preg_replace problems with mac vs pc

whaites

I'm trying to create a process for users to download content from a database, edit the content in MS Word as an HTML document (to maintain a minimum of formatting) and then upload the document back to the database. Needless to say, I would like to strip much of the Word generated HTML before posting the data back to our mysql db. I created a series of scripts to do just this and it worked fine with a Mac, but not with a PC. I compared the resulting MS Word HTML -- they look identical. In fact, if I use BBEdit to save the html generated on a PC as a Mac file type, everything works. I'm pretty confused, considering the server where the scripts are executing is running FreeBSD.

This is an example of an expression which does not work correctly on a PC but runs fine on a MAC (it should strip everything between the head tags, including the head tags):
$text = preg_replace("/<head[^{>]?>.?<\/head>/si","",$text);}

I also noticed that the strip_tags function behaves in a similar fashion (on a PC it does not seem to strip the tag attributes, but it does so on a Mac).

Any ideas on what might be going on? It really sucks to have code which only works on Macs when 99% of the office has PCs.

Sxooter

This is almost certainly a CR/LF versus LF problem.

On Macs and Unix boxen, a LF is used as an end of line marker alone, on Windows boxen, there's a CR/LF pair.

You should be able to use a replace function to replace /r/n with /n and get it to work

whaites

I did a text replace to print the CR/LF comands and compare the MAC vs. the DOS files--the Mac file only shows \r and the Dos file shows \r\n. But, when I do a replace and either turn all the \r\n to \n (since the Mac version works), or simply kill all the \n (or all the \n and all the \r), the preg_replace STILL only works on the Mac version. Are there other hiddens I can check for--I didn't find any \t chars.

My modification, which DID NOT work:

$text=str_replace("\r\n","\r",$text);
$text = preg_replace("/<head[^{>]?>.?<\/head>/si","",$text);}

the alsofails:
$text=str_replace("\r","",$text);
$text=str_replace("\n","",$text);
$text = preg_replace("/<head[^{>]?>.?<\/head>/si","",$text);}

$text=str_replace("\n","",$text);
$text = preg_replace("/<head[^{>]?>.?<\/head>/si","",$text);}

$text=str_replace("\n\r","\r",$text);
$text = preg_replace("/<head[^{>]?>.?<\/head>/si","",$text);}

whaites

HELP! This does not seem to be a line break problem.

Thanks,
Tom

whaites

It was a line break problem--just not with preg_replace() but fgets().

Sorry.

namor

Would you care to post how you fixed the problem? I am running into the same thing.
I cannot parse through my mac formatted text files properly.

whaites

Can you post some more specifics?
Maybe I can remember what was going on in my instance.