diff options
| author | Christian Schneider | 2025-05-09 14:31:12 +0200 |
|---|---|---|
| committer | Christian Schneider | 2025-05-09 14:37:31 +0200 |
| commit | f2225077d7450ea08f40ea45cc348184f1d30b97 (patch) | |
| tree | 00af524abe9f1e70a3fb0a534f318b351107b710 | |
| parent | 73b14dc5ea8c74fe17def89a31ab0eb5e53a4594 (diff) | |
| download | itools-f2225077d7450ea08f40ea45cc348184f1d30b97.tar.gz itools-f2225077d7450ea08f40ea45cc348184f1d30b97.tar.bz2 itools-f2225077d7450ea08f40ea45cc348184f1d30b97.zip | |
Handle mailto:-links and tags inside <a> tags in it_html::sanitize() (support request for https://search.ch/tel/biel-bienne/bahnhofstrasse-5/groupe-mutuel-4)HEADmaster
| -rw-r--r-- | it_html.class | 11 | ||||
| -rwxr-xr-x | test/it_html.t | 12 |
2 files changed, 18 insertions, 5 deletions
diff --git a/it_html.class b/it_html.class index e3053b6..9bb8e7a 100644 --- a/it_html.class +++ b/it_html.class @@ -393,32 +393,33 @@ static function sanitize($html) if ($charset == "utf-8") $html = it::any2utf8($html); $html = it::replace(array('[\0\s]+' => " "), $html); # \s also matches \r and \n - $urlpattern = 'https?://[^">]+'; + $urlpattern = '(?:https?://|mailto:)[^">]+'; + $placeholder = bin2hex(random_bytes(16)); if ($tag = it::match("(.*?)<(div|p|ol|ul|li|i|b|strong|h[1-6])\b[^>]*>(.*?)</\\2>(.*)", $html)) { # Simple tags with content, no attributes kept list($head, $tagname, $content, $tail) = $tag; $tagname = strtolower($tagname); - $result .= it_html::sanitize($head) . "<$tagname>" . it_html::sanitize($content) . "</$tagname>" . it_html::sanitize($tail); + $result .= it::replace([$placeholder => "<$tagname>" . it_html::sanitize($content) . "</$tagname>"], it_html::sanitize("$head$placeholder$tail")); } else if ($tag = it::match('(.*)<a\b[^>]+?\bhref\s*=\s*"(' . $urlpattern . ')"[^>]*?>(.*?)</a>(.*)', $html)) { # Link tags, keeps only href attribute list($head, $href, $content, $tail) = $tag; - $result .= it_html::sanitize($head) . '<a href="' . it_html::Q(it_html::U(html_entity_decode($href, ENT_COMPAT, $charset))) . '">' . it_html::sanitize($content) . "</a>" . it_html::sanitize($tail); + $result .= it::replace([$placeholder => '<a href="' . it_html::Q(it_html::U(html_entity_decode($href, ENT_COMPAT, $charset))) . '">' . it_html::sanitize($content) . "</a>"], it_html::sanitize("$head$placeholder$tail")); } else if ($tag = it::match('(.*)<img\b[^>]+?\bsrc\s*=\s*"(' . $urlpattern . ')"[^>]*?>(.*)', $html)) { # Image tags, keeps only src attribute list($head, $src, $tail) = $tag; - $result .= it_html::sanitize($head) . '<img src="' . it_html::Q(it_html::U(html_entity_decode($src, ENT_COMPAT, $charset))) . '" alt="" />' . it_html::sanitize($tail); + $result .= it::replace([$placeholder => '<img src="' . it_html::Q(it_html::U(html_entity_decode($src, ENT_COMPAT, $charset))) . '" alt="" />'], it_html::sanitize("$head$placeholder$tail")); } else if ($tag = it::match("(.*)<(br|/tr)\b[^>]*>(.*)", $html)) { # brs and table rows are converted so simple line breaks list($head, $tagname, $tail) = $tag; - $result .= it_html::sanitize($head) . "<br />" . it_html::sanitize($tail); + $result .= it::replace([$placeholder => "<br />"], it_html::sanitize("$head$placeholder$tail")); } else $result = it::replace(array('&(#\d+;)' => '&$1'), it_html::Q(html_entity_decode(strip_tags($html), ENT_COMPAT, $charset))); diff --git a/test/it_html.t b/test/it_html.t index 15f444d..20ab65f 100755 --- a/test/it_html.t +++ b/test/it_html.t @@ -260,6 +260,18 @@ is( ); is( + it_html::sanitize('<a href="http://search.ch/"><strong>foo</strong></a>'), + '<a href="http://search.ch/"><strong>foo</strong></a>', + 'it_html::sanitize handle nesting of tags inside <a>' +); + +is( + it_html::sanitize('<a href="mailto:neuman@example.com">foo</a>'), + '<a href="mailto:neuman@example.com">foo</a>', + 'it_html::sanitize handle mailto links' +); + +is( it_html::sanitize("<a href='http://search.ch/'>foo</a>"), '<a href="http://search.ch/">foo</a>', 'TODO it_html::sanitize handle anchors with single quotes at attribute value' |