> -----Original Message-----
> From: David B Funk [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, November 18, 2003 11:57 PM
> To: Chris Santerre
> Cc: ML-spamassassin-talk
> Subject: RE: Re[2]: [SAtalk] Sanity checking new uri rules? 
> 
> 
> On Tue, 18 Nov 2003, Chris Santerre wrote:
> 
> > > > uri      WLS_URI_1 /^http:.*\b0-go.org\b/i
> >
> > Regex confusion on my part! '\b' is bounding, but I thought 
> that meant bound
> > by space??? wouldn't this above regex _NOT_ hit :
> >
> > http://stuff.0-go.org/stuff
> >
> > Isn't it looking for:
> > http://stuff. 0-go.org
> >
> > I'm confused! (it's not the first time, won't be the last!)
> >
> > --Chris
> 
> The "\b" match operator is a bit special in that it does not
> match a specific character but the "gap" between two adjacent
> characters. Think of it like the "insertion" cursor of a word
> processor, it points between the characters, not on a character.
> Sort of like "^" points to the beginning of a line, not at the
> first character of the line.
> 
> If you know what the perl "\w" and "\W" character classes are,
> then \b points to the boundary between two characters that are
> matched by either the regex "\W\w" or the regex "\w\W"
> 
> See page 180 of the O'Reilly "Programming Perl" book (Third edition).
> (Good book, written by a guy named Larry Wall ;)
> 
> So that WLS_URI_1 regex is looking for:
> start-of-line, followed by the litteral character string "http:"
> Possibly followed by some number of unidentified characters, the
> last one of which -must- match the "\W" character class
> (note that there could be zero of the above critters as the ":"
> at the end of "http:" nicely matches the "\W" requirement).
> Followed by the litteral character string "0-go", followed by
> one random character (note that "." is a wildcard), followed by
> the litteral character string "org" followed by something that
> matches "\W". (Thus "orga" would not match here).
> 
> Boy, it takes a bunch of words to explain what that little jumble
> of regex does, powerful stuff these regexes ;)
> 
> Dave

All good info, nothing to *snip* :)

Thanks, I got it now. I updated my evilrules last night, and they tested
great overnight! I shall post them shortly. This should speed them up
greatly for everyone! Would this help even more?

/?:\bsomedomain\.com\b/i

would the addition of the ?: make it even faster? 

LOL, the only reason I recognise the name Larry Wall is because of Theo's
sigs! :)
I guess I need to go but that book and help support the man.

--Chris Santerre 


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk

Reply via email to