<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8">
<META NAME="GENERATOR" CONTENT="GtkHTML/3.18.3">
</HEAD>
<BODY>
On Tue, 2008-10-28 at 08:13 -0500, B. Estrade wrote:<BR>
<BR>
<BLOCKQUOTE TYPE=CITE>
<PRE>
> >
> > You do need the /g at the end, of course.
Do you or don't you? I am not familiar with using the "g" switch in a pure match - I usually just use it when doing global search and replaces.
</PRE>
</BLOCKQUOTE>
If /g was not used then it would only match once.
<BLOCKQUOTE TYPE=CITE>
<PRE>
David:
Is this for Apache? If so, you are treading on well-worn ground -
<A HREF="http://www.google.com/search?q=perl+parse+apache+log+file&ie=utf-8&oe=utf-8&aq=t&rls=com.ubuntu:en-US:unofficial&client=firefox-a">http://www.google.com/search?q=perl+parse+apache+log+file&ie=utf-8&oe=utf-8&aq=t&rls=com.ubuntu:en-US:unofficial&client=firefox-a</A>
</PRE>
</BLOCKQUOTE>
<BR>
I wish it was, that would be easier ;)<BR>
This is how the Astaro Firewall spits out the http proxy logs.
<BLOCKQUOTE TYPE=CITE>
<PRE>
Also, if you want to analyze your log files, you may want to check out AWStats - <A HREF="http://awstats.sourceforge.net/">http://awstats.sourceforge.net/</A>.
Lastly, you can try an approach that essentially parses parts of the file in parallel. I am not familiar with writing multi-threaded Perl scripts, but that would allow you to get further speed-up once you've found the magic regex to use. Of course, you might have to deal with bringing back the results in some ordered way, so it is a rather advanced approach to take.
</PRE>
</BLOCKQUOTE>
Yep, we use AWStats for the web sites on Plesk.<BR>
I'll be alright with the regex approach. Essentially, I wrote a parser in bash a while back that matches keywords and excludes certain url's in order to see who's browsing pron @ work. I've been too lazy to redo it right in Perl. Just getting around to it. ;)<BR>
<BR>
On Tue, 2008-10-28 at 08:16 -0500, B. Estrade wrote:<BR>
<BLOCKQUOTE TYPE=CITE>
<BR>
<BR>
You should check out a fairly old language called APL - <A HREF="http://en.wikipedia.org/wiki/APL_programming_language">http://en.wikipedia.org/wiki/APL_programming_language</A><BR>
<BR>
People are saying that Perl is becoming more LISP-like and more APL-like :).<BR>
<BR>
</BLOCKQUOTE>
Now that's intriguing. LISP to. I wonder how much LISP T2 has? ;)<BR>
<BR>
David
</BODY>
</HTML>