Added --restriction flag

[bonham79]

Added --restriction flag to limit 'big scrape' to individual languages. Can specify one or more languages to scrape by following flag with ISO codes of desired languages.

• Updated Unreleased in CHANGELOG.md to reflect the changes in code or data.

[kylebgorman]

Cool! I'm going to add @lfashby to review; looks basically right but maybe some style issues.

Could you type args when it's passed to main? E.g.:

def main(args: argparse.Namespace) -> None:

Could you add "Closes #153" to the PR itself? That'll cause GitHub to automatically close the issue when/if this is merged.

[kylebgorman]

Also do add this to the CHANGELOG as detailed above, and check the check-box when you do.

[kylebgorman]

Probably should add a line to data/wikipron/src/README.md to indicate this feature is available and you don't need to make a copy of languages.json to run a partial scrape.

[kylebgorman]

• Can you use a standard "# foo" comment style (space after #, just one per line, declarative sentence etc.)?
• If you don't specify False as the default for --restriction you'll just get None as the default, which seems desirable. I think the relevant conditionals should still work, though.

[lfashby]

I think this looks alright. There is only one thing I would add in addition to Kyle's comments.

For the record, I'm not convinced that the big scrape should have this sort of functionality. I guess if it makes our lives easier (with respect to testing how WikiPron works on a subset of languages) then it is fine to have, but it doesn't address the currently awkward part of running a big scrape -- having to manually find the languages WikiPron failed on during the big scrape and then restricting languages.json to just those languages. In the future we could automate the the process of building a languages.json-esque object for languages that WikiPron failed on during the big scrape so we don't have to manually modify languages.json to retry failed languages.

[bonham79]

Closes #153

[kylebgorman]

Two nits left then we're ready. I am looking forward to having this around.

To reflow on this project we use black -l79 scrape.py, then check flake8 scrape.py (for linting) and mypy scrape.py (for static typing errors). All three---black, flake8, and mypy---are on PyPi).

[kylebgorman]

Approving (tho note note about out-of-date comment). Thanks for this!