Skip to content
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

Add minimal 'mypy' run to the pre-commit hooks. #4176

Merged
merged 2 commits into from
Jun 20, 2020

Conversation

greschd
Copy link
Member

@greschd greschd commented Jun 15, 2020

Currently, only two files are checked: aiida/tools/groups/paths.py and aiida/engine/processes/calcjobs/calcjob.py. The option follow_imports is set to skip, because there are a lot of issues in other files, mostly because these lack type hints.
Even on the two checked files, the strictness level of mypy is minimal (see the commented options in mypy.ini).

From here, we can continue expanding the type hints one file at
a time:

  • add the file to files in the mypy pre-commit hook
  • fix all issues that mypy reports

@chrisjsewell I had to change some of the type hints in groups/paths.py - please check that these are correct.

@chrisjsewell
Copy link
Member

Yeh but you know we can't add these proper type annotations, until we drop python 3.5 support in October: https://devguide.python.org/#status-of-python-branches

@sphuber
Copy link
Contributor

sphuber commented Jun 15, 2020

Yeh but you know we can't add these proper type annotations, until we drop python 3.5 support in October: https://devguide.python.org/#status-of-python-branches

These are only method argument and return value annotations, which is supported by Python 3.5. We just cannot add variable annotations because that was only added in Python 3.6.

Anyway, I really cannot wait for pyproject.toml to catch on and we can do away with the hundreds of different configuration files in the repo root for all the different tools we run...

@greschd
Copy link
Member Author

greschd commented Jun 15, 2020

Only variable annotations and lazy annotations are not supported, right? Need to fix the lazy thing, though..

@chrisjsewell
Copy link
Member

We just cannot add variable annotations because that was only added in Python 3.6

Ah fair enough brilliant, for some reason I thought it was 3.5, otherwise I would have already added them!

@greschd greschd force-pushed the add_minimal_mypy_run branch 2 times, most recently from 75f85e1 to eb7701d Compare June 15, 2020 15:00
@chrisjsewell
Copy link
Member

Only variable annotations and lazy annotations are not supported, right?

Well the other big thing in the pipeline is TypedDict (introduced in 3.8). Having worked with interfaces in TypeScript, which are similar, these will be so helpful.

Other than that LGTM 👍

@greschd
Copy link
Member Author

greschd commented Jun 15, 2020

@chrisjsewell just changed something more: get_or_create_group returns Tuple[orm.Group, bool], not Tuple['GroupPath', bool], right?

@greschd
Copy link
Member Author

greschd commented Jun 15, 2020

Ah, and

def walk_nodes(self, filters=None, node_class=None, query_batch=None) -> Iterable[WalkNodeResult]:

to

def walk_nodes(self, filters=None, node_class=None, query_batch=None) -> Iterator[WalkNodeResult]:

aiida/tools/groups/paths.py Outdated Show resolved Hide resolved
aiida/tools/groups/paths.py Outdated Show resolved Hide resolved
aiida/tools/groups/paths.py Outdated Show resolved Hide resolved
@codecov
Copy link

codecov bot commented Jun 15, 2020

Codecov Report

Merging #4176 into develop will increase coverage by 0.01%.
The diff coverage is 100.00%.

Impacted file tree graph

@@             Coverage Diff             @@
##           develop    #4176      +/-   ##
===========================================
+ Coverage    78.94%   78.94%   +0.01%     
===========================================
  Files          467      467              
  Lines        34501    34510       +9     
===========================================
+ Hits         27232    27242      +10     
+ Misses        7269     7268       -1     
Flag Coverage Δ
#django 70.87% <100.00%> (+0.02%) ⬆️
#sqlalchemy 71.75% <100.00%> (+0.02%) ⬆️
Impacted Files Coverage Δ
aiida/tools/groups/paths.py 91.62% <100.00%> (+0.48%) ⬆️
aiida/transports/plugins/local.py 80.47% <0.00%> (+0.26%) ⬆️

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 41f5d5d...23e1eb8. Read the comment docs.

