - - -int URI_FUNC(ToString)(URI_CHAR * dest, const URI_TYPE(Uri) * uri, - int maxChars, int * charsWritten) { - return URI_FUNC(ToStringEngine)(dest, uri, maxChars, charsWritten, NULL); +int URI_FUNC(ToString)(URI_CHAR * dest, const URI_TYPE(Uri) * uri, int maxChars, + int * charsWritten) { + return URI_FUNC(ToStringEngine)(dest, uri, maxChars, charsWritten, NULL); } - - -static URI_INLINE int URI_FUNC(ToStringEngine)(URI_CHAR * dest, - const URI_TYPE(Uri) * uri, int maxChars, int * charsWritten, - int * charsRequired) { - int written = 0; - if ((uri == NULL) || ((dest == NULL) && (charsRequired == NULL))) { - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_NULL; - } - - if (maxChars < 1) { - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - maxChars--; /* So we don't have to subtract 1 for '\0' all the time */ - - /* [01/19] result = "" */ - if (dest != NULL) { - dest[0] = _UT('\0'); - } else { - (*charsRequired) = 0; - } - /* [02/19] if defined(scheme) then */ - if (uri->scheme.first != NULL) { - /* [03/19] append scheme to result; */ - const int charsToWrite - = (int)(uri->scheme.afterLast - uri->scheme.first); - if (dest != NULL) { - if (written + charsToWrite <= maxChars) { - memcpy(dest + written, uri->scheme.first, - charsToWrite * sizeof(URI_CHAR)); - written += charsToWrite; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += charsToWrite; - } - /* [04/19] append ":" to result; */ - if (dest != NULL) { - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT(":"), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 1; - } - /* [05/19] endif; */ - } - /* [06/19] if defined(authority) then */ - if (URI_FUNC(IsHostSet)(uri)) { - /* [07/19] append "//" to result; */ - if (dest != NULL) { - if (written + 2 <= maxChars) { - memcpy(dest + written, _UT("//"), - 2 * sizeof(URI_CHAR)); - written += 2; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 2; - } - /* [08/19] append authority to result; */ - /* UserInfo */ - if (uri->userInfo.first != NULL) { - const int charsToWrite = (int)(uri->userInfo.afterLast - uri->userInfo.first); - if (dest != NULL) { - if (written + charsToWrite <= maxChars) { - memcpy(dest + written, uri->userInfo.first, - charsToWrite * sizeof(URI_CHAR)); - written += charsToWrite; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT("@"), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += charsToWrite + 1; - } - } - - /* Host */ - if (uri->hostData.ip4 != NULL) { - /* IPv4 */ - int i = 0; - for (; i < 4; i++) { - const unsigned char value = uri->hostData.ip4->data[i]; - const int charsToWrite = (value > 99) ? 3 : ((value > 9) ? 2 : 1); - if (dest != NULL) { - if (written + charsToWrite <= maxChars) { - URI_CHAR text[4]; - if (value > 99) { - text[0] = _UT('0') + (value / 100); - text[1] = _UT('0') + ((value % 100) / 10); - text[2] = _UT('0') + (value % 10); - } else if (value > 9) { - text[0] = _UT('0') + (value / 10); - text[1] = _UT('0') + (value % 10); - } else { - text[0] = _UT('0') + value; - } - text[charsToWrite] = _UT('\0'); - memcpy(dest + written, text, charsToWrite * sizeof(URI_CHAR)); - written += charsToWrite; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - if (i < 3) { - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT("."), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } - } else { - (*charsRequired) += charsToWrite + ((i == 3) ? 0 : 1); - } - } - } else if (uri->hostData.ip6 != NULL) { - /* IPv6 */ - int i = 0; - if (dest != NULL) { - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT("["), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 1; - } - - for (; i < 16; i++) { - const unsigned char value = uri->hostData.ip6->data[i]; - if (dest != NULL) { - if (written + 2 <= maxChars) { - URI_CHAR text[3]; - text[0] = URI_FUNC(HexToLetterEx)(value / 16, URI_FALSE); - text[1] = URI_FUNC(HexToLetterEx)(value % 16, URI_FALSE); - text[2] = _UT('\0'); - memcpy(dest + written, text, 2 * sizeof(URI_CHAR)); - written += 2; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 2; - } - if (((i & 1) == 1) && (i < 15)) { - if (dest != NULL) { - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT(":"), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 1; - } - } - } - - if (dest != NULL) { - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT("]"), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 1; - } - } else if (uri->hostData.ipFuture.first != NULL) { - /* IPvFuture */ - const int charsToWrite = (int)(uri->hostData.ipFuture.afterLast - - uri->hostData.ipFuture.first); - if (dest != NULL) { - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT("["), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - - if (written + charsToWrite <= maxChars) { - memcpy(dest + written, uri->hostData.ipFuture.first, charsToWrite * sizeof(URI_CHAR)); - written += charsToWrite; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT("]"), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 1 + charsToWrite + 1; - } - } else if (uri->hostText.first != NULL) { - /* Regname */ - const int charsToWrite = (int)(uri->hostText.afterLast - uri->hostText.first); - if (dest != NULL) { - if (written + charsToWrite <= maxChars) { - memcpy(dest + written, uri->hostText.first, - charsToWrite * sizeof(URI_CHAR)); - written += charsToWrite; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += charsToWrite; - } - } - - /* Port */ - if (uri->portText.first != NULL) { - const int charsToWrite = (int)(uri->portText.afterLast - uri->portText.first); - if (dest != NULL) { - /* Leading ':' */ - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT(":"), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - - /* Port number */ - if (written + charsToWrite <= maxChars) { - memcpy(dest + written, uri->portText.first, - charsToWrite * sizeof(URI_CHAR)); - written += charsToWrite; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 1 + charsToWrite; - } - } - /* [09/19] endif; */ - } - /* [10/19] append path to result; */ - /* Slash needed here? */ - if (uri->absolutePath || ((uri->pathHead != NULL) - && URI_FUNC(IsHostSet)(uri))) { - if (dest != NULL) { - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT("/"), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 1; - } - } - - if (uri->pathHead != NULL) { - URI_TYPE(PathSegment) * walker = uri->pathHead; - do { - const int charsToWrite = (int)(walker->text.afterLast - walker->text.first); - if (dest != NULL) { - if (written + charsToWrite <= maxChars) { - memcpy(dest + written, walker->text.first, - charsToWrite * sizeof(URI_CHAR)); - written += charsToWrite; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += charsToWrite; - } - - /* Not last segment -> append slash */ - if (walker->next != NULL) { - if (dest != NULL) { - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT("/"), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 1; - } - } - - walker = walker->next; - } while (walker != NULL); - } - /* [11/19] if defined(query) then */ - if (uri->query.first != NULL) { - /* [12/19] append "?" to result; */ - if (dest != NULL) { - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT("?"), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 1; - } - /* [13/19] append query to result; */ - { - const int charsToWrite - = (int)(uri->query.afterLast - uri->query.first); - if (dest != NULL) { - if (written + charsToWrite <= maxChars) { - memcpy(dest + written, uri->query.first, - charsToWrite * sizeof(URI_CHAR)); - written += charsToWrite; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += charsToWrite; - } - } - /* [14/19] endif; */ - } - /* [15/19] if defined(fragment) then */ - if (uri->fragment.first != NULL) { - /* [16/19] append "#" to result; */ - if (dest != NULL) { - if (written + 1 <= maxChars) { - memcpy(dest + written, _UT("#"), - 1 * sizeof(URI_CHAR)); - written += 1; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += 1; - } - /* [17/19] append fragment to result; */ - { - const int charsToWrite - = (int)(uri->fragment.afterLast - uri->fragment.first); - if (dest != NULL) { - if (written + charsToWrite <= maxChars) { - memcpy(dest + written, uri->fragment.first, - charsToWrite * sizeof(URI_CHAR)); - written += charsToWrite; - } else { - dest[0] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = 0; - } - return URI_ERROR_TOSTRING_TOO_LONG; - } - } else { - (*charsRequired) += charsToWrite; - } - } - /* [18/19] endif; */ - } - /* [19/19] return result; */ - if (dest != NULL) { - dest[written++] = _UT('\0'); - if (charsWritten != NULL) { - *charsWritten = written; - } - } - return URI_SUCCESS; +static URI_INLINE int URI_FUNC(ToStringEngine)(URI_CHAR * dest, const URI_TYPE(Uri) * uri, + int maxChars, int * charsWritten, + int * charsRequired) { + int written = 0; + if ((uri == NULL) || ((dest == NULL) && (charsRequired == NULL))) { + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_NULL; + } + + if (maxChars < 1) { + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + maxChars--; /* So we don't have to subtract 1 for '\0' all the time */ + + /* NOTE: The curly brackets here force deeper indent (and that's all) */ + { + { + { + /* clang-format off */ + /* [01/19] result = "" */ + /* clang-format on */ + if (dest != NULL) { + dest[0] = _UT('\0'); + } else { + (*charsRequired) = 0; + } + /* clang-format off */ + /* [02/19] if defined(scheme) then */ + /* clang-format on */ + if (uri->scheme.first != NULL) { + /* clang-format off */ + /* [03/19] append scheme to result; */ + /* clang-format on */ + const size_t charsToWrite = uri->scheme.afterLast - uri->scheme.first; + if (dest != NULL) { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - written) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + if (written + charsToWrite <= (size_t)maxChars) { + // Detect and avoid integer overflow + if (charsToWrite > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + memcpy(dest + written, uri->scheme.first, + charsToWrite * sizeof(URI_CHAR)); + written += charsToWrite; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - *charsRequired) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + (*charsRequired) += charsToWrite; + } + /* clang-format off */ + /* [04/19] append ":" to result; */ + /* clang-format on */ + if (dest != NULL) { + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT(":"), 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + (*charsRequired) += 1; + } + /* clang-format off */ + /* [05/19] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [06/19] if defined(authority) then */ + /* clang-format on */ + if (URI_FUNC(HasHost)(uri)) { + /* clang-format off */ + /* [07/19] append "//" to result; */ + /* clang-format on */ + if (dest != NULL) { + if (written + 2 <= maxChars) { + memcpy(dest + written, _UT("//"), 2 * sizeof(URI_CHAR)); + written += 2; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + (*charsRequired) += 2; + } + /* clang-format off */ + /* [08/19] append authority to result; */ + /* clang-format on */ + /* UserInfo */ + if (uri->userInfo.first != NULL) { + const size_t charsToWrite = + uri->userInfo.afterLast - uri->userInfo.first; + if (dest != NULL) { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - written) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + if (written + charsToWrite <= (size_t)maxChars) { + // Detect and avoid integer overflow + if (charsToWrite > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + memcpy(dest + written, uri->userInfo.first, + charsToWrite * sizeof(URI_CHAR)); + written += charsToWrite; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT("@"), 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + // Detect and avoid integer overflow + if ((charsToWrite > (size_t)INT_MAX - 1) + || (charsToWrite + 1 + > (size_t)INT_MAX - *charsRequired)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + (*charsRequired) += charsToWrite + 1; + } + } + + /* Host */ + if (uri->hostData.ip4 != NULL) { + /* IPv4 */ + int i = 0; + for (; i < 4; i++) { + const unsigned char value = uri->hostData.ip4->data[i]; + const int charsToWrite = + (value > 99) ? 3 : ((value > 9) ? 2 : 1); + if (dest != NULL) { + if (written + charsToWrite <= maxChars) { + URI_CHAR text[4]; + if (value > 99) { + text[0] = _UT('0') + (value / 100); + text[1] = _UT('0') + ((value % 100) / 10); + text[2] = _UT('0') + (value % 10); + } else if (value > 9) { + text[0] = _UT('0') + (value / 10); + text[1] = _UT('0') + (value % 10); + } else { + text[0] = _UT('0') + value; + } + text[charsToWrite] = _UT('\0'); + memcpy(dest + written, text, + charsToWrite * sizeof(URI_CHAR)); + written += charsToWrite; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + if (i < 3) { + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT("."), + 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } + } else { + (*charsRequired) += charsToWrite + ((i == 3) ? 