diff options
author | Urban Müller | 2024-05-28 17:49:09 +0200 |
---|---|---|
committer | Urban Müller | 2024-05-28 17:49:09 +0200 |
commit | 25a946b8c2bf6638b7adfe2afa387fa26cb97e71 (patch) | |
tree | aa693ef50ac2531fc5399614e289c9fd8609d2af /it_html.class | |
parent | 937d04e90f85e332186c6c16fc6b06fa0ca400a7 (diff) | |
download | itools-25a946b8c2bf6638b7adfe2afa387fa26cb97e71.tar.gz itools-25a946b8c2bf6638b7adfe2afa387fa26cb97e71.tar.bz2 itools-25a946b8c2bf6638b7adfe2afa387fa26cb97e71.zip |
Revert "Improve handling of nested tags in it_html::sanitize": getting "Exceeded pcre.backtrack_limit of 1000000 bytes"
This reverts commit b484fab88a9229f7c87ea053564d0d8d3d2a565d.
Diffstat (limited to 'it_html.class')
-rw-r--r-- | it_html.class | 10 |
1 files changed, 3 insertions, 7 deletions
diff --git a/it_html.class b/it_html.class index effe920..9780d5d 100644 --- a/it_html.class +++ b/it_html.class @@ -408,15 +408,11 @@ static function sanitize($html) $html = it::replace(array('[\0\s]+' => " "), $html); # \s also matches \r and \n $urlpattern = 'https?://[^">]+'; - if ($tag = it::match('(<(div|p|ol|ul|li|i|b|strong|h[1-6])\b[^>]*>((?:(?!</?\2\b).|(?R))*?)</\2>)', $html, ['offset_capture' => 1])) + if ($tag = it::match("(.*?)<(div|p|ol|ul|li|i|b|strong|h[1-6])\b[^>]*>(.*?)</\\2>(.*)", $html)) { # Simple tags with content, no attributes kept - $offset = $tag[0][1]; - $length = strlen($tag[0][0]); - $head = substr($html, 0, $offset); - $tail = substr($html, $offset + $length); - $content = $tag[2][0]; - $tagname = strtolower($tag[1][0]); + list($head, $tagname, $content, $tail) = $tag; + $tagname = strtolower($tagname); $result .= it_html::sanitize($head) . "<$tagname>" . it_html::sanitize($content) . "</$tagname>" . it_html::sanitize($tail); } else if ($tag = it::match('(.*)<a\b[^>]+?\bhref\s*=\s*"(' . $urlpattern . ')"[^>]*?>(.*?)</a>(.*)', $html)) |