Fix reDOS in HTML tokenizer regex

Crozzers · Crozzers · commit 034d126c2084 · 2025-07-12T17:06:04.000+01:00
diff --git a/lib/markdown2.py b/lib/markdown2.py
@@ -1273,7 +1273,7 @@ def _run_span_gamut(self, text: str) -> str:
                     \s+                           # whitespace after tag
                     (?:[^\t<>"'=/]+:)?
                     [^<>"'=/]+=                   # attr name
-                    (?:".*?"|'.*?'|[^<>"'=/\s]+)  # value, quoted or unquoted. If unquoted, no spaces allowed
+                    (?:"[^"]*?"|'[^']*?'|[^<>"'=/\s]+)  # value, quoted or unquoted. If unquoted, no spaces allowed
                 )*
                 \s*/?>
                 |

Original file line number	Diff line number	Diff line change
`@@ -1273,7 +1273,7 @@ def _run_span_gamut(self, text: str) -> str:`
`1273`	`1273`	`\s+ # whitespace after tag`
`1274`	`1274`	`(?:[^\t<>"'=/]+:)?`
`1275`	`1275`	`[^<>"'=/]+= # attr name`
`1276`		`- (?:".?"\|'.?'\|[^<>"'=/\s]+) # value, quoted or unquoted. If unquoted, no spaces allowed`
	`1276`	`+ (?:"[^"]?"\|'[^']?'\|[^<>"'=/\s]+) # value, quoted or unquoted. If unquoted, no spaces allowed`
`1277`	`1277`	`)*`
`1278`	`1278`	`\s*/?>`
`1279`	`1279`	`\|`