0 : 1); + } + } + } else if (uri->hostData.ip6 != NULL) { + /* IPv6 */ + int i = 0; + if (dest != NULL) { + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT("["), 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + (*charsRequired) += 1; + } + + for (; i < 16; i++) { + const unsigned char value = uri->hostData.ip6->data[i]; + if (dest != NULL) { + if (written + 2 <= maxChars) { + URI_CHAR text[3]; + text[0] = + URI_FUNC(HexToLetterEx)(value / 16, URI_FALSE); + text[1] = + URI_FUNC(HexToLetterEx)(value % 16, URI_FALSE); + text[2] = _UT('\0'); + memcpy(dest + written, text, 2 * sizeof(URI_CHAR)); + written += 2; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + (*charsRequired) += 2; + } + if (((i & 1) == 1) && (i < 15)) { + if (dest != NULL) { + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT(":"), + 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + (*charsRequired) += 1; + } + } + } + + if (dest != NULL) { + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT("]"), 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + (*charsRequired) += 1; + } + } else if (uri->hostData.ipFuture.first != NULL) { + /* IPvFuture */ + const size_t charsToWrite = uri->hostData.ipFuture.afterLast + - uri->hostData.ipFuture.first; + if (dest != NULL) { + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT("["), 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - written) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + if (written + charsToWrite <= (size_t)maxChars) { + // Detect and avoid integer overflow + if (charsToWrite > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + memcpy(dest + written, uri->hostData.ipFuture.first, + charsToWrite * sizeof(URI_CHAR)); + written += charsToWrite; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT("]"), 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + // Detect and avoid integer overflow + if ((charsToWrite > (size_t)INT_MAX - 1 - 1) + || (1 + charsToWrite + 1 + > (size_t)INT_MAX - *charsRequired)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + (*charsRequired) += 1 + charsToWrite + 1; + } + } else if (uri->hostText.first != NULL) { + /* Regname */ + const size_t charsToWrite = + uri->hostText.afterLast - uri->hostText.first; + if (dest != NULL) { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - written) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + if (written + charsToWrite <= (size_t)maxChars) { + // Detect and avoid integer overflow + if (charsToWrite > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + memcpy(dest + written, uri->hostText.first, + charsToWrite * sizeof(URI_CHAR)); + written += charsToWrite; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - *charsRequired) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + (*charsRequired) += charsToWrite; + } + } + + /* Port */ + if (uri->portText.first != NULL) { + const size_t charsToWrite = + uri->portText.afterLast - uri->portText.first; + if (dest != NULL) { + /* Leading ':' */ + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT(":"), 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - written) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + /* Port number */ + if (written + charsToWrite <= (size_t)maxChars) { + // Detect and avoid integer overflow + if (charsToWrite > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + memcpy(dest + written, uri->portText.first, + charsToWrite * sizeof(URI_CHAR)); + written += charsToWrite; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + // Detect and avoid integer overflow + if ((charsToWrite > (size_t)INT_MAX - 1) + || (1 + charsToWrite + > (size_t)INT_MAX - *charsRequired)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + (*charsRequired) += 1 + charsToWrite; + } + } + /* clang-format off */ + /* [09/19] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [10/19] append path to result; */ + /* clang-format on */ + /* Slash needed here? */ + if (uri->absolutePath + || ((uri->pathHead != NULL) && URI_FUNC(HasHost)(uri))) { + if (dest != NULL) { + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT("/"), 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + (*charsRequired) += 1; + } + } + + if (uri->pathHead != NULL) { + URI_TYPE(PathSegment) * walker = uri->pathHead; + do { + const size_t charsToWrite = + walker->text.afterLast - walker->text.first; + if (dest != NULL) { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - written) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + if (written + charsToWrite <= (size_t)maxChars) { + // Detect and avoid integer overflow + if (charsToWrite > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + memcpy(dest + written, walker->text.first, + charsToWrite * sizeof(URI_CHAR)); + written += charsToWrite; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - *charsRequired) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + (*charsRequired) += charsToWrite; + } + + /* Not last segment -> append slash */ + if (walker->next != NULL) { + if (dest != NULL) { + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT("/"), + 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + (*charsRequired) += 1; + } + } + + walker = walker->next; + } while (walker != NULL); + } + /* clang-format off */ + /* [11/19] if defined(query) then */ + /* clang-format on */ + if (uri->query.first != NULL) { + /* clang-format off */ + /* [12/19] append "?" to result; */ + /* clang-format on */ + if (dest != NULL) { + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT("?"), 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + (*charsRequired) += 1; + } + /* clang-format off */ + /* [13/19] append query to result; */ + /* clang-format on */ + const size_t charsToWrite = uri->query.afterLast - uri->query.first; + if (dest != NULL) { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - written) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + if (written + charsToWrite <= (size_t)maxChars) { + // Detect and avoid integer overflow + if (charsToWrite > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + memcpy(dest + written, uri->query.first, + charsToWrite * sizeof(URI_CHAR)); + written += charsToWrite; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - *charsRequired) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + (*charsRequired) += charsToWrite; + } + /* clang-format off */ + /* [14/19] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [15/19] if defined(fragment) then */ + /* clang-format on */ + if (uri->fragment.first != NULL) { + /* clang-format off */ + /* [16/19] append "#" to result; */ + /* clang-format on */ + if (dest != NULL) { + if (written + 1 <= maxChars) { + memcpy(dest + written, _UT("#"), 1 * sizeof(URI_CHAR)); + written += 1; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + (*charsRequired) += 1; + } + /* clang-format off */ + /* [17/19] append fragment to result; */ + /* clang-format on */ + const size_t charsToWrite = + uri->fragment.afterLast - uri->fragment.first; + if (dest != NULL) { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - written) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + if (written + charsToWrite <= (size_t)maxChars) { + // Detect and avoid integer overflow + if (charsToWrite > SIZE_MAX / sizeof(URI_CHAR)) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + memcpy(dest + written, uri->fragment.first, + charsToWrite * sizeof(URI_CHAR)); + written += charsToWrite; + } else { + dest[0] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = 0; + } + return URI_ERROR_TOSTRING_TOO_LONG; + } + } else { + // Detect and avoid integer overflow + if (charsToWrite > (size_t)INT_MAX - *charsRequired) { + return URI_ERROR_TOSTRING_TOO_LONG; + } + + (*charsRequired) += charsToWrite; + } + /* clang-format off */ + /* [18/19] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [19/19] return result; */ + /* clang-format on */ + if (dest != NULL) { + dest[written++] = _UT('\0'); + if (charsWritten != NULL) { + *charsWritten = written; + } + } + return URI_SUCCESS; + } + } + } } - - #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriResolve.c b/cpp/src/arrow/vendored/uriparser/UriResolve.c index ee4cba818a5b..c951b7d1c05a 100644 --- a/cpp/src/arrow/vendored/uriparser/UriResolve.c +++ b/cpp/src/arrow/vendored/uriparser/UriResolve.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -41,289 +41,357 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriResolve.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriResolve.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriResolve.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriResolve.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -# include "UriCommon.h" -# include "UriMemory.h" -#endif - - +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriMemory.h" +# endif /* Appends a relative URI to an absolute. The last path segment of * the absolute URI is replaced. */ static URI_INLINE UriBool URI_FUNC(MergePath)(URI_TYPE(Uri) * absWork, - const URI_TYPE(Uri) * relAppend, UriMemoryManager * memory) { - URI_TYPE(PathSegment) * sourceWalker; - URI_TYPE(PathSegment) * destPrev; - if (relAppend->pathHead == NULL) { - return URI_TRUE; - } - - /* Replace last segment ("" if trailing slash) with first of append chain */ - if (absWork->pathHead == NULL) { - URI_TYPE(PathSegment) * const dup = memory->malloc(memory, sizeof(URI_TYPE(PathSegment))); - if (dup == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - dup->next = NULL; - absWork->pathHead = dup; - absWork->pathTail = dup; - } - absWork->pathTail->text.first = relAppend->pathHead->text.first; - absWork->pathTail->text.afterLast = relAppend->pathHead->text.afterLast; - - /* Append all the others */ - sourceWalker = relAppend->pathHead->next; - if (sourceWalker == NULL) { - return URI_TRUE; - } - destPrev = absWork->pathTail; - - for (;;) { - URI_TYPE(PathSegment) * const dup = memory->malloc(memory, sizeof(URI_TYPE(PathSegment))); - if (dup == NULL) { - destPrev->next = NULL; - absWork->pathTail = destPrev; - return URI_FALSE; /* Raises malloc error */ - } - dup->text = sourceWalker->text; - destPrev->next = dup; - - if (sourceWalker->next == NULL) { - absWork->pathTail = dup; - absWork->pathTail->next = NULL; - break; - } - destPrev = dup; - sourceWalker = sourceWalker->next; - } - - return URI_TRUE; + const URI_TYPE(Uri) * relAppend, + UriMemoryManager * memory) { + URI_TYPE(PathSegment) * sourceWalker; + URI_TYPE(PathSegment) * destPrev; + if (relAppend->pathHead == NULL) { + return URI_TRUE; + } + + /* Replace last segment ("" if trailing slash) with first of append chain */ + if (absWork->pathHead == NULL) { + URI_TYPE(PathSegment) * const dup = + memory->malloc(memory, sizeof(URI_TYPE(PathSegment))); + if (dup == NULL) { + return URI_FALSE; /* Raises malloc error */ + } + dup->next = NULL; + absWork->pathHead = dup; + absWork->pathTail = dup; + } + absWork->pathTail->text.first = relAppend->pathHead->text.first; + absWork->pathTail->text.afterLast = relAppend->pathHead->text.afterLast; + + /* Append all the others */ + sourceWalker = relAppend->pathHead->next; + if (sourceWalker == NULL) { + return URI_TRUE; + } + destPrev = absWork->pathTail; + + for (;;) { + URI_TYPE(PathSegment) * const dup = + memory->malloc(memory, sizeof(URI_TYPE(PathSegment))); + if (dup == NULL) { + destPrev->next = NULL; + absWork->pathTail = destPrev; + return URI_FALSE; /* Raises malloc error */ + } + dup->text = sourceWalker->text; + destPrev->next = dup; + + if (sourceWalker->next == NULL) { + absWork->pathTail = dup; + absWork->pathTail->next = NULL; + break; + } + destPrev = dup; + sourceWalker = sourceWalker->next; + } + + return URI_TRUE; } - static int URI_FUNC(ResolveAbsolutePathFlag)(URI_TYPE(Uri) * absWork, - UriMemoryManager * memory) { - if (absWork == NULL) { - return URI_ERROR_NULL; - } - - if (URI_FUNC(IsHostSet)(absWork) && absWork->absolutePath) { - /* Empty segment needed, instead? */ - if (absWork->pathHead == NULL) { - URI_TYPE(PathSegment) * const segment = memory->malloc(memory, sizeof(URI_TYPE(PathSegment))); - if (segment == NULL) { - return URI_ERROR_MALLOC; - } - segment->text.first = URI_FUNC(SafeToPointTo); - segment->text.afterLast = URI_FUNC(SafeToPointTo); - segment->next = NULL; - - absWork->pathHead = segment; - absWork->pathTail = segment; - } - - absWork->absolutePath = URI_FALSE; - } - - return URI_SUCCESS; + UriMemoryManager * memory) { + if (absWork == NULL) { + return URI_ERROR_NULL; + } + + if (URI_FUNC(HasHost)(absWork) && absWork->absolutePath) { + /* Empty segment needed, instead? */ + if (absWork->pathHead == NULL) { + URI_TYPE(PathSegment) * const segment = + memory->malloc(memory, sizeof(URI_TYPE(PathSegment))); + if (segment == NULL) { + return URI_ERROR_MALLOC; + } + segment->text.first = URI_FUNC(SafeToPointTo); + segment->text.afterLast = URI_FUNC(SafeToPointTo); + segment->next = NULL; + + absWork->pathHead = segment; + absWork->pathTail = segment; + } + + absWork->absolutePath = URI_FALSE; + } + + return URI_SUCCESS; } - static int URI_FUNC(AddBaseUriImpl)(URI_TYPE(Uri) * absDest, - const URI_TYPE(Uri) * relSource, - const URI_TYPE(Uri) * absBase, - UriResolutionOptions options, UriMemoryManager * memory) { - UriBool relSourceHasScheme; - - if (absDest == NULL) { - return URI_ERROR_NULL; - } - URI_FUNC(ResetUri)(absDest); - - if ((relSource == NULL) || (absBase == NULL)) { - return URI_ERROR_NULL; - } - - /* absBase absolute? */ - if (absBase->scheme.first == NULL) { - return URI_ERROR_ADDBASE_REL_BASE; - } - - /* [00/32] -- A non-strict parser may ignore a scheme in the reference */ - /* [00/32] -- if it is identical to the base URI's scheme. */ - /* [00/32] if ((not strict) and (R.scheme == Base.scheme)) then */ - relSourceHasScheme = (relSource->scheme.first != NULL) ? URI_TRUE : URI_FALSE; - if ((options & URI_RESOLVE_IDENTICAL_SCHEME_COMPAT) - && (absBase->scheme.first != NULL) - && (relSource->scheme.first != NULL) - && (0 == URI_FUNC(CompareRange)(&(absBase->scheme), &(relSource->scheme)))) { - /* [00/32] undefine(R.scheme); */ - relSourceHasScheme = URI_FALSE; - /* [00/32] endif; */ - } - - /* [01/32] if defined(R.scheme) then */ - if (relSourceHasScheme) { - /* [02/32] T.scheme = R.scheme; */ - absDest->scheme = relSource->scheme; - /* [03/32] T.authority = R.authority; */ - if (!URI_FUNC(CopyAuthority)(absDest, relSource, memory)) { - return URI_ERROR_MALLOC; - } - /* [04/32] T.path = remove_dot_segments(R.path); */ - if (!URI_FUNC(CopyPath)(absDest, relSource, memory)) { - return URI_ERROR_MALLOC; - } - if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest, memory)) { - return URI_ERROR_MALLOC; - } - /* [05/32] T.query = R.query; */ - absDest->query = relSource->query; - /* [06/32] else */ - } else { - /* [07/32] if defined(R.authority) then */ - if (URI_FUNC(IsHostSet)(relSource)) { - /* [08/32] T.authority = R.authority; */ - if (!URI_FUNC(CopyAuthority)(absDest, relSource, memory)) { - return URI_ERROR_MALLOC; - } - /* [09/32] T.path = remove_dot_segments(R.path); */ - if (!URI_FUNC(CopyPath)(absDest, relSource, memory)) { - return URI_ERROR_MALLOC; - } - if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest, memory)) { - return URI_ERROR_MALLOC; - } - /* [10/32] T.query = R.query; */ - absDest->query = relSource->query; - /* [11/32] else */ - } else { - /* [28/32] T.authority = Base.authority; */ - if (!URI_FUNC(CopyAuthority)(absDest, absBase, memory)) { - return URI_ERROR_MALLOC; - } - /* [12/32] if (R.path == "") then */ - if (relSource->pathHead == NULL && !relSource->absolutePath) { - /* [13/32] T.path = Base.path; */ - if (!URI_FUNC(CopyPath)(absDest, absBase, memory)) { - return URI_ERROR_MALLOC; - } - /* [14/32] if defined(R.query) then */ - if (relSource->query.first != NULL) { - /* [15/32] T.query = R.query; */ - absDest->query = relSource->query; - /* [16/32] else */ - } else { - /* [17/32] T.query = Base.query; */ - absDest->query = absBase->query; - /* [18/32] endif; */ - } - /* [19/32] else */ - } else { - /* [20/32] if (R.path starts-with "/") then */ - if (relSource->absolutePath) { - int res; - /* [21/32] T.path = remove_dot_segments(R.path); */ - if (!URI_FUNC(CopyPath)(absDest, relSource, memory)) { - return URI_ERROR_MALLOC; - } - res = URI_FUNC(ResolveAbsolutePathFlag)(absDest, memory); - if (res != URI_SUCCESS) { - return res; - } - if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest, memory)) { - return URI_ERROR_MALLOC; - } - /* [22/32] else */ - } else { - /* [23/32] T.path = merge(Base.path, R.path); */ - if (!URI_FUNC(CopyPath)(absDest, absBase, memory)) { - return URI_ERROR_MALLOC; - } - if (!URI_FUNC(MergePath)(absDest, relSource, memory)) { - return URI_ERROR_MALLOC; - } - /* [24/32] T.path = remove_dot_segments(T.path); */ - if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest, memory)) { - return URI_ERROR_MALLOC; - } - - if (!URI_FUNC(FixAmbiguity)(absDest, memory)) { - return URI_ERROR_MALLOC; - } - /* [25/32] endif; */ - } - /* [26/32] T.query = R.query; */ - absDest->query = relSource->query; - /* [27/32] endif; */ - } - URI_FUNC(FixEmptyTrailSegment)(absDest, memory); - /* [29/32] endif; */ - } - /* [30/32] T.scheme = Base.scheme; */ - absDest->scheme = absBase->scheme; - /* [31/32] endif; */ - } - /* [32/32] T.fragment = R.fragment; */ - absDest->fragment = relSource->fragment; - - return URI_SUCCESS; - + const URI_TYPE(Uri) * relSource, + const URI_TYPE(Uri) * absBase, + UriResolutionOptions options, + UriMemoryManager * memory) { + UriBool relSourceHasScheme; + + if (absDest == NULL) { + return URI_ERROR_NULL; + } + URI_FUNC(ResetUri)(absDest); + + if ((relSource == NULL) || (absBase == NULL)) { + return URI_ERROR_NULL; + } + + /* absBase absolute? */ + if (absBase->scheme.first == NULL) { + return URI_ERROR_ADDBASE_REL_BASE; + } + + /* NOTE: The curly brackets here force deeper indent (and that's all) */ + { + { + { + /* clang-format off */ + /* [00/32] -- A non-strict parser may ignore a scheme in the reference */ + /* [00/32] -- if it is identical to the base URI's scheme. */ + /* [00/32] if ((not strict) and (R.scheme == Base.scheme)) then */ + /* clang-format on */ + relSourceHasScheme = + (relSource->scheme.first != NULL) ? URI_TRUE : URI_FALSE; + if ((options & URI_RESOLVE_IDENTICAL_SCHEME_COMPAT) + && (absBase->scheme.first != NULL) + && (relSource->scheme.first != NULL) + && (URI_FUNC(RangeEquals)(&(absBase->scheme), + &(relSource->scheme)))) { + /* clang-format off */ + /* [00/32] undefine(R.scheme); */ + /* clang-format on */ + relSourceHasScheme = URI_FALSE; + /* clang-format off */ + /* [00/32] endif; */ + /* clang-format on */ + } + + /* clang-format off */ + /* [01/32] if defined(R.scheme) then */ + /* clang-format on */ + if (relSourceHasScheme) { + /* clang-format off */ + /* [02/32] T.scheme = R.scheme; */ + /* clang-format on */ + absDest->scheme = relSource->scheme; + /* clang-format off */ + /* [03/32] T.authority = R.authority; */ + /* clang-format on */ + if (!URI_FUNC(CopyAuthority)(absDest, relSource, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [04/32] T.path = remove_dot_segments(R.path); */ + /* clang-format on */ + if (!URI_FUNC(CopyPath)(absDest, relSource, memory)) { + return URI_ERROR_MALLOC; + } + if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [05/32] T.query = R.query; */ + /* clang-format on */ + absDest->query = relSource->query; + /* clang-format off */ + /* [06/32] else */ + /* clang-format on */ + } else { + /* clang-format off */ + /* [07/32] if defined(R.authority) then */ + /* clang-format on */ + if (URI_FUNC(HasHost)(relSource)) { + /* clang-format off */ + /* [08/32] T.authority = R.authority; */ + /* clang-format on */ + if (!URI_FUNC(CopyAuthority)(absDest, relSource, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [09/32] T.path = remove_dot_segments(R.path); */ + /* clang-format on */ + if (!URI_FUNC(CopyPath)(absDest, relSource, memory)) { + return URI_ERROR_MALLOC; + } + if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [10/32] T.query = R.query; */ + /* clang-format on */ + absDest->query = relSource->query; + /* clang-format off */ + /* [11/32] else */ + /* clang-format on */ + } else { + /* clang-format off */ + /* [28/32] T.authority = Base.authority; */ + /* clang-format on */ + if (!URI_FUNC(CopyAuthority)(absDest, absBase, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [12/32] if (R.path == "") then */ + /* clang-format on */ + if (relSource->pathHead == NULL && !relSource->absolutePath) { + /* clang-format off */ + /* [13/32] T.path = Base.path; */ + /* clang-format on */ + if (!URI_FUNC(CopyPath)(absDest, absBase, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [14/32] if defined(R.query) then */ + /* clang-format on */ + if (relSource->query.first != NULL) { + /* clang-format off */ + /* [15/32] T.query = R.query; */ + /* clang-format on */ + absDest->query = relSource->query; + /* clang-format off */ + /* [16/32] else */ + /* clang-format on */ + } else { + /* clang-format off */ + /* [17/32] T.query = Base.query; */ + /* clang-format on */ + absDest->query = absBase->query; + /* clang-format off */ + /* [18/32] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [19/32] else */ + /* clang-format on */ + } else { + /* clang-format off */ + /* [20/32] if (R.path starts-with "/") then */ + /* clang-format on */ + if (relSource->absolutePath) { + int res; + /* clang-format off */ + /* [21/32] T.path = remove_dot_segments(R.path); */ + /* clang-format on */ + if (!URI_FUNC(CopyPath)(absDest, relSource, memory)) { + return URI_ERROR_MALLOC; + } + res = URI_FUNC(ResolveAbsolutePathFlag)(absDest, memory); + if (res != URI_SUCCESS) { + return res; + } + if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest, + memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [22/32] else */ + /* clang-format on */ + } else { + /* clang-format off */ + /* [23/32] T.path = merge(Base.path, R.path); */ + /* clang-format on */ + if (!URI_FUNC(CopyPath)(absDest, absBase, memory)) { + return URI_ERROR_MALLOC; + } + if (!URI_FUNC(MergePath)(absDest, relSource, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [24/32] T.path = remove_dot_segments(T.path); */ + /* clang-format on */ + if (!URI_FUNC(RemoveDotSegmentsAbsolute)(absDest, + memory)) { + return URI_ERROR_MALLOC; + } + + if (!URI_FUNC(FixAmbiguity)(absDest, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [25/32] endif; */ + } + /* clang-format off */ + /* [26/32] T.query = R.query; */ + /* clang-format on */ + absDest->query = relSource->query; + /* clang-format off */ + /* [27/32] endif; */ + /* clang-format on */ + } + URI_FUNC(FixEmptyTrailSegment)(absDest, memory); + /* clang-format off */ + /* [29/32] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [30/32] T.scheme = Base.scheme; */ + /* clang-format on */ + absDest->scheme = absBase->scheme; + /* clang-format off */ + /* [31/32] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [32/32] T.fragment = R.fragment; */ + /* clang-format on */ + absDest->fragment = relSource->fragment; + } + } + } + return URI_SUCCESS; } - - -int URI_FUNC(AddBaseUri)(URI_TYPE(Uri) * absDest, - const URI_TYPE(Uri) * relSource, const URI_TYPE(Uri) * absBase) { - const UriResolutionOptions options = URI_RESOLVE_STRICTLY; - return URI_FUNC(AddBaseUriEx)(absDest, relSource, absBase, options); +int URI_FUNC(AddBaseUri)(URI_TYPE(Uri) * absDest, const URI_TYPE(Uri) * relSource, + const URI_TYPE(Uri) * absBase) { + const UriResolutionOptions options = URI_RESOLVE_STRICTLY; + return URI_FUNC(AddBaseUriEx)(absDest, relSource, absBase, options); } - - -int URI_FUNC(AddBaseUriEx)(URI_TYPE(Uri) * absDest, - const URI_TYPE(Uri) * relSource, const URI_TYPE(Uri) * absBase, - UriResolutionOptions options) { - return URI_FUNC(AddBaseUriExMm)(absDest, relSource, absBase, options, NULL); +int URI_FUNC(AddBaseUriEx)(URI_TYPE(Uri) * absDest, const URI_TYPE(Uri) * relSource, + const URI_TYPE(Uri) * absBase, UriResolutionOptions options) { + return URI_FUNC(AddBaseUriExMm)(absDest, relSource, absBase, options, NULL); } +int URI_FUNC(AddBaseUriExMm)(URI_TYPE(Uri) * absDest, const URI_TYPE(Uri) * relSource, + const URI_TYPE(Uri) * absBase, UriResolutionOptions options, + UriMemoryManager * memory) { + int res; + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ -int URI_FUNC(AddBaseUriExMm)(URI_TYPE(Uri) * absDest, - const URI_TYPE(Uri) * relSource, const URI_TYPE(Uri) * absBase, - UriResolutionOptions options, UriMemoryManager * memory) { - int res; - - URI_CHECK_MEMORY_MANAGER(memory); /* may return */ - - res = URI_FUNC(AddBaseUriImpl)(absDest, relSource, absBase, options, memory); - if ((res != URI_SUCCESS) && (absDest != NULL)) { - URI_FUNC(FreeUriMembersMm)(absDest, memory); - } - return res; + res = URI_FUNC(AddBaseUriImpl)(absDest, relSource, absBase, options, memory); + if ((res != URI_SUCCESS) && (absDest != NULL)) { + URI_FUNC(FreeUriMembersMm)(absDest, memory); + } + return res; } - - #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetFragment.c b/cpp/src/arrow/vendored/uriparser/UriSetFragment.c new file mode 100644 index 000000000000..3e310146dc2b --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetFragment.c @@ -0,0 +1,174 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetFragment.