very quick link regex

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

Replies are listed 'Best First'.
Re: very quick link regex by Thelonius (Priest) on Aug 03, 2007 at 09:09 UTC
You have `[a..z]` where you should have `[a-z]`. You also have 'my' in front of `$content2`, which is wrong if you are trying to match a string which is already in $content2. If you are trying to match against $_ and put the results in $content2, you should write: `my $content2 = m#(yourregexhere)#i;` [download] or, maybe clearer, `my $content2 = ($_ =~ m#(yourregexhere)#i);` [download] The construct (.+)+ doesn't really make sense. Just .+ would be okay, but you have to worry about matching too much. One possibility is `.+?`, but that's not the most efficient code and you could still get false matches in some cases, such as if the source text was, e.g.: `http://www.page.com/files4/i/90c/93898.gif is not a .jpg file` [download] You might want `m#(www\.page\.com/files\d+/[a-z]/\d\S+\.jpg)#i` [download] or, if this is a real HTML link, `m#(www\.page\.com/files\d+/[a-z]/[^"']+\.jpg)#i` [download]	[reply] [d/l] [select]
Re: very quick link regex by Zaxo (Archbishop) on Aug 03, 2007 at 04:35 UTC
If you're parsing HTML, try HTML::LinkExtor. If text, Regexp::Common::URI. The former is a parser, the latter a well-tested regex. After Compline, Zaxo	[reply]
Re: very quick link regex by GrandFather (Saint) on Aug 03, 2007 at 04:55 UTC
Are you trying to match, validate or extract parts? `my $url = 'http://www.page.com/files4/i/90c/1493010807t1.jpg'; my ($domain, $path, $file) = $url =~ m'^(\w+://[\w.]+/)(.*?/)([\w.]+)$ +'; print "$domain\n$path\n$file\n";` [download] Prints: `http://www.page.com/ files4/i/90c/ 1493010807t1.jpg` [download] which may or may not be anything like what you need to do. DWIM is Perl's answer to Gödel	[reply] [d/l] [select]
Re: very quick link regex by Anno (Deacon) on Aug 03, 2007 at 11:40 UTC
Other monks have provided possible solutions to your problem. I only want to comment on the subject of your query. I may be lacking in monkish indulgence, but I resent it when a question is presented as "very quick" (or "easy", or whatever). Since you don't have a solution, how do you know it's "quick"? Anno	[reply]
Re: very quick link regex by Anonymous Monk on Aug 03, 2007 at 04:31 UTC
actually after doing a little more research I found that the link is MORE dynamic than I thought. I need to match anything after page.com/files\d+... until after the .jpg . No other part of the page has /files\d+ so if it finds a match, it'll always be the right one. Can someone help with this?	[reply]
Re^2: very quick link regex by varian (Chaplain) on Aug 03, 2007 at 07:40 UTC
You were very close to result already. `/(.+)+\.jpg` [download] In this last part of your regexp the dot-plus will match all up to the end of line. The remaining '.jpg' part will not be found and therefore the regexp never matches. The solution is to make the dot-plus part of the regexp less greedy, change the part above to: `/(.+?)\.jpg` [download]	[reply] [d/l] [select]
Re^2: very quick link regex by atemon (Chaplain) on Aug 03, 2007 at 05:16 UTC
try `my $content2 =~ m#(www\.page\.com/files\d+/.+\.jpg)#i;` [download] Note: the code is NOT tested. Cheers ! --VC There are three sides to any argument..... your side, my side and the right side.	[reply] [d/l]