Bug #20879 for Algorithm-C3: Bug in Algorithm-C3

Mon Aug 07 15:38:02 2006 tcarnahan [...] ideaworks.com - Ticket created

Subject:	Bug in Algorithm-C3
Date:	Mon, 07 Aug 2006 14:37:22 -0500
To:	bug-Algorithm-C3 [...] rt.cpan.org
From:	Ted Carnahan <tcarnahan [...] ideaworks.com>

The upgrade from Algorithm-C3 v0.01 to v0.02 causes problems for my particular inheritance tree. A complete set of testing output is attached in the zip file. If you don't receive it, please email me for a copy. Thanks, -- Ted Carnahan Idea Works, Inc. 100 West Briarwood Columbia, Missouri 65203 USA P: (573) 445-4554 F: (573) 446-2199 tcarnahan@ideaworks.com www.ideaworks.com

Download bug.zip
application/octet-stream 14.1k

Message body not shown because it is not plain text.

Tue Aug 08 15:49:55 2006 blblack [...] gmail.com - Taken

Tue Aug 08 16:04:08 2006 blblack [...] gmail.com - Correspondence added

On Mon Aug 07 15:38:02 2006, tcarnahan@ideaworks.com wrote: Show quoted text

> The upgrade from Algorithm-C3 v0.01 to v0.02 causes problems for my > particular inheritance tree. A complete set of testing output is > attached in the zip file. If you don't receive it, please email me for > a copy. > > Thanks, >

The .dot and .png files in the "v0.01" and "upgraded class c3 only" subdirectories of your zip file are identical (md5sum is identical). Is that intentional? -- Brandon

Tue Aug 08 16:04:09 2006 The RT System itself - Status changed from 'new' to 'open'

Tue Aug 08 16:25:22 2006 tcarnahan [...] ideaworks.com - Correspondence added

Subject:	Re: [rt.cpan.org #20879] Bug in Algorithm-C3
Date:	Tue, 08 Aug 2006 15:25:07 -0500
To:	bug-Algorithm-C3 [...] rt.cpan.org
From:	Ted Carnahan <tcarnahan [...] ideaworks.com>

Yes - I should have been more clear - that was to demonstrate that it is not the Class::C3 module upgrade that causes the problem - we isolated it to Algorithm::C3. Sorry for the confusion. Ted Carnahan Idea Works, Inc. 100 West Briarwood Columbia, Missouri 65203 USA P: (573) 445-4554 F: (573) 446-2199 tcarnahan@ideaworks.com www.ideaworks.com via RT wrote: Show quoted text

> <URL: http://rt.cpan.org/Ticket/Display.html?id=20879 > > > On Mon Aug 07 15:38:02 2006, tcarnahan@ideaworks.com wrote: >

>>The upgrade from Algorithm-C3 v0.01 to v0.02 causes problems for my >>particular inheritance tree. A complete set of testing output is >>attached in the zip file. If you don't receive it, please email me for >>a copy. >> >>Thanks, >>

> > > The .dot and .png files in the "v0.01" and "upgraded class c3 only" > subdirectories of your zip file are identical (md5sum is identical). Is > that intentional? > > -- Brandon > > > > . >

Tue Aug 08 17:52:03 2006 blblack [...] gmail.com - Correspondence added

I believe I've found the issue. It will take me a bit longer to work up some test cases and a proper fix for 0.03, but in the meantime, could you verify that the attached patch makes things work correctly for you, so that I know I'm on the right track?

=== lib/Algorithm/C3.pm ================================================================== --- lib/Algorithm/C3.pm (revision 13160) +++ lib/Algorithm/C3.pm (local) @@ -55,7 +55,9 @@ # http://www.python.org/2.3/mro.html :) # Initial set - my @seqs = ([$current_root], @$recurse_mergeout, $current_parents); + my @deepcopy_recurse_mergeout; + push(@deepcopy_recurse_mergeout, [@$_]) for (@$recurse_mergeout); + my @seqs = ([$current_root], @deepcopy_recurse_mergeout, [@$current_parents]); # Construct the tail-checking hash my %tails;

Tue Aug 08 23:35:40 2006 blblack [...] gmail.com - Correspondence added

I've gone ahead and released an 0.03 to CPAN that has a fix similar to the posted patch, plus a new automated test based on the @ISA tree in your bug report. You can grab it at http://dtmf.com/Algorithm-C3-0.03.tar.gz if you don't want to wait for CPAN to sync it around.