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetFragment.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriMemory.h" +# include "UriSets.h" +# endif + +# include + +UriBool URI_FUNC(IsWellFormedFragment)(const URI_CHAR * first, + const URI_CHAR * afterLast) { + if ((first == NULL) || (afterLast == NULL)) { + return URI_FALSE; + } + + /* The related part of the grammar in RFC 3986 reads: + * + * fragment = *( pchar / "/" / "?" ) + * pchar = unreserved / pct-encoded / sub-delims / ":" / "@" + */ + while (first < afterLast) { + switch (first[0]) { + case URI_SET_PCHAR_WITHOUT_PERCENT(_UT): + break; + + /* pct-encoded */ + case _UT('%'): + if (afterLast - first < 3) { + return URI_FALSE; + } + switch (first[1]) { + case URI_SET_HEXDIG(_UT): + break; + default: + return URI_FALSE; + } + switch (first[2]) { + case URI_SET_HEXDIG(_UT): + break; + default: + return URI_FALSE; + } + first += 2; + break; + + case _UT('/'): + case _UT('?'): + break; + + default: + return URI_FALSE; + } + + first++; + } + return URI_TRUE; +} + +int URI_FUNC(SetFragmentMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + /* Input validation (before making any changes) */ + if ((uri == NULL) || ((first == NULL) != (afterLast == NULL))) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + if ((first != NULL) + && (URI_FUNC(IsWellFormedFragment)(first, afterLast) == URI_FALSE)) { + return URI_ERROR_SYNTAX; + } + + /* Clear old value */ + if ((uri->owner == URI_TRUE) && (uri->fragment.first != uri->fragment.afterLast)) { + memory->free(memory, (URI_CHAR *)uri->fragment.first); + } + uri->fragment.first = NULL; + uri->fragment.afterLast = NULL; + + /* Already done? */ + if (first == NULL) { + return URI_SUCCESS; + } + + assert(first != NULL); + + /* Ensure owned */ + if (uri->owner == URI_FALSE) { + const int res = URI_FUNC(MakeOwnerMm)(uri, memory); + if (res != URI_SUCCESS) { + return res; + } + } + + assert(uri->owner == URI_TRUE); + + /* Apply new value */ + URI_TYPE(TextRange) sourceRange; + sourceRange.first = first; + sourceRange.afterLast = afterLast; + + if (URI_FUNC(CopyRangeAsNeeded)(&uri->fragment, &sourceRange, memory) == URI_FALSE) { + return URI_ERROR_MALLOC; + } + + return URI_SUCCESS; +} + +int URI_FUNC(SetFragment)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetFragmentMm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetHostAuto.c b/cpp/src/arrow/vendored/uriparser/UriSetHostAuto.c new file mode 100644 index 000000000000..d2d4cfaa5b58 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetHostAuto.c @@ -0,0 +1,124 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetHostAuto.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetHostAuto.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriSetHostBase.h" +# include "UriSetHostCommon.h" +# include "UriMemory.h" +# endif + +# include + +int URI_FUNC(SetHostAutoMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + if ((uri == NULL) || ((first == NULL) != (afterLast == NULL))) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + if ((first == NULL) || (first >= afterLast)) { + return URI_FUNC(SetHostRegNameMm)(uri, first, afterLast, memory); + } + + /* Auto-detect type and then apply */ + UriHostType hostType; + + /* IPv6 or IPvFuture? */ + if (first[0] == _UT('[')) { + if ((afterLast - first < 2) || (afterLast[-1] != _UT(']'))) { + return URI_ERROR_SYNTAX; + } + + /* Drop the bracket wrap (for InternalSetHostMm call below) */ + first++; + afterLast--; + + if (first >= afterLast) { + return URI_ERROR_SYNTAX; + } + + switch (first[0]) { + case _UT('v'): + case _UT('V'): + hostType = URI_HOST_TYPE_IPFUTURE; + break; + default: + hostType = URI_HOST_TYPE_IP6; + break; + } + /* IPv4? */ + } else if (URI_FUNC(IsWellFormedHostIp4)(first, afterLast)) { + hostType = URI_HOST_TYPE_IP4; + } else { + /* RegName! */ + hostType = URI_HOST_TYPE_REGNAME; + } + + return URI_FUNC(InternalSetHostMm)(uri, hostType, first, afterLast, memory); +} + +int URI_FUNC(SetHostAuto)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetHostAutoMm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetHostBase.h b/cpp/src/arrow/vendored/uriparser/UriSetHostBase.h new file mode 100644 index 000000000000..081db36f219d --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetHostBase.h @@ -0,0 +1,49 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef URI_SET_HOST_BASE_H +# define URI_SET_HOST_BASE_H 1 + +typedef enum UriHostTypeEnum { + URI_HOST_TYPE_IP4, + URI_HOST_TYPE_IP6, + URI_HOST_TYPE_IPFUTURE, + URI_HOST_TYPE_REGNAME +} UriHostType; + +#endif /* URI_SET_HOST_BASE_H */ diff --git a/cpp/src/arrow/vendored/uriparser/UriSetHostCommon.c b/cpp/src/arrow/vendored/uriparser/UriSetHostCommon.c new file mode 100644 index 000000000000..beeb95068a33 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetHostCommon.c @@ -0,0 +1,248 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/** + * @file UriSetHostCommon.c + * Holds code used by multiple SetHost* functions. + * NOTE: This source file includes itself twice. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetHostCommon.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetHostCommon.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriIp4.h" +# include "UriCommon.h" +# include "UriMemory.h" +# include "UriSetHostBase.h" +# include "UriSetHostCommon.h" +# endif + +# include + +int URI_FUNC(InternalSetHostMm)(URI_TYPE(Uri) * uri, UriHostType hostType, + const URI_CHAR * first, const URI_CHAR * afterLast, + UriMemoryManager * memory) { + /* Superficial input validation (before making any changes) */ + if ((uri == NULL) || ((first == NULL) != (afterLast == NULL))) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + /* The RFC 3986 grammar reads: + * authority = [ userinfo "@" ] host [ ":" port ] + * So no user info or port without a host. */ + if (first == NULL) { + if (uri->userInfo.first != NULL) { + return URI_ERROR_SETHOST_USERINFO_SET; + } else if (uri->portText.first != NULL) { + return URI_ERROR_SETHOST_PORT_SET; + } + } + + /* Syntax-check the new value */ + if (first != NULL) { + switch (hostType) { + case URI_HOST_TYPE_IP4: + if (URI_FUNC(IsWellFormedHostIp4)(first, afterLast) == URI_FALSE) { + return URI_ERROR_SYNTAX; + } + break; + case URI_HOST_TYPE_IP6: { + const int res = URI_FUNC(IsWellFormedHostIp6Mm)(first, afterLast, memory); + assert((res == URI_SUCCESS) || (res == URI_ERROR_SYNTAX) + || (res == URI_ERROR_MALLOC)); + if (res != URI_SUCCESS) { + return res; + } + } break; + case URI_HOST_TYPE_IPFUTURE: { + const int res = + URI_FUNC(IsWellFormedHostIpFutureMm)(first, afterLast, memory); + assert((res == URI_SUCCESS) || (res == URI_ERROR_SYNTAX) + || (res == URI_ERROR_MALLOC)); + if (res != URI_SUCCESS) { + return res; + } + } break; + case URI_HOST_TYPE_REGNAME: + if (URI_FUNC(IsWellFormedHostRegName)(first, afterLast) == URI_FALSE) { + return URI_ERROR_SYNTAX; + } + break; + default: + assert(0 && "Unsupported URI host type"); + } + } + + /* Clear old value */ + const UriBool hadHostBefore = URI_FUNC(HasHost)(uri); + if (uri->hostData.ipFuture.first != NULL) { + /* NOTE: .hostData.ipFuture holds the very same range pointers + * as .hostText; we must not free memory twice. */ + uri->hostText.first = NULL; + uri->hostText.afterLast = NULL; + + if ((uri->owner == URI_TRUE) + && (uri->hostData.ipFuture.first != uri->hostData.ipFuture.afterLast)) { + memory->free(memory, (URI_CHAR *)uri->hostData.ipFuture.first); + } + uri->hostData.ipFuture.first = NULL; + uri->hostData.ipFuture.afterLast = NULL; + } else if (uri->hostText.first != NULL) { + if ((uri->owner == URI_TRUE) + && (uri->hostText.first != uri->hostText.afterLast)) { + memory->free(memory, (URI_CHAR *)uri->hostText.first); + } + uri->hostText.first = NULL; + uri->hostText.afterLast = NULL; + } + + if (uri->hostData.ip4 != NULL) { + memory->free(memory, uri->hostData.ip4); + uri->hostData.ip4 = NULL; + } else if (uri->hostData.ip6 != NULL) { + memory->free(memory, uri->hostData.ip6); + uri->hostData.ip6 = NULL; + } + + /* Already done setting? */ + if (first == NULL) { + /* Yes, but disambiguate as needed */ + if (hadHostBefore == URI_TRUE) { + if (uri->pathHead != NULL) { + uri->absolutePath = URI_TRUE; + } + + const UriBool success = + URI_FUNC(EnsureThatPathIsNotMistakenForHost)(uri, memory); + return (success == URI_TRUE) ? URI_SUCCESS : URI_ERROR_MALLOC; + } + + return URI_SUCCESS; + } + + assert(first != NULL); + + /* Ensure owned */ + if (uri->owner == URI_FALSE) { + const int res = URI_FUNC(MakeOwnerMm)(uri, memory); + if (res != URI_SUCCESS) { + return res; + } + } + + assert(uri->owner == URI_TRUE); + + /* Apply new value; NOTE that .hostText is set for all four host types */ + URI_TYPE(TextRange) sourceRange; + sourceRange.first = first; + sourceRange.afterLast = afterLast; + + if (URI_FUNC(CopyRangeAsNeeded)(&uri->hostText, &sourceRange, memory) == URI_FALSE) { + return URI_ERROR_MALLOC; + } + + uri->absolutePath = URI_FALSE; /* always URI_FALSE for URIs with host */ + + /* Fill .hostData as needed */ + switch (hostType) { + case URI_HOST_TYPE_IP4: { + uri->hostData.ip4 = memory->malloc(memory, sizeof(UriIp4)); + if (uri->hostData.ip4 == NULL) { + return URI_ERROR_MALLOC; + } + + const int res = + URI_FUNC(ParseIpFourAddress)(uri->hostData.ip4->data, first, afterLast); +# if defined(NDEBUG) + (void)res; /* i.e. mark as unused */ +# else + assert(res == URI_SUCCESS); /* because checked for well-formedness earlier */ +# endif + } break; + case URI_HOST_TYPE_IP6: { + uri->hostData.ip6 = memory->malloc(memory, sizeof(UriIp6)); + if (uri->hostData.ip6 == NULL) { + return URI_ERROR_MALLOC; + } + + const int res = + URI_FUNC(ParseIpSixAddressMm)(uri->hostData.ip6, first, afterLast, memory); + assert((res == URI_SUCCESS) + || (res == URI_ERROR_MALLOC)); /* because checked for + well-formedness earlier */ + if (res != URI_SUCCESS) { + return res; + } + } break; + case URI_HOST_TYPE_IPFUTURE: + uri->hostData.ipFuture.first = uri->hostText.first; + uri->hostData.ipFuture.afterLast = uri->hostText.afterLast; + break; + case URI_HOST_TYPE_REGNAME: + break; + default: + assert(0 && "Unsupported URI host type"); + } + + return URI_SUCCESS; +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetHostCommon.h b/cpp/src/arrow/vendored/uriparser/UriSetHostCommon.h new file mode 100644 index 000000000000..5b2a348cc274 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetHostCommon.h @@ -0,0 +1,74 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#if (defined(URI_PASS_ANSI) && !defined(URI_SET_HOST_COMMON_H_ANSI)) \ + || (defined(URI_PASS_UNICODE) && !defined(URI_SET_HOST_COMMON_H_UNICODE)) \ + || (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* What encodings are enabled? */ +# include "UriDefsConfig.h" +# if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetHostCommon.h" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetHostCommon.h" +# undef URI_PASS_UNICODE +# endif +/* Only one pass for each encoding */ +# elif (defined(URI_PASS_ANSI) && !defined(URI_SET_HOST_COMMON_H_ANSI) \ + && defined(URI_ENABLE_ANSI)) \ + || (defined(URI_PASS_UNICODE) && !defined(URI_SET_HOST_COMMON_H_UNICODE) \ + && defined(URI_ENABLE_UNICODE)) +# ifdef URI_PASS_ANSI +# define URI_SET_HOST_COMMON_H_ANSI 1 +# include "UriDefsAnsi.h" +# else +# define URI_SET_HOST_COMMON_H_UNICODE 1 +# include "UriDefsUnicode.h" +# endif + +int URI_FUNC(InternalSetHostMm)(URI_TYPE(Uri) * uri, UriHostType hostType, + const URI_CHAR * first, const URI_CHAR * afterLast, + UriMemoryManager * memory); + +# endif +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetHostIp4.c b/cpp/src/arrow/vendored/uriparser/UriSetHostIp4.c new file mode 100644 index 000000000000..949e6a595887 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetHostIp4.c @@ -0,0 +1,91 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetHostIp4.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetHostIp4.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriIp4.h" +# include "UriMemory.h" +# include "UriSetHostBase.h" +# include "UriSetHostCommon.h" +# endif + +UriBool URI_FUNC(IsWellFormedHostIp4)(const URI_CHAR * first, + const URI_CHAR * afterLast) { + if ((first == NULL) || (afterLast == NULL)) { + return URI_FALSE; + } + + unsigned char octetOutput[4]; + return (URI_FUNC(ParseIpFourAddress)(octetOutput, first, afterLast) == URI_SUCCESS) + ? URI_TRUE + : URI_FALSE; +} + +int URI_FUNC(SetHostIp4Mm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + return URI_FUNC(InternalSetHostMm)(uri, URI_HOST_TYPE_IP4, first, afterLast, memory); +} + +int URI_FUNC(SetHostIp4)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetHostIp4Mm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetHostIp6.c b/cpp/src/arrow/vendored/uriparser/UriSetHostIp6.c new file mode 100644 index 000000000000..c1f4cd4a3e08 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetHostIp6.c @@ -0,0 +1,158 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetHostIp6.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetHostIp6.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriMemory.h" +# include "UriSetHostBase.h" +# include "UriSetHostCommon.h" +# endif + +# include +# include /* for memcpy */ + +# define URI_MAX_IP6_LEN \ + (8 * 4 + 7 * 1) /* i.e. 8 full quads plus 7 colon separators \ + */ + +int URI_FUNC(ParseIpSixAddressMm)(UriIp6 * output, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + /* NOTE: output is allowed to be NULL */ + if ((first == NULL) || (afterLast == NULL)) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + /* Are we dealing with potential IPvFuture input? */ + if (first < afterLast) { + switch (first[0]) { + case _UT('v'): + case _UT('V'): + return URI_ERROR_SYNTAX; + default: + break; + } + } + + /* Are we dealing with IPv6 input? */ + /* Assemble "//[..]" input wrap for upcoming parse as a URI + * NOTE: If the input contains closing "]" on its own, the resulting + * string will not be valid URI syntax, and hence there is + * no risk of false positives from "bracket injection". */ + URI_CHAR candidate[3 + URI_MAX_IP6_LEN + 1 + 1] = _UT("//["); + const size_t inputLenChars = (afterLast - first); + + /* Detect overflow */ + if (inputLenChars > URI_MAX_IP6_LEN) { + return URI_ERROR_SYNTAX; + } + + memcpy(candidate + 3, first, inputLenChars * sizeof(URI_CHAR)); + memcpy(candidate + 3 + inputLenChars, _UT("]"), + 2 * sizeof(URI_CHAR)); /* includes zero terminator */ + + /* Parse as an RFC 3986 URI */ + const size_t candidateLenChars = 3 + inputLenChars + 1; + URI_TYPE(Uri) uri; + const int res = URI_FUNC(ParseSingleUriExMm)( + &uri, candidate, candidate + candidateLenChars, NULL, memory); + + assert((res == URI_SUCCESS) || (res == URI_ERROR_SYNTAX) + || (res == URI_ERROR_MALLOC)); + + if (res == URI_SUCCESS) { + assert(uri.hostData.ip6 != NULL); + + if (output != NULL) { + memcpy(output->data, uri.hostData.ip6->data, sizeof(output->data)); + } + + URI_FUNC(FreeUriMembersMm)(&uri, memory); + } + + return res; +} + +int URI_FUNC(ParseIpSixAddress)(UriIp6 * output, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(ParseIpSixAddressMm)(output, first, afterLast, NULL); +} + +int URI_FUNC(IsWellFormedHostIp6Mm)(const URI_CHAR * first, const URI_CHAR * afterLast, + UriMemoryManager * memory) { + return URI_FUNC(ParseIpSixAddressMm)(NULL, first, afterLast, memory); +} + +int URI_FUNC(IsWellFormedHostIp6)(const URI_CHAR * first, const URI_CHAR * afterLast) { + return URI_FUNC(IsWellFormedHostIp6Mm)(first, afterLast, NULL); +} + +int URI_FUNC(SetHostIp6Mm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + return URI_FUNC(InternalSetHostMm)(uri, URI_HOST_TYPE_IP6, first, afterLast, memory); +} + +int URI_FUNC(SetHostIp6)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetHostIp6Mm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetHostIpFuture.c b/cpp/src/arrow/vendored/uriparser/UriSetHostIpFuture.c new file mode 100644 index 000000000000..8393e05e5761 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetHostIpFuture.c @@ -0,0 +1,157 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetHostIpFuture.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetHostIpFuture.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriMemory.h" +# include "UriSetHostBase.h" +# include "UriSetHostCommon.h" +# endif + +# include +# include /* for memcpy */ + +int URI_FUNC(IsWellFormedHostIpFutureMm)(const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + if ((first == NULL) || (afterLast == NULL)) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + /* Are we dealing with potential IPv6 input? */ + if (first < afterLast) { + switch (first[0]) { + case _UT('v'): + case _UT('V'): + break; + default: + return URI_ERROR_SYNTAX; + } + } + + /* Are we dealing with IPvFuture input? */ + /* Assemble "//[..]" input wrap for upcoming parse as a URI + * NOTE: If the input contains closing "]" on its own, the resulting + * string will not be valid URI syntax, and hence there is + * no risk of false positives from "bracket injection". */ + const size_t inputLenChars = (afterLast - first); + const size_t MAX_SIZE_T = (size_t)-1; + + /* Detect overflow */ + if (MAX_SIZE_T - inputLenChars < 3 + 1 + 1) { + return URI_ERROR_MALLOC; + } + + const size_t candidateLenChars = 3 + inputLenChars + 1; + + /* Detect overflow */ + if (MAX_SIZE_T / sizeof(URI_CHAR) < candidateLenChars + 1) { + return URI_ERROR_MALLOC; + } + + URI_CHAR * const candidate = + memory->malloc(memory, (candidateLenChars + 1) * sizeof(URI_CHAR)); + + if (candidate == NULL) { + return URI_ERROR_MALLOC; + } + + memcpy(candidate, _UT("//["), 3 * sizeof(URI_CHAR)); + memcpy(candidate + 3, first, inputLenChars * sizeof(URI_CHAR)); + memcpy(candidate + 3 + inputLenChars, _UT("]"), + 2 * sizeof(URI_CHAR)); /* includes zero terminator */ + + /* Parse as an RFC 3986 URI */ + URI_TYPE(Uri) uri; + const int res = URI_FUNC(ParseSingleUriExMm)( + &uri, candidate, candidate + candidateLenChars, NULL, memory); + + assert((res == URI_SUCCESS) || (res == URI_ERROR_SYNTAX) + || (res == URI_ERROR_MALLOC)); + + if (res == URI_SUCCESS) { + assert(uri.hostData.ipFuture.first != NULL); + URI_FUNC(FreeUriMembersMm)(&uri, memory); + } + + memory->free(memory, candidate); + + return res; +} + +int URI_FUNC(IsWellFormedHostIpFuture)(const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(IsWellFormedHostIpFutureMm)(first, afterLast, NULL); +} + +int URI_FUNC(SetHostIpFutureMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + return URI_FUNC(InternalSetHostMm)(uri, URI_HOST_TYPE_IPFUTURE, first, afterLast, + memory); +} + +int URI_FUNC(SetHostIpFuture)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetHostIpFutureMm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetHostRegName.c b/cpp/src/arrow/vendored/uriparser/UriSetHostRegName.c new file mode 100644 index 000000000000..d25b4aa1026b --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetHostRegName.c @@ -0,0 +1,124 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetHostRegName.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetHostRegName.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriMemory.h" +# include "UriSetHostBase.h" +# include "UriSetHostCommon.h" +# include "UriSets.h" +# endif + +UriBool URI_FUNC(IsWellFormedHostRegName)(const URI_CHAR * first, + const URI_CHAR * afterLast) { + if ((first == NULL) || (afterLast == NULL)) { + return URI_FALSE; + } + + /* reg-name = *( unreserved / pct-encoded / sub-delims ) */ + while (first < afterLast) { + switch (first[0]) { + case URI_SET_UNRESERVED(_UT): + break; + + /* pct-encoded */ + case _UT('%'): + if (afterLast - first < 3) { + return URI_FALSE; + } + switch (first[1]) { + case URI_SET_HEXDIG(_UT): + break; + default: + return URI_FALSE; + } + switch (first[2]) { + case URI_SET_HEXDIG(_UT): + break; + default: + return URI_FALSE; + } + first += 2; + break; + + case URI_SET_SUB_DELIMS(_UT): + break; + + default: + return URI_FALSE; + } + + first++; + } + return URI_TRUE; +} + +int URI_FUNC(SetHostRegNameMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + return URI_FUNC(InternalSetHostMm)(uri, URI_HOST_TYPE_REGNAME, first, afterLast, + memory); +} + +int URI_FUNC(SetHostRegName)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetHostRegNameMm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetPath.c b/cpp/src/arrow/vendored/uriparser/UriSetPath.c new file mode 100644 index 000000000000..e8ffe357244c --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetPath.c @@ -0,0 +1,343 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetPath.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetPath.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriMemory.h" +# include "UriSets.h" +# endif + +# include + +UriBool URI_FUNC(IsWellFormedPath)(const URI_CHAR * first, const URI_CHAR * afterLast, + UriBool hasHost) { + if ((first == NULL) || (afterLast == NULL)) { + return URI_FALSE; + } + + if ((hasHost == URI_TRUE) && ((first >= afterLast) || (first[0] != _UT('/')))) { + return URI_FALSE; + } + + /* The related part of the grammar in RFC 3986 (section 3.3) reads: + * + * path = path-abempty ; begins with "/" or is empty + * / path-absolute ; begins with "/" but not "//" + * / path-noscheme ; begins with a non-colon segment + * / path-rootless ; begins with a segment + * / path-empty ; zero characters + * + * path-abempty = *( "/" segment ) + * path-absolute = "/" [ segment-nz *( "/" segment ) ] + * path-noscheme = segment-nz-nc *( "/" segment ) + * path-rootless = segment-nz *( "/" segment ) + * path-empty = 0 + * + * segment = *pchar + * segment-nz = 1*pchar + * segment-nz-nc = 1*( unreserved / pct-encoded / sub-delims / "@" ) + * ; non-zero-length segment without any colon ":" + * + * pchar = unreserved / pct-encoded / sub-delims / ":" / "@" + * + * The check below simplifies this to .. + * + * path = *( unreserved / pct-encoded / sub-delims / ":" / "@" / "/" ) + * + * .. and leaves the rest to pre-return removal of ambiguity + * from cases like "path1:/path2" and "//path1/path2" inside SetPath. + */ + while (first < afterLast) { + switch (first[0]) { + case URI_SET_PCHAR_WITHOUT_PERCENT(_UT): + break; + + /* pct-encoded */ + case _UT('%'): + if (afterLast - first < 3) { + return URI_FALSE; + } + switch (first[1]) { + case URI_SET_HEXDIG(_UT): + break; + default: + return URI_FALSE; + } + switch (first[2]) { + case URI_SET_HEXDIG(_UT): + break; + default: + return URI_FALSE; + } + first += 2; + break; + + case _UT('/'): + break; + + default: + return URI_FALSE; + } + + first++; + } + return URI_TRUE; +} + +static void URI_FUNC(DropEmptyFirstPathSegment)(URI_TYPE(Uri) * uri, + UriMemoryManager * memory) { + assert(uri != NULL); + assert(memory != NULL); + assert(uri->pathHead != NULL); + assert(uri->pathHead->text.first == uri->pathHead->text.afterLast); + + URI_TYPE(PathSegment) * const originalHead = uri->pathHead; + + uri->pathHead = uri->pathHead->next; + + originalHead->text.first = NULL; + originalHead->text.afterLast = NULL; + memory->free(memory, originalHead); +} + +/* URIs without a host encode a leading slash in the path as .absolutePath == URI_TRUE. + * This function checks for a leading empty path segment (that would have the "visual + * effect" of a leading slash during stringification) and transforms it into .absolutePath + * == URI_TRUE instead, if present. */ +static void URI_FUNC(TransformEmptyLeadPathSegments)(URI_TYPE(Uri) * uri, + UriMemoryManager * memory) { + assert(uri != NULL); + assert(memory != NULL); + + if ((URI_FUNC(HasHost)(uri) == URI_TRUE) || (uri->pathHead == NULL) + || (uri->pathHead->text.first != uri->pathHead->text.afterLast)) { + return; /* i.e. nothing to do */ + } + + assert(uri->absolutePath == URI_FALSE); + + URI_FUNC(DropEmptyFirstPathSegment)(uri, memory); + + uri->absolutePath = URI_TRUE; +} + +static int URI_FUNC(InternalSetPath)(URI_TYPE(Uri) * destUri, const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + assert(destUri != NULL); + assert(first != NULL); + assert(afterLast != NULL); + assert(memory != NULL); + assert(destUri->pathHead == NULL); /* set by SetPathMm right before */ + assert(destUri->pathTail == NULL); /* set by SetPathMm right before */ + assert(destUri->absolutePath == URI_FALSE); /* set by SetPathMm right before */ + + /* Skip the leading slash from target URIs with a host (so that we can + * transfer the path 1:1 further down) */ + if (URI_FUNC(HasHost)(destUri) == URI_TRUE) { + /* NOTE: This is because SetPathMm called IsWellFormedPath earlier: */ + assert((afterLast - first >= 1) && (first[0] == _UT('/'))); + first++; + } else if (first == afterLast) { + /* This