Opened 3 years ago
Closed 2 years ago
#29061 closed enhancement (fixed)
Upgrade to symmetrica-3.0.1
Reported by: | mjo | Owned by: | |
---|---|---|---|
Priority: | blocker | Milestone: | sage-9.2 |
Component: | packages: standard | Keywords: | |
Cc: | arojas, fbissey, saraedum, gh-timokau | Merged in: | |
Authors: | Michael Orlitzky, Dima Pasechnik | Reviewers: | Matthias Koeppe |
Report Upstream: | N/A | Work issues: | |
Branch: | 68ee17e (Commits, GitHub, GitLab) | Commit: | 68ee17e4e1a52e4988623dda05fbc2d1ab27c975 |
Dependencies: | Stopgaps: |
Description
I tagged a new release of symmetrica with all of the build/crash fixes from over the years:
- https://gitlab.com/sagemath/symmetrica/-/releases
- https://gitlab.com/sagemath/symmetrica/uploads/b3d8e1ad5ab2449c30bbc3147e7a5e53/symmetrica-3.0.1.tar.xz
A standard ./configure && make && make install should suffice, with no patching.
Change History (25)
comment:1 Changed 3 years ago by
Cc: | arojas fbissey saraedum gh-timokau added |
---|
comment:2 follow-up: 3 Changed 3 years ago by
comment:3 Changed 3 years ago by
Replying to gh-timokau:
The README states "The original developers of symmetrica at the University of Bayreuth are no longer active, but symmetrica is still in use in SageMath and elsewhere". Is there some source to this claim? Did anyone ever try contacting upstream?
Yes, the primary maintainer passed away a few years ago and as far as I know there is no one else at Bayreuth working on symmetrica. A colleague of his, Michael Kiermaier, has given us his blessing insofar as is possible. And the other distribution maintainers are on board.
comment:4 Changed 3 years ago by
Ah, now I remember the sage-devel discussion. That is sad to hear. Thanks for clarifying.
comment:5 follow-up: 6 Changed 3 years ago by
Is there anything in the new release that provides version information to programs linking to the library?
comment:6 Changed 3 years ago by
Replying to mkoeppe:
Is there anything in the new release that provides version information to programs linking to the library?
You can detect this exact version by looking at the soname (which was bumped) or some of the files we rearranged. We also took a few things out of the public API that were there by mistake (and so can be detected).
But long term, this is the way to go: https://gitlab.com/sagemath/symmetrica/-/issues/10
comment:7 Changed 3 years ago by
Milestone: | sage-9.1 → sage-9.2 |
---|
comment:8 Changed 2 years ago by
Authors: | → Michael Orlitzky, Dima Pasechnik |
---|---|
Branch: | → public/packages/symmetrica/symmetrica301 |
Commit: | → 98b9e75673ea3607d444be8a30bd1d269cb33807 |
Priority: | major → blocker |
Status: | new → needs_review |
here is a branch, please test. It builds with SAGE_CHECK=yes on macOS 10.15 with Xcode 12.
New commits:
98b9e75 | symmetrica 3.0.1 from the GitLab repo
|
comment:9 follow-up: 12 Changed 2 years ago by
also works on Fedora 30 - but for some reason I need to download the tarball mentioned in checksum.ini manually.
comment:10 Changed 2 years ago by
CFLAGS="-O2"
should suffice. The preprocessor definitions are made automatically now in src/Makefile.am
:
# -DFAST disables some runtime checks in src/macro.h to speed things up. # # -DALLTRUE defines a bunch of other constants in def.h that ultimately # enable every symmetrica "subsystem." libsymmetrica_la_CPPFLAGS = -DFAST -DALLTRUE
The cd src
can also be dropped. Right now, src
is the only subdirectory with a test suite, but make check
at the top level runs is what runs them all.
comment:12 Changed 2 years ago by
Replying to dimpase:
also works on Fedora 30 - but for some reason I need to download the tarball mentioned in checksum.ini manually.
We had the same problems on another upgrade ticket that used a gitlab URL. Perhaps best to put the tarball somewhere else to allow for automatic testing...
comment:13 Changed 2 years ago by
Commit: | 98b9e75673ea3607d444be8a30bd1d269cb33807 → 68ee17e4e1a52e4988623dda05fbc2d1ab27c975 |
---|
Branch pushed to git repo; I updated commit sha1. New commits:
68ee17e | better tarball location
|
comment:19 Changed 2 years ago by
Status: | needs_review → needs_work |
---|
Patches don't apply - see for example https://github.com/mkoeppe/sage/runs/1152418779
comment:21 follow-up: 23 Changed 2 years ago by
comment:22 Changed 2 years ago by
Status: | needs_work → needs_review |
---|
comment:23 Changed 2 years ago by
Replying to jhpalmieri:
Oh, don't use both #30608 and this ticket, use just this one. If this one gets positively reviewed, that one can be closed as invalid. If this one runs into problems or takes too long to get reviewed on enough platforms, then #30608 will be a stopgap.
Ah, my mistake, sorry for the noise. I had merged all the Xcode 12 related tickets that were in positive review or needs review status.
comment:24 Changed 2 years ago by
Reviewers: | → Matthias Koeppe |
---|---|
Status: | needs_review → positive_review |
comment:25 Changed 2 years ago by
Branch: | public/packages/symmetrica/symmetrica301 → 68ee17e4e1a52e4988623dda05fbc2d1ab27c975 |
---|---|
Resolution: | → fixed |
Status: | positive_review → closed |
The README states "The original developers of symmetrica at the University of Bayreuth are no longer active, but symmetrica is still in use in SageMath and elsewhere". Is there some source to this claim? Did anyone ever try contacting upstream?