note 113979 added to function.file

file() seems to have a problem in handling UTF-16 with or without BOM. file() is likely to think "\n"=LF (0A) as a line-ending. So, not only "000A" but also "010A, 020A,...,FE0A, FF0A,..." are regarded as line-endings. Moreover, file() causes a serious problem in UTF-16LE. file() loses first "0A" (the first half of "0A00")! And the next line begins with "00" (the rest of "0A00"). So lines after the first "0A" are totally different. To avoid this phenomena, eg. in case (php_script : UTF-8 , file : UTF-16 with line-ending "\r\n"), <?php mb_regex_encoding('UTF-16'); // mb_ereg_search_..() can work properly $str = file_get_contents($file_path); $pattern = mb_convert_encoding('[^\r]*\r\n', 'UTF-16', 'UTF-8'); mb_ereg_search_init($str, $pattern); while ($res = mb_ereg_search_regs()) { $file[] = $res[0]; } ?> instead of $file = file($file_path); If line-ending is "\n", $pattern = mb_convert_encoding('[^\n]*\n', 'UTF-16', 'UTF-8'); ---- Server IP: 49.212.134.217 Probable Submitter: 101.141.141.233 ---- Manual Page -- http://php.net/manual/en/function.file.php Edit -- https://master.php.net/note/edit/113979 Del: integrated -- https://master.php.net/note/delete/113979/integrated Del: useless -- https://master.php.net/note/delete/113979/useless Del: bad code -- https://master.php.net/note/delete/113979/bad+code Del: spam -- https://master.php.net/note/delete/113979/spam Del: non-english -- https://master.php.net/note/delete/113979/non-english Del: in docs -- https://master.php.net/note/delete/113979/in+docs Del: other reasons-- https://master.php.net/note/delete/113979 Reject -- https://master.php.net/note/reject/113979 Search -- https://master.php.net/manage/user-notes.php