[Debian-science-sagemath] Updating fplll in unstable

[Tobias Hansen]

Hi,

fplll in experimental FTBFS on armel, armhf and mips due to

FAIL: test_lll_gram
===================

The gram GSO-object is not LLL-reduced after calling LLL
FAIL test_lll_gram (exit status: 255)

fpylll FTBFS on i386.

Could you have a look please?

Best,
Tobias



On 11/08/2017 10:58 AM, Julien Puydt wrote:
> Hi,
>
> there has been a new version of fplll, 5.2.0 in september ; I updated
> the repository then, but it wasn't a good idea to push it as sagemath
> was still using 5.1.0 (through fpylll).
>
> The situation is now different, since sagemath updated to this version
> of fplll two weeks ago :
> https://trac.sagemath.org/ticket/24042
>
> Notice :
> * there was a soname bump, so it means a transition ;
> * I checked reverse depends using my tentative 5.2.0-1 package :
>     - fpylll:
>         + it would FTBFS with the new fplll ;
>         + it was updated in sagemath (same ticket as above) ;
>         + I updated the DPMT git repo
>     - gap-float:
>         + it would FTBFS with the new fplll ;
>         + I tried to see if I could update it, but it looked
>           a bit non-trivial so it might be better and faster
>           to let Jérôme poke it.
>     - sollya: good to go
>
> Cheers,
>
> Snark on irc.debian.org
>