feat: surface libxml2's parse error in REQBODY_ERROR_MSG

src/request_body_processor/xml.cc

@@ -15,6 +15,7 @@
 
 #include "src/request_body_processor/xml.h"
 
+#include <cctype>
 #include <list>
 #include <iostream>
 #include <string>
@@ -151,6 +152,25 @@
     }
 }
 
+
+/* Build the error text from libxml2's last error on this context, so the
+ * actual parse failure ("Extra content at the end of the document", a tag
+ * mismatch, ...) reaches REQBODY_ERROR_MSG and the debug log instead of a
+ * fixed string. Falls back to "XML: <what>." when libxml2 has no message.
+ * xmlCtxtGetLastError() keeps working under XML_PARSE_NOERROR. */
+static std::string parseError(xmlParserCtxtPtr ctx, const std::string &what) {
+    const xmlError *e = (ctx != NULL) ? xmlCtxtGetLastError(ctx) : NULL;
+    if (e == NULL || e->message == NULL || *e->message == '\0') {
+        return "XML: " + what + ".";
+    }
+    std::string detail(e->message);
+    while (!detail.empty() && isspace(static_cast<unsigned char>(detail.back()))) {
+        detail.pop_back();
+    }
+    return "XML: " + what + ": " + detail + " (line " + std::to_string(e->line)
+        + ", column " + std::to_string(e->int2) + ")";
+}
+
 XML::XML(Transaction *transaction)
     : m_transaction(transaction) {
     m_data.doc = NULL;
@@ -287,8 +307,9 @@
         xmlParseChunk(m_data.parsing_ctx, buf, size, 0);
         m_data.xml_parser_state->parsing_ctx_arg = m_data.parsing_ctx_arg;
         if (m_data.parsing_ctx->wellFormed != 1) {
-            error->assign("XML: Failed to parse document.");
-            ms_dbg_a(m_transaction, 4, "XML: Failed to parse document.");
+            error->assign(parseError(m_data.parsing_ctx,
+                "Failed to parse document"));
+            ms_dbg_a(m_transaction, 4, *error);
             return false;
         }
     }
@@ -303,8 +324,9 @@
         ) {
         xmlParseChunk(m_data.parsing_ctx_arg, buf, size, 0);
         if (m_data.parsing_ctx_arg->wellFormed != 1) {
-            error->assign("XML: Failed to parse document for ARGS.");
-            ms_dbg_a(m_transaction, 4, "XML: Failed to parse document for ARGS.");
+            error->assign(parseError(m_data.parsing_ctx_arg,
+                "Failed to parse document for ARGS"));
+            ms_dbg_a(m_transaction, 4, *error);
             return false;
         }
     }
@@ -325,6 +347,11 @@
             /* Preserve the results for our reference. */
             m_data.well_formed = m_data.parsing_ctx->wellFormed;
             m_data.doc = m_data.parsing_ctx->myDoc;
+            std::string parse_error;
+            if (m_data.well_formed != 1) {
+                parse_error = parseError(m_data.parsing_ctx,
+                    "Failed to parse document");
+            }
 
             /* Clean up everything else. */
             xmlFreeParserCtxt(m_data.parsing_ctx);
@@ -333,8 +360,8 @@
                 + std::to_string(m_data.well_formed) + ").");
 
             if (m_data.well_formed != 1) {
-                error->assign("XML: Failed to parse document.");
-                ms_dbg_a(m_transaction, 4, "XML: Failed to parse document.");
+                error->assign(parse_error);
+                ms_dbg_a(m_transaction, 4, *error);
                 return false;
             }
         }
@@ -352,7 +379,8 @@
                     error->assign(m_data.xml_error);
                 }
                 else {
-                    error->assign("XML: Failed to parse document for ARGS.");
+                    error->assign(parseError(m_data.parsing_ctx_arg,
+                        "Failed to parse document for ARGS"));
                 }
                 xmlFreeParserCtxt(m_data.parsing_ctx_arg);
                 m_data.parsing_ctx_arg = NULL;

test/test-cases/regression/variable-REQBODY_PROCESSOR_ERROR.json

@@ -140,5 +140,84 @@
       "SecRule REQBODY_PROCESSOR_ERROR \"@contains test\" \"phase:2,id:1,pass,t:trim\"",
       "SecRule REQBODY_PROCESSOR_ERROR_MSG \"@contains test\" \"phase:2,id:2,pass,t:trim\""
     ]
+  },
+  {
+    "enabled": 1,
+    "version_min": 300000,
+    "title": "Testing Variables :: REQBODY_ERROR_MSG carries the libxml2 parse error",
+    "resource": "libxml2",
+    "client": {
+      "ip": "200.249.12.31",
+      "port": 123
+    },
+    "server": {
+      "ip": "200.249.12.31",
+      "port": 80
+    },
+    "request": {
+      "headers": {
+        "Host": "localhost",
+        "User-Agent": "curl/7.38.0",
+        "Accept": "*/*",
+        "Content-Type": "text/xml",
+        "Content-Length": "736"
+      },
+      "uri": "/?key=value&key=other_value",
+      "method": "POST",
+      "http_version": 1.1,
+      "body": [
+        "<?xml version=\"1.0\" encoding=\"UTF-8\"?>",
+        "<bookstore>",
+        "<book category=\"COOKING\">",
+        "<title lang=\"en\">Everyday Italian</title>",
+        "<author>Giada De Laurentiis</author>",
+        "<year>2005</year>",
+        "<price>30.00</price>",
+        "</book>",
+        "<book category=\"CHILDREN\">",
+        "<title lang=\"en\">Harry Potter</title>",
+        "<author>J K. Rowling</author>",
+        "<year>2005</year>",
+        "<price>29.99</price>",
+        "</book>",
+        "<book category=\"WEB\">",
+        "<title lang=\"en\">XQuery Kick Start</title>",
+        "<author>James McGovern</author>",
+        "<author>Per Bothner</author>",
+        "<author>Kurt Cagle</author>",
+        "<author>James Linn</author>",
+        "<author>Vaidyanathan Nagarajan</author>",
+        "<year>2003</year>",
+        "<price>49.99</price>",
+        "</book>",
+        "<book category=\"WEB\">",
+        "<title lang=\"en\">Learning XML</title>",
+        "<author>Erik T. Ray</author>",
+        "<year>2003</year>",
+        "<price>39.95</price>",
+        "</bookasdf>",
+        "</bookstore>"
+      ]
+    },
+    "response": {
+      "headers": {
+        "Date": "Mon, 13 Jul 2015 20:02:41 GMT",
+        "Last-Modified": "Sun, 26 Oct 2014 22:33:37 GMT",
+        "Content-Type": "text/html",
+        "Content-Length": "8"
+      },
+      "body": [
+        "no need."
+      ]
+    },
+    "expected": {
+      "debug_log": "XML: Failed to parse document: Opening and ending tag mismatch: book line 1 and bookasdf \\(line 1, column [0-9]+\\)",
+      "http_code": 403
+    },
+    "rules": [
+      "SecRuleEngine On",
+      "SecRule REQUEST_HEADERS:Content-Type \"^text/xml$\" \"id:500005,phase:1,t:none,t:lowercase,nolog,pass,ctl:requestBodyProcessor=XML\"",
+      "SecRule REQBODY_ERROR_MSG \"@contains Opening and ending tag mismatch: book line 1 and bookasdf\" \"phase:2,id:3,deny,status:403\""
+    ]
   }
 ]