Bug #81468 for Config-AutoConf: Win64

Mon Nov 26 16:22:38 2012 feratilbeau [...] hotmail.com - Ticket created

Subject:

Win64

C.f. http://code.activestate.com/ppm/Marpa-R2/ Config::AutoConf does not install on Win64, something in the t/*t part. Would you mind to make some of the tests non critical, so that it installs? Thank you so much, Great tiny module, I recommend it

Tue Nov 27 06:25:07 2012 ambs [...] cpan.org - Correspondence added

On Mon Nov 26 16:22:38 2012, jddfr74 wrote: Show quoted text

> C.f. http://code.activestate.com/ppm/Marpa-R2/ > > Config::AutoConf does not install on Win64, something in the t/*t part. > Would you mind to make some of the tests non critical, so that it

installs? Hello Instead of marking tests as TODO or making them con critical, I would prefer to make the module work as expected for Windows 64. Is there any chance you do a resume of what is failing (so I do not need to look through all the reports)? Cheers ambs

Tue Nov 27 06:25:08 2012 The RT System itself - Status changed from 'new' to 'open'

Fri Nov 30 19:36:43 2012 feratilbeau [...] hotmail.com - Correspondence added

Subject:	RE: [rt.cpan.org #81468] Win64
Date:	Sat, 1 Dec 2012 01:36:32 +0100
To:	<bug-config-autoconf [...] rt.cpan.org>
From:	un soleil pour tous <feratilbeau [...] hotmail.com>

Hello, First, an apology, the correct url is: http://code.activestate.com/ppm/Config-AutoConf/ The errors are always about atan() not in -lm (-lm on windows? Seems to work on activestate's win32, my guess is we are lucky an pick an -lm coming from somewhere else but native windows stuff). I note the comment in 03.link.t: ## OK, we really hope people have -lm around -lm is not everywhere, though. Cheers, JD. # Failed test 'atan() in -lm' # at t/03.link.t line 17. # Failed test 'searching lib for atan()' # at t/03.link.t line 21. # Failed test 'library for atan() found (or none required)' # at t/03.link.t line 22. # got: '0' # expected: anything else # Looks like you failed 3 tests of 6. t/03.link.t ....... 1..6 ok 1 - Instantiating Config::AutoConf for check_lib() tests ok 2 - Instantiating Config::AutoConf for search_libs() tests Checking for atan in -lm... no not ok 3 - atan() in -lm Checking for foobar in -lm... no ok 4 - foobar() not in -lm Checking for library containing atan... no not ok 5 - searching lib for atan() not ok 6 - library for atan() found (or none required)

Sun Dec 02 10:10:38 2012 ambs [...] perl-hackers.net - Correspondence added

Subject:	Re: [rt.cpan.org #81468] Win64
Date:	Sun, 02 Dec 2012 15:10:09 +0000
To:	bug-Config-AutoConf [...] rt.cpan.org
From:	Alberto Simoes <ambs [...] perl-hackers.net>

Hello, Can you please test the version in attach? Thanks Alberto On 01/12/12 00:36, jddurand via RT wrote: Show quoted text

> Queue: Config-AutoConf > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=81468 > > > > Hello, > > First, an apology, the correct url is: http://code.activestate.com/ppm/Config-AutoConf/ > The errors are always about atan() not in -lm (-lm on windows? Seems to work on activestate's win32, my guess is we are lucky an pick an -lm coming from somewhere else but native windows stuff). > > I note the comment in 03.link.t: > ## OK, we really hope people have -lm around > > > -lm is not everywhere, though. > > Cheers, JD. > > # Failed test 'atan() in -lm' > # at t/03.link.t line 17. > > # Failed test 'searching lib for atan()' > # at t/03.link.t line 21. > > # Failed test 'library for atan() found (or none required)' > # at t/03.link.t line 22. > # got: '0' > # expected: anything else > # Looks like you failed 3 tests of 6. > t/03.link.t ....... > 1..6 > ok 1 - Instantiating Config::AutoConf for check_lib() tests > ok 2 - Instantiating Config::AutoConf for search_libs() tests > Checking for atan in -lm... no > not ok 3 - atan() in -lm > Checking for foobar in -lm... no > ok 4 - foobar() not in -lm > Checking for library containing atan... no > not ok 5 - searching lib for atan() > not ok 6 - library for atan() found (or none required) > > >

Download Config-AutoConf-0.21.tar.gz
application/x-gzip 17.3k

Message body not shown because it is not plain text.

