Skip Menu |
 
rt.cpan.org will be shut down on March 1st, 2021.

This queue is for tickets about the Algorithm-Merge CPAN distribution.

Report information
The Basics
Id: 11184
Status: resolved
Worked: 15 min
Priority: 0/
Queue: Algorithm-Merge

People
Owner: cpan [...] jamesmith.com
Requestors: cpan_rt [...] account.loath.org
Cc:
AdminCc:

Bug Information
Severity: Normal
Broken in: 0.05
Fixed in: (no value)



Subject: diff3 confused by change in middle of deletion
Download (untitled) / with headers
text/plain 1.2k
Thanks for Algorithm::Merge. I'm finding it extremely useful. However, I've found a situation where it gives very odd output. I'm using this version of Algorithm::Merge: $VERSION = '0.05'; $REVISION = (qw$Revision: 1.11 $)[-1]; And this version of Algorithm::Diff: $VERSION = 1.19_01; With Perl: This is perl, v5.8.4 built for i386-linux-thread-multi On Debian testing: Linux sheol.chthonic.net 2.4.25-1-386 #2 Wed Apr 14 19:38:08 EST 2004 i686 GNU/Linux diff3 seems to get confused when you diff a deletion against a change in the middle of that deletion. Instead of 'c', it gives 'r', and the left change is a misplaced element. Here's a sample: #!/usr/bin/perl use Algorithm::Merge qw(diff3); @a = qw(1 2 3 4 5 6 7); @b = qw(1 2 6 7); @c = qw(1 2 3 0 5 6 7); my @diff = diff3(\@a, \@b, \@c); foreach $row (@diff) { print join(" ", map { defined($_) ? $_ : '-' } @$row) . "\n"; } This code gives: # yields: u 1 1 1 u 2 2 2 l 3 - 3 r 4 1 0 <- shouldn't this be 'c 4 - 0'? why the 1? l 5 - 5 u 6 6 6 u 7 7 7 I can see the argument that the change to get from @a to @c is 4 -> 0, and with @a to @b being -3 -4 -5, that suggests: l 3 - 3 o 4 - - r - - 0 l 5 - 5 In any case, the presence of that stray 1 bothers me.
Should be fixed in 0.06.


This service is sponsored and maintained by Best Practical Solutions and runs on Perl.org infrastructure.

Please report any issues with rt.cpan.org to rt-cpan-admin@bestpractical.com.