Closed rocco8773 closed 4 years ago
failure is most probably unrelated and I think is present in master, too. I would first try to limit the pytest version to be <5.4 as we had issues with it elsewhere.
Merging #100 into master will increase coverage by
0.03%
. The diff coverage is75.00%
.
@@ Coverage Diff @@
## master #100 +/- ##
==========================================
+ Coverage 87.15% 87.19% +0.03%
==========================================
Files 5 5
Lines 693 695 +2
==========================================
+ Hits 604 606 +2
Misses 89 89
Impacted Files | Coverage Δ | |
---|---|---|
sphinx_automodapi/automodapi.py | 91.13% <ø> (ø) |
|
sphinx_automodapi/smart_resolver.py | 78.78% <75.00%> (+0.66%) |
:arrow_up: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update 30953a9...01add30. Read the comment docs.
It looks like the error was with coverage
, so I limited that to coverage<5.0
.
I also set all the builds to run on the latest versions of macOS, Windows, and Ubuntu. This caused the mac build to actually run. This setup currently works, but I'll leave it up to you to which environments you actually want to test on.
When will a new release be uploaded to pypi with this fix?
When Spinx updated to
v3.0.0
,sphinx-automodapi
started raisingKeyErrors
on'refexplicit'
and'refdoc'
in thesmart_resolver.missing_reference_handler()
function. I could not find the change in Sphinx that caused this, but the update caused the originalnode
to not be populated with these attributes anymore. To resolve this I replaced the key callnode['refexplicit']
with a get callnode.get('refexplicit')
, which applies a default value if the key does not exist.Based on this very helpful comment https://github.com/sphinx-doc/sphinx/issues/1572#issuecomment-68590981, I added docstrings and comments to better inform what the function does.
Note: I noticed these lines of code were not hit by tests, thus, the tests never failed. I do not know enough about the architecture to write these tests, so I'm open to suggestions.
Closes #99