#42312 [Fbk->Opn]: BOM mark interpreted as output

ID: 42312 User updated by: draconpern at hotmail dot com Reported By: draconpern at hotmail dot com -Status: Feedback +Status: Open Bug Type: I18N and L10N related Operating System: FreeBSD PHP Version: 5.2.3 New Comment: Anyway to enable that option without doing a recompile? Also, isn't it still experimental? Hence not on by default. Also, different BOM gives different results. Using utf16 encoded files results in a correct bom but then unable to parse the code. Where as utf8 just gives the output warning. Previous Comments: ------------------------------------------------------------------------ [2007-08-16 06:25:19] jani@php.net And using --enable-zend-multibyte does not work for you? ------------------------------------------------------------------------ [2007-08-15 19:49:52] draconpern at hotmail dot com All the other bugs suggested ignoring/discarding the output which is incorrect. The correct action should be to read the script, and save the BOM. Do the header processing, and then output the BOM as part of the output content. ------------------------------------------------------------------------ [2007-08-15 15:40:45] jani@php.net And how isn't this exactly same what bug #22247 is about? Or bug #22108 ? ------------------------------------------------------------------------ [2007-08-15 14:59:34] draconpern at hotmail dot com Description: ------------ BOM marks on utf files are interpreted as output. So any php script that has to output before any file content has problems. This is not a duplicate of bug #22247. Instead, BOM should be passed through to output, but not interpreted as output. Expected result: ---------------- Any php file with BOM. ------------------------------------------------------------------------ -- Edit this bug report at http://bugs.php.net/?id=42312&edit=1