avoids (1) all the expensive but unnecessary work below + * and also (2) mis-encoding as single empty path segment + * that would need (detection and) repair further down otherwise */ + return URI_SUCCESS; + } + + /* Assemble "///.." input wrap for upcoming parse as a URI */ + const size_t inputLenChars = (afterLast - first); + const size_t MAX_SIZE_T = (size_t)-1; + + /* Detect overflow */ + if (MAX_SIZE_T - inputLenChars < 3 + 1) { + return URI_ERROR_MALLOC; + } + + const size_t candidateLenChars = 3 + inputLenChars; + + /* Detect overflow */ + if (MAX_SIZE_T / sizeof(URI_CHAR) < candidateLenChars + 1) { + return URI_ERROR_MALLOC; + } + + URI_CHAR * const candidate = + memory->malloc(memory, (candidateLenChars + 1) * sizeof(URI_CHAR)); + + if (candidate == NULL) { + return URI_ERROR_MALLOC; + } + + memcpy(candidate, _UT("///"), 3 * sizeof(URI_CHAR)); + memcpy(candidate + 3, first, inputLenChars * sizeof(URI_CHAR)); + candidate[3 + inputLenChars] = _UT('\0'); + + /* Parse as an RFC 3986 URI */ + URI_TYPE(Uri) tempUri; + int res = URI_FUNC(ParseSingleUriExMm)(&tempUri, candidate, + candidate + candidateLenChars, NULL, memory); + assert((res == URI_SUCCESS) || (res == URI_ERROR_SYNTAX) + || (res == URI_ERROR_MALLOC)); + if (res != URI_SUCCESS) { + memory->free(memory, candidate); + return res; + } + + /* Nothing but path and host is supposed to be set by the parse, in + * particular not: */ + assert(tempUri.query.first == NULL); + assert(tempUri.fragment.first == NULL); + + /* Ensure that the strings in the path segments are all owned by + * `tempUri` because we want to (1) rip out and keep the full path + * list further down and (2) be able to free the parsed string + * (`candidate`) also. */ + res = URI_FUNC(MakeOwnerMm)(&tempUri, memory); + assert((res == URI_SUCCESS) || (res == URI_ERROR_MALLOC)); + if (res != URI_SUCCESS) { + URI_FUNC(FreeUriMembersMm)(&tempUri, memory); + memory->free(memory, candidate); + return res; + } + assert(tempUri.owner == URI_TRUE); + + /* Move path to destination URI */ + assert(tempUri.absolutePath == URI_FALSE); /* always URI_FALSE for URIs with host */ + destUri->pathHead = tempUri.pathHead; + destUri->pathTail = tempUri.pathTail; + destUri->absolutePath = URI_FALSE; + + tempUri.pathHead = NULL; + tempUri.pathTail = NULL; + + /* Free the rest of the temp URI */ + URI_FUNC(FreeUriMembersMm)(&tempUri, memory); + memory->free(memory, candidate); + + /* Restore use of .absolutePath as needed */ + URI_FUNC(TransformEmptyLeadPathSegments)(destUri, memory); + + /* Disambiguate as needed */ + UriBool success = URI_FUNC(FixPathNoScheme)(destUri, memory); + if (success == URI_FALSE) { + return URI_ERROR_MALLOC; + } + + success = URI_FUNC(EnsureThatPathIsNotMistakenForHost)(destUri, memory); + if (success == URI_FALSE) { + return URI_ERROR_MALLOC; + } + + return URI_SUCCESS; +} + +int URI_FUNC(SetPathMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + /* Input validation (before making any changes) */ + if ((uri == NULL) || ((first == NULL) != (afterLast == NULL))) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + if ((first != NULL) + && (URI_FUNC(IsWellFormedPath)(first, afterLast, URI_FUNC(HasHost)(uri)) + == URI_FALSE)) { + return URI_ERROR_SYNTAX; + } + + /* Clear old value */ + int res = URI_FUNC(FreeUriPath)(uri, memory); + if (res != URI_SUCCESS) { + return res; + } + uri->absolutePath = URI_FALSE; + + /* Already done? */ + if (first == NULL) { + return URI_SUCCESS; + } + + assert(first != NULL); + + /* Ensure owned */ + if (uri->owner == URI_FALSE) { + res = URI_FUNC(MakeOwnerMm)(uri, memory); + if (res != URI_SUCCESS) { + return res; + } + } + + assert(uri->owner == URI_TRUE); + + /* Apply new value */ + res = URI_FUNC(InternalSetPath)(uri, first, afterLast, memory); + assert((res == URI_SUCCESS) || (res == URI_ERROR_SYNTAX) + || (res == URI_ERROR_MALLOC)); + return res; +} + +int URI_FUNC(SetPath)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetPathMm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetPort.c b/cpp/src/arrow/vendored/uriparser/UriSetPort.c new file mode 100644 index 000000000000..7018396785ae --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetPort.c @@ -0,0 +1,149 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetPort.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetPort.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriMemory.h" +# include "UriSets.h" +# endif + +# include + +UriBool URI_FUNC(IsWellFormedPort)(const URI_CHAR * first, const URI_CHAR * afterLast) { + if ((first == NULL) || (afterLast == NULL)) { + return URI_FALSE; + } + + /* NOTE: Grammar reads "port = *DIGIT" which includes the empty string. */ + while (first < afterLast) { + switch (first[0]) { + case URI_SET_DIGIT(_UT): + break; + default: + return URI_FALSE; + } + first++; + } + return URI_TRUE; +} + +int URI_FUNC(SetPortTextMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + /* Input validation (before making any changes) */ + if ((uri == NULL) || ((first == NULL) != (afterLast == NULL))) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + /* The RFC 3986 grammar reads: + * authority = [ userinfo "@" ] host [ ":" port ] + * So no port without a host. */ + if ((first != NULL) && (URI_FUNC(HasHost)(uri) == URI_FALSE)) { + return URI_ERROR_SETPORT_HOST_NOT_SET; + } + + if ((first != NULL) && (URI_FUNC(IsWellFormedPort)(first, afterLast) == URI_FALSE)) { + return URI_ERROR_SYNTAX; + } + + /* Clear old value */ + if ((uri->owner == URI_TRUE) && (uri->portText.first != uri->portText.afterLast)) { + memory->free(memory, (URI_CHAR *)uri->portText.first); + } + uri->portText.first = NULL; + uri->portText.afterLast = NULL; + + /* Already done? */ + if (first == NULL) { + return URI_SUCCESS; + } + + assert(first != NULL); + + /* Ensure owned */ + if (uri->owner == URI_FALSE) { + const int res = URI_FUNC(MakeOwnerMm)(uri, memory); + if (res != URI_SUCCESS) { + return res; + } + } + + assert(uri->owner == URI_TRUE); + + /* Apply new value */ + URI_TYPE(TextRange) sourceRange; + sourceRange.first = first; + sourceRange.afterLast = afterLast; + + if (URI_FUNC(CopyRangeAsNeeded)(&uri->portText, &sourceRange, memory) == URI_FALSE) { + return URI_ERROR_MALLOC; + } + + return URI_SUCCESS; +} + +int URI_FUNC(SetPortText)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetPortTextMm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetQuery.c b/cpp/src/arrow/vendored/uriparser/UriSetQuery.c new file mode 100644 index 000000000000..9daa5a2af3bd --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetQuery.c @@ -0,0 +1,172 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetQuery.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetQuery.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriMemory.h" +# include "UriSets.h" +# endif + +# include + +UriBool URI_FUNC(IsWellFormedQuery)(const URI_CHAR * first, const URI_CHAR * afterLast) { + if ((first == NULL) || (afterLast == NULL)) { + return URI_FALSE; + } + + /* The related part of the grammar in RFC 3986 reads: + * + * query = *( pchar / "/" / "?" ) + * pchar = unreserved / pct-encoded / sub-delims / ":" / "@" + */ + while (first < afterLast) { + switch (first[0]) { + case URI_SET_PCHAR_WITHOUT_PERCENT(_UT): + break; + + /* pct-encoded */ + case _UT('%'): + if (afterLast - first < 3) { + return URI_FALSE; + } + switch (first[1]) { + case URI_SET_HEXDIG(_UT): + break; + default: + return URI_FALSE; + } + switch (first[2]) { + case URI_SET_HEXDIG(_UT): + break; + default: + return URI_FALSE; + } + first += 2; + break; + + case _UT('/'): + case _UT('?'): + break; + + default: + return URI_FALSE; + } + + first++; + } + return URI_TRUE; +} + +int URI_FUNC(SetQueryMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + /* Input validation (before making any changes) */ + if ((uri == NULL) || ((first == NULL) != (afterLast == NULL))) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + if ((first != NULL) && (URI_FUNC(IsWellFormedQuery)(first, afterLast) == URI_FALSE)) { + return URI_ERROR_SYNTAX; + } + + /* Clear old value */ + if ((uri->owner == URI_TRUE) && (uri->query.first != uri->query.afterLast)) { + memory->free(memory, (URI_CHAR *)uri->query.first); + } + uri->query.first = NULL; + uri->query.afterLast = NULL; + + /* Already done? */ + if (first == NULL) { + return URI_SUCCESS; + } + + assert(first != NULL); + + /* Ensure owned */ + if (uri->owner == URI_FALSE) { + const int res = URI_FUNC(MakeOwnerMm)(uri, memory); + if (res != URI_SUCCESS) { + return res; + } + } + + assert(uri->owner == URI_TRUE); + + /* Apply new value */ + URI_TYPE(TextRange) sourceRange; + sourceRange.first = first; + sourceRange.afterLast = afterLast; + + if (URI_FUNC(CopyRangeAsNeeded)(&uri->query, &sourceRange, memory) == URI_FALSE) { + return URI_ERROR_MALLOC; + } + + return URI_SUCCESS; +} + +int URI_FUNC(SetQuery)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetQueryMm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetScheme.c b/cpp/src/arrow/vendored/uriparser/UriSetScheme.c new file mode 100644 index 000000000000..c40d6edc2156 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetScheme.c @@ -0,0 +1,168 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetScheme.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetScheme.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriMemory.h" +# include "UriSets.h" +# endif + +# include + +UriBool URI_FUNC(IsWellFormedScheme)(const URI_CHAR * first, const URI_CHAR * afterLast) { + if ((first == NULL) || (afterLast == NULL)) { + return URI_FALSE; + } + + /* The related part of the grammar in RFC 3986 reads: + * + * scheme = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." ) + */ + if (first >= afterLast) { + return URI_FALSE; + } + + switch (first[0]) { + case URI_SET_ALPHA(_UT): + break; + + default: + return URI_FALSE; + } + + first++; + + while (first < afterLast) { + switch (first[0]) { + case URI_SET_ALPHA(_UT): + case URI_SET_DIGIT(_UT): + case _UT('+'): + case _UT('-'): + case _UT('.'): + break; + + default: + return URI_FALSE; + } + + first++; + } + return URI_TRUE; +} + +int URI_FUNC(SetSchemeMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + /* Input validation (before making any changes) */ + if ((uri == NULL) || ((first == NULL) != (afterLast == NULL))) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + if ((first != NULL) + && (URI_FUNC(IsWellFormedScheme)(first, afterLast) == URI_FALSE)) { + return URI_ERROR_SYNTAX; + } + + /* Clear old value */ + if ((uri->owner == URI_TRUE) && (uri->scheme.first != uri->scheme.afterLast)) { + memory->free(memory, (URI_CHAR *)uri->scheme.first); + } + uri->scheme.first = NULL; + uri->scheme.afterLast = NULL; + + /* Already done setting? */ + if (first == NULL) { + /* Yes, but disambiguate as needed */ + const UriBool success = URI_FUNC(FixPathNoScheme)(uri, memory); + return (success == URI_TRUE) ? URI_SUCCESS : URI_ERROR_MALLOC; + } + + assert(first != NULL); + + /* Ensure owned */ + if (uri->owner == URI_FALSE) { + const int res = URI_FUNC(MakeOwnerMm)(uri, memory); + if (res != URI_SUCCESS) { + return res; + } + } + + assert(uri->owner == URI_TRUE); + + /* Apply new value */ + URI_TYPE(TextRange) sourceRange; + sourceRange.first = first; + sourceRange.afterLast = afterLast; + + if (URI_FUNC(CopyRangeAsNeeded)(&uri->scheme, &sourceRange, memory) == URI_FALSE) { + return URI_ERROR_MALLOC; + } + + return URI_SUCCESS; +} + +int URI_FUNC(SetScheme)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetSchemeMm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSetUserInfo.c b/cpp/src/arrow/vendored/uriparser/UriSetUserInfo.c new file mode 100644 index 000000000000..9a4930d986c6 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSetUserInfo.c @@ -0,0 +1,180 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriSetUserInfo.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriSetUserInfo.