Add pre-commit hooks

added 1 commit

• 692798ec - update versions of hooks

Compare with previous version

added 1 commit

• f5dc121f - add documentation

Compare with previous version

marked this merge request as ready

requested review from @hans.fangohr

added 1 commit

• 1603ed87 - update development file link

Compare with previous version

added 1 commit

• 760070da - correct reference to development.rst

Compare with previous version

added 1 commit

• 54109267 - remove link to development.rst

Compare with previous version

added 1 commit

• 557f2d02 - add pre-commit dev-dependency to pyproject.toml

Compare with previous version

This is an RST file (not markdown).

I didnt want to add an md file when the readme is in rst. Is there a reason why you chose to work on rst ( given that it has a strange syntax for a markup language)

RST is different from markdown, and you have learned markdown first, so perhaps that is why you consider it strange ;-). I agree the links are more intuitive in markdown.

RST is more powerful, and the language used in sphinx and (often?) python docstrings.

One feature I like in particular for README.rst files is that you can have a table of content created (which is not supported by markdown).

I am all in favour of sticking to RST. I think I have fixed most of this file already?

Looks fine to me; apart from requested changes.

It is an extra step to set up the repo for devolpment. So it is good this is documented. Should check that every possible developer is familiar with this (in Python).

added 1 commit

• 5bf66265 - Apply changes from Hanss changes

Compare with previous version

added 1 commit

• 9c609105 - remove unused file checks in precommit

Compare with previous version

added 1 commit

• 117d781f - update codeblock syntax

Compare with previous version

If we assume every developer runs the pre-commit hooks, then no violation of that should ever be committed. But of course we cannot enforce that.

How about we add a new workflow (or a command to our existing style checks) that runs the pre-commit checks and fails if it detects a violation? This way, we maintain the standard even if somebody doesn't use the pre-commit hooks. (As a downside: if you don't want to install the pre-commit hooks, and then the CI complains, then it is getting increasingly harder to fix this. On the positive: this will prevent that violations of the standards are detected by the next developer. )

Ok so ill replace the current style ci with pre-commit checks ( since it includes ruff and black )?

If all the checks are covered by the new pre-commit, and the output is as useful (I think it should be by default but you may want to check), then that seems fine. Perhaps do this in a new MR?

sure, ill do it once this is merged

added 2 commits

• e4061935 - remove reaquirements txt check
• cfd060a7 - run precommit on all files

Compare with previous version

Im trying to add rstcheck to check the rst files we have. I fixed a few errors that it showed but i cant make sense of one of the error:

rstcheck.................................................................Failed
- hook id: rstcheck
- exit code: 1

README.rst:2: (INFO/1) Enumerated list start value not ordinal-1: "3" (ordinal 3)
Error! Issues detected.

is this one obvious to you @hans.fangohr? I have reduced the verbosity for now to get that check passing

Is this still an open question? @akarnad

It is, I have just ignored the INFO errors

added 4 commits

• 328fafa8 - update precommit vers
• 273dfdcf - add rstcheck precommit
• 2e23d013 - reduce verbosity of rstcheck
• 490c853b - fix issues mentioned by rstcheck

Compare with previous version

approved this merge request

merged

mentioned in commit b322f8f3

mentioned in issue #17 (closed)