Thu Dec 06 13:15:38 2012 jkegl [...] cpan.org - Cc JKEGL added

Thu Dec 06 13:28:53 2012 jkegl [...] cpan.org - Correspondence added

Great module, one which has opened a lot of doors for Marpa::R2. I now use it for Windows, Sun and OpenBSD. I've just asked the Marpa mailing list if anyone on it can do this test. thanks, jeffrey On Sun Dec 02 10:10:38 2012, ambs@perl-hackers.net wrote: Show quoted text

> Hello, > > Can you please test the version in attach? > > Thanks > Alberto > > On 01/12/12 00:36, jddurand via RT wrote:

> > Queue: Config-AutoConf > > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=81468 > > > > > > > Hello, > > > > First, an apology, the correct url is:

> http://code.activestate.com/ppm/Config-AutoConf/

> > The errors are always about atan() not in -lm (-lm on windows? Seems

> to work on activestate's win32, my guess is we are lucky an pick an > -lm coming from somewhere else but native windows stuff).

> > > > I note the comment in 03.link.t: > > ## OK, we really hope people have -lm around > > > > > > -lm is not everywhere, though. > > > > Cheers, JD. > > > > # Failed test 'atan() in -lm' > > # at t/03.link.t line 17. > > > > # Failed test 'searching lib for atan()' > > # at t/03.link.t line 21. > > > > # Failed test 'library for atan() found (or none required)' > > # at t/03.link.t line 22. > > # got: '0' > > # expected: anything else > > # Looks like you failed 3 tests of 6. > > t/03.link.t ....... > > 1..6 > > ok 1 - Instantiating Config::AutoConf for check_lib() tests > > ok 2 - Instantiating Config::AutoConf for search_libs() tests > > Checking for atan in -lm... no > > not ok 3 - atan() in -lm > > Checking for foobar in -lm... no > > ok 4 - foobar() not in -lm > > Checking for library containing atan... no > > not ok 5 - searching lib for atan() > > not ok 6 - library for atan() found (or none required) > > > > > >

>

Thu Dec 06 14:46:49 2012 feratilbeau [...] hotmail.com - Correspondence added

From:

feratilbeau [...] hotmail.com

t/03.link.t ....... Bareword found where operator expected at t/03.link.t line 19, near "ok" (Missing semicolon on previous line?) syntax error at t/03.link.t line 19, near "ok" Execution of t/03.link.t aborted due to compilation errors. Btw I can test only on: - linux 32 bits debian/unstable i.e. gcc based - win32 strawberry i.e. gcc based - win32 home-made with nmake in other words I do not own a win64. Cheers, JD.

Thu Dec 06 16:11:41 2012 ambs [...] perl-hackers.net - Correspondence added

Subject:	Re: [rt.cpan.org #81468] Win64
Date:	Thu, 06 Dec 2012 21:11:18 +0000
To:	bug-Config-AutoConf [...] rt.cpan.org
From:	Alberto Simoes <ambs [...] perl-hackers.net>

Stupid me, missing a semicolon. New version in attach. If anybody owns a Win64, that would help :)

Download Config-AutoConf-0.21.tar.gz
application/x-gzip 17.3k

Message body not shown because it is not plain text.

Fri Dec 07 16:28:12 2012 feratilbeau [...] hotmail.com - Correspondence added

From:

feratilbeau [...] hotmail.com

Win32 perl-5.14 nmake/cl/link based: ok Win32 perl-5.16 strawberry i.e. dmake/gcc based: ok linux32 perl-5.14 debian/unstable i.e. make/gcc based: ok Mega Thanks! Regards, Jean-Damien.

