Bug #73969 for Text-Transliterator: Fails since bleadperl v5.15.6-282-gce43265

Fri Jan 13 00:42:30 2012 ANDK [...] cpan.org - Ticket created

CC:	"Karl Williamson" <public [...] khwilliamson.com>
Subject:	Fails since bleadperl v5.15.6-282-gce43265

This is the commit: commit ce4326557fd52ed5f0dbfaaf93f76209db291470 Author: Karl Williamson <public@khwilliamson.com> Date: Fri Dec 30 09:11:23 2011 -0700 mktables: Change function name This is in preparation for Unicode 6.1 for which the new name will make more sense. Regards,

Mon Jan 16 05:02:42 2012 DAMI [...] cpan.org - Correspondence added

Subject:	Text::Transliterator::Unaccent fails since bleadperl v5.15.6-282-gce43265
RT-Send-CC:	public [...] khwilliamson.com

Le Ven 13 Jan 2012 00:42:30, ANDK a écrit : Show quoted text

> This is the commit: > > commit ce4326557fd52ed5f0dbfaaf93f76209db291470 > Author: Karl Williamson <public@khwilliamson.com> > Date: Fri Dec 30 09:11:23 2011 -0700

Thanks Andreas for the ticket. Unfortunately I don't have access to 5.15.6, and just by looking at the source code I can't find out which Unicode changes broke Text::Transliterator::Unaccent. So could you please run the commands below and post me the output ? Thanks in advance, Laurent D. perl -MUnicode::UCD=charscript -MYAML -e "print Dump charscript ('Latin')" perl -MUnicode::Normalize=getCanon -e "print substr(getCanon(0xe9), 0, 1)"

Mon Jan 16 05:02:43 2012 The RT System itself - Status changed from 'new' to 'open'

Mon Jan 16 11:44:22 2012 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Correspondence added

On Mon Jan 16 05:02:42 2012, DAMI wrote: Show quoted text

> Le Ven 13 Jan 2012 00:42:30, ANDK a écrit :

> > This is the commit: > > > > commit ce4326557fd52ed5f0dbfaaf93f76209db291470 > > Author: Karl Williamson <public@khwilliamson.com> > > Date: Fri Dec 30 09:11:23 2011 -0700

> > Thanks Andreas for the ticket. Unfortunately I don't have access to > 5.15.6, and just by looking at the source code I can't find out which > Unicode changes broke Text::Transliterator::Unaccent. So could you > please run the commands below and post me the output ? > > Thanks in advance, Laurent D. > > perl -MUnicode::UCD=charscript -MYAML -e "print Dump charscript > ('Latin')"

--- - - 65 - 90 - Latin - - 97 - 122 - Latin - - 170 - 170 - Latin - - 186 - 186 - Latin - - 192 - 214 - Latin - - 216 - 246 - Latin - - 248 - 696 - Latin - - 736 - 740 - Latin - - 7424 - 7461 - Latin - - 7468 - 7516 - Latin - - 7522 - 7525 - Latin - - 7531 - 7543 - Latin - - 7545 - 7614 - Latin - - 7680 - 7935 - Latin - - 8305 - 8305 - Latin - - 8319 - 8319 - Latin - - 8336 - 8348 - Latin - - 8490 - 8491 - Latin - - 8498 - 8498 - Latin - - 8526 - 8526 - Latin - - 8544 - 8584 - Latin - - 11360 - 11391 - Latin - - 42786 - 42887 - Latin - - 42891 - 42894 - Latin - - 42896 - 42897 - Latin - - 42912 - 42921 - Latin - - 43002 - 43007 - Latin - - 64256 - 64262 - Latin - - 65313 - 65338 - Latin - - 65345 - 65370 - Latin Show quoted text

> perl -MUnicode::Normalize=getCanon -e "print substr(getCanon(0xe9), > 0, 1)"

That just gives me ‘e’. This output is with a recent bleadperl (84f93674).

