Minor addition

[Purdy]

Just to make a minor point (plus get my first post in for 2003! ;)), the original poster's data has semicolons as the separating character instead of the default comma, which can easily be specified in Text::CSV_XS's sep_char attribute in the new() method. Hope I'm not doing anyone's homework:

#!/usr/bin/perl -w

use strict;
use Text::CSV_XS;

my ( $csv, $xml );

$csv = Text::CSV_XS->new( {
                            'sep_char'    => ';'
        } );

$xml = '';

while( <DATA> ) {
    chomp;
    if ( $csv->parse( $_ ) ) {
        my ( $line, $n, @fields, $field );
        $line = '<row>';
        $n = 1;
        @fields = $csv->fields();
        foreach $field ( @fields ) {
            $line .= "<col$n>$field</col$n>";
            $n++;
        }
        $xml .= $line . "</row>\n";

    } else {
        print "parse() failed on this line: " . $csv->error_input() . 
+"\n";
        # die?
    }
}

print $xml;

__DATA__
a;b;c;d;e
f;g;h;i
j;k
l
m;n;o

Output:

$ ./main.pl
<row><col1>a</col1><col2>b</col2><col3>c</col3><col4>d</col4><col5>e</col5></row>
<row><col1>f</col1><col2>g</col2><col3>h</col3><col4>i</col4></row>
<row><col1>j</col1><col2>k</col2></row>
<row><col1>l</col1></row>
<row><col1>m</col1><col2>n</col2><col3>o</col3></row>

There's probably an XML package out there should your required output become more complex or you want to guarantee that you are using a standardized and optimized solution.

Peace,

Purdy

[jeffa]

Here is another approach which uses Text::CSV_XS and, gulp, CGI:

use strict;
use warnings;

use Text::CSV_XS;
use CGI::Pretty qw(-any);

my $i;
my $csv = Text::CSV_XS->new({sep_char => ';'});

while (<DATA>) {
   $i = 0;
   warn and next unless $csv->parse($_);
   print CGI::row(map eval "CGI::col@{[++$i]}('$_')", $csv->fields);
}

__DATA__
colContents;nextColContents;lastColContents
colContents2;nextColContents2;lastColContents2
colContents3;nextColContents3;lastColContents3
a;b;c;d;e
f;g;h;i
j;k
l
m;n;o

The -any pragma in CGI.pm is pretty nifty, you can actually use it to create XML. No guarantees on valid XML, of course. Just be sure and either append the CGI package name or use the OO interface (else Perl will complain that main has no such method). I was able to increment the <colN> tags by using an evil eval trick. That was the toughest part of this code. Drop that requirement and utilize CGI.pm's distributive shortcuts feature thingy:

use Text::CSV_XS;
use CGI::Pretty qw(-any);

my $csv = Text::CSV_XS->new({sep_char => ';'});
$csv->parse($_) and print CGI::row(CGI::col([$csv->fields]))
      while <DATA>;

jeffa

L-LL-L--L-LL-L--L-LL-L--
-R--R-RR-R--R-RR-R--R-RR
B--B--B--B--B--B--B--B--
H---H---H---H---H---H---
(the triplet paradiddle with high-hat)