Use type-hints for all function arguments & return values

[bsdci]

Use type-hints for all function arguments & return values

[araujobsd]

The first goal of mypy was to be an alternative to Python language, something like pypy but with different purpose. The main design wasn't to be a test tool! ;)

I would discourage you to use it, it will add more complexity to the project as well as additional confusion.
Also, I can't see a real benefit of it, and plus it is an experimental project.

Stick with pure python is the right thing to do, force contributors to implement something that is not standard in the market is just a big "no go".

I would revert this commit too: ba62e1f

[igalic]

a first big slew, or rather, the ground-work! has now been laid with the merging of #54.
Thanks to everyone who's helped out here — in particular, @gronke, who did a lot of debugging with me.

You too can help out with this issue! take a look at mypy --strict libiocage/ and choosing, and following a thread. start adding types, validate with mypy libiocage until it comes back clean.

[gronke]

PEP 3107 -- Function Annotations

[igalic]

oh, interesting! i thought (PEP 484 -- Type Hints)[https://www.python.org/dev/peps/pep-0484/) was all we needed

[gronke]

Same syntax! Let's do it.

[skarekrow]

They work in concert :D and yay yes this was on my table for "eventually", very happy to see this being worked on 👍

[igalic]

how about formally extending CONTRIBUTING.md with #42, to say that no functions are to be added without a type signature?

[skarekrow]

ba62e1f done

[igalic]

💜

[igalic]

i thought that part of the reason for choosing python 3.6 was so that the project could make full use of the new language features, including types. mypy is a crutch here, and i would rather revert adding that, than adding types

[araujobsd]

You can do full use of Python 3.6; but, include an external dependency that is in experimental stage, doesn't sounds good for several reasons: maintainability, interdependence, stability and plus I can't see what improvements it can bring to the project, but it only adds more complexity.

For me it is not clear what problem you are trying to solve here using mypy. So, I'm totally against to include mypy in this project, if there is not a technical debt or a serious problem to be solved.

[gronke]

Strong typing is established in the entire library.