Mon Jan 16 14:36:13 2012 public [...] khwilliamson.com - Correspondence added

CC:	ANDK [...] cpan.org
Subject:	Re: [rt.cpan.org #73969] Fails since bleadperl v5.15.6-282-gce43265
Date:	Mon, 16 Jan 2012 12:35:56 -0700
To:	bug-Text-Transliterator [...] rt.cpan.org
From:	Karl Williamson <public [...] khwilliamson.com>

On 01/16/2012 09:44 AM, Father Chrysostomos via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=73969> > > On Mon Jan 16 05:02:42 2012, DAMI wrote:

>> Le Ven 13 Jan 2012 00:42:30, ANDK a écrit :

>>> This is the commit: >>> >>> commit ce4326557fd52ed5f0dbfaaf93f76209db291470 >>> Author: Karl Williamson<public@khwilliamson.com> >>> Date: Fri Dec 30 09:11:23 2011 -0700

>> >> Thanks Andreas for the ticket. Unfortunately I don't have access to >> 5.15.6, and just by looking at the source code I can't find out which >> Unicode changes broke Text::Transliterator::Unaccent. So could you >> please run the commands below and post me the output ? >> >> Thanks in advance, Laurent D.

It's hard for me to imagine that the specified commit is the culprit. FYI, the outputs for charscript and the getCanon are identical before and after the commit. Show quoted text

>> >> perl -MUnicode::UCD=charscript -MYAML -e "print Dump charscript >> ('Latin')"

> > --- > - > - 65 > - 90 > - Latin > - > - 97 > - 122 > - Latin > - > - 170 > - 170 > - Latin > - > - 186 > - 186 > - Latin > - > - 192 > - 214 > - Latin > - > - 216 > - 246 > - Latin > - > - 248 > - 696 > - Latin > - > - 736 > - 740 > - Latin > - > - 7424 > - 7461 > - Latin > - > - 7468 > - 7516 > - Latin > - > - 7522 > - 7525 > - Latin > - > - 7531 > - 7543 > - Latin > - > - 7545 > - 7614 > - Latin > - > - 7680 > - 7935 > - Latin > - > - 8305 > - 8305 > - Latin > - > - 8319 > - 8319 > - Latin > - > - 8336 > - 8348 > - Latin > - > - 8490 > - 8491 > - Latin > - > - 8498 > - 8498 > - Latin > - > - 8526 > - 8526 > - Latin > - > - 8544 > - 8584 > - Latin > - > - 11360 > - 11391 > - Latin > - > - 42786 > - 42887 > - Latin > - > - 42891 > - 42894 > - Latin > - > - 42896 > - 42897 > - Latin > - > - 42912 > - 42921 > - Latin > - > - 43002 > - 43007 > - Latin > - > - 64256 > - 64262 > - Latin > - > - 65313 > - 65338 > - Latin > - > - 65345 > - 65370 > - Latin > >

>> perl -MUnicode::Normalize=getCanon -e "print substr(getCanon(0xe9), >> 0, 1)"

> > That just gives me ‘e’. > > This output is with a recent bleadperl (84f93674). > >

Mon Jan 16 15:04:52 2012 laurent.dami [...] free.fr - Correspondence added

Subject:	Re: [rt.cpan.org #73969] Fails since bleadperl v5.15.6-282-gce43265
Date:	Mon, 16 Jan 2012 21:04:43 +0100
To:	bug-Text-Transliterator [...] rt.cpan.org
From:	laurent dami <laurent.dami [...] free.fr>

Le 16.01.2012 20:36, karl williamson via RT a écrit : Show quoted text

> It's hard for me to imagine that the specified commit is the culprit. > FYI, the outputs for charscript and the getCanon are identical before > and after the commit.

OK, thanks for answering; indeed I don't see anything wrong in the outputs for charscript and getCanon, so the problem must be somewhere else ... but I don't know what it could be. I'll try to find a machine where I can install bleadperl and give a try at perl -d.

Mon Jan 16 15:40:52 2012 public [...] khwilliamson.com - Correspondence added

CC:	ANDK [...] cpan.org
Subject:	Re: [rt.cpan.org #73969] Fails since bleadperl v5.15.6-282-gce43265
Date:	Mon, 16 Jan 2012 13:40:36 -0700
To:	bug-Text-Transliterator [...] rt.cpan.org
From:	Karl Williamson <public [...] khwilliamson.com>

On 01/16/2012 01:04 PM, laurent.dami@free.fr via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=73969> > > Le 16.01.2012 20:36, karl williamson via RT a écrit :

>> It's hard for me to imagine that the specified commit is the culprit. >> FYI, the outputs for charscript and the getCanon are identical before >> and after the commit.

