search

conda-forge / osqp-feedstock

A conda-smithy repository for osqp.
BSD 3-Clause "New" or "Revised" License
0 stars 13 forks source link

Rebuild for python39 #24

Closed regro-cf-autotick-bot closed 3 years ago

regro-cf-autotick-bot commented 4 years ago

This PR has been triggered in an effort to update python39.

Notes and instructions for merging this PR:

  1. Please merge the PR only after the tests have passed.
  2. Feel free to push to the bot's branch to update this PR if needed.

Please note that if you close this PR we presume that the feedstock has been rebuilt, so if you are going to perform the rebuild yourself don't close this PR until the your rebuild has been merged.

This package has the following downstream children:

And potentially more.

If this PR was opened in error or needs to be updated please add the bot-rerun label to this PR. The bot will close this PR and schedule another one. If you do not have permissions to add this label, you can use the phrase code>@<space/conda-forge-admin, please rerun bot in a PR comment to have the conda-forge-admin add it for you.

This PR was created by the regro-cf-autotick-bot. The regro-cf-autotick-bot is a service to automatically track the dependency graph, migrate packages, and propose package version updates for conda-forge. If you would like a local version of this bot, you might consider using rever. Rever is a tool for automating software releases and forms the backbone of the bot's conda-forge PRing capability. Rever is both conda (conda install -c conda-forge rever) and pip (pip install re-ver) installable. Finally, feel free to drop us a line if there are any issues! This PR was generated by , please use this URL for debugging

conda-forge-linter commented 4 years ago

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

github-actions[bot] commented 4 years ago

Hi! This is the friendly conda-forge automerge bot!

I considered the following status checks when analyzing this PR:

Thus the PR was not passing and not merged.

h-vetinari commented 4 years ago

@bstellato @gbanjac Interestingly, the same recipe as before is now failing with 23 test errors, also on versions not affected by the 3.9 migration. Any idea what's happening on osqp-side here?

I did see that we could update the qdldl-version from 0.1.4 to 0.1.5 (and eventually, we could move to depending on qdldl-python?), but I strongly doubt that that's the reason for the failure.

h-vetinari commented 4 years ago

Also blocked on https://github.com/conda-forge/scipy-feedstock/issues/139

Anthchirp commented 3 years ago

The scipy Windows build is now available.

@conda-forge-admin, please restart ci

h-vetinari commented 3 years ago

@conda-forge-admin, please rerender

h-vetinari commented 3 years ago

@bstellato @gbanjac @conda-forge/osqp I'm going to need help with the test failures here, I think (see my comment 3 weeks ago)...

h-vetinari commented 3 years ago

@conda-forge-admin, please rerender

h-vetinari commented 3 years ago

@bstellato @gbanjac

Coming back to this since it is blocking (e.g.) cvxpy on python 3.9...

Adding setuptools as a runtime dependency has reduced the failures quite a bit. We shouldn't be doing that, but I guess for the current state of things it's unavoidable (though arguably it could be just host- & test- dependency if the codegen stuff is not expected to work without; or we make a separate package osqp-codegen that depends on setuptools?)

However, there are still some test failures remaining. In particular, it seems that all the matrix-update tests (and only those) are failing: 

============================= test session starts ==============================
platform linux -- Python 3.9.1, pytest-6.1.2, py-1.9.0, pluggy-0.13.1
rootdir: $SRC_DIR
collected 49 items

tests/basic_test.py ...........                                          [ 22%]
tests/codegen_matrices_test.py .........                                 [ 40%]
tests/codegen_vectors_test.py .....                                      [ 51%]
tests/dual_infeasibility_test.py ...                                     [ 57%]
tests/feasibility_test.py .                                              [ 59%]
tests/mkl_pardiso_test.py .                                              [ 61%]
tests/non_convex_test.py ...                                             [ 67%]
tests/polishing_test.py ...                                              [ 73%]
tests/primal_infeasibility_test.py ..                                    [ 77%]
tests/unconstrained_test.py .                                            [ 79%]
tests/update_matrices_test.py FFFFFFFFF                                  [ 97%]
tests/warm_start_test.py .                                               [100%]

Those tests have hard-coded expected values, and those are violated now, but very consistently across all versions & platforms. Can we be sure that these are not maybe wrong for whatever upstream bug?

bstellato commented 3 years ago

Sorry about the delay. We just released 0.6.2. It should fix these errors. @h-vetinari the bot should open a new PR for 0.6.2.

h-vetinari commented 3 years ago

Not sure which package hasn't been built here yet, but it seems to push the resolver to python 2 for some reason:

Package ca-certificates conflicts for:
python[version='>=3.9,<3.10.0a0'] -> openssl[version='>=1.1.1i,<1.1.2a'] -> ca-certificates
setuptools -> python[version='>=2.7,<2.8.0a0'] -> ca-certificates
pytest -> python[version='>=2.7,<2.8.0a0'] -> ca-certificates
cvxpy -> python[version='>=2.7,<2.8.0a0'] -> ca-certificates
scipy -> python[version='>=2.7,<2.8.0a0'] -> ca-certificates
conda-forge-linter commented 3 years ago

Hi! This is the friendly automated conda-forge-linting service.

I wanted to let you know that I linted all conda-recipes in your PR (recipe) and found some lint.

Here's what I've got...

For recipe:

conda-forge-linter commented 3 years ago

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

h-vetinari commented 3 years ago

Ah, now I understand. There's a circular (test-)dependency of osqp on cvxpy, which itself depends on osqp. Will have to deactivate the tests for python 3.9 for the time being....

Package python conflicts for:
cvxpy[version='>=1.1.7'] -> python[version='>=3.6,<3.7.0a0|>=3.7,<3.8.0a0|>=3.8,<3.9.0a0']
[...]
bstellato commented 3 years ago

Thanks @h-vetinari for fixing the merge. I have just made a post-release where I switched from cvxpy to cvxopt to run the tests. It should no longer have the circular dependency