Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Hello,
Turns out there was more to be done. The tl;dr; is that we still need a shim
setup.py
, and a few other things changed.When I tried to install pywfa for a package feature I'm testing, the github action that tests my package wasn't able to successfully install pywfa==0.4.2. I was getting an error that pywfa was named "UNKNOWN-0.0.0"
full error
I also needed to fix the
MANIFEST.in
to include the_custom_build.py
and Makefile inpywfa/WFA2_lib/*
. Also, the new_custom_build.py
has a check to ensure that the make command works. However, the make command never exited non-zero due to only returning the status of the final command (wascd pywfa/WFA2_lib; make .. ; cd ../
).I believe these changes should work better because I was able to get
python3 setup.py sdist
to create a valid, usablepython3 -m pip install dist/pywfa-0.5.0.dev1.tar.gz
. And by having the shim, things likepip install -e
seem to work. Furthermore, I successfully tested by editing my github action to pull directly via:python3 -m pip install 'pywfa @ git+https://github.com/ACEnglish/pywfa'
. Which is great, but I don't know if this is a great emulation of what pypi will make.Finally, I bumped the version to 0.5.0-dev. At least once trying to install 0.4.2 failed and pip slyly installed 0.4.1 instead (which had caused the old
undefined symbol: wavefront_align
). I'm guessing that bumping the minor version instead of the patch may prevent this.I don't know your workflow for distributing to pypi, but if possible, could you try a test.pypi.org upload before pushing to regular pypi? If there's more problems, we'll at least not be filling it up with my mistakes lol.
Have a great day,
~/Adam English