Re: [sage-release] Sage 9.7.beta7 released

I've opened https://trac.sagemath.org/ticket/34268 (Reduce R to a dummy package, upgrade rpy2 to 3.4.5) On Tuesday, August 2, 2022 at 11:34:22 AM UTC-7 Matthias Köppe wrote: > Yes, our R package is wildly outdated, which is problematic given that it > is a standard package. > > > > On Tuesday,

Re: [sage-release] Sage 9.7.beta7 released

And also https://trac.sagemath.org/ticket/22349 Deprecate sorting of Graph vertices and edges by default This ticket touches many files and so can lead to conflicts. Best, David. > Le 3 août 2022 à 13:33, Kwankyu Lee a écrit : > > It would make a sense to release sage beta8 quickly that just

Re: [sage-release] Sage 9.7.beta7 released

Yes, our R package is wildly outdated, which is problematic given that it is a standard package. On Tuesday, August 2, 2022 at 11:23:08 AM UTC-7 isu...@gmail.com wrote: > Is there an issue for the R build failure? This seems like an upstream R > issue and maybe updating to a newer R version

Re: [sage-release] Sage 9.7.beta7 released

Is there an issue for the R build failure? This seems like an upstream R issue and maybe updating to a newer R version will help. (R-3.6 is 3 years old) ``` In file included from gdi.c:26: /Users/runner/work/sage/sage/.tox/local-conda-forge-macos-minimal/conda/include/omp.h:492:39: error:

Re: [sage-release] Sage 9.7.beta7 released

On Tuesday, August 2, 2022 at 10:51:04 AM UTC-7 isu...@gmail.com wrote: > We build conda packages regularly with non-Xcode clang on macOS. > Yes, we do also test the Sage distribution build on conda-forge-macos, which uses clang. conda-forge-macos-standard succeeds, using lots of conda

Re: [sage-release] Sage 9.7.beta7 released

I've added the sentence "An alternative is to use a sufficiently recent version of clang (LLVM)." But having more details would be better. On Tuesday, August 2, 2022 at 10:46:32 AM UTC-7 Dima Pasechnik wrote: > > > On Tue, 2 Aug 2022, 18:33 Matthias Köppe, wrote: > >> On Tuesday, August 2,

Re: [sage-release] Sage 9.7.beta7 released

We build conda packages regularly with non-Xcode clang on macOS. Isuru On Tue, Aug 2, 2022 at 12:33 PM Matthias Köppe wrote: > On Tuesday, August 2, 2022 at 10:24:39 AM UTC-7 Dima Pasechnik wrote: > >> >> there is a bit of vagueness there: >> Support for system GCC older than 6.3 dropped ¶ >>

Re: [sage-release] Sage 9.7.beta7 released

Do you know which versions of clang work if the user is using clang? Our compiler requirements are currently coming from https://docs.scipy.org/doc/scipy/dev/toolchain.html#c-language-standards, which unfortunately does not talk about clang. On Tuesday, August 2, 2022 at 10:46:32 AM UTC-7 Dima

Re: [sage-release] Sage 9.7.beta7 released

On Tue, 2 Aug 2022, 18:33 Matthias Köppe, wrote: > On Tuesday, August 2, 2022 at 10:24:39 AM UTC-7 Dima Pasechnik wrote: > >> >> there is a bit of vagueness there: >> Support for system GCC older than 6.3 dropped ¶ >>

Re: [sage-release] Sage 9.7.beta7 released

On Tuesday, August 2, 2022 at 10:24:39 AM UTC-7 Dima Pasechnik wrote: > > there is a bit of vagueness there: > Support for system GCC older than 6.3 dropped ¶ > > > Building Sage from source now

Re: [sage-release] Sage 9.7.beta7 released

On Tue, 2 Aug 2022, 18:17 Matthias Köppe, wrote: > On Tuesday, August 2, 2022 at 10:13:05 AM UTC-7 david@gmail.com wrote: > >> With this new beta, I see that the source code and the compiled code are >> mixed. >> For instance >> sapristi:sage dcoudert$ ll src/sage/graphs/ >> total 76264 >>

Re: [sage-release] Sage 9.7.beta7 released

I see. Thank you for the answers. So I will have to live with it. Best. > Le 2 août 2022 à 19:19, Dima Pasechnik a écrit : > > it's the -- enable-editable option which is now the default. > > One advantage is that after changing a .py file, one does not need ./sage -b > > One can turn it off

Re: [sage-release] Sage 9.7.beta7 released

it's the -- enable-editable option which is now the default. One advantage is that after changing a .py file, one does not need ./sage -b One can turn it off by running ./configure On Tue, 2 Aug 2022, 18:13 David Coudert, wrote: > With this new beta, I see that the source code and the

Re: [sage-release] Sage 9.7.beta7 released

On Tuesday, August 2, 2022 at 10:13:05 AM UTC-7 david@gmail.com wrote: > With this new beta, I see that the source code and the compiled code are > mixed. > For instance > sapristi:sage dcoudert$ ll src/sage/graphs/ > total 76264 > drwxr-xr-x 123 dcoudert coati 3936 2 aoû 16:10 . >

Re: [sage-release] Sage 9.7.beta7 released

With this new beta, I see that the source code and the compiled code are mixed. For instance sapristi:sage dcoudert$ ll src/sage/graphs/ total 76264 drwxr-xr-x 123 dcoudert coati 3936 2 aoû 16:10 . drwxr-xr-x 69 dcoudert coati 2208 1 aoû 23:18 .. drwxr-xr-x 19 dcoudert coati

Re: [sage-release] Sage 9.7.beta7 released

On Debian TESTING (Amd 64): -installation ok. - make ptestlong: 2 tests failed: 67134 src/sage/combinat/nu_dyck_word.py [179 tests, 1.17 s] sage -t --long --warn-long 34.1 --random-seed=3139455004425827870060396354068704 67134 src/sage/libs/gap/test_long.py [6 tests, 34.43 s]

Re: [sage-release] Sage 9.7.beta7 released

This is on Apple Silicon Macs with Homebrew up to date (make and make pytest and make ptestlong). Nothing new: On macOS 12.5 (Monterey) with Xcode 13.4.1: no problems. On macOS 11.6.8 (Big Sur) with Xcode 13.2.1: make ptestlong gives the usual matrix_gf2e_dense.pyx error (see

[sage-release] Sage 9.7.beta7 released

As always, you can get the latest beta version from the "develop" git branch. Alternatively, the self-contained source tarball is at http://www.sagemath.org/download-latest.html cd1e2b13af (HEAD -> develop, tag: 9.7.beta7, trac/develop) Updated SageMath version to 9.7.beta7 c94df55711 Trac