-
Notifications
You must be signed in to change notification settings - Fork 184
replace use of scripts
with entry_points
#311
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
replace use of scripts
with entry_points
#311
Conversation
Codecov Report
@@ Coverage Diff @@
## master #311 +/- ##
===========================================
- Coverage 100.00% 99.29% -0.71%
===========================================
Files 26 26
Lines 2115 2135 +20
===========================================
+ Hits 2115 2120 +5
- Misses 0 15 +15
Continue to review full report at Codecov.
|
I'll work to get it added to the test suite for coverage. |
Thanks for your PR, that seems like a good idea.
Great! |
This is actually quite useful data: windows does not support all of the signal handlers!
|
Oops. I could swear I tested it on Windows... That's indeed a good thing to also test the few lines launching the script... |
Moving the import got further on the |
moved them back into respective files, seems to skip correctly. |
I think it just needed the |
well, i guess integrated in the existing test cases is probably better, long term, than something special |
Any idea why none of the lines in the |
it looks like codecov hasn't updated, perhaps? |
I just tried it on my machine and I have the same result than codecov. Running
What command are you using to show the coverage? |
|
I tried with another conda env with your same python version and progressively modified/added my dependencies to match your pip list but I still have the same coverage detected and missing the lines 419-448 I am still on Ubuntu Linux 20.04.3 LTS |
As am i. Very strange. It might be a a bit before I can look back into this... trying to really run everything in-process might take a lot more mocks, what with another option might be to skip |
Looks like linux py38 is reporting what i'm seeing... I'll try adding a console report to the coverage job so we can see a little more... |
I also note that the coverage job doesn't have |
The lint issue is likely psf/black#2964 |
I am going to update black in a separate PR as it changes the formatting a bit and a lot of files are affected. |
since coverage is working under tox, what about adding a dedicated |
Looks like it did the thing?
|
Yes, it worked! I had the same behavior on my machine with 100% coverage with tox and not with make tests and was able to pinpoint one of the difference being using I have no idea why that's the case but it works... Codecov is not updating the comment but sometimes it can take some time. |
For all the These wheels can also be cached more effectively in CI (which is a separate PR, probably). As my point in doing all of this is getting to a more-cacheable and -portable state of installing |
Alright! |
Thanks again for your inputs. |
References
scripts
with cross-platformentry_points
#310Changes
scripts/gql-cli
to agql_cli
function ingql/cli.py
scripts/gql-cli
setup.py
MANIFEST.in
Makefile