note that gap_package installed alters code paths of GAP code which might not even depend explicitly on it.
On Thu, 12 Aug 2021, 17:07 Emmanuel Charpentier, < emanuel.charpent...@gmail.com> wrote: > Sorry for the late answer... > > Le lundi 9 août 2021 à 18:16:23 UTC+2, Samuel Lelievre a écrit : > >> 2021-08-09 12:00 UTC, Emmanuel Charpentier: >> > >> > On Debian testing running on core i5 + 8 GB RAM, >> > upgrading 9.4.rc0 to 9.4.rc1 and running ptestlong >> > gives three permanent failures : >> > >> > sage -t --long --warn-long 92.9 --random-seed=0 >> src/sage/doctest/test.py # 7 doctests failed >> > sage -t --long --warn-long 92.9 --random-seed=0 >> src/sage/tests/cmdline.py # 2 doctests failed >> > sage -t --long --warn-long 92.9 --random-seed=0 >> src/sage/groups/perm_gps/permgroup.py # 3 doctests failed >> > >> > As far as I can tell now, these failures are identical to those already >> reported for a couple previous releases. >> >> Regarding the last one, I opened >> >> - Sage Trac ticket 32294 >> Fix failing doctests in groups/perm_gps/permgroup.py >> https://trac.sagemath.org/ticket/32294 >> >> Is it the same doctests as in that ticket you see failing? >> > > Yep. But on another system (Debian testing running on core I7 + 16 GB RAM) > it passes. Go figure... > > It seems to depend on atmospheric pressure and temperature, phase of the > moon AND Dow Jones closing... > > >> Do you have any optional packages installed? >> > > Yep : cmdstanpy cypari2 dot2tex fricas gap_jupyter gap_package iniconfig > json5 jupyter_kernel_gap jupyter_kernel_singular kenzo libsemigroups pandas > pycodestyle pyflakes pysingular pytest python_dateutil pyyaml relint saclib > sagemath_standard singular_jupyter tqdm ujson > > >> Can you reproduce on a fresh clone? >> > > I'll have to install a new tree... > > >> I observed the doctest failures on Debian 10 "buster" >> for Sage with many optional packages installed. That Sage >> was built a few releases ago, and those tests have been >> failing for a while on it, including after upgrading to 9.4.rc1. >> >> On the same machine, another Sage, also with many optional >> packages, but installed from a fresh clone more recently, >> passed those doctests, both for Sage 9.4.rc0 and Sage 9.4.rc1. >> >> I guess I should inspect the config.log files to try and spot >> the key difference between these two Sage installations. > > -- > You received this message because you are subscribed to the Google Groups > "sage-release" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to sage-release+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/sage-release/94a41093-7ca7-4894-8d28-bb9e540e5fb8n%40googlegroups.com > <https://groups.google.com/d/msgid/sage-release/94a41093-7ca7-4894-8d28-bb9e540e5fb8n%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "sage-release" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-release+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-release/CAAWYfq0OY9Z%2B7deYYT6i9S-j1R0Ufkf_Wp5YgP6Fq%2BaSEKDApQ%40mail.gmail.com.