@greschd
Copy link
Member Author

greschd commented Jun 15, 2020

 sqlalchemy.exc.TimeoutError: QueuePool limit of size 5 overflow 0 reached, connection timed out, timeout 1 (Background on this error at: http://sqlalche.me/e/3o7r)

Is this a known flaky test?

aiida/tools/groups/paths.py Outdated Show resolved Hide resolved
chrisjsewell
chrisjsewell previously approved these changes Jun 15, 2020
chrisjsewell
chrisjsewell previously approved these changes Jun 15, 2020
@sphuber
Copy link
Contributor

sphuber commented Jun 16, 2020

 sqlalchemy.exc.TimeoutError: QueuePool limit of size 5 overflow 0 reached, connection timed out, timeout 1 (Background on this error at: http://sqlalche.me/e/3o7r)

Is this a known flaky test?

I haven't seen it fail yet, but it is relatively new. This looks like it comes from a test that was put in to check the REST API cleanly closing the SqlAlchemy sessions so we don't get a timeout because the queuepool is exhausted. Clearly this was triggered here where we run with reduced pool size and overflow, but don't see how your changes should have affected it. So I would just chalk this up to bad luck for now and ignore it until we start to see it more often

setup.json Outdated Show resolved Hide resolved
@sphuber
Copy link
Contributor

sphuber commented Jun 16, 2020

BTW, I have just been reading up on a 2-year long discussion on mypy adding support for pyproject.toml. Seems it might be getting close to getting there: python/mypy#5208
pytest also just merged it and should be releasing support with v6.0

@greschd
Copy link
Member Author

greschd commented Jun 16, 2020

FYI, there is a potential for confusion because our orm and the stdlib typing share some names, specifically Dict and List.

In my own packages I solved this by

import typing as ty
<...>
ty.Dict

I'm open for better suggestions, but since this is a problem that's bound to happen in many places in our code base, it's probably a good idea to have a consistent style.

Should probably be changed in paths.py once we agree on the style.

chrisjsewell
chrisjsewell previously approved these changes Jun 16, 2020
Currently, only two files are checked: aiida/tools/groups/paths.py
and aiida/engine/processes/calcjobs/calcjob.py. The option
'follow_imports' is set to 'skip', because there are a lot of
issues in other files, mostly because these lack type hints.

From here, we can continue expanding the type hints one file at
a time:
- add the file to 'files' in the 'mypy' pre-commit hook
- fix all issues that mypy reports

Co-authored-by: Chris Sewell <chrisj_sewell@hotmail.com>
Comment on lines +8 to +31
; Strictness settings
; disallow_any_unimported = True
; disallow_any_expr = True
; disallow_any_decorated = True
; disallow_any_explicit = True
; disallow_any_generics = True
; disallow_subclassing_any = True

; disallow_untyped_calls = True
; disallow_untyped_defs = True
; disallow_incomplete_defs = True
; disallow_untyped_decorators = True

; no_implicit_optional = True
; no_strict_optional = False

; Enable all warnings
; warn_redundant_casts = True
; warn_unused_ignores = True
; warn_return_any = True
; warn_unreachable = True

; allow_untyped_globals = False
; strict_equality = True
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can we get rid of all these if we are not using them?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, although it might be nice seeing them listed here since (I think) the goal would be to slowly enable some of them.

ignore_missing_imports = True

[mypy-plumpy.*]
ignore_missing_imports = True
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this similar to pylint issuing missing import warnings if it is not being run in the "system" environment? If so, would it make sense to also run mypy under system?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Mypy.. has its own "import system" - the problem here is that when dependencies don't have type annotations, that will be reported as a failure.

For now I'm not sure this makes a difference since we also use skip for the follow_imports option, but we definitely want to get rid of that one eventually.

The import mechanism is described here: https://mypy.readthedocs.io/en/stable/running_mypy.html#how-mypy-handles-imports

@sphuber sphuber merged commit e9d3153 into aiidateam:develop Jun 20, 2020
@sphuber sphuber deleted the add_minimal_mypy_run branch June 20, 2020 10:59
@sphuber
Copy link
Contributor

sphuber commented Jun 20, 2020

Thanks @greschd

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants