[dokuwiki] Re: preg_match() Compilation failed: regular expression too large...

Hi all,
I didn't remember doing something weird with the lexer. Anyways, the lexer is always called with that function p_get_instructions($text). In Jaume's case, $text is always filled by rawWiki($id). But I guess, the stinky thing is for each link found in the main document p_get_instructions(rawWiki($id)) is called with the link id. 

I believe that I didn't clean up the instructions results set, properly.

Few month ago a ask for that :
http://www.freelists.org/archives/dokuwiki/09-2006/msg00299.html

Harry, what do you think about that ?

Cordialement,
Danjer

Harry Fuecks wrote:

From a brief scan of the code (found at
http://danjer.doudouke.org/tech/dokutexit) in particular
class.texitrender.php, it invoking the parser directly and _may_ have
passed something to the lexer which depends on the input document. I'd
discuss this with the author...

On 5/22/07, Jaume Obrador <obrador@xxxxxxxxxxxx> wrote:

The only plugin I have installed and it's used when this error occurs is
dokutexit, which parses dokuwiki syntax into LaTeX to generate a PDF
file.

Thnks.
Jaume Obrador.


El dl 21 de 05 del 2007 a les 23:47 +0200, en/na Harry Fuecks va
escriure:
> On 5/21/07, Jaume Obrador <obrador@xxxxxxxxxxxx> wrote:
> > Hi, I try to use dokuwiki together with dokutexit plugin to generate a 
> > PDF file from some pages. I may say that it works great with a few
> > pages, but the problem comes when I try to generate a PDF from a larger > > number of pages, total Kb of those pages together are 209K. I get the 
> > following error 22 times:
> >
> > Warning: preg_match() [function.preg-match]: Compilation failed: 
> >         regular expression too large at offset 0
> > in /var/www/lledoner/dokuwiki/inc/parser/lexer.php on line 115 
> >
> > I increased the default php memory_limit from 8M to 32M, with no good 
> > results.
>
> That's a limitation of PCRE not PHP - see http://www.pcre.org/pcre.txt
> section "LIMITATIONS": "The maximum length of a compiled pattern is
> 65539 (sic) bytes  if  PCRE
>        is compiled with the default internal linkage size of 2."
>
> It goes on to say you can compile PCRE differently to get round this
> but that will probably get interesting with PHP, where PCRE is part of
> the core distribution.
>
> Without knowing more I'd say this problem _isn't_ directly caused by
> trying to generate a PDF from many pages, because the size of the
> parser regex should not be dependent on the size of the document(s)
> you are parsing.
>
> To have this error possible causes might be;
>
> - a very large smileys or acronyms file - last time I checked these
> get turned into a single regular expression so a big input file could
> lead to a regex > 65539 bytes. Normally though if the problem was
> here, you'd see it on every page, not just the PDF version
>
> - a plugin which _IS_ using the document to build further regular
> expressions, in some way: list your plugins...
>
> Hope that helps

--
DokuWiki mailing list - more info at
http://wiki.splitbrain.org/wiki:mailinglist


--
DokuWiki mailing list - more info at
http://wiki.splitbrain.org/wiki:mailinglist

Other related posts: