Code cleanups from ruff and pyupgrade

[nsoranzo]

• Enable flake8-comprehensions ruff rules, but allow dict() calls that make use of keyword arguments (e.g., dict(a=1, b=2)).
• Move out from the galaxy-util package a file that requires galaxy.files
• Code cleanups from pyupgrade

How to test the changes?

(Select all options that apply)

• I've included appropriate automated tests.
• This is a refactoring of components with existing test coverage.
• Instructions for manual testing are as follows:
  1. [add testing steps and prerequisites here if you didn't write automated tests covering all your changes]

License

• I agree to license these and all my past contributions to the core galaxy codebase under the MIT license.

[jmchilton]

Thats for the galaxy.files fix. That looks right to me!

[jdavcs]

Would you mind waiting for #17180 to be merged first? I'm afraid this might cause conflicts and some painful rebasing.

[nsoranzo]

Would you mind waiting for #17180 to be merged first? I'm afraid this might cause conflicts and some painful rebasing.

@jdavcs I've compared the 2 PRs side-by-side and I think the only conflict would be in lib/galaxy_test/selenium/framework.py (pretty easy to solve).

[jdavcs]

thanks for checking, @nsoranzo, then it'll be fine.

[jdavcs]

...and I merged too soon! This: return [wid for wid in sa_session.scalars(stmt)] does not translate into this: list(sa_session.scalars(stmt)): the latter will raise an error.

I'll go through this line by line and will submit a fix shortly.

[nsoranzo]

...and I merged too soon! This: return [wid for wid in sa_session.scalars(stmt)] does not translate into this: list(sa_session.scalars(stmt)): the latter will raise an error.

I'll go through this line by line and will submit a fix shortly.

Is this a SqlAlchemy 2.0 change or are these lines not tested?
The scalars results are iterable, so the syntax should be equivalent, but I am not at the computer so cannot test right now.

[jdavcs]

It's not a SQLAlchemy thing, it applies to sqlite3 as well. I've tested it - they are not directly iterable, here's one example:

-> for i, row in enumerate(results):
(Pdb) row
<sqlite3.Row object at 0x7ff558d996c0>
(Pdb) list(row)
*** Error in argument: '(row)'
(Pdb) [x for x in row]
[1, 'john', 'davis', 'sagd']

Same for SQLAlchemy's Result object. Why exactly these are not iterable when passed to list, I don't know.

[nsoranzo]

list is a command in Pdb, as you may guess from the error. Try print(list(row))

[jdavcs]

Now, this is truly embarrassing 😆 After getting tripped by args in pdb in the past, I didn't expect to step on the same rake twice. Closing the fix PR. Thank you! :)

[nsoranzo]

No worries, I was able to notice this because I have been confused by these Pdb weirdnesses too many times!