Fri Dec 07 16:31:23 2012 ambs [...] perl-hackers.net - Correspondence added

Subject:	Re: [rt.cpan.org #81468] Win64
Date:	Fri, 07 Dec 2012 21:30:56 +0000
To:	bug-Config-AutoConf [...] rt.cpan.org
From:	Alberto Simoes <ambs [...] perl-hackers.net>

On 07/12/12 21:28, jddurand via RT wrote: Show quoted text

> Queue: Config-AutoConf > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=81468 > > > Win32 perl-5.14 nmake/cl/link based: ok > Win32 perl-5.16 strawberry i.e. dmake/gcc based: ok > linux32 perl-5.14 debian/unstable i.e. make/gcc based: ok > > Mega Thanks! > > Regards, Jean-Damien. >

Should I release it? :)

Fri Dec 07 18:12:47 2012 feratilbeau [...] hotmail.com - Correspondence added

From:

feratilbeau [...] hotmail.com

You're the owner of the package. Looking to the changes this look safe to me nonetheless.

Sat Dec 08 09:00:40 2012 ambs [...] cpan.org - Correspondence added

On Fri Dec 07 18:12:47 2012, jddfr74 wrote: Show quoted text

> You're the owner of the package. > Looking to the changes this look safe to me nonetheless.

Hahaha. I meant, the bug was introduced as Win64 related, and nobody tested in Win64 yet, as far as I could understand. What I would like to know is if this fixed any bug in any system where the previous version didn't work. Cheers ambs

Sat Dec 08 18:53:21 2012 jkegl [...] cpan.org - Correspondence added

What's going on is that Jean-Damien and I are basing our conclusions about Marpa::R2 and Windows 64 on the Activestate test matrix: http://code.activestate.com/ppm/Marpa-R2/ Unfortunately, Activestate only tests indexed CPAN releases. I know that, if the situation was reversed, and to get testing I had to be put a change into an Marpa::R2 indexed release, I would only do so if I could convince myself that the change added value in other respects (cleaner code, more in conformance with best practices, a situation likely to arise in other cases, etc., etc.). So I'd certainly understand if you decide you need to see a test result first. An alternative is that I could make a personal plea on the Marpa::R2 mailing list for someone with a Windows 64 box, explaining the situation. I've already asked, but can do so with more explanation, etc., etc. -- jeffrey On Sat Dec 08 09:00:40 2012, AMBS wrote: Show quoted text

> On Fri Dec 07 18:12:47 2012, jddfr74 wrote:

> > You're the owner of the package. > > Looking to the changes this look safe to me nonetheless.

> > Hahaha. > > I meant, the bug was introduced as Win64 related, and nobody tested in > Win64 yet, as far as I could understand. What I would like to know is if > this fixed any bug in any system where the previous version didn't work. > > Cheers > ambs

Sun Dec 09 08:06:39 2012 ambs [...] perl-hackers.net - Correspondence added

Subject:	Re: [rt.cpan.org #81468] Win64
Date:	Sun, 09 Dec 2012 13:06:22 +0000
To:	bug-Config-AutoConf [...] rt.cpan.org
From:	Alberto Simoes <ambs [...] perl-hackers.net>

Hello Got it. Uploading :) On 08/12/12 23:53, Jeffrey Kegler via RT wrote: Show quoted text

> Queue: Config-AutoConf > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=81468 > > > What's going on is that Jean-Damien and I are basing our conclusions about Marpa::R2 and > Windows 64 on the Activestate test matrix: http://code.activestate.com/ppm/Marpa-R2/ > > Unfortunately, Activestate only tests indexed CPAN releases. I know that, if the situation was > reversed, and to get testing I had to be put a change into an Marpa::R2 indexed release, I > would only do so if I could convince myself that the change added value in other respects > (cleaner code, more in conformance with best practices, a situation likely to arise in other > cases, etc., etc.). So I'd certainly understand if you decide you need to see a test result first. > > An alternative is that I could make a personal plea on the Marpa::R2 mailing list for someone > with a Windows 64 box, explaining the situation. I've already asked, but can do so with more > explanation, etc., etc. > > -- jeffrey > > On Sat Dec 08 09:00:40 2012, AMBS wrote:

>> On Fri Dec 07 18:12:47 2012, jddfr74 wrote:

>>> You're the owner of the package. >>> Looking to the changes this look safe to me nonetheless.

