Upgrading NumPy and friends
Hi Guix! I would like to open a conversation on upgrading NumPy, Pandas, Numba and SciPY. It would be reasonable to refresh that core scientific packages for Python after merging python-team which is nearly finished building. Stats: - branch :: master - repository URL :: https://git.savannah.gnu.org/git/guix.git - commit :: 48097f511929053468ce6f09e0a24644c90fe670 | Package | Dependents | Guix version | Upstream version | Checked | |---++--+--+| | python-numba |136 | 0.56.4 | 0.60.0 | <2024-10-19 Sat 08:42> | | python-numpy | 3394 | 1.23.2 |2.1.2 | <2024-10-19 Sat 08:38> | | python-pandas |423 |2.1.1 |2.2.3 | <2024-10-19 Sat 08:43> | | python-scipy |334 | 1.12.0 | 1.14.1 | <2024-10-19 Sat 08:42> | I've started noticing that more and more projects require a fresh version of NumPy during Astro upgrade monthly, sometimes it's captured on sanity check phase, but most of the time the package is just not work especially when it depends on python-astropy which heavily depends on NumPy. Does anyone work on that packages or would be interested in taking the flag? -- Thanks, Oleg signature.asc Description: PGP signature
Re: Seemingly unintentional near-world rebuild
Hello Ian, Am Sat, Oct 19, 2024 at 09:53:06AM -0700 schrieb Ian Eure: > > When my Cuirass pulled commit 5794926bed6fad4598bb565fb7f49be4205b11a1 > > this morning, it started rebuilding every package in my channel. This > > includes a package with zero inputs other than what cmake-build-system > > needs[1]. > > > > ci.guix has been evaluating this commit for 90 minutes[2] at the time > > of this writing, but hasn’t started any builds (it’s also 504ing > > around half the times I try to load it). I think it’s going to do a > > world rebuild, or very close to it. > > > > Was this expected? Should anything be done about it? difficult to do anything now, there are a bunch of commits that would need to be reverted. Normally the world rebuild has been done on bayfront for the fonts split branch. I have just rebased it on master. So indeed what is new on master since October 7 may need rebuilding. But I tried with my home profile, and all of the packages were available. Also, QA says that there is above 90% availability on ci and bordeaux for the master branch on x86_64, and about 90% on i686. Close to 100% on aarch64 on bordeaux as well. (Hm, this seems to be for the revision before the merge. So I do not know; the numbers for the fonts split branch were similar, however.) So the only situation where I would see a lot of rebuilding is aarch64 for persons who have disabled substitutes from bordeaux. Or does this "my channel" refer to a private channel? Then indeed, the merge was a world rebuild change; so external channels will only now start to be built. Andreas
Re: Seemingly unintentional near-world rebuild
Fixing broken cc to Andreas. Ian Eure writes: When my Cuirass pulled commit 5794926bed6fad4598bb565fb7f49be4205b11a1 this morning, it started rebuilding every package in my channel. This includes a package with zero inputs other than what cmake-build-system needs[1]. ci.guix has been evaluating this commit for 90 minutes[2] at the time of this writing, but hasn’t started any builds (it’s also 504ing around half the times I try to load it). I think it’s going to do a world rebuild, or very close to it. Was this expected? Should anything be done about it? Thanks, — Ian [1]: https://codeberg.org/ieure/atomized-guix/src/branch/main/atomized/packages/floppy-disk.scm#L54-L77 [2]: https://ci.guix.gnu.org/eval/1733664
