On the O/S versions at my current disposal, the unix and linux diff have the same bug in them which causes false differences to appear in the output.

Unfortunately, Text::Diff is non-core (which rules it out of the particular environments I am trying to compare files across) and warns of a nasty memory leak problem when calling it many times using Perl v5.6.1 or earlier (just my bad luck that all of those conditions manage to apply). I also have to find a solution that takes a matter of minutes of my time or otherwise the time spent by the people who tried and failed to do the job with unix diff and korn shell cannot be recovered and there will be a lot of hassle from the bean counters.

So it occurred to me that it really ought to be just a few lines of code to hand-roll this. My immediate thought was to write something like:

sub Diff {
    my @a;
    my $dif = undef();
    for my $fdx (0..1) {
        open my $fh, $_[$fdx] or die "$!: $_[$fdx]\n";
        my @tmp = <$fh>;
        $a[0] = \@tmp;
        close $fh;
    }
    while ( $a[0] -> [0] and $a[1] -> [0] ) {
        if ( $a[0] -> [0] eq $a[1] -> [0] ) {
            shift @$a[0]; shift @$a[1];
        }
        elsif ( $a[0] -> [0] ) {
            $dif .= '< ' . shift( @$a[0] );  
        }
        else {
            $dif .= '> ' . shift( @$a[1] );
        }
    }
    return $dif;
}
[download]

In a blunt sort of way this works, but I didn't even see fit to test this because it seems to miss the opportunity of handling insertions to the right (no problem with insertions to the left hand file). Instead, as soon as the right hand file has an insertion, the rest of each file would get dumped out in turn as difference sets.

It looks like I need to search ahead for a match on the right hand side before the point where the left hand side gets treated as different. But this could cause bad performance for large and different files, being proportional in iteration count to the square of the line count. Does anyone have a brighter idea for a difference algorithm that improves on such performance metrics? Or can perhaps this one be modified simply enough?

Thanks in advance for any suggestions.

Update: This is what it looks like with that RH lookahead, which by the way makes it function the same as Algorithm::Diff

sub Diff {
    my @a;
    my $dif = undef();
    my $found;
    for my $fdx (0..1) {
        open my $fh, $_[$fdx] or die "$!: $_[$fdx]\n";
        my @tmp = <$fh>;
        $a[0] = \@tmp;
        close $fh;
    }
    while ( $a[0] -> [0] and $a[1] -> [0] ) {
        if ( $a[0] -> [0] eq $a[1] -> [0] ) {
            shift @$a[0]; shift @$a[1];
        }
        elsif ( $found = Search( @a )) {
            $found--;
            for my $idx ( 0..$found ) {
                $dif .= '> ' . shift( @$a[1] );
            }
        }
        elsif ( $a[0] -> [0] and not Search( @a ) {
            $dif .= '< ' . shift( @$a[0] );  
        }
        else {
            $dif .= '> ' . shift( @$a[1] );
        }
    }
    return $dif;
}
sub Search {
    my $found = 0;
    my $max = $#$_[1];
    while ( $_[0] -> [0] ne $_[1] -> [$found] ) {
        $found++;
        ( $found >= $max ) and return 0;
    }
    return $found;
}
[download]

---

---

Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!

Titles consisting of a single word are discouraged, and in most cases are disallowed outright.

Read Where should I post X? if you're not absolutely sure you're posting in the right place.

Please read these before you post! —

• How do I compose an effective node title?
• How do I post a question effectively?
• Markup in the Monastery

Posts may use any of the Perl Monks Approved HTML tags:

a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr

You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)

	For:		Use:
	&		&
	<		&lt;
	>		&gt;
	[		&#91;
	]		&#93;

Link using PerlMonks shortcuts! What shortcuts can I use for linking?

See Writeup Formatting Tips and other pages linked from there for more info.