gmp 6.3.0, mpc 1.3.1, mpfr 4.2.1, mpfrcx 0.6.3, gmpy2 2.2.0a1#36351
gmp 6.3.0, mpc 1.3.1, mpfr 4.2.1, mpfrcx 0.6.3, gmpy2 2.2.0a1#36351vbraun merged 19 commits intosagemath:developfrom
gmp 6.3.0, mpc 1.3.1, mpfr 4.2.1, mpfrcx 0.6.3, gmpy2 2.2.0a1#36351Conversation
|
FWIW, we are shipping 2.2.0a1 in void linux since september. This supports python 3.12 and is available on pypi: https://pypi.org/project/gmpy2/2.2.0a1/ |
|
Thanks @tornaria, I'll try switching to this release! |
|
Full tests at https://github.com/mkoeppe/sage/actions/runs/7205490817
|
|
we can as we often do get the MPC version from mpc.h - akin to what we do for NTL 😭 /* Define MPC version number */
#define MPC_VERSION_MAJOR 1
#define MPC_VERSION_MINOR 3
#define MPC_VERSION_PATCHLEVEL 1
#define MPC_VERSION_STRING "1.3.1"
/* Macros dealing with MPC VERSION */
#define MPC_VERSION_NUM(a,b,c) (((a) << 16L) | ((b) << 8) | (c))
#define MPC_VERSION \
MPC_VERSION_NUM(MPC_VERSION_MAJOR,MPC_VERSION_MINOR,MPC_VERSION_PATCHLEVEL)a bit shorter than NTL code will use |
|
I'm just copying the test from gmpy.h |
|
Do you mean |
|
No |
|
This is a compile-time test |
|
It's Also not clear how you make sure to test for |
|
I'll get back to it later when I have a bit more time. Feel free to push a fix |
|
ready now, I think |
|
Thanks. I've switched the lib test to use a newer function, like I did previously in 109671e |
|
Thanks. |
sagemathgh-36351: build/pkgs/gmpy2: Update to 2.2.0a1 <!-- ^^^^^ Please provide a concise, informative and self-explanatory title. Don't put issue numbers in there, do this in the PR body below. For example, instead of "Fixes sagemath#1234" use "Introduce new method to calculate 1+1" --> <!-- Describe your changes here in detail --> <!-- Why is this change required? What problem does it solve? --> The unreleased 2.2.x series has work on support for Python 3.12 https://github.com/aleaxit/gmpy The new version makes it necessary to tighten the requirements for MPC and MPFR. We do the version check without pkg-config as before; changing it to pkg-config is sagemath#31348. <!-- If this PR resolves an open issue, please link to it here. For example "Fixes sagemath#12345". --> <!-- If your change requires a documentation PR, please link it appropriately. --> ### 📝 Checklist <!-- Put an `x` in all the boxes that apply. --> <!-- If your change requires a documentation PR, please link it appropriately --> <!-- If you're unsure about any of these, don't hesitate to ask. We're here to help! --> <!-- Feel free to remove irrelevant items. --> - [x] The title is concise, informative, and self-explanatory. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### ⌛ Dependencies <!-- List all open PRs that this PR logically depends on - sagemath#12345: short description why this is a dependency - sagemath#34567: ... --> - Depends on sagemath#36775 (merged here) <!-- If you're unsure about any of these, don't hesitate to ask. We're here to help! --> URL: sagemath#36351 Reported by: Matthias Köppe Reviewer(s):
|
Conda tests fail for this PR. |
|
I explained the principle to you in #35593 (comment) |
|
Since when do we have the principle to just ignore tests when they don't suit you? There is no need for sagelib to require a unreleased version of gmpy2. |
|
Tobias, stop this nonsense. |
|
if conda tests fail, we need to make sure this gmpy2 version (released on PyPI) makes it to Conda quickly. I am not sure it is feasible to maintain support for two seriously different versions of gmpy2 and a bunch of libraries - unless it all just works on Conda, and the test in question breaks it for no good reason. |
|
Tests fail because of missing hypothesis, is this intentional? |
…ctually accepts our current version
|
With the gmpy2 upgrade here, passes all tests on the same machine where I reproduced the failures reported in #36775 (comment) with the old gmpy2. |
|
Documentation preview for this PR (built with commit 4bdb76d; changes) is ready! 🎉 |
|
All tests passed, let's get this in. |
sagemathgh-36351: `gmp` 6.3.0, `mpc` 1.3.1, `mpfr` 4.2.1, `mpfrcx` 0.6.3, `gmpy2` 2.2.0a1 <!-- ^^^^^ Please provide a concise, informative and self-explanatory title. Don't put issue numbers in there, do this in the PR body below. For example, instead of "Fixes sagemath#1234" use "Introduce new method to calculate 1+1" --> <!-- Describe your changes here in detail --> <!-- Why is this change required? What problem does it solve? --> The unreleased 2.2.x series has work on support for Python 3.12 https://github.com/aleaxit/gmpy The new version makes it necessary to tighten the requirements for MPC and MPFR. We do the version check without pkg-config as before; changing it to pkg-config is sagemath#31348. (The upgrade of the packages was previously done in sagemath#36775, but has now been merged here.) <!-- If this PR resolves an open issue, please link to it here. For example "Fixes sagemath#12345". --> <!-- If your change requires a documentation PR, please link it appropriately. --> ### 📝 Checklist <!-- Put an `x` in all the boxes that apply. --> <!-- If your change requires a documentation PR, please link it appropriately --> <!-- If you're unsure about any of these, don't hesitate to ask. We're here to help! --> <!-- Feel free to remove irrelevant items. --> - [x] The title is concise, informative, and self-explanatory. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### ⌛ Dependencies <!-- List all open PRs that this PR logically depends on - sagemath#12345: short description why this is a dependency - sagemath#34567: ... --> <!-- If you're unsure about any of these, don't hesitate to ask. We're here to help! --> URL: sagemath#36351 Reported by: Matthias Köppe Reviewer(s):
Note that conda usually does not package alphas. Is the alpha absolutely necessary now? Does anything break badly if I just install sagelib with gmpy 2.1.x? |
|
Check what's declared in |
|
It says gmpy2 ~=2.1.b999. Which means yes. Though that |
|
This should accept all 2.1.x versions, or at least so I claimed in the 4bdb76d commit. |
Right, that's not normalized |
The unreleased 2.2.x series has work on support for Python 3.12
https://github.com/aleaxit/gmpy
The new version makes it necessary to tighten the requirements for MPC and MPFR.
We do the version check without pkg-config as before; changing it to pkg-config is #31348.
(The upgrade of the packages was previously done in #36775, but has now been merged here.)
📝 Checklist
⌛ Dependencies