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriMemory.h" +# include "UriSets.h" +# endif + +# include + +UriBool URI_FUNC(IsWellFormedUserInfo)(const URI_CHAR * first, + const URI_CHAR * afterLast) { + if ((first == NULL) || (afterLast == NULL)) { + return URI_FALSE; + } + + /* userinfo = *( unreserved / pct-encoded / sub-delims / ":" ) */ + while (first < afterLast) { + switch (first[0]) { + case URI_SET_UNRESERVED(_UT): + break; + + /* pct-encoded */ + case _UT('%'): + if (afterLast - first < 3) { + return URI_FALSE; + } + switch (first[1]) { + case URI_SET_HEXDIG(_UT): + break; + default: + return URI_FALSE; + } + switch (first[2]) { + case URI_SET_HEXDIG(_UT): + break; + default: + return URI_FALSE; + } + first += 2; + break; + + case URI_SET_SUB_DELIMS(_UT): + break; + + /* ":" */ + case _UT(':'): + break; + + default: + return URI_FALSE; + } + + first++; + } + return URI_TRUE; +} + +int URI_FUNC(SetUserInfoMm)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast, UriMemoryManager * memory) { + /* Input validation (before making any changes) */ + if ((uri == NULL) || ((first == NULL) != (afterLast == NULL))) { + return URI_ERROR_NULL; + } + + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ + + /* The RFC 3986 grammar reads: + * authority = [ userinfo "@" ] host [ ":" port ] + * So no user info without a host. */ + if ((first != NULL) && (URI_FUNC(HasHost)(uri) == URI_FALSE)) { + return URI_ERROR_SETUSERINFO_HOST_NOT_SET; + } + + if ((first != NULL) + && (URI_FUNC(IsWellFormedUserInfo)(first, afterLast) == URI_FALSE)) { + return URI_ERROR_SYNTAX; + } + + /* Clear old value */ + if ((uri->owner == URI_TRUE) && (uri->userInfo.first != uri->userInfo.afterLast)) { + memory->free(memory, (URI_CHAR *)uri->userInfo.first); + } + uri->userInfo.first = NULL; + uri->userInfo.afterLast = NULL; + + /* Already done? */ + if (first == NULL) { + return URI_SUCCESS; + } + + assert(first != NULL); + + /* Ensure owned */ + if (uri->owner == URI_FALSE) { + const int res = URI_FUNC(MakeOwnerMm)(uri, memory); + if (res != URI_SUCCESS) { + return res; + } + } + + assert(uri->owner == URI_TRUE); + + /* Apply new value */ + URI_TYPE(TextRange) sourceRange; + sourceRange.first = first; + sourceRange.afterLast = afterLast; + + if (URI_FUNC(CopyRangeAsNeeded)(&uri->userInfo, &sourceRange, memory) == URI_FALSE) { + return URI_ERROR_MALLOC; + } + + return URI_SUCCESS; +} + +int URI_FUNC(SetUserInfo)(URI_TYPE(Uri) * uri, const URI_CHAR * first, + const URI_CHAR * afterLast) { + return URI_FUNC(SetUserInfoMm)(uri, first, afterLast, NULL); +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/UriSets.h b/cpp/src/arrow/vendored/uriparser/UriSets.h new file mode 100644 index 000000000000..a6a2c46a14d7 --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriSets.h @@ -0,0 +1,174 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/** + * @file UriSets.h + * Holds character set definitions. + */ + +// NOTE: We cannot use a regular include-once guard here because the +// file must support being included twice, e.g. from file UriParse.c. +#if !defined(URI_SET_DIGIT) + +// clang-format off +# define URI_SET_DIGIT(ut) \ + ut('0'): \ + case ut('1'): \ + /* clang-format on */ \ + case ut('2'): \ + case ut('3'): \ + case ut('4'): \ + case ut('5'): \ + case ut('6'): \ + case ut('7'): \ + case ut('8'): \ + case ut('9') + +// clang-format off +# define URI_SET_HEX_LETTER_LOWER(ut) \ + ut('a'): \ + case ut('b'): \ + /* clang-format on */ \ + case ut('c'): \ + case ut('d'): \ + case ut('e'): \ + case ut('f') + +// clang-format off +# define URI_SET_HEX_LETTER_UPPER(ut) \ + ut('A'): \ + case ut('B'): \ + /* clang-format on */ \ + case ut('C'): \ + case ut('D'): \ + case ut('E'): \ + case ut('F') + +// clang-format off +# define URI_SET_HEXDIG(ut) \ + URI_SET_DIGIT(ut): \ + case URI_SET_HEX_LETTER_LOWER(ut): \ + /* clang-format on */ \ + case URI_SET_HEX_LETTER_UPPER(ut) + +// clang-format off +# define URI_SET_ALPHA(ut) \ + URI_SET_HEX_LETTER_UPPER(ut): \ + case URI_SET_HEX_LETTER_LOWER(ut): \ + /* clang-format on */ \ + case ut('g'): \ + case ut('G'): \ + case ut('h'): \ + case ut('H'): \ + case ut('i'): \ + case ut('I'): \ + case ut('j'): \ + case ut('J'): \ + case ut('k'): \ + case ut('K'): \ + case ut('l'): \ + case ut('L'): \ + case ut('m'): \ + case ut('M'): \ + case ut('n'): \ + case ut('N'): \ + case ut('o'): \ + case ut('O'): \ + case ut('p'): \ + case ut('P'): \ + case ut('q'): \ + case ut('Q'): \ + case ut('r'): \ + case ut('R'): \ + case ut('s'): \ + case ut('S'): \ + case ut('t'): \ + case ut('T'): \ + case ut('u'): \ + case ut('U'): \ + case ut('v'): \ + case ut('V'): \ + case ut('w'): \ + case ut('W'): \ + case ut('x'): \ + case ut('X'): \ + case ut('y'): \ + case ut('Y'): \ + case ut('z'): \ + case ut('Z') + +// clang-format off +# define URI_SET_SUB_DELIMS(ut) \ + ut('!'): \ + case ut('$'): \ + /* clang-format on */ \ + case ut('&'): \ + case ut('\''): \ + case ut('('): \ + case ut(')'): \ + case ut('*'): \ + case ut('+'): \ + case ut(','): \ + case ut(';'): \ + case ut('=') + +// clang-format off +# define URI_SET_UNRESERVED(ut) \ + URI_SET_ALPHA(ut): \ + case URI_SET_DIGIT(ut): \ + /* clang-format on */ \ + case ut('-'): \ + case ut('.'): \ + case ut('_'): \ + case ut('~') + +// clang-format off +# define URI_SET_PCHAR_WITHOUT_PERCENT(ut) \ + URI_SET_UNRESERVED(ut): \ + case URI_SET_SUB_DELIMS(ut): \ + /* clang-format on */ \ + case ut(':'): \ + case ut('@') + +// clang-format off +# define URI_SET_PCHAR(ut) \ + URI_SET_PCHAR_WITHOUT_PERCENT(ut): \ + case ut('%') +/* clang-format on */ + +#endif // ! defined(URI_SET_DIGIT) diff --git a/cpp/src/arrow/vendored/uriparser/UriShorten.c b/cpp/src/arrow/vendored/uriparser/UriShorten.c index f00b05f59457..eb34ab88618b 100644 --- a/cpp/src/arrow/vendored/uriparser/UriShorten.c +++ b/cpp/src/arrow/vendored/uriparser/UriShorten.c @@ -5,32 +5,32 @@ * Copyright (C) 2007, Sebastian Pipping * All rights reserved. * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions * are met: * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED @@ -41,284 +41,385 @@ #include "UriDefsConfig.h" #if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) /* Include SELF twice */ -# ifdef URI_ENABLE_ANSI -# define URI_PASS_ANSI 1 -# include "UriShorten.c" -# undef URI_PASS_ANSI -# endif -# ifdef URI_ENABLE_UNICODE -# define URI_PASS_UNICODE 1 -# include "UriShorten.c" -# undef URI_PASS_UNICODE -# endif +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriShorten.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriShorten.c" +# undef URI_PASS_UNICODE +# endif #else -# ifdef URI_PASS_ANSI -# include "UriDefsAnsi.h" -# else -# include "UriDefsUnicode.h" -# include -# endif - - - -#ifndef URI_DOXYGEN -# include "Uri.h" -# include "UriCommon.h" -# include "UriMemory.h" -#endif - - +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# include "UriCommon.h" +# include "UriMemory.h" +# endif static URI_INLINE UriBool URI_FUNC(AppendSegment)(URI_TYPE(Uri) * uri, - const URI_CHAR * first, const URI_CHAR * afterLast, - UriMemoryManager * memory) { - /* Create segment */ - URI_TYPE(PathSegment) * segment = memory->malloc(memory, 1 * sizeof(URI_TYPE(PathSegment))); - if (segment == NULL) { - return URI_FALSE; /* Raises malloc error */ - } - segment->next = NULL; - segment->text.first = first; - segment->text.afterLast = afterLast; - - /* Put into chain */ - if (uri->pathTail == NULL) { - uri->pathHead = segment; - } else { - uri->pathTail->next = segment; - } - uri->pathTail = segment; - - return URI_TRUE; + const URI_CHAR * first, + const URI_CHAR * afterLast, + UriMemoryManager * memory) { + /* Create segment */ + URI_TYPE(PathSegment) * segment = + memory->malloc(memory, 1 * sizeof(URI_TYPE(PathSegment))); + if (segment == NULL) { + return URI_FALSE; /* Raises malloc error */ + } + segment->next = NULL; + segment->text.first = first; + segment->text.afterLast = afterLast; + + /* Put into chain */ + if (uri->pathTail == NULL) { + uri->pathHead = segment; + } else { + uri->pathTail->next = segment; + } + uri->pathTail = segment; + + return URI_TRUE; } - - static URI_INLINE UriBool URI_FUNC(EqualsAuthority)(const URI_TYPE(Uri) * first, - const URI_TYPE(Uri) * second) { - /* IPv4 */ - if (first->hostData.ip4 != NULL) { - return ((second->hostData.ip4 != NULL) - && !memcmp(first->hostData.ip4->data, - second->hostData.ip4->data, 4)) ? URI_TRUE : URI_FALSE; - } - - /* IPv6 */ - if (first->hostData.ip6 != NULL) { - return ((second->hostData.ip6 != NULL) - && !memcmp(first->hostData.ip6->data, - second->hostData.ip6->data, 16)) ? URI_TRUE : URI_FALSE; - } - - /* IPvFuture */ - if (first->hostData.ipFuture.first != NULL) { - return ((second->hostData.ipFuture.first != NULL) - && !URI_FUNC(CompareRange)(&first->hostData.ipFuture, - &second->hostData.ipFuture)) ? URI_TRUE : URI_FALSE; - } - - return !URI_FUNC(CompareRange)(&first->hostText, &second->hostText) - ? URI_TRUE : URI_FALSE; + const URI_TYPE(Uri) * second) { + /* IPv4 */ + if (first->hostData.ip4 != NULL) { + return ((second->hostData.ip4 != NULL) + && !memcmp(first->hostData.ip4->data, second->hostData.ip4->data, 4)) + ? URI_TRUE + : URI_FALSE; + } + + /* IPv6 */ + if (first->hostData.ip6 != NULL) { + return ((second->hostData.ip6 != NULL) + && !memcmp(first->hostData.ip6->data, second->hostData.ip6->data, 16)) + ? URI_TRUE + : URI_FALSE; + } + + /* IPvFuture */ + if (first->hostData.ipFuture.first != NULL) { + return ((second->hostData.ipFuture.first != NULL) + && URI_FUNC(RangeEquals)(&first->hostData.ipFuture, + &second->hostData.ipFuture)) + ? URI_TRUE + : URI_FALSE; + } + + return URI_FUNC(RangeEquals)(&first->hostText, &second->hostText) ? URI_TRUE + : URI_FALSE; } - - static int URI_FUNC(RemoveBaseUriImpl)(URI_TYPE(Uri) * dest, - const URI_TYPE(Uri) * absSource, - const URI_TYPE(Uri) * absBase, - UriBool domainRootMode, UriMemoryManager * memory) { - if (dest == NULL) { - return URI_ERROR_NULL; - } - URI_FUNC(ResetUri)(dest); - - if ((absSource == NULL) || (absBase == NULL)) { - return URI_ERROR_NULL; - } - - /* absBase absolute? */ - if (absBase->scheme.first == NULL) { - return URI_ERROR_REMOVEBASE_REL_BASE; - } - - /* absSource absolute? */ - if (absSource->scheme.first == NULL) { - return URI_ERROR_REMOVEBASE_REL_SOURCE; - } - - /* [01/50] if (A.scheme != Base.scheme) then */ - if (URI_FUNC(CompareRange)(&absSource->scheme, &absBase->scheme)) { - /* [02/50] T.scheme = A.scheme; */ - dest->scheme = absSource->scheme; - /* [03/50] T.authority = A.authority; */ - if (!URI_FUNC(CopyAuthority)(dest, absSource, memory)) { - return URI_ERROR_MALLOC; - } - /* [04/50] T.path = A.path; */ - if (!URI_FUNC(CopyPath)(dest, absSource, memory)) { - return URI_ERROR_MALLOC; - } - /* [05/50] else */ - } else { - /* [06/50] undef(T.scheme); */ - /* NOOP */ - /* [07/50] if (A.authority != Base.authority) then */ - if (!URI_FUNC(EqualsAuthority)(absSource, absBase)) { - /* [08/50] T.authority = A.authority; */ - if (!URI_FUNC(CopyAuthority)(dest, absSource, memory)) { - return URI_ERROR_MALLOC; - } - /* [09/50] T.path = A.path; */ - if (!URI_FUNC(CopyPath)(dest, absSource, memory)) { - return URI_ERROR_MALLOC; - } - /* [10/50] else */ - } else { - /* [11/50] if domainRootMode then */ - if (domainRootMode == URI_TRUE) { - /* [12/50] undef(T.authority); */ - /* NOOP */ - /* [13/50] if (first(A.path) == "") then */ - /* GROUPED */ - /* [14/50] T.path = "/." + A.path; */ - /* GROUPED */ - /* [15/50] else */ - /* GROUPED */ - /* [16/50] T.path = A.path; */ - /* GROUPED */ - /* [17/50] endif; */ - if (!URI_FUNC(CopyPath)(dest, absSource, memory)) { - return URI_ERROR_MALLOC; - } - dest->absolutePath = URI_TRUE; - - if (!URI_FUNC(FixAmbiguity)(dest, memory)) { - return URI_ERROR_MALLOC; - } - /* [18/50] else */ - } else { - const URI_TYPE(PathSegment) * sourceSeg = absSource->pathHead; - const URI_TYPE(PathSegment) * baseSeg = absBase->pathHead; - /* [19/50] bool pathNaked = true; */ - UriBool pathNaked = URI_TRUE; - /* [20/50] undef(last(Base.path)); */ - /* NOOP */ - /* [21/50] T.path = ""; */ - dest->absolutePath = URI_FALSE; - /* [22/50] while (first(A.path) == first(Base.path)) do */ - while ((sourceSeg != NULL) && (baseSeg != NULL) - && !URI_FUNC(CompareRange)(&sourceSeg->text, &baseSeg->text) - && !