Parallel tests

[genedan]

Summary of Changes

Experimental PR, let's see how fast the updated workflows execute.

I guess this is also an excuse for me to use rich, which is an awesome library. But more seriously, the tests run in 1/4th the time when developing on my machine, and take just over a minute to run.

rich is optional. Without it, the script outputs plaintext. But you should test it out locally with rich because it's colorful.

run: .python .github/scripts/pytest_parallel.py --nbmake docs chainladder

Screencast.From.2026-06-30.18-16-22.mp4

Related GitHub Issue(s)

#1080

Additional Context for Reviewers

• I passed tests locally for both code (uv run pytest) and documentation changes (uv run jb build docs --builder=custom --custom-builder=doctest)

---

Note

Medium Risk
Changes how all unit CI jobs execute tests (parallel xdist + combined coverage); library code is untouched but flaky or coverage-merge issues could affect CI signal.

Overview
CI and local test runs now go through a new .github/scripts/pytest_parallel.py wrapper instead of invoking pytest directly. The script always runs pytest -n auto --dist=loadfile with default --cov=chainladder, captures output over a PTY, and reprints results as per-file dot lines (optional Rich live progress when rich is installed).

pytest.yml and pytest_upstream_nightly.yml were updated to call this script (with existing flags like --nbmake, --cov-report=xml, and path/markers forwarded). pyproject.toml adds pytest-xdist to the test extra and a dev dependency group with rich; uv.lock reflects those pins (including a pandas upper bound tweak in metadata).

^{Reviewed by Cursor Bugbot for commit 6cd5599. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

Pyright Type Completeness

View the full pyright --verifytypes output for this commit

Project (full chainladder package, at this PR's head): 14.1% of exported symbols fully typed (175 / 1239)

	Known	Ambiguous	Unknown	Total
Project (head)	175	107	957	1239

Other symbols referenced but not exported by chainladder: 13

	Known	Ambiguous	Unknown	Total
Other (head)	3	1	9	13

Symbols without documentation:

• Functions without docstring: 314
• Functions without default param: 0
• Classes without docstring: 10

Patch (exported symbols added or changed by this PR): no exported symbol type-completeness changes detected.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 89.55%. Comparing base (921bfca) to head (6cd5599).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1081      +/-   ##
==========================================
- Coverage   89.75%   89.55%   -0.20%     
==========================================
  Files          91       91              
  Lines        5408     5286     -122     
  Branches      696      671      -25     
==========================================
- Hits         4854     4734     -120     
+ Misses        389      387       -2     
  Partials      165      165              

Flag	Coverage Δ
unittests	89.55% <ø> (-0.20%)	⬇️

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

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit f72cde6. Configure here.}

[genedan]

They ran a little faster. GitHub actions only allocates 2 workers per job, a far cry from the 16 I have on my machine. So it's like 3-4 minutes per job instead of 4-5.

I think this script is too heavy to maintain for using the CI/CD, but it's still very helpful to me (and I hope to you) for local development due to the speed increase there. My vote is:

• Remove script from CI/CD
• Keep it as a utility for local development

If you don't want it in, I'm ok with saving the script outside the repo and just using it personally.