pcre regex

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

Replies are listed 'Best First'.
Re: pcre regex by GrandFather (Saint) on Apr 06, 2012 at 10:30 UTC
Show us some sample data that fails and some test code. If the problem is real you shouldn't need more than a few lines of data and a few lines of code to reproduce the problem. Are you sure the pattern you want to match is at the start of the line? Could it be that you are reading the entire file into a string and are then trying to match against that string? If so you probably need to use the m (multiple line match) switch. Consider: `use strict; use warnings; my $str = <<TEXT; First line 1234X last line TEXT print "Matched\n" if $str =~ /^\d{4,4}[A-Z0-9]$/; print "Multi-matched\n" if $str =~ /^\d{4,4}[A-Z0-9]$/m;` [download] prints: `Multi-matched` [download] True laziness is hard work	[reply] [d/l] [select]
Re^2: pcre regex by Anonymous Monk on Apr 06, 2012 at 10:41 UTC
Yes have to search in multilines so i used /m but it still does not give any matches. I think the problem is with ^ and $ as only regex having these identifiers are giving me problem rest ase working fine such as for email validation /[a-z0-9!#$%&'+/=?^_`{\|}~-]+(?:\.[a-z0-9!#$%&'+/=?^_`{\|}~-]+)@(?:[a-z0-9](?:[a-z0-9-][a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?/g	[reply] [d/l]
Re: pcre regex by moritz (Cardinal) on Apr 06, 2012 at 10:32 UTC
The ^ and $ around the regex mean that it only matches a whole string. If you want to search for matches inside a string, leave out those anchors. See perlretut for more details. Perl 6 - second systems done right	[reply]
Re^2: pcre regex by Anonymous Monk on Apr 06, 2012 at 10:56 UTC
how do i substitute ^ and $ in my regex	[reply]
Re^3: pcre regex by moritz (Cardinal) on Apr 06, 2012 at 11:28 UTC
Well, what semantics do you want? If you just want to remove the restriction that the regex has to match the whole string, remove them. Perl 6 - second systems done right	[reply]
Re^4: pcre regex by Anonymous Monk on Apr 06, 2012 at 11:36 UTC
Re^5: pcre regex by moritz (Cardinal) on Apr 06, 2012 at 11:40 UTC
Re^5: pcre regex by bart (Canon) on Apr 06, 2012 at 12:00 UTC
Re: pcre regex by nemesdani (Friar) on Apr 06, 2012 at 10:22 UTC
Post your code with which you read the text file, maybe we can work out someting. Oh, and `d{4,4} == d{4}` I'm too lazy to be proud of being impatient.	[reply] [d/l]
Re^2: pcre regex by Anonymous Monk on Apr 06, 2012 at 10:28 UTC
When i give it user input it matches the regex `print "Please enter the string "; chomp($string=<STDIN>); $regex= "^\d{4,4}[A-Z0-9]$ "; if ($string =~ m!($regex)!g) { print 'match'; print "$regex"; } else { print 'no match'; print "$regex"; }` [download] It matches the regex in this case but when the input is a text file it fails to recognize any patterns	[reply] [d/l]
Re: pcre regex by Anonymous Monk on Apr 06, 2012 at 12:48 UTC
The lines from the text files may be ending with '\r\n' and '$' matches before an optional '\n'. Opening the file in the '<:crlf' mode, or adding an optional "\r" into regex (`/^\d{4}[A-Z0-9]\r?$/`), may help.	[reply] [d/l]