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

The problem of sorting a file full of nothing but HTML mailto: links came across my desk, and armed with Perl I naturally solved it a few minutes (a miracle for me, this stuff is finally starting to stick). Then I noticed that not all of the links in the file actually contained mailto: even if they were linking to email addresses. So now I'm stuck trying to change my regex to match either the entire word "mailto:" or nothing at all. My best guess is commented below. Can anyone offer me advice? (And yes, I have read Death to Dot Star! even if I don't fully comprehend it yet.) 
#!/usr/bin/perl -w
use strict;

my %names;

while ( <> ) {
    s/\s+/ /ig;
    /<a href="(.*)">(.*)<\/a><br>/;
    # /<a href="(mailto:)??(.*)">(.*)<\/a><br>/;
    $names{"$2($1)"} = $1;
}

open OUT, ">sortboater.txt" or die "Cannot open file: $!";

for ( sort keys %names ) {
    /^(.*)\(.*\)$/;
    print OUT qq(<a href="mailto:$names{$_}">$1</a><br />\n);
}

close OUT or die "Cannot close file: $!";

[download]
-- More than perfect! Let us engage the Concord!

Replies are listed 'Best First'.
Re: Problems with regex grouping
by abstracts (Hermit) on Aug 17, 2001 at 21:45 UTC
    Hello

    The regexp you want is: 

    my @ar = (); # empty array
while(<>){
   #for every match in the line 
   #(if there are more than one email in a line)
   while(/<a href="(mailto:)?([^"]*)">([^<]*)<\/a>/g){ 
      push @ar, "$3\t$2";
   }
}
open ...
for(sort @ar){
   my ($name, $email) = split/\t/;
   print OUT qq(<a href="mailto:$email">$name</a><br />\n);
}

    [download]

    Update: If you have only one link perl line, you can do this: 

    while(<>){
   if(/<a href="(mailto:)?([^"]*)">([^<]*)<\/a>/){
      push @ar, "$3\t$2";
   } else {
      warn "Err: $_\n";
   }
}

    [download]

    Hope this helps,,,

    Aziz,,,

[reply]
[d/l]
[select]
      To avoid throwing stuff in $1 when you don't need to, there is a great feature called clustering in REs with the (?:stuff here) stuff :). Just change 
      /<a href="(mailto:)?([^"]*)">([^<"]*)<\/a>/g

      [download]
      to

      /<a href="(?:mailto:)?([^"]*)">([^<"]*)<\/a>/g

      And that way nothing will be put in $1.

      You may also want to consider chomping  $2 and $3 because a newline could get in there somewhere and cause some trouble.

      Hope I helped out. :):):) Later 

      $_.=($=+(6<<1));print(chr(my$a=$_));$^H=$_+$_;$_=$^H;
print chr($_-39); # Easy but its ok.

      [download]
[reply]
[d/l]
[select]
(dkubb) Re: (1) Problems with regex grouping
by dkubb (Deacon) on Aug 17, 2001 at 23:29 UTC

    Have you thought about taking a different approach and using CPAN modules to do the work for you? Since you are searching for email addresses in a string, this looks like the perfect job for Email::Find.

    Here is some working sample code to demonstrate how it's used.

[reply]

Back to Seekers of Perl Wisdom