Split tests into a separate package, plus rattler-build

[mgorny]

Checklist

• Used a personal fork of the feedstock to propose changes
• Bumped the build number (if the version is unchanged)
• Reset the build number to 0 (if the version changed)
• Re-rendered with the latest conda-smithy (Use the phrase @conda-forge-admin, please rerender in a comment in this PR for automated rerendering)
• Ensured the license file is being packaged.

• split tests into a separate package as suggested in Recipe improvements #22, reducing the main package size
• removed obsolete code for CUDA < 12

[conda-forge-admin]

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/meta.yaml) and found it was in an excellent condition.

I do have some suggestions for making it better though...

For recipe/meta.yaml:

• ℹ️ The recipe is not parsable by parser conda-souschef (grayskull). This parser is not currently used by conda-forge, but may be in the future. We are collecting information to see which recipes are compatible with grayskull.
• ℹ️ The recipe is not parsable by parser conda-recipe-manager. The recipe can only be automatically migrated to the new v1 format if it is parseable by conda-recipe-manager.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/12484953648. Examine the logs at this URL for more detail.}

[hmaarrfk]

Thanks I think reduced download sizes are definitely welcome!

[conda-forge-admin]

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

I failed to even lint the recipe, probably because of a conda-smithy bug 😢. This likely indicates a problem in your meta.yaml, though. To get a traceback to help figure out what's going on, install conda-smithy and run conda smithy recipe-lint --conda-forge . from the recipe directory. You can also examine the workflow logs for more detail.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/12495694887. Examine the logs at this URL for more detail.}

[mgorny]

I've added a rattler-build conversion, and filed output request in conda-forge/admin-requests#1248. FWICS conda-smithy crashes on it, I'm going to look into that later today.

[mgorny]

I'm going to guess that the osx failure is a fluke.

[mgorny]

The conda-smithy problems are conda-forge/conda-smithy#2165 and conda-forge/conda-smithy#2196. The former I've been able to workaround by avoiding nested conditions.

[hmaarrfk]

I came to check in, and saw

test_decode_gif[True-treescap]

failed due to connection issues. I always feel bad since conda-forge hammers webservers which are meant to be cached locally for data.

Should we skip this test?

[conda-forge-admin]

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

I considered the following status checks when analyzing this PR:

• linter: failed
• azure: passed

Thus the PR was not passing and not merged.

[mgorny]

Should we skip this test?

Yeah, FWICS test_decode_gif goes straight for requests.get() with no caching.

[hmaarrfk]

I guess our cis clear cache unless you work really hard not to. So any access to the internet during testing should be discouraged

[mgorny]

I'm going to do a test run with pytest-socket to check for other cases of Internet access.

[mgorny]

Ok, so the failures are test_decode_gif (as expected) and a few tests directly related to checking whether Internet-related API works. Should I skip the latter too, or leave them?

 │ FAILED test/test_image.py::test_decode_gif[True-gifgrid] - pytest_socket.Sock...
 │ FAILED test/test_image.py::test_decode_gif[True-fire] - pytest_socket.SocketB...
 │ FAILED test/test_image.py::test_decode_gif[True-porsche] - pytest_socket.Sock...
 │ FAILED test/test_image.py::test_decode_gif[True-treescap] - pytest_socket.Soc...
 │ FAILED test/test_image.py::test_decode_gif[True-treescap-interlaced] - pytest...
 │ FAILED test/test_image.py::test_decode_gif[True-solid2] - pytest_socket.Socke...
 │ FAILED test/test_image.py::test_decode_gif[True-x-trans] - pytest_socket.Sock...
 │ FAILED test/test_image.py::test_decode_gif[True-earth] - pytest_socket.Socket...
 │ FAILED test/test_image.py::test_decode_gif[False-gifgrid] - pytest_socket.Soc...
 │ FAILED test/test_image.py::test_decode_gif[False-fire] - pytest_socket.Socket...
 │ FAILED test/test_image.py::test_decode_gif[False-porsche] - pytest_socket.Soc...
 │ FAILED test/test_image.py::test_decode_gif[False-treescap] - pytest_socket.So...
 │ FAILED test/test_image.py::test_decode_gif[False-treescap-interlaced] - pytes...
 │ FAILED test/test_image.py::test_decode_gif[False-solid2] - pytest_socket.Sock...
 │ FAILED test/test_image.py::test_decode_gif[False-x-trans] - pytest_socket.Soc...
 │ FAILED test/test_image.py::test_decode_gif[False-earth] - pytest_socket.Socke...
 │ FAILED test/test_internet.py::TestDatasetUtils::test_download_url[True] - pyt...
 │ FAILED test/test_internet.py::TestDatasetUtils::test_download_url[False] - py...
 │ FAILED test/test_internet.py::TestDatasetUtils::test_download_url_retry_http[True]
 │ FAILED test/test_internet.py::TestDatasetUtils::test_download_url_retry_http[False]
 │ FAILED test/test_internet.py::TestDatasetUtils::test_download_url_dont_exist[True]
 │ FAILED test/test_internet.py::TestDatasetUtils::test_download_url_dont_exist[False]
 │ FAILED test/test_internet.py::TestDatasetUtils::test_download_url_dispatch_download_from_google_drive[True]
 │ FAILED test/test_internet.py::TestDatasetUtils::test_download_url_dispatch_download_from_google_drive[False]

[hmaarrfk]

Yeah. Any test that needs a download should be skipped

google drive will definitely boot you….