New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Bleadperl v5.27.2-30-gdf6b4bd565 breaks SYBER/Panda-Lib-1.3.2.tar.gz #16123
Comments
From @andkDiscovered by Slaven, hvala lijepa! bisect commit df6b4bd give REGEXP SVs the POK flag again cpantesters http://www.cpantesters.org/cpan/report/a0526a7e-74e4-11e7-b839-7de6a299d00c perl -V Summary of my perl5 (revision 5 version 27 subversion 3) configuration: Characteristics of this binary (from libperl): -- |
From @jkeenanOn 08/23/2017 05:50 PM, (Andreas J. Koenig) (via RT) wrote:
In each case, the failing unit test is one that tests the This will probably require consultation with the module author in order Thank you very much. |
The RT System itself - Status changed from 'new' to 'open' |
From @jkeenanOn Thu, 24 Aug 2017 01:04:30 GMT, jkeenan@pobox.com wrote:
Compare the result of 'perl Makefile.PL && make' on perl-5.26.0 and blead, respectively: ##### $ ls -ltr | tail -2 $ ls -ltr | tail -2 Much larger Lib.o file when built against blead. Would the fact that blead has a newer version of ExtUtils::ParseXS make any difference? #####
Thank you very much. -- |
From @cpansproutOn Wed, 23 Aug 2017 18:40:01 -0700, jkeenan wrote:
Debugging, or non-debugging? That may explain the .o file size difference.
-- Father Chrysostomos |
From @iabynOn Wed, Aug 23, 2017 at 09:04:14PM -0400, James E Keenan wrote:
Its a bug in the author's code. In _elem_cmp(), there is this: if (SvOK(f) & SvOK(s)) { // both are not undefs which is expecting that SvOK() returns a boolean (0/1) value. It's That code needs to be if (SvOK(f) && SvOK(s)) { // both are not undefs and probably similar things done with ^, | and & in other places too. -- |
From @jkeenanOn Thu, 24 Aug 2017 03:15:44 GMT, sprout wrote:
Ah, yes. For the purpose of looking at *other* BBC tickets, my blead build was -DDEBUGGING. So the difference in object sizes is not relevant to this ticket. Thanks for the catch.
-- |
From @dur-randirFixed with 1.3.5 |
From @xsawyerxOn Mon, 15 Jan 2018 10:38:19 -0800, randir wrote:
Can anyone confirm? |
From @eserteDana Mon, 15 Jan 2018 12:22:32 -0800, xsawyerx@cpan.org reče:
Fails to build on most platforms I tried (FreeBSD 10,11,12, CentOS6, Debian/jessie, Ubuntu 16.04), but probably due to other reasons (fails also with 5.26.x). But I had success with 5.27.7 on Debian/stretch. |
From @dur-randirOn Mon, 15 Jan 2018 12:41:30 -0800, slaven@rezic.de wrote:
Have you tried 1.3.4 or 1.3.5? 1.3.4 failed to build on threaded perls. |
From @eserteDana Mon, 15 Jan 2018 12:56:05 -0800, randir reče:
I tried 1.3.5. The failures were all compilation failures in the dependencies. |
From @dur-randirOn Mon, 15 Jan 2018 13:02:52 -0800, slaven@rezic.de wrote:
Can you provide a full install log for any of the failing cases? Cpantesters shows no reports for 1.3.5 yet. What I can think of - one of this module's prerequisite, Panda::XS, requires a c++14 compliant compiler as of version 2.1.8. |
From @eserteDana Mon, 15 Jan 2018 13:14:10 -0800, randir reče:
This is Panda-specific, so I put the log into a new ticket: |
From @dur-randirOn Tue, 16 Jan 2018 11:01:50 -0800, slaven@rezic.de wrote:
Thanks, I'll look further into it. |
From @iabynOn Tue, Jan 16, 2018 at 03:23:49PM -0800, Sergey Aleynikov via RT wrote:
Closing this perl ticket, as it's no longer a perl issue. -- |
@iabyn - Status changed from 'open' to 'resolved' |
Migrated from rt.perl.org#131953 (status was 'resolved')
Searchable as RT131953$
The text was updated successfully, but these errors were encountered: