Re: Unexplained Issue Using Regex

On Fri, Mar 6, 2009 at 4:01 PM, Nitsan Bin-Nun <nitsan@binnun.co.il> wrote: > On Fri, Mar 6, 2009 at 11:53 PM, haliphax <haliphax@gmail.com> wrote: >> >> On Fri, Mar 6, 2009 at 3:44 PM, Nitsan Bin-Nun <nitsan@binnun.co.il> >> wrote: >> > I'm not looking for other ideas, the main thing here is that I have >> > about >> > 30-100 regex's in the database and the script fetches them and applies >> > them >> > to the string. I can't build again the engine and I'm not going to do >> > that. >> > I'm trying to solve my problem ;) If you have any ideas regarding my >> > issue >> > and not going in another way this would be very appreciated. >> >> Nitsan, >> >> I think it's because you're referencing the capture group with index >> instead of index 2. Also, I don't understand why you have the pipe >> ("|") character in your regex string... is that part of your engine? >> >> This code: >> >> $orig = >> 'http://www.zshare.net/video/541070871c7a8d9c'; >> $matches = array(); >> >> preg_match('#http://(www\.)zshare\.net/video/([^/]+)#', >> $orig, $matches); >> echo $matches[2]; >> >> Grabs the correct match: >> >> 541070871c7a8d9c >> >> The regex pattern works with the pipe char, but it is unnecessary and >> may lead to some strange behavior. > > Thank you Todd, I also want to capture when I don't have the www in the > beginning of the URL. > For instance, try to execute your code with > $orig = 'http://zshare.net/video/541070871c7a8d9c'; > > That's why I used (www\.|), but I'm not a regex expert and I'm sure there a > way better solutions to this problem. http://www.regular-expressions.info is your best friend. Spend an afternoon playing around on it... that's really the only advantage I have over someone who hasn't. Anyway, you can make that entire group optional with the ? character like so: #http://(www\.)?zshare\.net/video/([^/]+)# And if you don't want it to be captured, making the URL suffix index 1 instead of index 2, do this: #http://(?:www\.)?zshare\.net/video/([^/]+)# Any group that begins with "?:" will not be captured in a match index. To recap: $pattern = '#http://(?:www\.)?zshare\.net/video/([^/]+)#'; $orig = 'http://zshare.net/video/541070871c7a8d9c'; $matches = array(); preg_match($pattern, $orig, $matches); echo $matches[1] . "\n"; $orig = 'http://www.zshare.net/video/541070871c7a8d9c'; preg_match($pattern, $orig, $matches); echo $matches[1] . "\n"; Produces this output: 541070871c7a8d9c 541070871c7a8d9c Hope this helps, -- // Todd