use strict;
use Data::Dumper;

my ($i,$j) = (0,0);
my $tab;
for (0..10) {
   push @{$tab->[$j]},$_;
   $j++ unless ++$i % 4;
}
print Dumper $tab;
[download]

$VAR1 = [
          [
            0,
            1,
            2,
            3
          ],
          [
            4,
            5,
            6,
            7
          ],
          [
            8,
            9,
            10
          ]
        ];
[download]

I asked a similar question, and got some good responses. My problem was slightly different; I was trying to make a series of tables with two rows each and each having no more than X columns. I suspect the solutions I received are adaptable, and I know they worked well for me.

--
$rattusillegitimus = Eliza::PerlMonks::Robot::new()

use HTML::Template;
my $template = HTML::Template->new(filename => "test.tmpl");

my @data = (1 .. 30);
my $items_per_row = 5;

my $x = 0;
my @loop;
foreach (@data) {
    my %row = ( data => $_ );
    $row{break} = 1 unless $x % $items_per_row or $x == 0;
    push @loop, \%row;
    $x++;
}
$template->param(loop => \@loop);
print $template->output();
[download]

And the corresponding template:

<table>
  <tr>
  <tmpl_loop loop>
    <tmpl_if break></tr><tr></tmpl_if>
    <td><tmpl_var data></td>
  </tmpl_loop>
  </tr>
</table>
[download]

Does that help?

-sam

Greetings atcroft,

One of the things I've noticed about HTML tables is that in some browsers you have to specify blank cells with " " or else something yucky looking gets put into the final display. So the code of my attempt to answer your question inserts blank cell data for whatever remaining cells are in the table.

use strict;
my @set_of_stuff = (1,3,4,6,7,8,10,12,13,14,15,19,23);
my $items_per_row = 4;
my @table_data;

for (0 .. scalar @set_of_stuff / $items_per_row) {
  my @row_data = map {$_={cell_data=>$_ }} splice(@set_of_stuff,0,$ite
+ms_per_row);
  push @row_data, {cell_data=>'&nbsp'} while scalar @row_data < $items
+_per_row;
  push @table_data, { row_data => \@row_data };
}
[download]

Also, I really like letting the brilliance of the module handle what data goes where in what type of table. I lean away from mucking around with the template itself (i.e. inserting "break" variables to set the rows) and rather prefer to let the module setup the table via the data. Gryphon's personal rule of HTML::Template: Put as much of the "thinking" into the Perl as possible leaving a simpler template.

use HTML::Template;
my $template = q{
<TABLE border=1><TMPL_LOOP name="table_data">
  <TR><TMPL_LOOP name="row_data">
    <TD><TMPL_VAR name="cell_data"></TD>
  </TMPL_LOOP></TR>
</TMPL_LOOP></TABLE>
};

my $template = HTML::Template->new(scalarref => \$template, option => 
+'value');
$template->param(table_data => \@table_data);
print $template->output();
[download]

General aside: When I was first playing around with HTML::Template, I was really annoyed with having to deal with all the references required. However, once I got used to it, I've come to realize that the elegance and brilliance of the module is all about putting all the "thinking" in your Perl and only the HTML in the templates. I'm amazed at what the author's done just about every time I use it.

gryphon
code('Perl') || die;