You need to specify that the string you look for should not appear in the part you try to extract, meaning instead of .*? you should be looking for (not abc)*? In perl, we have the negative lookahed for that: (?!...): m/abc((.(?!abc))*?)xyz/
However, this would fail if you have a string abc abcabc toto toto... because it's a "zero-length" test and therefore it would check the first character after the abc abc and see that it's 1: a, and 2: is followed by a "bc " which is OK. So we need to place the negative lookahead on both sides of the character: m/abc(((?!abc).(?!abc))*?)xyz/ Not a very nice looking regex, but does the trick: $st = "abc abcabc toto toto xyz xyz xyz"; print $1 if $st =~ m/abc(((?!abc).(?!abc))*?)xyz/ PS C:\> perl .\test.pl toto toto On 27 April 2010 10:52, HolyNoob <nhagiaubungbu2...@gmail.com> wrote: > Hi, > > I'm trying to make a regexp to match the last appearance of a word (lets say > 'abc') until the first appearance of another word (for ex: 'xyz'), and I > still cannot do it. > > For example: with a string like this "abc abc abc toto toto xyz xyz xyz" , > which regexp I have to use to get "abc toto toto xyz". > I can extract this "abc abc abc toto toto xyz" with m/abc(.*?)xyz/ > but still got a problem to have exactly what I need. > > Would you guys please help me? > > Thank you very much > -- Erez Support the New Israeli Fund - http://www.nif.org/ -- To unsubscribe, e-mail: beginners-unsubscr...@perl.org For additional commands, e-mail: beginners-h...@perl.org http://learn.perl.org/
