Bug #2942 [Com]: Problems with files with over 200 row / or 30kb of filesize

From: Date: Wed, 01 Apr 2009 08:38:29 +0000
Subject: Bug #2942 [Com]: Problems with files with over 200 row / or 30kb of filesize
References: 1  Groups: php.pear.bugs 
Request: Send a blank email to pear-bugs+get-287@lists.php.net to get a copy of this message
Edit report at http://pear.php.net/bugs/bug.php?id=2942&edit=1 ID: 2942 Comment by: tomasz.rup@gmail.com Reported By: alex dot petri at gmx dot de Summary: Problems with files with over 200 row / or 30kb of filesize Status: Open Type: Bug Package: Spreadsheet_Excel_Writer Operating System: Linux PHP Version: 4.3.9 Roadmap Versions: New Comment: Test file: require_once "Spreadsheet/Excel/Writer.php"; class xls { /** * @var Spreadsheet_Excel_Writer */ private $xls; /** * @var Spreadsheet_Excel_Writer_Worksheet */ private $sheet; private $row = 0; private $colDateMasks; private function init($filename) { $this->xls = new Spreadsheet_Excel_Writer($filename); $this->xls->setVersion(8); $this->sheet =& $this->xls->addWorksheet('Test'); $this->sheet->setInputEncoding('iso-8859-1'); } private function createRow($cols) { for($col = 0; $col < $cols; $col++) { $content = date($this->colDateMasks[$col], strtotime('2009-01-01 00:00:00') + ($this->row + $col)); $this->sheet->writeString($this->row, $col, $content); } $this->row++; } private function finalize() { $this->xls->close(); } public function test($filename, $cols, $rows, $colDateMasks) { $this->colDateMasks = $colDateMasks; $this->init($filename); for($row = 0; $row < $rows; $row++) { $this->createRow($cols); } $this->finalize(); } } $xls = new xls; $xls->test('test1.xls', 1, 201, array('Y-m-d H:i:s')); unset($xls); $xls = new xls; $xls->test('test2.xls', 2, 137, array('Y-m-d H:i:s', 'H:i:s')); This code produce 2 files without letters. First file broken on row 201. Second on row 137. Both files have a size above 14k My configuration: Linux PHP 5.2.9 PEAR::Spreadsheet_Excel_Writer 0.9.1 Previous Comments: ------------------------------------------------------------------------ [2008-11-24 14:00:20] mek7 valli, thank you very much, you saved me a day and my hair, wow! :) ------------------------------------------------------------------------ [2007-11-20 05:25:32] valli Try out the the patch in http://pear.php.net/bugs/bug.php?id=1572 (Miso's patch) You can also download the patched Workbook.php from http://bugs.gentoo.org/attachment.cgi?id=136412 ------------------------------------------------------------------------ [2007-09-28 05:34:42] vtsuper I got the same problem with the others, I believe the setVersion and setInputEncoding('utf-8') is not work all the time when I set $file->setVersion(8); $sheet->setInputEncoding('utf-8'); and the excel content with Chinese character then it works. but if the excel with only English character then it didn't works. will it has a update version soon? ------------------------------------------------------------------------ [2007-08-23 00:58:55] johnnz Oops, further to my previous comment. I see that _calculateSharedStringSizes in ExcelWorkbook does consider the split for the continue record, but its either not yet implemented or not correctly implemented. I am still looking at that bit. Whichever way, at the moment the result is not correct. ------------------------------------------------------------------------ [2007-08-23 00:44:49] johnnz I have narrowed down the problem. At this stage still trying to figure out the fix. When using unicode and when the string table (BIFF8 0xFC STT) grows bigger than the size limit it will require a continue record (BIFF8 0x3C CONTINUE). The Excel format has some additional rules about splitting strings and these are not considered by the BIFFWrite _addContinue function. It splits the data and breaks the rules. From the OpenOffice.org Documention of the Excel File Format ... Unicode strings are split in a special way. At the beginning of each CONTINUE record the option flags byte is repeated. Only the character size flag will be set in this flags byte, the Rich-Text flag and the Far-East flag are set to zero. In each CONTINUE record it is possible that the character size changes from 8-bit characters to 16-bit characters and vice versa. !Never a Unicode string is split until and including the first character. That means, all header fields (string length,option flags, optional Rich-Text size, and optional Far-East data size) and the first character of the string have tooccur together in the leading record, or have to be moved completely into the CONTINUE record. !Formatting runs (&#10140;2.5.1) cannot be split between their components (character index and FONT record index). If a string is split between two formatting runs, the option flags field will not be repeated in the CONTINUE record ... If you use non-unicode I guess you will avoid this problem. ------------------------------------------------------------------------ The remainder of the comments for this report are too long. To view the rest of the comments, please view the bug report online at http://pear.php.net/bugs/bug.php?id=2942 -- Edit this bug report at http://pear.php.net/bugs/bug.php?id=2942&edit=1

« previous php.pear.bugs (#287) next »