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

Hi Monks, Here's yet another simple task that is confounding me. I sent out an emailer to a list of clients. It had been a while sine the owner had used the list, so not suprisingly, there were a lot of bounces. I congealed the bounced addresses into a file bad_emails from the address book, and wrote a little ditty to turn the paragraph style collage into a nice neat column. I also will have to do dup removals, once I get the (explicitive) thing to match more than one email address per line.

The next task is to compare the list to the original list and remove the matches: raw list - bad list = good list.

But before I can do that, I need to fix my regex, as it seems to only match the first one in each line. I obviously don't have the grasp of /g I thought I did. 

 push (@list, "$1\n") and $count++
  if /\b([\w]+@[\w]+\.[\w]{3})\b/g;

[download]
Why am I only getting 1 match per line? Heres a smaple line from the data file:

wings@yahoo.com <wings@yahoo.com>; wine@hotmail.com <wine@hotmail.com>;

Replies are listed 'Best First'.
Re: scripty email address regex stuff
by chromatic (Archbishop) on Oct 15, 2003 at 20:59 UTC

    I think you mean while and not if, since there's a loop in the former and not in the latter. You might have more luck with the Mail::Address module, though.

[reply]
[d/l]
[select]
      Thanks, chromatic. You were right on both counts. mail::address is pretty easy to use, though the output is not coming out the way I'd hoped. Am I doing this right? 
      #!/usr/local/bin/perl
use Mail::Address;
my $file = shift;
open (FILE, $file) or die "$@ Couldn't open file";
my @contents = <FILE>;


foreach my $line (@contents) {

              my @addrs = Mail::Address->parse($line);
    foreach $addr (@addrs) {
        print $count++, $addr->address,  "\n";
    }
 }

      [download]

      I'm looking for just the address (and the count), but names, commas, and semicolons are coming out as individual results.

[reply]
[d/l]
Re: scripty email address regex stuff
by Abigail-II (Bishop) on Oct 15, 2003 at 21:30 UTC
    I typically create dedicated addresses when I sign up to something of the form $name$@abigail.nl. Perfectly legal address and it seldom leads to problems.

    But your regex doesn't match it.

    Abigail

[reply]
[d/l]
Re: scripty email address regex stuff
by Beechbone (Friar) on Oct 16, 2003 at 11:38 UTC
    Do you want to parse the addresses or just extract them from the file? In the second case, why not just: 
    local $/ = undef;
my $raw = <FILE>;
$raw =~ s/[\r\n]//g;
$raw =~ s/;\s+//g;
my @list = split /;/, $raw;

    [download]
    and maybe: 
    foreach (@list) {
  s/^[^<>]+<([^<>]+)>$/$1/;
  # or:
  s/^[^<>]+<([^<>\s]+@[^<>\s]+)>$/$1/;
}

    [download]
    but the second regex also throws away some legal addesses like "user!host1!host2!host3"
    Search, Ask, Know
[reply]
[d/l]
[select]
      Did you actually try this
[reply]
        Sed and uniq were really helpful for creating the final list
[reply]

Back to Seekers of Perl Wisdom