Excluding a Whole Word with Regular Expression

[gratuitous]

[Usacomp2k3]

Do you mean like a google search? Then it would just be "-fox" however, that's too easy, so I guess I don't understand the question.

[gratuitous]

[Usacomp2k3]

Quite a bit over my head then. Good luck

[notfred]

Probably do better in Developer's Den.

If you can call out to grep then do it as a "grep -v" that's probably the easiest way.

[morphine]

(.*?)dog(.*)

This you capture everything before "dog" and after "dog", essentialy giving you the same result. I presume you're using this inside some sort of program for text search/matching.

[gratuitous]

[morphine]

The given regular expression was meant to be used in some sort of program (it's a Perl-compatible regular expression, although this particular case is not Perl-specific).

If you're looking for a usenet post in some page, then I'm betting that page has some sort of specific form of regexp query, which is useless for us trying to guess what it is

What's the specific site?

[gratuitous]

[morphine]

Someone named Jeff on ICQ left this to me, so I thought I'd share it with you (maybe Jeff is the poster, I don't know )

In answer to this thread that
you posted in:

viewtopic.php?t=34782

Here is the regex that solves the problem:

http://www.codinghorror.com/blog/archives/000425.html

Could you post it in the thread?

Thanks,

Jeff

[thegleek]

cat filename.ext | grep -v yourwordgoeshere

[gratuitous]

[JavaDog]

Just code your own usenet reader...

[Captain Ned]

Hmm, this is really looking like code jockey stuff, so off to The Developer's Den it goes.

[morphine]

Well, that's life

But yeah, you should use your own usenet reader instead of a third-party solution.

[gratuitous]

[SecretSquirrel]

The perlish regexp would be:

Code: Select all

$string="The quick brown fox jumped over the lazy dog.";
$search="fox"
$string=~s/\s$search(\s)/$1/g;


If will search for the search term surrounded by a whitespace character and strip it out. The parens around the second whitespace save it and it is actually used as the replacement source. This keeps you from getting "brownjumped" by removing both whitespace elements. You should actually extend it to take into account it being the first in the sentance or being followed by punctuation or stuff.

--SS

[jimveta]

the standard way of specifying a "word" regex is

Code: Select all

\<fox\>

(it will match "fox" surrounded by spaces, punctuation, etc)
.. now all you need is a way to negate that which unfortunately isn't so standardized.. does the webform page you're using this in have a manual?