#31649 [Opn]: urldecode does not follow ECMA standard or standard browser practice

ID: 31649 Updated by: derick@php.net Reported By: james at gogo dot co dot nz Status: Open -Bug Type: URL related +Bug Type: Feature/Change Request Operating System: All PHP Version: 4.3.10 New Comment: PHP doesnt support unicode in a whole lot of places. Marking this as a feature request instead. Previous Comments: ------------------------------------------------------------------------ [2005-01-21 22:29:07] james at gogo dot co dot nz Description: ------------ urldecode() does not understand the %uxxxx format for escaping unicode characters above 0xFF. This is a very old bug, originally reported as bug #15027 and declared bogus, I believe erroneously, and here is the reasoning... In all modern browsers (including Mozilla), JavaScript's escape() function uses %HH for Unicode codepoints below 0x0100, but %uHHHH for codepoints above there. From ECMA-262: -------------- For characters whose Unicode encoding is 0xFF or less, a two-digit escape sequence of the form %xx is used in accordance with RFC1738. For characters whose Unicode encoding is greater than 0xFF, a four-digit escape sequence of the form %uxxxx is used. -------------- I believe this is a bug, PHP is unable to urldecode the valid escape()d values from modern browsers when those escape()d strings contain unicode characters greater than 0xFF. Declaring it not a bug because it is not in the RFCs, but rather defined by ECMA is a poor decision. Reproduce code: --------------- echo urldecode('%u2013'); Expected result: ---------------- A string containing the three characters comprising the unicode character 0x2013 (En Dash) in utf-8, namely 0xE2 0x80 and 0x93. Actual result: -------------- The literal string "%u2013". ------------------------------------------------------------------------ -- Edit this bug report at http://bugs.php.net/?id=31649&edit=1