> OK, thanks for answering; indeed I don't see anything wrong in the > outputs for charscript and getCanon, so the problem must be somewhere > else ... but I don't know what it could be. I'll try to find a machine > where I can install bleadperl and give a try at perl -d. > >

Actually the outputs have been stable throughout all of the 5.15 series

Mon Jan 16 15:53:29 2012 public [...] khwilliamson.com - Correspondence added

CC:	ANDK [...] cpan.org
Subject:	Re: [rt.cpan.org #73969] Fails since bleadperl v5.15.6-282-gce43265
Date:	Mon, 16 Jan 2012 13:53:16 -0700
To:	bug-Text-Transliterator [...] rt.cpan.org
From:	Karl Williamson <public [...] khwilliamson.com>

On 01/16/2012 01:04 PM, laurent.dami@free.fr via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=73969> > > Le 16.01.2012 20:36, karl williamson via RT a écrit :

>> It's hard for me to imagine that the specified commit is the culprit. >> FYI, the outputs for charscript and the getCanon are identical before >> and after the commit.

> OK, thanks for answering; indeed I don't see anything wrong in the > outputs for charscript and getCanon, so the problem must be somewhere > else ... but I don't know what it could be. I'll try to find a machine > where I can install bleadperl and give a try at perl -d. > >

I don't see any details in the bug report as to how it is failing. I undertook to try it on my system with the latest blead. I get this from make test PERL_DL_NONLAZY=1 /home/khw/cpanbroken/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t t/00-load.t ............ 1/2 # Testing Text::Transliterator 1.00, Perl 5.015006, /home/khw/cpanbroken/bin/perl t/00-load.t ............ ok t/01-transliterator.t .. ok t/02-unaccent.t ........ ok t/pod-coverage.t ....... skipped: Test::Pod::Coverage 1.08 required for testing POD coverage t/pod.t ................ skipped: Test::Pod 1.22 required for testing POD All tests successful. Files=5, Tests=5, 0 wallclock secs ( 0.06 usr 0.02 sys + 0.56 cusr 0.02 csys = 0.66 CPU) Result: PASS

Mon Jan 16 16:08:18 2012 andreas.koenig.7os6VVqR [...] franz.ak.mind.de - Correspondence added

CC:	bug-Text-Transliterator [...] rt.cpan.org, ANDK [...] cpan.org
Subject:	Re: [rt.cpan.org #73969] Fails since bleadperl v5.15.6-282-gce43265
Date:	Mon, 16 Jan 2012 22:08:05 +0100
To:	Karl Williamson <public [...] khwilliamson.com>
From:	andreas.koenig.7os6VVqR [...] franz.ak.mind.de (Andreas J. Koenig)

Show quoted text

>>>>> On Mon, 16 Jan 2012 12:35:56 -0700, Karl Williamson <public@khwilliamson.com> said:

Show quoted text

> It's hard for me to imagine that the specified commit is the culprit.

Wow, bisecting is getting harder. I'm sorry for the false claim. I can't explain how I came to it. All I can do is to increase the sanity checks in the future before I come to conclusions and, above all, to put more evidence on the table so that the source for errors is easier to spot later. I do see the history of my commands and some tables of the outcomes but not enough to understand what exactly went wrong. Sorry for that. Bottom line: on the most recent bleadperl versions on my box I see Text-Transliterator recovered (that is: v5.15.6-556-g86e2f32, v5.15.6-550-g838cf71, v5.15.6-546-gd3a16da, v5.15.6-540-g711a390, v5.15.6-519-g8c85651 all have it installed with a PASS). I'll close this ticket in a minute. Sorry for the inconvenience, -- andreas

Mon Jan 16 16:13:01 2012 ANDK [...] cpan.org - Correspondence added

Closing. If I find more evidence, what was getting me on the wrong track, I'll let you know.

Mon Jan 16 16:13:01 2012 ANDK [...] cpan.org - Status changed from 'open' to 'resolved'