Regex Needed

Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

I'm in need of a regex and hopefully someone can help me out. I'm looping through a text file that contains lines such as the following:

some_text1 some_text2 some_text3 some_text4
some_text1 some_text2 some_text3 some_text4 FOUND

I need a regex that will return some_text4 whenever it finds the word FOUND. In other words, I need the word that comes before a given word in a line of text.

One more thing that might help. Given the example above, some_text3 will always end with a colon (:).

Any suggestions would be much appreciated.

Comment on Regex Needed

Replies are listed 'Best First'.
Re: Regex Needed by Happy-the-monk (Canon) on Mar 25, 2004 at 01:44 UTC
`m/(\w+) (?=FOUND)/g;` have a look at `perldoc perlre` where look-ahead is mentioned. The example there quite matches your problem.	[reply] [d/l]
Re: Re: Regex Needed by TilRMan (Friar) on Mar 25, 2004 at 02:15 UTC
I think the OP has a different definition of "word" than Perl does -- in particular, a word can contain a colon. My guess is that the OP wants a word to be any non-whitespace. `m/(\S+)\s+FOUND/; # The word is now in $1` I don't see the need to use lookahead for this regex. -- LP^>	[reply] [d/l]
Re: Re: Regex Needed by ysth (Canon) on Mar 25, 2004 at 02:07 UTC
That would work, but if the goal is to have the entire match ($&) be the desired word, it should be `m/\w+(?= FOUND)/g` [download] If capturing into $1 is the goal, `m/(\w+) FOUND/g` [download] is all that is needed.	[reply] [d/l] [select]