>> >> Hahaha. >> >> I meant, the bug was introduced as Win64 related, and nobody tested in >> Win64 yet, as far as I could understand. What I would like to know is if >> this fixed any bug in any system where the previous version didn't work. >> >> Cheers >> ambs

> >

Mon Dec 10 20:32:21 2012 jkegl [...] cpan.org - Correspondence added

The Activestate test is in. Still a failure, alas, but perhaps the remaining issue is straightforward: a failing test of the library for atan(). The build report is http://ppm4.activestate.com/MSWin32-x64/5.12/1200/J/JK/JKEGL/Marpa-R2-2.032000.d/log-20121209T212103.txt -- jeffrey On Sun Dec 09 08:06:39 2012, ambs@perl-hackers.net wrote: Show quoted text

> Hello > > Got it. Uploading :) > > On 08/12/12 23:53, Jeffrey Kegler via RT wrote:

> > Queue: Config-AutoConf > > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=81468 > > > > > What's going on is that Jean-Damien and I are basing our conclusions

> about Marpa::R2 and

> > Windows 64 on the Activestate test matrix:

> http://code.activestate.com/ppm/Marpa-R2/

> > > > Unfortunately, Activestate only tests indexed CPAN releases. I know

> that, if the situation was

> > reversed, and to get testing I had to be put a change into an

> Marpa::R2 indexed release, I

> > would only do so if I could convince myself that the change added

> value in other respects

> > (cleaner code, more in conformance with best practices, a situation

> likely to arise in other

> > cases, etc., etc.). So I'd certainly understand if you decide you

> need to see a test result first.

> > > > An alternative is that I could make a personal plea on the Marpa::R2

> mailing list for someone

> > with a Windows 64 box, explaining the situation. I've already

> asked, but can do so with more

> > explanation, etc., etc. > > > > -- jeffrey > > > > On Sat Dec 08 09:00:40 2012, AMBS wrote:

> >> On Fri Dec 07 18:12:47 2012, jddfr74 wrote:

> >>> You're the owner of the package. > >>> Looking to the changes this look safe to me nonetheless.

> >> > >> Hahaha. > >> > >> I meant, the bug was introduced as Win64 related, and nobody tested

> in

> >> Win64 yet, as far as I could understand. What I would like to know

> is if

> >> this fixed any bug in any system where the previous version didn't

> work.

> >> > >> Cheers > >> ambs

> > > >

>

Tue Dec 11 10:26:11 2012 ambs [...] perl-hackers.net - Correspondence added

Subject:	Re: [rt.cpan.org #81468] Win64
Date:	Tue, 11 Dec 2012 15:25:55 +0000
To:	bug-Config-AutoConf [...] rt.cpan.org
From:	Alberto Simoes <ambs [...] perl-hackers.net>

On 11/12/12 01:32, Jeffrey Kegler via RT wrote: Show quoted text

> Queue: Config-AutoConf > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=81468 > > > The Activestate test is in. Still a failure, alas, but perhaps the > remaining issue is straightforward: a failing test of the library for > atan(). The build report is > http://ppm4.activestate.com/MSWin32-x64/5.12/1200/J/JK/JKEGL/Marpa-R2-2.032000.d/log-20121209T212103.txt

My bad. Added some more "todo" tests, to help solving this issue for now. Version 0.22 hitting CPAN mirrors soon. cheers ambs

Sat Dec 22 20:57:38 2012 jkegl [...] cpan.org - Correspondence added

Just an update: still no retest, and my guess is there will be no test on Windows 64 until I create a new indexed Marpa::R2 release, at which point Activestate will do one for its matrix. All signs seem to be that Windows 64 is not very popular. Nobody seems to have a Windows 64 box, and the very active effort currently devoted to expanding Windows testing for CPAN does not even mention the existence of Windows 64. I'll let you know. And, by the way, Alberto, thanks for Config::AutoConf. It's made a lot of Marpa's installation problems just disappear. -- jeffrey On Tue Dec 11 10:26:11 2012, ambs@perl-hackers.net wrote: Show quoted text

> On 11/12/12 01:32, Jeffrey Kegler via RT wrote:

> > Queue: Config-AutoConf > > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=81468 > > > > > The Activestate test is in. Still a failure, alas, but perhaps the > > remaining issue is straightforward: a failing test of the library

