Your code ignores the first two columns, because the columns are numbered from zero.

I don't quite understand your overall goal, but this should help:

#!/usr/bin/perl -w
use strict;
use Text::CSV::Simple;

my $parser = Text::CSV::Simple->new;
$parser->want_fields(1, 2, 3, 4);
my @data = $parser->read_file(\*DATA);
foreach my $line (@data)
{
   if ($line->[0] =~ m/SMTP:/i)
   {
      print "here are the parts of the first column:\n";
      my @parts_of_first_column = split(';', $line->[0]);
      foreach my $part (@parts_of_first_column)
      {
         print " * $part\n";
      }
   }
}
__DATA__
"CN=MBX_MANA,OU= Mailboxes,DC=doman,DC=com",SMTP:Manager@domain.com;sm
+tp:MBX_MANA@domain.com;FAX:MBX_MANA@domain.com;X400:c=us\;a= \;p=doma
+in\;o=Exchange\;s=MANAGER\;,MBX_MANA,/o=Exchange Org/ou=First Adminis
+trative Group/cn=Configuration/cn=Servers/cn=SERVER
"CN=Guest,CN=Users,DC=domain,DC=com",,Guest,
[download]

Which prints out:

here are the parts of the first column:
 * SMTP:Manager@domain.com
 * smtp:MBX_MANA@domain.com
 * FAX:MBX_MANA@domain.com
 * X400:c=us\
 * a= \
 * p=domain\
 * o=Exchange\
 * s=MANAGER\
[download]

So Text::CSV::Simple splits up the CSV in a smart way, and then the code splits up the first (remaining) column by semicolons. From there on you should be able to test each part for the stuff you want.

Apologies for the typo with my list of fields - I mistakenly pasted from old code - I've got the correct (1,2,3) in my new stuff, although I was pulling my hair out there for a while.

That is helpful, thank you. So, the parts of the line are line[0], line[1], line[2] when they're divvied up by Text::CSV::Simple. You've dumped line[0] into an array to break up its components, that's great.

So I can manipulate line[1] and line[2] to grab the other parts I need. In that loop, I should be able to output something line-by-line like (pseudocode) "@parts[0]    line[1]@line[2]", which is effectively "SMTP_ADDRESS    USER@SERVER". That's fantastic, I was imagining all kinds of wierd structures in my head, not a simple and understandable one.

I was imagining all kinds of wierd structures in my head, not a simple and understandable one.

That's why they invented Data::Dumper.

I'd be lost without it.

Just add use Data::Dumper; and do print Dumper(\$something); and it makes things a lot easier. That's how I figured out what you data was doing.

---




($_='kkvvttuu bbooppuuiiffss qqffssmm iibbddllffss')
=~y~b-v~a-z~s; print

The @data that you get from Text::CSV::Simple is an array of array references. That means that each item in the array is itself a reference to an array composed of three fields. That means you need to derefernce it to get at the fields like this:

for my $row(@data){
    my($smtp,$user,$server) = @$row;
    # munge $stmp
    next unless $smtp;
    # munge $user and $server
    # recombine them into a string
    # write the string to a file
}
[download]

No, that's correct - I was getting my knickers in a knot trying to figure out when I do the munging and output to a file. So what I essentially need to do here is read a row at a time, fiddle around with it and output it before working on the next line. I knew that logically, but couldn't figure out the forest for looking at the trees.

Thank you for breaking down what I need to do structurally - I have real problems with that aspect of coding (infrequent use). I think I have enough from you and Cody Pendant to be able to figure it out from here.

"The @data that you get from Text::CSV::Simple is an array of array references. That means that each item in the array is itself a reference to an array composed of three fields. That means you need to derefernce it..."
This is need to know information that isn't made that clear in the module docs. I've annotated the CPAN module with a paraphrase of your explanation.
Thanks,
I have been butting my head for a few minutes trying to get some code using this module to work. Update: this reply was intended for jZed, I fat fingered the post.

#! /usr/bin/perl
#parsing Exchange output file to create address aliases
use strict;
use warnings;

my($infile) = ("C:\\recipient.txt");
my($outfile) = ("C:\\outfile.txt");

use Text::CSV::Simple;

open(OUTPUT, ">$outfile") or die "Can't open $outfile : $!";

my $parser = Text::CSV::Simple->new;
$parser->want_fields(1, 2, 3);

my @data = $parser->read_file($infile);
foreach my $line (@data)  {

# process only accounts that contain SMTP addesses                    
+               
    if ($line->[0] =~ m/SMTP:/i)    {    
        my $user = $line->[1];        
        my $server = substr(($line->[2]), -6);
    
#split up the line that contains all the email addresses
              my @smtp = split(';', $line->[0]);        

        foreach my $part (@smtp)    {
        
# do not process anything that is not an SMTP address
# (exclude fax & X400)
            if ($part =~ m/SMTP:/i)   {
                $part = substr($part, 5);
                print OUTPUT ("$user\@$server \t $part\n");
            }
        }
    }
 }
 
 
 
# CSVDE command used on domain controller to 
# generate recipient information for users in domain
# Dumps the user name, all proxy addresses 
# and the home mailbox server - it WON'T exclude the DN.
 
# csvde -m -n -g -f "C:\recipient.txt" -r "(|(&(objectClass=user)(obje
+ctCategory=person))(objectClass=groupOfNames)(objectClass=msExchDynam
+icDistributionList))" -o "distinguishedName, DN" -L "name,proxyAddres
+ses,msExchHomeServerName"
[download]

Here is my code that uses Text::CSV::Simple; the goal was to parse a NYSE price list from the Wall Street Journal.

use strict;
use Text::CSV::Simple;

#usage wsj2csv AMEX091806.txt

my ($filename) = shift;
print $filename."\n";
print "What is the trading date(MMDDYY)?";

my $traddt = <STDIN>;
print $traddt."\n";

my $parser = Text::CSV::Simple->new;
$parser->want_fields(1, 2, 3, 4, 5, 6, 8);

       my @data = $parser->read_file($filename);
       my $datalines = @data;
       print $datalines."is number of rows.\n";
       for my $row(@data) {
       my @line = @$row;
       foreach my $row1(@line) {
       print $row1.",";
       }
    print "\n";
    }
[download]


I intend to add more functionality to this; but this thread was helpful to me in getting the parser to work