#!/usr/bin/perl -w
#
# comparesnaps - Daniel Smith, dls at daniel, org - 2003 or so...
#
# this takes two files with lines of the form: <md5sum> <filename>
#
use strict;
no strict 'refs';

print "starting...";

sub get_checksums;
sub dup_check;
sub change_check;

my (%first_filenames, %second_filenames);
my (%first_dups, %second_dups);
my (%first_mds, %second_mds);
my (%first_cs, %second_cs);

my ($first, $second);

if ($#ARGV != 1) {
    die "usage: $0 prev_snapshot cur_snapshot\n";
}

$first = $ARGV[0];
$second = $ARGV[1];

get_cksums($first, \%first_filenames, \%first_cs, \%first_dups);
get_cksums($second, \%second_filenames, \%second_cs, \%second_dups);

print "duplicates from snapshot: $first\n";
dup_check(\%first_filenames, \%first_dups);

print "duplicates from snapshot: $second\n";
dup_check(\%second_filenames, \%second_dups);

change_check(\%first_cs, \%second_cs);

exit;

sub get_cksums {
    my $fn = shift;
    my $fn_ref = shift;
    my $cs_ref = shift;
    my $dup_ref = shift;

    my ($cksum, $filename);

    open(CKSUM_FILE, "< $fn") ||
        die "unable to read checksum file \"$fn\": $!\n";

     while (<CKSUM_FILE>) {
        chop;
        ($cksum, $filename) = split;
        $$fn_ref{$cksum} .= "$cksum $filename\n";
        $$cs_ref{$filename} .= "$cksum $filename\n";
        $$dup_ref{$cksum}++;
    }
}



sub dup_check {
    my $compare_ref = shift;
    my $dup_ref = shift;
    my ($i);

    foreach $i (keys %$dup_ref) {
        if ($$dup_ref{$i} > 1) {
            print "$$compare_ref{$i}\n";
        }
    }
}

sub change_check {
    my $first_cs_ref = shift;
    my $second_cs_ref = shift;
    my ($i);

    foreach $i (keys %$second_cs_ref) {
        if (! $$first_cs_ref{$i})  {
            print "new file: $$second_cs_ref{$i}\n";
            next;
        }
        if ($$second_cs_ref{$i} !~ $$first_cs_ref{$i}) {
            print "changed file: $$second_cs_ref{$i}\n";
        }
    }
}