Use ruff as a linter as a replacement for flake8/isort

[Tom-Willemsen]

Use ruff as a linter, as a replacement for flake8 & isort.

ruff is much faster (Somewhere around 100x) than flake8+isort, and seems to be the direction which a lot of modern python projects are taking. It is pretty much 1:1 compatible with flake8 + isort, with very few exceptions.

To test this:

• Run python -m pip install -e .[dev] or an equivalent command to get ruff installed.
• Run pre-commit or ruff ., verify that ruff runs successfully
• Introduce a linting fail in a .py file in src or tests, verify that ruff picks this up
  • running ruff .
  • running pre-commit
• Launching vscode from this directory, verify that:
  • Ruff lints are reported (but not auto-fixed by default) in editor window. ruff can fix quite a lot of problems automatically but I found this somewhat surprising to run "on save", happy to hear other opinions about this?
  • Ruff automatically changes import order on save.
  • Note: may need to accept installing suggested extensions on first load of vscode on this branch.
• Verify that relevant docs have been updated to mention ruff instead of flake8 or isort.

[gilesknap]

@Tom-Willemsen looks good. I'll try this out against the ibek project.

@coretl I'm pretty sure we want to go with this - what do you think?

[codecov[bot]]

Codecov Report

Merging #147 (199f346) into main (f02b12e) will not change coverage. Report is 2 commits behind head on main. The diff coverage is n/a.

@@            Coverage Diff            @@
##              main      #147   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files            2         2           
  Lines           16        16           
=========================================
  Hits            16        16           

:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more

[gilesknap]

I have adopted the PR branch into ibek and all looks well.

ruff converted an 'unecesary generator to a set comprehension which was nice! :-)

One issue I had with docs is probably just because of Pydantic. See below.

I find that the conf.py

autoclass_content = "both"

results in lots of errors like the following

/scratch/hgv27681/work/ibek/src/ibek/globals.py:docstring of ibek.globals.BaseSettings:5: WARNING: 'any' reference target not found: ValidationError
/scratch/hgv27681/work/ibek/src/ibek/globals.py:docstring of ibek.globals.BaseSettings:8: WARNING: 'any' reference target not found: __init__
/scratch/hgv27681/work/ibek/src/ibek/globals.py:docstring of ibek.globals.BaseSettings:8: WARNING: 'any' reference target not found: __pydantic_self__
/scratch/hgv27681/work/ibek/src/ibek/globals.py:docstring of ibek.globals.BaseSettings:8: WARNING: 'any' reference target not found: self
/scratch/hgv27681/work/ibek/src/ibek/globals.py:docstring of ibek.globals.BaseSettings:8: WARNING: 'any' reference target not found: self
/scratch/hgv27681/work/ibek/src/ibek/support.py:docstring of ibek.support.Arg:5: WARNING: 'any' reference target not found: ValidationError
/scratch/hgv27681/work/ibek/src/ibek/support.py:docstring of ibek.support.Arg:8: WARNING: 'any' reference target not found: __init__
/scratch/hgv27681/work/ibek/src/ibek/support.py:docstring of ibek.support.Arg:8: WARNING: 'any' reference target not found: __pydantic_self__
/scratch/hgv27681/work/ibek/src/ibek/support.py:docstring of ibek.support.Arg:8: WARNING: 'any' reference target not found: self
/scratch/hgv27681/work/ibek/src/ibek/support.py:docstring of ibek.support.Arg:8: WARNING: 'any' reference target not found: self

restoring

autoclass_content = "clas"

gets rid of these.

Pydantic seems to upset Sphinx quite a bit, so I don't see this as an issue.

[coretl]

This seems fine, shall I merge?

[Tom-Willemsen]

If you're happy with it then yes please? Think I've addressed all the review comments that have been left.