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

Hi Perlmonks.
I'm very confused with regex..:@

The partial of the input file looks like the line below: 
>sp|P49929|SC52_SHEEP 
DITCAEPQSVRGLRRLGRKIAHGVKKYG
>gi|1771590|emb|CAA70562.1| temporin B precursor MFTLKKSLLLLFFLGTINLSL
+CEEERNAEEERRDEPDERDVQVEKRLLPI

[download]

I only want to get first 2 values separated by a |
Output expected: 
sp == P49929
gi == 1771592

[download]
I try using the regex below: 
 ##$lines contains the string to be matched 
 if(/^>(\w+)\|(.*)?\|.*?$line.*\n/i){
         $flag=1;
         print OUT "$_";
         print "$1 == $2\n";    
}

[download]
It will print
sp == P49929 as expected
gi == 1771592|emb|CAA70564.1 unexpected
Any suggestion??
Thank you in advanced.

Replies are listed 'Best First'.
Re: get some part of the string using regex
by davis (Vicar) on May 17, 2002 at 09:10 UTC
    Hi,
    If I were you, I wouldn't use a regex to split the input data: 
    #!/usr/bin/perl -w

use strict;
while(<>) {
        chomp;
        next unless($_ =~ /\|/);                   #Skip if no pipe sy
+mbol
        my @fields = split '\|', $_;
        $fields[0] =~ s/^>//;                      #Remove leading ang
+le-bracket
        print $fields[0], " == ", $fields[1], "\n";
}

    [download]
    Cheers

    davis
    Is this going out live?
    No, Homer, very few cartoons are broadcast live - it's a terrible strain on the animator's wrist
[reply]
[d/l]
      Of course when using split you can always limit the number times the string gets split by adding the third parameter: 
      my $foo = 'abc|def|ghi|jkl|mno';
my @bar = split('|', $foo, 3);

      [download]
      Since you are only interested in the first two fields.
[reply]
[d/l]
Re: get some part of the string using regex
by jmcnamara (Monsignor) on May 17, 2002 at 09:11 UTC

    The problem is that your regex is "greedy": it is matching as far ahead as possible. A negated character class in the regex is one way of dealing with this: 
        #!usr/bin/perl -w

    use strict;

    my @lines = qw(
        >sp|P49929|SC52_SHEEP 
        DITCAEPQSVRGLRRLGRKIAHGVKKYG
        >gi|1771590|emb|CAA70562.1| temporin B precursor
    );

    foreach (@lines) {
        if (/^>(\w+)\|([^|]+)/){
            print "$1 == $2\n";    
        }
    }
    
    __END__
    Prints:

    sp == P49929
    gi == 1771590

    [download]

    Also, if the second field only contains word charcters you could use a simpler match:     /^>(\w+)\|(\w+)/

    --
    John.

[reply]
[d/l]
[select]
Re: get some part of the string using regex
by Dog and Pony (Priest) on May 17, 2002 at 09:13 UTC
    If you want a non-greedy dot-star, you need to replace (.*)? with (.*?).

    Maybe you should even consider something like \|([^|]*)\| to be sure you can't cross the boundary of the pipes (|).

    You have moved into a dark place.
    It is pitch black. You are likely to be eaten by a grue.
[reply]
[d/l]
[select]
      Thanks...
      You save me...
[reply]

Back to Seekers of Perl Wisdom