added boolean optional argument support

[jspaezp]

Improved the toml parsing so it supports BooleanOptionalAction arguments in it.
(also added tests for it ...)

[tool.pydocstringformatter]
write = false
style = ["numpydoc"]
strip-whitespaces = true  # <<<<<- previously unsupported
split-summary-body = false # <<<<<- previously unsupported
no-numpydoc-section-hyphen-length = true # <<<<<- previously unsupported

.... where should I document this?

[Pierre-Sassoulas]

Looks good, thank you for adding this !

[codecov]

Codecov Report

Merging #166 (0242e5e) into main (9268fd6) will not change coverage.
The diff coverage is 100.00%.

@@            Coverage Diff            @@
##              main      #166   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           21        21           
  Lines          591       604   +13     
=========================================
+ Hits           591       604   +13     

Impacted Files	Coverage Δ
...ydocstringformatter/_configuration/toml_parsing.py	100.00% <100.00%> (ø)

[jspaezp]

@Pierre-Sassoulas I think I should also add this to the documentation on the toml configuration, do you happen to know where that should be added?

[Pierre-Sassoulas]

This is the full doc on configuration: https://pydocstringformatter.readthedocs.io/en/latest/?badge=latest#configuration

[jspaezp]

Oh i see... so that is getting auto generated from the readme ...

Should we modify:

[tool.pydocstringformatter]
write = true
exclude = "**/my_dir/**,**/my_other_dir/**"
# Or:
exclude = ["**/my_dir/**", "**/my_other_dir/**"]

To this:

[tool.pydocstringformatter]
write = true
exclude = "**/my_dir/**,**/my_other_dir/**"
# Or:
exclude = ["**/my_dir/**", "**/my_other_dir/**"]
style = ["numpydoc"]
strip-whitespaces = true
split-summary-body = false
no-numpydoc-section-hyphen-length = true

to exemplify how to set other argument types.
LMK what you think

[Pierre-Sassoulas]

Maybe we could autogenerate the full list of possible options with default values ?

[jspaezp]

I like that idea but it would not exemplify the use of 'no-xxxxx = true' style configs. ... Maybe also have comments showing the equivalent config on the negative form ??

This could be a feature for a later improvement in documentation (and probably not inside the readme ... So it does not get too clutered)

[Pierre-Sassoulas]

I think the no-x style is only used in CLI (or should) this is given by the --help command so we don't need to document it imo'

[DanielNoord]

The no- options are needed to disable default checks. I think they should indeed be shown as a comment in the example config.

For pylint we use tomlkit. This is one of the few parsers for toml that saves comments and style.
A solution where we initialise all options and then write a file with tomlkit would be nice.
Perhaps it would be even better if we could create a --create-config command so that we can also offer it as a cli
tool?

[Pierre-Sassoulas]

The no- options are needed to disable default checks

Isn't it possible to do x=false or no-x=true ?

[DanielNoord]

The no- options are needed to disable default checks

Isn't it possible to do x=false or no-x=true ?

It depends on how we handle that in this PR. We could just disallow them completely as well?

What do you think makes most sense @jspaezp?

[jspaezp]

Hello there!

I thought about it for a while...
and I think the most consistent behavior would be to allow "--no" version. I really like the idea of the config being as close as possible to copy/pasting the command line arguments.

regarding: #166 (comment)

Isn't it possible to do x=false or no-x=true ?

its easy but not SUUUUPER EASY, just having to consider it robust to arguments that in the future might start with --no- but are not of the same "optional boolean" type.

I will add the mods to the readme in the meantime

[Pierre-Sassoulas]

I meant it's possible to do it as a user (so you should always be able to use simply 'x' which is simpler both for the user and for us). Imo we should not permit to use no-x in configuration, there's very little benefit and a lot of headaches to expect.

[jspaezp]

Alright, I added the enforcement of the positive form of the arguments (+testing, +updated readme, +informative error message).

[Pierre-Sassoulas]

I like the clear error message ! LGTM, I'll let Daniel merge.

[DanielNoord]

Thanks once again @jspaezp 🎉 😄

[github-actions]

According to the primer, this change has no effect on the checked open source code. 🤖🎉