|
|
php.net | support | documentation | report a bug | advanced search | search howto | statistics | random bug | login |
[2015-07-14 13:14 UTC] ongun dot kanat at gmail dot com
Description: ------------ Using many languages in one page(at the same time) is possible when your clients work globally/multilingual. And sometimes developers like me are struggling in implement a page uses/modifies strings from many languages at the same time. And sometimes it's just meaningless to change locale of whole page for just one string. It would be awesome a parameter to change language in mbstring upper/lowercase operations. Example case I am displaying a page in English but also displaying strings in languages below: ß -> SS in German, ı->I in Turkish Unicode already has mappings for chars and special cases. http://unicode.org/faq/casemap_charprop.html#1 Test script: --------------- <?php mb_internal_encoding("UTF-8"); echo mb_strtoupper("Maß\n"); // "MAß" echo mb_strtolower("MAẞ\n"); // "maß" echo mb_strtoupper("ilaç\n"); // "ILAÇ" echo mb_strtolower("IŞIK\n"); // "iŞik" ?> Expected result: ---------------- MASS // Eszett maß İLAÇ // Turkish I issue ışık Actual result: -------------- MAß maß ILAÇ iŞik PatchesPull RequestsHistoryAllCommentsChangesGit/SVN commits
|
|||||||||||||||||||||||||||||||||||||
Copyright © 2001-2025 The PHP GroupAll rights reserved. |
Last updated: Sun Jul 20 02:00:03 2025 UTC |
This is still relevant in PHP 8.3.7 Behavior is very unpredictable according to the position of the letters and according to the neighboring letters in a word. I mean if the resulting string was consistent somehow, we could accept what was commented by PHP team before; but since the resulting string is not predictable and not consistent, this, IMHO is a legit bug. echo mb_convert_case('içecek', MB_CASE_LOWER); // prints normally echo mb_convert_case('İçecek', MB_CASE_LOWER); // prints something over "ç" which looks like a cedilla but it's not. I can't paste it here because this form strips that character. In fact it is a multibyte character attached to "i" and not to letter next to it. echo mb_convert_case('ıi', MB_CASE_UPPER); // prints: II This is already well-known. echo mb_convert_case('İÇİŞLERİ', MB_CASE_LOWER); // prints that strange character everywhere distorting the whole word.