Fix it_html::sanitize with b/br combination (tag prefix of other tag bug)

1 files changed, 4 insertions, 4 deletions

diff --git a/it_html.class b/it_html.class
index 9debd34..3ccc084 100644
--- a/it_html.class
+++ b/it_html.class
@@ -378,26 +378,26 @@ function sanitize($html)
 	$urlpattern = 'https?://[^">]+';
 	$charset = $GLOBALS['it_html']->p['charset'] ? $GLOBALS['it_html']->p['charset'] : 'iso-8859-1';
 
-	if ($tag = it::match("(.*)<(div|p|i|b)[^>]*>(.*?)</\\2>(.*)", $html))
+	if ($tag = it::match("(.*)<(div|p|i|b)\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);
 	}
-	else if ($tag = it::match('(.*)<a[^>]+?href="(' . $urlpattern . ')"[^>]*?>(.*?)</a>(.*)', $html))
+	else if ($tag = it::match('(.*)<a\b[^>]+?href="(' . $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(html_entity_decode($href), ENT_COMPAT, $charset) . '">' . it_html::sanitize($content) . "</a>" . it_html::sanitize($tail);
 	}
-	else if ($tag = it::match('(.*)<img[^>]+?src="(' . $urlpattern . ')"[^>]*?>(.*)', $html))
+	else if ($tag = it::match('(.*)<img\b[^>]+?src="(' . $urlpattern . ')"[^>]*?>(.*)', $html))
 	{
 		# Image tags, keeps only src attribute
 		list($head, $src, $tail) = $tag;
 		$result .= it_html::sanitize($head) . '<img src="' . it_html::Q(html_entity_decode($src, ENT_COMPAT, $charset)) . '" alt="" />' . it_html::sanitize($tail);
 	}
-	else if ($tag = it::match("(.*)<(br|/tr)[^>]*>(.*)", $html))
+	else if ($tag = it::match("(.*)<(br|/tr)\b[^>]*>(.*)", $html))
 	{
 		# brs and table rows are converted so simple line breaks
 		list($head, $tagname, $tail) = $tag;