Bug #65072 [Opn->Nab]: strlen result wrong after passing through html_entity_decode

Edit report at https://bugs.php.net/bug.php?id=65072&edit=1 ID: 65072 Updated by: nikic@php.net Reported by: kstirn at gmail dot com Summary: strlen result wrong after passing through html_entity_decode -Status: Open +Status: Not a bug Type: Bug Package: Unknown/Other Function Operating System: Windows 7 PHP Version: 5.4.16 Block user comment: N Private report: N New Comment: html_entity_decode('&nbsp;') returns a U+00A0 NO-BREAK SPACE, which encoded in UTF8 is "\xc2\xa0". strlen() returns the length in bytes (not code points), as such strlen("\xc2\xa0") will be 2, not 1. If you want to have the output in a different encoding (e.g. ISO-8859-1) you can specify it via the last argument to the function. Previous Comments: ------------------------------------------------------------------------ [2013-06-20 16:47:46] kstirn at gmail dot com Description: ------------ If you run string "X&nbsp;X" through html_entity_decode then check its' length it returns 4 instead of 3. PHP 5.3.x gives correct result. PHP 5.4.x and 5.5.x give incorrect result. Possibly the same problem with other HTML entities, such as &raquo; etc... Test script: --------------- <?php echo 'strlen(a) = ' .strlen("X X") . "<br>\n"; echo 'strlen(b) = ' .strlen(html_entity_decode('X&nbsp;X')) . "<br>\n"; ?> Expected result: ---------------- strlen(a) = 3 strlen(b) = 3 Actual result: -------------- strlen(a) = 3 strlen(b) = 4 ------------------------------------------------------------------------ -- Edit this bug report at https://bugs.php.net/bug.php?id=65072&edit=1