Try using setup.cfg declarative metadata

[MarcoGorelli]

Trying to imitate asottile https://github.com/asottile/pyupgrade

Seems I'd put loads of extra things (taken from some cookiecutter template) which might not have been needed?

@s-weigand as mentioned on gitter, I think you should be included as co-author - if you agree I'll do so

@girip11 if I've done this correctly, it lets us have the default configs as a separate file

[codecov-io]

Codecov Report

Merging #403 into master will not change coverage.
The diff coverage is 100.00%.

@@            Coverage Diff            @@
##            master      #403   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           11        11           
  Lines          505       507    +2     
=========================================
+ Hits           505       507    +2     

Impacted Files	Coverage Δ
nbqa/config.py	100.00% <100.00%> (ø)

---

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 e07baad...611a875. Read the comment docs.

[s-weigand]

Problems with bump2version and setup.cfg

bump2version sadly doesn't support the setup.cfg only layout properly (see c4urself/bump2version#83 and c4urself/bump2version#87).
The other issue with bump2version would be that it most likely will have a fight with setup-cfg-fmt and pre-commit-hooks/trailing-whitespace, since it adds trailing whitespaces (see c4urself/bump2version#161) and also rewrites the whole setup.cfg in config parser style (space indent -> tab indent).

I guess the only workaround to still use bump2version would be to use .bumpversion.cfg file.

Proposed setup.cfg changes

Version

The version can be read automagically if we set setuptools>=46.4.0 (released: May 17, 2020) as a requirement (see the packaging guide)

[metadata]
version = attr: package.__version__

Project Urls

Having links to the docs and issues on PyPi is a nice to have (should look something like this)

[metadata]
project_urls =
        Documentation = https://nbQA.readthedocs.io/en/latest/
        Source = https://github.com/nbQA-dev/nbQA
        Tracker = https://github.com/nbQA-dev/nbQA/issues

It renders like this on PyPi

Additional classifiers and keywords

I would also add the following classifiers (full list here)

[metadata]
classifiers =
    Framework :: Jupyter
    Operating System :: OS Independent
    Environment :: Console
    Topic :: Software Development :: Quality Assurance

keywords

[metadata]
keywords = jupyter, notebook, format, lint, ...

[sourcery-ai]

Sourcery Code Quality Report

❌  Merging this PR will decrease code quality in the affected files by 1.13%.

Quality metrics	Before	After	Change
Complexity	1.78 ⭐	2.21 ⭐	0.43 👎
Method Length	46.30 ⭐	47.93 ⭐	1.63 👎
Working memory	7.84 🙂	7.75 🙂	-0.09 👍
Quality	81.51% ⭐	80.38% ⭐	-1.13% 👎

Other metrics	Before	After	Change
Lines	918	763	-155

Changed files	Quality Before	Quality After	Quality Change
setup.py	53.75% 🙂	100.00% ⭐	46.25% 👍
nbqa/main.py	79.19% ⭐	79.19% ⭐	0.00%
nbqa/config_parser.py	79.39% ⭐	79.39% ⭐	0.00%
nbqa/toml_parser.py	84.92% ⭐	84.92% ⭐	0.00%

Here are some functions in these files that still need a tune-up:

File	Function	Complexity	Length	Working Memory	Quality	Recommendation
nbqa/main.py	_run_on_one_root_dir	12 🙂	198 😞	19 ⛔	35.44% 😞	Try splitting into smaller methods. Extract out complex expressions
nbqa/main.py	_run_command	0	109 🙂	15 😞	59.92% 🙂	Extract out complex expressions
nbqa/main.py	_map_python_line_to_nb_lines	1 ⭐	79 🙂	10 😞	71.00% 🙂	Extract out complex expressions
nbqa/main.py	_replace_temp_python_file_references_in_out_err	1 ⭐	52 ⭐	10 😞	75.17% ⭐	Extract out complex expressions
nbqa/main.py	_get_command_not_found_msg	0	37 ⭐	11 😞	76.70% ⭐	Extract out complex expressions

Legend and Explanation

The emojis denote the absolute quality of the code:

• ⭐ excellent
• 🙂 good
• 😞 poor
• ⛔ very poor

The 👍 and 👎 indicate whether the quality has improved or gotten worse with this pull request.

---

Please see our documentation here for details on how these metrics are calculated.

We are actively working on this report - lots more documentation and extra metrics to come!

Let us know what you think of it by mentioning @sourcery-ai in a comment.

[MarcoGorelli]

Thanks for your suggestions!

It's not too much effort to change version numbers, and there's usually some small changes that need to be done anyway, so I don't mind that not being automated just yet