Small cleanups for the new subtraction operator

[oleksbabieiev]

What do these changes do?

Small cleanups for the new subtraction operator.

Are there changes in behavior for the user?

N/A

Related issue number

Resolves #1377

Checklist

• I think the code is well written
• Unit tests for the changes exist
• Documentation reflects the changes

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.06%. Comparing base (3a14e46) to head (dfb457e).
Report is 5 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1384   +/-   ##
=======================================
  Coverage   96.05%   96.06%           
=======================================
  Files          31       31           
  Lines        5752     5765   +13     
  Branches      360      361    +1     
=======================================
+ Hits         5525     5538   +13     
  Misses        201      201           
  Partials       26       26           

Flag	Coverage Δ
CI-GHA	96.06% <100.00%> (+<0.01%)	⬆️
MyPy	46.05% <64.70%> (+0.09%)	⬆️
OS-Linux	99.55% <100.00%> (+<0.01%)	⬆️
OS-Windows	99.61% <100.00%> (+<0.01%)	⬆️
OS-macOS	99.30% <100.00%> (+<0.01%)	⬆️
Py-3.10.11	99.28% <100.00%> (+<0.01%)	⬆️
Py-3.10.15	99.51% <100.00%> (+<0.01%)	⬆️
Py-3.11.10	99.51% <100.00%> (+<0.01%)	⬆️
Py-3.11.9	99.28% <100.00%> (+<0.01%)	⬆️
Py-3.12.7	99.51% <100.00%> (+<0.01%)	⬆️
Py-3.13.0	99.51% <100.00%> (+<0.01%)	⬆️
Py-3.9.13	99.23% <100.00%> (+<0.01%)	⬆️
Py-3.9.20	99.47% <100.00%> (+<0.01%)	⬆️
Py-pypy7.3.16	99.52% <100.00%> (+<0.01%)	⬆️
Py-pypy7.3.17	99.55% <100.00%> (+<0.01%)	⬆️
VM-macos-latest	99.30% <100.00%> (+<0.01%)	⬆️
VM-ubuntu-latest	99.55% <100.00%> (+<0.01%)	⬆️
VM-windows-latest	99.61% <100.00%> (+<0.01%)	⬆️
pytest	99.55% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[codspeed-hq]

CodSpeed Performance Report

Merging #1384 will not alter performance

_{Comparing oleksbabieiev:relpath-cleanup (dfb457e) with master (3a14e46)}

Summary

✅ 84 untouched benchmarks

[bdraco]

FAILED tests/test_url.py::test_sub_combined_with_join - hypothesis.errors.FailedHealthCheck: Data generation is extremely slow: Only produced 9 valid examples in 1.05 seconds (0 invalid ones and 7 exceeded maximum size). Try decreasing size of the data you're generating (with e.g. max_size or max_leaves parameters).

It looks like its failing on performance. The performance is nearly quadratic unfortunately so any long paths are going to be very slow.

[bdraco]

I tried to solve the performance issues in #1388, however in the process I became aware that there are other considerations that are missing.

Its clear this isn't ready to go yet and I merged #1340 too soon (apologies to everyone involved) so I've opened up #1391 to revert it so we can start over with a new approach.

I'm not sure if #1388 is the way to go or not but if it is, its probably a good start as its looks like using the built-in path logic in Python is going to be a non-starter and its likely we need to implement it ourselves since the built-in path logic doesn't translate well for URLs as the requirements have subtle differences.

[bdraco]

Apologizes for not noticing the issues before merging #1340. I sincerely hope that will not dissuade a second attempt.

[webknjaz]

Side note: it would probably be useful to have own hypothesis strategy under yarl.testing that could be utilized both within yarl's own test suite and any downstream/end-user projects. That strategy would need to produce various URL edge cases.
This is not in the scope of this PR, however, I just wanted to record the idea.

[oleksbabieiev]

Well, I think it would make sense if I create a new pull request to rework this feature @bdraco

[oleksbabieiev]

#1392