#1067539 Causes FTBFS for gap-polymaking by failing tests 2

Package:
polymake
Source:
polymake
Description:
Tool for algorithmic discrete geometry
Submitter:
Joachim Zobel
Date:
2024-04-20 03:45:03 UTC
Severity:
normal
#1067539#5
Date:
2024-03-23 10:55:01 UTC
From:
To:
Hi.

I am seeing FTBFS for my packages gap-polymaking and gap-hapcryst.
The error message is
multi/auto/Polymake/Ext/Ext.so' for module Polymake::Ext:
libflint.so.18: cannot open shared object file: No such file or
directory at /usr/lib/x86_64-linux-gnu/perl-base/DynaLoader.pm line
201.

This is most likely caused by the latest version of libflint beeing
libflint18t64. The bug is similiar to #1053316, just the library is
different.

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1067319
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1067296

Sincerely,
Joachim
#1067539#16
Date:
2024-03-23 11:29:22 UTC
From:
To:
I just ran a pbuilder build of experimental for gap polymaking. The
error message persists.
#1067539#21
Date:
2024-03-23 11:55:08 UTC
From:
To:
Joachim Zobel <jz-2017@heute-morgen.de> writes:

I see the NMUed "t64" version of polymake has not been uploaded to
unstable yet, so I assume that transition is still in progress for
polymake.
#1067539#26
Date:
2024-03-24 11:23:54 UTC
From:
To:
Joachim Zobel <jzobel@heute-morgen.de> writes:

The version of flint in unstable has an uncoordinated transition to
SONAME libflint19. As far as I can tell this is from Julien's commit
711f501dec6ed05ac5c6d4b21eb428b5cfc48da3.

There is a certain amount of confusion due to the t64 transition, but I
think there is an RC bug about this already for flint.

In summary I don't think this is a polymake bug. Or at least there is an
RC bug in flint that should probably be fixed first, before we can debug
polymake.
#1067539#31
Date:
2024-04-20 03:38:42 UTC
From:
To:
Hi.

My packages gap-polymaking and gap-hapcryst are now using autopkgtest.
In both cases the FTBFS was caused by tests running as part of the
build, so my problem is solved.

Sincerely,
Joachim