[ticket/9933] Adjust word censor regex for non-unicode mode.

PHPBB3-9933
This commit is contained in:
rxu 2011-01-08 12:11:02 +07:00 committed by Andreas Fischer
parent 1359cf6547
commit 8c1866bc0c

View file

@ -3451,11 +3451,11 @@ function get_censor_preg_expression($word)
$unicode = ((version_compare(PHP_VERSION, '5.1.0', '>=') || (version_compare(PHP_VERSION, '5.0.0-dev', '<=') && version_compare(PHP_VERSION, '4.4.0', '>='))) && @preg_match('/\p{L}/u', 'a') !== false) ? true : false; $unicode = ((version_compare(PHP_VERSION, '5.1.0', '>=') || (version_compare(PHP_VERSION, '5.0.0-dev', '<=') && version_compare(PHP_VERSION, '4.4.0', '>='))) && @preg_match('/\p{L}/u', 'a') !== false) ? true : false;
} }
if ($unicode)
{
// Unescape the asterisk to simplify further conversions // Unescape the asterisk to simplify further conversions
$word = str_replace('\*', '*', preg_quote($word, '#')); $word = str_replace('\*', '*', preg_quote($word, '#'));
if ($unicode)
{
// Replace asterisk(s) inside the pattern, at the start and at the end of it with regexes // Replace asterisk(s) inside the pattern, at the start and at the end of it with regexes
$word = preg_replace(array('#(?<=[\p{Nd}\p{L}_])\*+(?=[\p{Nd}\p{L}_])#iu', '#^\*+#', '#\*+$#'), array('([\x20]*?|[\p{Nd}\p{L}_-]*?)', '[\p{Nd}\p{L}_-]*?', '[\p{Nd}\p{L}_-]*?'), $word); $word = preg_replace(array('#(?<=[\p{Nd}\p{L}_])\*+(?=[\p{Nd}\p{L}_])#iu', '#^\*+#', '#\*+$#'), array('([\x20]*?|[\p{Nd}\p{L}_-]*?)', '[\p{Nd}\p{L}_-]*?', '[\p{Nd}\p{L}_-]*?'), $word);
@ -3464,7 +3464,11 @@ function get_censor_preg_expression($word)
} }
else else
{ {
$preg_expr = '#(?<!\S)(' . str_replace('\*', '\S*?', preg_quote($word, '#')) . ')(?!\S)#iu'; // Replace the asterisk inside the pattern, at the start and at the end of it with regexes
$word = preg_replace(array('#(?<=\S)\*+(?=\S)#iu', '#^\*+#', '#\*+$#'), array('(\x20*?\S*?)', '\S*?', '\S*?'), $word);
// Generate the final substitution
$preg_expr = '#(?<!\S)(' . $word . ')(?!\S)#iu';
} }
return $preg_expr; return $preg_expr;