((sourceSeg->text.first == sourceSeg->text.afterLast) - && ((sourceSeg->next == NULL) != (baseSeg->next == NULL)))) { - /* [23/50] A.path++; */ - sourceSeg = sourceSeg->next; - /* [24/50] Base.path++; */ - baseSeg = baseSeg->next; - /* [25/50] endwhile; */ - } - /* [26/50] while defined(first(Base.path)) do */ - while ((baseSeg != NULL) && (baseSeg->next != NULL)) { - /* [27/50] Base.path++; */ - baseSeg = baseSeg->next; - /* [28/50] T.path += "../"; */ - if (!URI_FUNC(AppendSegment)(dest, URI_FUNC(ConstParent), - URI_FUNC(ConstParent) + 2, memory)) { - return URI_ERROR_MALLOC; - } - /* [29/50] pathNaked = false; */ - pathNaked = URI_FALSE; - /* [30/50] endwhile; */ - } - /* [31/50] while defined(first(A.path)) do */ - while (sourceSeg != NULL) { - /* [32/50] if pathNaked then */ - if (pathNaked == URI_TRUE) { - /* [33/50] if (first(A.path) contains ":") then */ - UriBool containsColon = URI_FALSE; - const URI_CHAR * ch = sourceSeg->text.first; - for (; ch < sourceSeg->text.afterLast; ch++) { - if (*ch == _UT(':')) { - containsColon = URI_TRUE; - break; - } - } - - if (containsColon) { - /* [34/50] T.path += "./"; */ - if (!URI_FUNC(AppendSegment)(dest, URI_FUNC(ConstPwd), - URI_FUNC(ConstPwd) + 1, memory)) { - return URI_ERROR_MALLOC; - } - /* [35/50] elseif (first(A.path) == "") then */ - } else if (sourceSeg->text.first == sourceSeg->text.afterLast) { - /* [36/50] T.path += "/."; */ - if (!URI_FUNC(AppendSegment)(dest, URI_FUNC(ConstPwd), - URI_FUNC(ConstPwd) + 1, memory)) { - return URI_ERROR_MALLOC; - } - /* [37/50] endif; */ - } - /* [38/50] endif; */ - } - /* [39/50] T.path += first(A.path); */ - if (!URI_FUNC(AppendSegment)(dest, sourceSeg->text.first, - sourceSeg->text.afterLast, memory)) { - return URI_ERROR_MALLOC; - } - /* [40/50] pathNaked = false; */ - pathNaked = URI_FALSE; - /* [41/50] A.path++; */ - sourceSeg = sourceSeg->next; - /* [42/50] if defined(first(A.path)) then */ - /* NOOP */ - /* [43/50] T.path += + "/"; */ - /* NOOP */ - /* [44/50] endif; */ - /* NOOP */ - /* [45/50] endwhile; */ - } - /* [46/50] endif; */ - } - /* [47/50] endif; */ - } - /* [48/50] endif; */ - } - /* [49/50] T.query = A.query; */ - dest->query = absSource->query; - /* [50/50] T.fragment = A.fragment; */ - dest->fragment = absSource->fragment; - - return URI_SUCCESS; + const URI_TYPE(Uri) * absSource, + const URI_TYPE(Uri) * absBase, + UriBool domainRootMode, + UriMemoryManager * memory) { + if (dest == NULL) { + return URI_ERROR_NULL; + } + URI_FUNC(ResetUri)(dest); + + if ((absSource == NULL) || (absBase == NULL)) { + return URI_ERROR_NULL; + } + + /* absBase absolute? */ + if (absBase->scheme.first == NULL) { + return URI_ERROR_REMOVEBASE_REL_BASE; + } + + /* absSource absolute? */ + if (absSource->scheme.first == NULL) { + return URI_ERROR_REMOVEBASE_REL_SOURCE; + } + + /* NOTE: The curly brackets here force deeper indent (and that's all) */ + { + { + { + /* clang-format off */ + /* [01/50] if (A.scheme != Base.scheme) then */ + /* clang-format on */ + if (!URI_FUNC(RangeEquals)(&absSource->scheme, &absBase->scheme)) { + /* clang-format off */ + /* [02/50] T.scheme = A.scheme; */ + /* clang-format on */ + dest->scheme = absSource->scheme; + /* clang-format off */ + /* [03/50] T.authority = A.authority; */ + /* clang-format on */ + if (!URI_FUNC(CopyAuthority)(dest, absSource, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [04/50] T.path = A.path; */ + /* clang-format on */ + if (!URI_FUNC(CopyPath)(dest, absSource, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [05/50] else */ + /* clang-format on */ + } else { + /* clang-format off */ + /* [06/50] undef(T.scheme); */ + /* clang-format on */ + /* NOOP */ + /* clang-format off */ + /* [07/50] if (A.authority != Base.authority) then */ + /* clang-format on */ + if (!URI_FUNC(EqualsAuthority)(absSource, absBase)) { + /* clang-format off */ + /* [08/50] T.authority = A.authority; */ + /* clang-format on */ + if (!URI_FUNC(CopyAuthority)(dest, absSource, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [09/50] T.path = A.path; */ + /* clang-format on */ + if (!URI_FUNC(CopyPath)(dest, absSource, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [10/50] else */ + /* clang-format on */ + } else { + /* clang-format off */ + /* [11/50] if domainRootMode then */ + /* clang-format on */ + if (domainRootMode == URI_TRUE) { + /* clang-format off */ + /* [12/50] undef(T.authority); */ + /* clang-format on */ + /* NOOP */ + /* clang-format off */ + /* [13/50] if (first(A.path) == "") then */ + /* clang-format on */ + /* GROUPED */ + /* clang-format off */ + /* [14/50] T.path = "/." + A.path; */ + /* clang-format on */ + /* GROUPED */ + /* clang-format off */ + /* [15/50] else */ + /* clang-format on */ + /* GROUPED */ + /* clang-format off */ + /* [16/50] T.path = A.path; */ + /* clang-format on */ + /* GROUPED */ + /* clang-format off */ + /* [17/50] endif; */ + /* clang-format on */ + if (!URI_FUNC(CopyPath)(dest, absSource, memory)) { + return URI_ERROR_MALLOC; + } + dest->absolutePath = URI_TRUE; + + if (!URI_FUNC(FixAmbiguity)(dest, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [18/50] else */ + /* clang-format on */ + } else { + const URI_TYPE(PathSegment) * sourceSeg = absSource->pathHead; + const URI_TYPE(PathSegment) * baseSeg = absBase->pathHead; + /* clang-format off */ + /* [19/50] bool pathNaked = true; */ + /* clang-format on */ + UriBool pathNaked = URI_TRUE; + /* clang-format off */ + /* [20/50] undef(last(Base.path)); */ + /* clang-format on */ + /* NOOP */ + /* clang-format off */ + /* [21/50] T.path = ""; */ + /* clang-format on */ + dest->absolutePath = URI_FALSE; + /* clang-format off */ + /* [22/50] while (first(A.path) == first(Base.path)) do */ + /* clang-format on */ + while ( + (sourceSeg != NULL) && (baseSeg != NULL) + && URI_FUNC(RangeEquals)(&sourceSeg->text, &baseSeg->text) + && !((sourceSeg->text.first == sourceSeg->text.afterLast) + && ((sourceSeg->next == NULL) + != (baseSeg->next == NULL)))) { + /* clang-format off */ + /* [23/50] A.path++; */ + /* clang-format on */ + sourceSeg = sourceSeg->next; + /* clang-format off */ + /* [24/50] Base.path++; */ + /* clang-format on */ + baseSeg = baseSeg->next; + /* clang-format off */ + /* [25/50] endwhile; */ + /* clang-format on */ + } + /* clang-format off */ + /* [26/50] while defined(first(Base.path)) do */ + /* clang-format on */ + while ((baseSeg != NULL) && (baseSeg->next != NULL)) { + /* clang-format off */ + /* [27/50] Base.path++; */ + /* clang-format on */ + baseSeg = baseSeg->next; + /* clang-format off */ + /* [28/50] T.path += "../"; */ + /* clang-format on */ + if (!URI_FUNC(AppendSegment)(dest, URI_FUNC(ConstParent), + URI_FUNC(ConstParent) + 2, + memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [29/50] pathNaked = false; */ + /* clang-format on */ + pathNaked = URI_FALSE; + /* clang-format off */ + /* [30/50] endwhile; */ + /* clang-format on */ + } + /* clang-format off */ + /* [31/50] while defined(first(A.path)) do */ + /* clang-format on */ + while (sourceSeg != NULL) { + /* clang-format off */ + /* [32/50] if pathNaked then */ + /* clang-format on */ + if (pathNaked == URI_TRUE) { + /* clang-format off */ + /* [33/50] if (first(A.path) contains ":") then */ + /* clang-format on */ + UriBool containsColon = URI_FALSE; + const URI_CHAR * ch = sourceSeg->text.first; + for (; ch < sourceSeg->text.afterLast; ch++) { + if (*ch == _UT(':')) { + containsColon = URI_TRUE; + break; + } + } + + if (containsColon) { + /* clang-format off */ + /* [34/50] T.path += "./"; */ + /* clang-format on */ + if (!URI_FUNC(AppendSegment)( + dest, URI_FUNC(ConstPwd), + URI_FUNC(ConstPwd) + 1, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [35/50] elseif (first(A.path) == "") then */ + /* clang-format on */ + } else if (sourceSeg->text.first + == sourceSeg->text.afterLast) { + /* clang-format off */ + /* [36/50] T.path += "/."; */ + /* clang-format on */ + if (!URI_FUNC(AppendSegment)( + dest, URI_FUNC(ConstPwd), + URI_FUNC(ConstPwd) + 1, memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [37/50] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [38/50] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [39/50] T.path += first(A.path); */ + /* clang-format on */ + if (!URI_FUNC(AppendSegment)(dest, sourceSeg->text.first, + sourceSeg->text.afterLast, + memory)) { + return URI_ERROR_MALLOC; + } + /* clang-format off */ + /* [40/50] pathNaked = false; */ + /* clang-format on */ + pathNaked = URI_FALSE; + /* clang-format off */ + /* [41/50] A.path++; */ + /* clang-format on */ + sourceSeg = sourceSeg->next; + /* clang-format off */ + /* [42/50] if defined(first(A.path)) then */ + /* clang-format on */ + /* NOOP */ + /* clang-format off */ + /* [43/50] T.path += + "/"; */ + /* clang-format on */ + /* NOOP */ + /* clang-format off */ + /* [44/50] endif; */ + /* clang-format on */ + /* NOOP */ + /* clang-format off */ + /* [45/50] endwhile; */ + /* clang-format on */ + } + /* clang-format off */ + /* [46/50] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [47/50] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [48/50] endif; */ + /* clang-format on */ + } + /* clang-format off */ + /* [49/50] T.query = A.query; */ + /* clang-format on */ + dest->query = absSource->query; + /* clang-format off */ + /* [50/50] T.fragment = A.fragment; */ + /* clang-format on */ + dest->fragment = absSource->fragment; + } + } + } + return URI_SUCCESS; } - - -int URI_FUNC(RemoveBaseUri)(URI_TYPE(Uri) * dest, - const URI_TYPE(Uri) * absSource, - const URI_TYPE(Uri) * absBase, - UriBool domainRootMode) { - return URI_FUNC(RemoveBaseUriMm)(dest, absSource, absBase, - domainRootMode, NULL); +int URI_FUNC(RemoveBaseUri)(URI_TYPE(Uri) * dest, const URI_TYPE(Uri) * absSource, + const URI_TYPE(Uri) * absBase, UriBool domainRootMode) { + return URI_FUNC(RemoveBaseUriMm)(dest, absSource, absBase, domainRootMode, NULL); } +int URI_FUNC(RemoveBaseUriMm)(URI_TYPE(Uri) * dest, const URI_TYPE(Uri) * absSource, + const URI_TYPE(Uri) * absBase, UriBool domainRootMode, + UriMemoryManager * memory) { + int res; + URI_CHECK_MEMORY_MANAGER(memory); /* may return */ -int URI_FUNC(RemoveBaseUriMm)(URI_TYPE(Uri) * dest, - const URI_TYPE(Uri) * absSource, - const URI_TYPE(Uri) * absBase, - UriBool domainRootMode, UriMemoryManager * memory) { - int res; - - URI_CHECK_MEMORY_MANAGER(memory); /* may return */ - - res = URI_FUNC(RemoveBaseUriImpl)(dest, absSource, - absBase, domainRootMode, memory); - if ((res != URI_SUCCESS) && (dest != NULL)) { - URI_FUNC(FreeUriMembersMm)(dest, memory); - } - return res; + res = URI_FUNC(RemoveBaseUriImpl)(dest, absSource, absBase, domainRootMode, memory); + if ((res != URI_SUCCESS) && (dest != NULL)) { + URI_FUNC(FreeUriMembersMm)(dest, memory); + } + return res; } - - #endif diff --git a/cpp/src/arrow/vendored/uriparser/UriVersion.c b/cpp/src/arrow/vendored/uriparser/UriVersion.c new file mode 100644 index 000000000000..89abaa49cafd --- /dev/null +++ b/cpp/src/arrow/vendored/uriparser/UriVersion.c @@ -0,0 +1,81 @@ +/* + * uriparser - RFC 3986 URI parsing library + * + * Copyright (C) 2025, Sebastian Pipping + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * 2. Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of + * its contributors may be used to endorse or promote products + * derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/** + * @file UriVersion.c + * Implements a runtime version getter. + * NOTE: This source file includes itself twice. + */ + +/* What encodings are enabled? */ +#include "UriDefsConfig.h" +#if (!defined(URI_PASS_ANSI) && !defined(URI_PASS_UNICODE)) +/* Include SELF twice */ +# ifdef URI_ENABLE_ANSI +# define URI_PASS_ANSI 1 +# include "UriVersion.c" +# undef URI_PASS_ANSI +# endif +# ifdef URI_ENABLE_UNICODE +# define URI_PASS_UNICODE 1 +# include "UriVersion.c" +# undef URI_PASS_UNICODE +# endif +#else +# ifdef URI_PASS_ANSI +# include "UriDefsAnsi.h" +# else +# include "UriDefsUnicode.h" +# include +# endif + +# ifndef URI_DOXYGEN +# include "Uri.h" +# endif + +const URI_CHAR * URI_FUNC(BaseRuntimeVersion)(void) { +# if defined(URI_PASS_ANSI) + return URI_VER_ANSI; +# elif defined(URI_PASS_UNICODE) + return URI_VER_UNICODE; +# else +# error Either URI_PASS_ANSI or URI_PASS_UNICODE must be defined +# endif +} + +#endif diff --git a/cpp/src/arrow/vendored/uriparser/config.h b/cpp/src/arrow/vendored/uriparser/config.h deleted file mode 100644 index 97f0605da723..000000000000 --- a/cpp/src/arrow/vendored/uriparser/config.h +++ /dev/null @@ -1,47 +0,0 @@ -/* - * uriparser - RFC 3986 URI parsing library - * - * Copyright (C) 2018, Sebastian Pipping - * All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * * Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * * Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * * Neither the name of the nor the names of its - * contributors may be used to endorse or promote products - * derived from this software without specific prior written - * permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR - * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, - * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) - * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED - * OF THE POSSIBILITY OF SUCH DAMAGE. - */ - -#if !defined(URI_CONFIG_H) -# define URI_CONFIG_H 1 - -#define PACKAGE_VERSION "0.9.3" - -// #define HAVE_WPRINTF -// #define HAVE_REALLOCARRAY - -#endif /* !defined(URI_CONFIG_H) */