Algorithm::Diff is pretty good for producing diff like output for two variables in Perl, but I would definitly just use: `/usr/bin/diff $file[1] $file[2]` to get the diff of files, its simpler. Also I'd suggest File::Find with a wanted funciton that checks the date range in order to pick your files

Algorithm::Diff useage for difflike output looks something like:

sub diffy {
  my $output;
  my @one = split(/\n/, $_[0]);            
  my @two = split(/\n/, $_[1]);
  my $equ = $_[2] || "==";   
  my $sub = $_[3] || "--";
  my $add = $_[4] || "++";
  my $eqX = $_[5] || "";
  my $suX = $_[6] || "";
  my $adX = $_[7] || "";

  use Algorithm::Diff qw(traverse_sequences);

  traverse_sequences(\@one, \@two, {
    MATCH     => sub { $output .= "$equ$one[shift]$eqX\n"},
    DISCARD_A => sub { $output .= "$sub$one[shift]$suX\n"},
    DISCARD_B => sub { $output .= "$add$two[shift,shift]$adX\n"},
  });

  return $output;
}
[download]

Which is sorta odd looking, but now that you have that code, you don't really have to think about it. Here you can pass in $one and $two as variables holding multi-line input, and optionally pass in start tags for matching lines, deleted lines, and added lines, as well as the end tags for the same.

-Daniel

Edit: 2001-03-03 by neshura