Wed Aug 09 11:42:37 2006 tcarnahan [...] ideaworks.com - Correspondence added

Subject:	Re: [rt.cpan.org #20879] Bug in Algorithm-C3
Date:	Wed, 09 Aug 2006 10:42:02 -0500
To:	bug-Algorithm-C3 [...] rt.cpan.org
From:	Ted Carnahan <tcarnahan [...] ideaworks.com>

We'll try it out - thanks for fixing this so quickly! Ted Carnahan Idea Works, Inc. 100 West Briarwood Columbia, Missouri 65203 USA P: (573) 445-4554 F: (573) 446-2199 tcarnahan@ideaworks.com www.ideaworks.com via RT wrote: Show quoted text

> <URL: http://rt.cpan.org/Ticket/Display.html?id=20879 > > > > I've gone ahead and released an 0.03 to CPAN that has a fix similar to > the posted patch, plus a new automated test based on the @ISA tree in > your bug report. You can grab it at > http://dtmf.com/Algorithm-C3-0.03.tar.gz if you don't want to wait for > CPAN to sync it around. > > >

Wed Aug 09 11:45:58 2006 tcarnahan [...] ideaworks.com - Correspondence added

Subject:	Re: [rt.cpan.org #20879] Bug in Algorithm-C3
Date:	Wed, 09 Aug 2006 10:44:28 -0500
To:	bug-Algorithm-C3 [...] rt.cpan.org
From:	Ted Carnahan <tcarnahan [...] ideaworks.com>

We can confirm that it works in our application - thank you. Ted Carnahan Idea Works, Inc. 100 West Briarwood Columbia, Missouri 65203 USA P: (573) 445-4554 F: (573) 446-2199 tcarnahan@ideaworks.com www.ideaworks.com via RT wrote: Show quoted text

> <URL: http://rt.cpan.org/Ticket/Display.html?id=20879 > > > > I've gone ahead and released an 0.03 to CPAN that has a fix similar to > the posted patch, plus a new automated test based on the @ISA tree in > your bug report. You can grab it at > http://dtmf.com/Algorithm-C3-0.03.tar.gz if you don't want to wait for > CPAN to sync it around. > > >

Wed Aug 09 11:54:52 2006 tcarnahan [...] ideaworks.com - Correspondence added

Subject:	Re: [rt.cpan.org #20879] Bug in Algorithm-C3
Date:	Wed, 09 Aug 2006 10:53:55 -0500
To:	bug-Algorithm-C3 [...] rt.cpan.org
From:	Ted Carnahan <tcarnahan [...] ideaworks.com>

FYI - Our application's test suite runs 12% faster overall using Algorithm-C3 0.03 over 0.01. Good job! Ted Carnahan Idea Works, Inc. 100 West Briarwood Columbia, Missouri 65203 USA P: (573) 445-4554 F: (573) 446-2199 tcarnahan@ideaworks.com www.ideaworks.com via RT wrote: Show quoted text

> <URL: http://rt.cpan.org/Ticket/Display.html?id=20879 > > > > I've gone ahead and released an 0.03 to CPAN that has a fix similar to > the posted patch, plus a new automated test based on the @ISA tree in > your bug report. You can grab it at > http://dtmf.com/Algorithm-C3-0.03.tar.gz if you don't want to wait for > CPAN to sync it around. > > >

Wed Aug 09 12:15:01 2006 blblack [...] gmail.com - Status changed from 'open' to 'resolved'

Wed Aug 09 12:15:02 2006 blblack [...] gmail.com - Severity Critical added

Wed Aug 09 12:15:02 2006 blblack [...] gmail.com - Fixed in 0.03 added

Wed Aug 09 12:15:03 2006 blblack [...] gmail.com - Broken in 0.02 added