> for

> > atan(). The build report is > > http://ppm4.activestate.com/MSWin32-x64/5.12/1200/J/JK/JKEGL/Marpa-

> R2-2.032000.d/log-20121209T212103.txt > > My bad. > Added some more "todo" tests, to help solving this issue for now. > Version 0.22 hitting CPAN mirrors soon. > > cheers > ambs >

Thu Dec 27 07:56:15 2012 jkegl [...] cpan.org - Correspondence added

The latest testing is in and the entire Windows 64 row is now "green", indicating success. There are some Build log wierdnesses, but I believe that these are old build logs that the Activestate database is slow to clear. If they do turn out to be real issues I can open a new ticket. I do not see any real obstacle to marking this ticket as resolved. Thanks, jeffrey On Sat Dec 22 20:57:38 2012, JKEGL wrote: Show quoted text

> Just an update: still no retest, and my guess is there will be no test > on Windows 64 until I create a new indexed Marpa::R2 release, at which > point Activestate will do one for its matrix. > > All signs seem to be that Windows 64 is not very popular. Nobody seems > to have a Windows 64 box, and the very active effort currently devoted > to expanding Windows testing for CPAN does not even mention the > existence of Windows 64. > > I'll let you know. > > And, by the way, Alberto, thanks for Config::AutoConf. It's made a lot > of Marpa's installation problems just disappear. > > -- jeffrey > > On Tue Dec 11 10:26:11 2012, ambs@perl-hackers.net wrote:

> > On 11/12/12 01:32, Jeffrey Kegler via RT wrote:

> > > Queue: Config-AutoConf > > > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=81468 > > > > > > > The Activestate test is in. Still a failure, alas, but perhaps the > > > remaining issue is straightforward: a failing test of the library

> > for

> > > atan(). The build report is > > > http://ppm4.activestate.com/MSWin32-x64/5.12/1200/J/JK/JKEGL/Marpa-

> > R2-2.032000.d/log-20121209T212103.txt > > > > My bad. > > Added some more "todo" tests, to help solving this issue for now. > > Version 0.22 hitting CPAN mirrors soon. > > > > cheers > > ambs > >

> >

Tue Jan 01 09:25:19 2013 ambs [...] cpan.org - Correspondence added

On Thu Dec 27 07:56:15 2012, JKEGL wrote: Show quoted text

> The latest testing is in and the entire Windows 64 row is now "green", > indicating success.

Great. Thanks for letting me know. Show quoted text

> I do not see any real obstacle to marking this ticket as resolved.

OK, doing that. Poke if necessary :) Happy 2013! Alberto Show quoted text

> > Thanks, jeffrey > > On Sat Dec 22 20:57:38 2012, JKEGL wrote:

> > Just an update: still no retest, and my guess is there will be no test > > on Windows 64 until I create a new indexed Marpa::R2 release, at which > > point Activestate will do one for its matrix. > > > > All signs seem to be that Windows 64 is not very popular. Nobody seems > > to have a Windows 64 box, and the very active effort currently devoted > > to expanding Windows testing for CPAN does not even mention the > > existence of Windows 64. > > > > I'll let you know. > > > > And, by the way, Alberto, thanks for Config::AutoConf. It's made a lot > > of Marpa's installation problems just disappear. > > > > -- jeffrey > > > > On Tue Dec 11 10:26:11 2012, ambs@perl-hackers.net wrote:

> > > On 11/12/12 01:32, Jeffrey Kegler via RT wrote:

> > > > Queue: Config-AutoConf > > > > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=81468 > > > > > > > > > The Activestate test is in. Still a failure, alas, but perhaps the > > > > remaining issue is straightforward: a failing test of the library

> > > for

> > > > atan(). The build report is > > > > http://ppm4.activestate.com/MSWin32-x64/5.12/1200/J/JK/JKEGL/Marpa-

> > > R2-2.032000.d/log-20121209T212103.txt > > > > > > My bad. > > > Added some more "todo" tests, to help solving this issue for now. > > > Version 0.22 hitting CPAN mirrors soon. > > > > > > cheers > > > ambs > > >

> > > >

> >

Tue Jan 01 09:25:22 2013 ambs [...] cpan.org - Status changed from 'open' to 'resolved'