Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 9 additions & 3 deletions sql_metadata/sql_cleaner.py
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,7 @@
import re
from typing import NamedTuple

from sqlglot import Dialect
from sqlglot.errors import TokenError
from sqlglot.tokens import Tokenizer, TokenType

Expand All @@ -35,11 +36,16 @@ def _is_wrapped(text: str) -> bool:
"""
if len(text) < 2 or text[0] != "(" or text[-1] != ")":
return False
dialect = "mysql" if "`" in text else "tsql" if "[" in text else None
try:
tokens = Dialect.get_or_raise(dialect).tokenize(text[1:-1])
except TokenError:
return False
depth = 0
for c in text[1:-1]:
if c == "(":
for token in tokens:
if token.token_type == TokenType.L_PAREN:
depth += 1
elif c == ")":
elif token.token_type == TokenType.R_PAREN:
depth -= 1
if depth < 0:
return False
Expand Down
48 changes: 48 additions & 0 deletions test/test_edge_cases.py
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,8 @@
are imported.
"""

import pytest

from sql_metadata import Parser
from sql_metadata.sql_cleaner import SqlCleaner
from sql_metadata.utils import UniqueList
Expand Down Expand Up @@ -65,3 +67,49 @@ def test_strip_outer_parens_unbalanced_middle():
# intact so both SELECT branches parse.
parser = Parser("(SELECT a FROM t1) UNION (SELECT b FROM t2)")
assert parser.tables == ["t1", "t2"]


@pytest.mark.parametrize("depth", [1, 2])
@pytest.mark.parametrize("literal", ["')'", "'left)right'", "'it'')s'", "'('"])
@pytest.mark.parametrize(
"template",
[
"UPDATE target SET value = {literal} WHERE id = 1",
"INSERT INTO target (value) VALUES ({literal})",
"DELETE FROM target WHERE value = {literal}",
],
)
def test_outer_parens_ignore_quoted_parentheses(
depth: int, literal: str, template: str
) -> None:
query = template.format(literal=literal)
wrapped = "(" * depth + query + ")" * depth
assert SqlCleaner.clean(wrapped).sql == query
expected = Parser(query)
actual = Parser(wrapped)
assert actual.query_type == expected.query_type
assert actual.tables == expected.tables
assert actual.columns == expected.columns


def test_outer_parens_preserve_unterminated_literal() -> None:
query = "((UPDATE target SET value = 'unterminated))"
assert SqlCleaner.clean(query).sql == query
with pytest.raises(ValueError):
Parser(query).tables


@pytest.mark.parametrize("name", ["we're", 'say"hello', "a)b"])
def test_outer_parens_preserve_mysql_quoted_names(name: str) -> None:
query = f"UPDATE `{name}` SET value = 1"
expected = Parser(query)
actual = Parser(f"(({query}))")
assert actual.tables == expected.tables


@pytest.mark.parametrize("name", ["we're", 'say"hello', "a)b"])
def test_outer_parens_preserve_tsql_quoted_names(name: str) -> None:
query = f"UPDATE [{name}] SET value = 1"
expected = Parser(query)
actual = Parser(f"(({query}))")
assert actual.tables == expected.tables