Please provide the prolematic cvs file and shortest possible script that can be used to replicate the problem.

Hi,

I just wanted to send you some additional information about the way I tested:

I tested importing a File on 3 Systems:
- 2 times php4.3.3
- 1 time php4.3.5-dev

I used one file on all these tests and every time I got the problem with fgetcsv which got solved using gets and explode.

Using another system with php4.3.3, same data structure but other data in 30,000 rows, I have never had a problem. So it could be the data or maybe fgetcsv is not binary safe.

I didn't have the problem some days ago under same circumstances with nearly the same data and size. I spent hours in trying to reproduce why the error occurs, but I can't.

Maybe a look into the PHP source code helps you more than I can at the moment.

I can't give you the original file as I already wrote.

But I'll spend some minutes in creating a test script.

Best regards,
Thomas

#!/usr/local/bin/php
<?php

$file_name="pks.dat";

	$fp=fopen($file_name,"r");
	while(feof($fp)===False)
	{
		// The comment-lines should work instead of fgetcsv.
		// $new = fgets($fp,4096);
		// $new_field_values=explode("|",$new);
		$new_field_values=fgetcsv($fp,4096,"|");
		$zaehler++;
	}
	fclose ($fp);
	print $zaehler."\n";
?>

Hi,

using fget and explode (like written in the script above) $zaehler returns 20967.

Using fgetcsv it returns 17861 reading the same file.

I am really sorry, I can't provide the file.

Best regards
Thomas

Hi, me again...

in case I consider it correctly, fgetcsv is sometimes
- reading and processing a line, and twice jumping to the next line 
- or ignoring one line and processing the next one.

best regards
thomas

fgetcvs() can support multi-line rows, which may explain why a file with X number of lines when read with fgetcvs() would return less lines (but equal amount of total data).

If you cannot provide the original cvs file, could you provide an equivalent that can replicate the problem. Without a cvs file it's nearly impossible for us to resolve this problem.

Hi iliaa,

one thing, I did was to do a count on the array of fgetcsv.

In result it was alwas 18 values per line, so there is less data returned than data existing in the file, i think.

best regards
thomas

The reason you get unequal amount of lines is because you use an un-escaped delimiter ("), which causes multiple lines to be parsed a single value.

There is an unrelated bug in fgetcsv() in PHP 4.X, to which I am testing a patch now.

This bug has been fixed in CVS.

In case this was a PHP problem, snapshots of the sources are packaged
every three hours; this change will be in the next snapshot. You can
grab the snapshot at http://snaps.php.net/.
 
In case this was a documentation problem, the fix will show up soon at
http://www.php.net/manual/.

In case this was a PHP.net website problem, the change will show
up on the PHP.net site and on the mirror sites in short time.
 
Thank you for the report, and for helping us make PHP better.