Add "attr.validators.disabled" context manager

[sscherfke]

Pull Request Check List

This is just a friendly reminder about the most common mistakes. Please make sure that you tick all boxes. But please read our contribution guide at least once, it will save you unnecessary review cycles!

If an item doesn't apply to your pull request, check it anyway to make it apparent that there's nothing left to do. If your pull request is a documentation fix or a trivial typo, feel free to delete the whole thing.

• Added tests for changed code.
• [xx] New features have been added to our Hypothesis testing strategy.
• Changes or additions to public APIs are reflected in our type stubs (files ending in .pyi).
  • ...and used in the stub test file tests/typing_example.py.
• Updated documentation for changed code.
  • New functions/classes have to be added to docs/api.rst by hand.
  • Changes to the signature of @attr.s() have to be added by hand too.
  • Changed/added classes/methods/functions have appropriate versionadded, versionchanged, or deprecated directives. Find the appropriate next version in our __init__.py file.
• Documentation in .rst files is written using semantic newlines.
• Changes (and possible deprecations) have news fragments in changelog.d.

If you have any questions to any of the points above, just submit and ask! This checklist is here to help you, not to deter you from contributing!

[sscherfke]

My first idea was to name the function out_validators():

with out_validators():
    ...

🤪

[hynek]

hmmm the problem here is that it's extremely not thread-safe? 😅

But I guess @glyph asked for it already in #188 (2017!) so I guess if we document it properly…?

[hynek]

I think I like his attr.validators.disabled() naming btw!

[sscherfke]

hmmm the problem here is that it's extremely not thread-safe? 😅

It’s not more thread-(un)safe than the underlying function set_run_validators() 🙃

I think I like his attr.validators.disabled() naming btw!

I’d have no problem with that, but set_run_validators() and get_run_validators() are defined in _config.py and are made available directly under the attr namespace. So putting no_run_validators() to validators.disable() would be a bit inconsistent, wouldn’t it?

[hynek]

hmmm the problem here is that it's extremely not thread-safe? 😅
It’s not more thread-(un)safe than the underlying function set_run_validators() 🙃

100% correct, it's just that a context manager looks like it might be so we should nail it home in the doc strings – ideally with a warning block.

I think I like his attr.validators.disabled() naming btw!

I’d have no problem with that, but set_run_validators() and get_run_validators() are defined in _config.py and are made available directly under the attr namespace. So putting no_run_validators() to validators.disable() would be a bit inconsistent, wouldn’t it?

It would. But OTOH I don't expect anyone to use set_run_validators() and get_run_validators() after disable() landed. :D I wouldn't even mind to create better named aliases in the validators namespace and deprecate the old ones (without ever removing them – but not taking them with us into the attrs namespace).

[sscherfke]

It has been done.

[sscherfke]

Is there anything left to be done?

[hynek]

• a bunch of suggestions
• we need types hints! :)
  • also please typing_example.py
• add a news fragment please: https://github.com/python-attrs/attrs/blob/main/.github/CONTRIBUTING.rst#changelog

[sscherfke]

Done. Not sure, what kind of example you think of for the typing examples, though.

[sscherfke]

"Just a few docs and typing changes. What could go wrong", he thought.

Gonna fix it tomorrow.

[hynek]

Not sure, what kind of example you think of for the typing examples, though.

Just use the API and see if MyPy complains. :)

[sscherfke]

Okay, fixed it.

[hynek]

Thanks!

I've done some minor fixes myself. The gh cli is really neat.