Re: [sage-release] Re: Sage 9.5.beta9 released

2022-01-08 Thread Samuel Lelièvre
2021-12-25 04:59 UTC, Samuel Lelievre on sage-release: > > Thanks Dima and Matthias for your feedback. > I opened three tickets related to my report: > > - #33078: Use system Python 3 on Cygwin > - #33079: Fix python3 build failure on Cygwin > - #33080: Fix scipy-3.7.2 build failure on Cygwin > >

Re: [sage-release] Re: Sage 9.5.beta9 released

2022-01-04 Thread Samuel Lelièvre
>From a fresh git clone, one of the following two options is needed to build Sage: ./bootstrap && ./configure && make make configure && ./configure && make >From a fresh tarball, the "make configure" is supposed not to be necessary, as tarballs ship an appropriate version of configure. You may

[sage-release] Re: Sage 9.5.beta9 released

2022-01-04 Thread John H Palmieri
I am having intermittent problems building Sage on OS X from a fresh tarball: it's worked sometimes, failed others. When it fails, it seems to be because there is no script "sage" in venv/bin. (There may be other problems, too, but this one is a major symptom.) I think this is the pattern: if

[sage-release] Re: Sage 9.5.beta9 released

2021-12-29 Thread 'axio...@yahoo.de' via sage-release
The findstat failure is, unfortunately, an upstream problem, which I cannot fix myself. Currently, the depth parameter is ignored in map queries. I notified the person that is hopefully able to fix it. On Wednesday, 29 December 2021 at 11:38:41 UTC+1 Sébastien Labbé wrote: > On Ubuntu 20.04,

[sage-release] Re: Sage 9.5.beta9 released

2021-12-29 Thread Sébastien Labbé
On Ubuntu 20.04, make build works fine and running tests with few optional and external packages I get: Using

Re: [sage-release] Re: Sage 9.5.beta9 released

2021-12-24 Thread Samuel Lelievre
Thanks Dima and Matthias for your feedback. I opened three tickets related to my report: - #33078: Use system Python 3 on Cygwin - #33079: Fix python3 build failure on Cygwin - #33080: Fix scipy-3.7.2 build failure on Cygwin Tried applying #33055, which did not seem to help with the Python 3

Re: [sage-release] Re: Sage 9.5.beta9 released

2021-12-24 Thread Matthias Köppe
On Friday, December 24, 2021 at 1:48:45 PM UTC-8 Samuel Lelievre wrote: > Now that Python 3.9 is packaged for Cygwin, maybe > `build/pkgs/python3/distros/cygwin.txt` should have > `python3` instead of `python38`. I'll check if that helps > detect Cygwin's Python 3.9.9 as a usable Python. +1,

Re: [sage-release] Re: Sage 9.5.beta9 released

2021-12-24 Thread Dima Pasechnik
On Fri, Dec 24, 2021 at 9:48 PM Samuel Lelièvre wrote: > > Thanks Volker for this new release. > > On macOS 10.14.6 Mojave with many packages from > Homebrew: "All tests passed!" > > On Cygwin 3.3.3-1 on Windows 10, with many packages: > > - Sage 9.5.beta9 fails to build python3-3.9.7 unless >

Re: [sage-release] Re: Sage 9.5.beta9 released

2021-12-24 Thread Samuel Lelièvre
Thanks Volker for this new release. On macOS 10.14.6 Mojave with many packages from Homebrew: "All tests passed!" On Cygwin 3.3.3-1 on Windows 10, with many packages: - Sage 9.5.beta9 fails to build python3-3.9.7 unless we configure `--with-python3=/usr/bin/python3` and fails to build

[sage-release] Re: Sage 9.5.beta9 released

2021-12-24 Thread Emmanuel Charpentier
FWIW, on Debian testing running on core i7 + 16 GB RAM, ptestlong gives only two (permanent) failures, both new IIRC : charpent@zen-book-flip:/usr/local/sage-9$ sage -t --long --random-seed=181521996982838512350283316361467808663 src/sage/quadratic_forms/binary_qf.py # 1 doctest failed