TooTallToby tutorials: unbreak and test.

[fischman]

• Unbreak the three broken tutorials (fixes Fix TTT tutorials: 23-T-24 Curved Support, 01-02-post-cap, 01-10-light-cap #848)
  • This involved a rewrite of PPP-01-10 because I already had my own solution to that one and I couldn't easily tell what was going wrong with the previous solution.
• Add assertions to all the tutorials so that non-raising means success
• Add the TTT examples to test_examples.py added recently for Examples should be kept working by being run in a test #909
• Also dos2unix'd the TTT .py files to match the rest of the codebase.
• Drop the copies of TTT examples from test_benchmarks.py and read them directly from the docs directory now.

[fischman]

FYI @jdegenstein. LMK if you'd rather I not replace your PPP0110 solution (and if you have a fix in mind already).

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.93%. Comparing base (dce0c5d) to head (4027664).
Report is 11 commits behind head on dev.

Additional details and impacted files

@@           Coverage Diff           @@
##              dev     #912   +/-   ##
=======================================
  Coverage   96.93%   96.93%           
=======================================
  Files          32       32           
  Lines        9460     9471   +11     
=======================================
+ Hits         9170     9181   +11     
  Misses        290      290           

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

[fischman]

@gumyr WDYT, worth testing TTT examples as part of automated testing?

[jdegenstein]

@fischman FYI TTT examples are (sort of) tested through the benchmarking workflow (in a separate python script)

[fischman]

@fischman FYI TTT examples are (sort of) tested through the benchmarking workflow (in a separate python script)

Thanks for the heads-up, I was unaware of that before.
Seems suboptimal that the examples' code is duplicated in the tests/ subdirectory (because of the likelihood of drift as one or the other place is edited, and the other is forgotten, and also because edits must be duplicated). Also ttt-23-t-24-curved_support is missing from the test.

Do you prefer to keep the structure at HEAD, or are you suggesting I keep this PR moving forward and delete the test_benchmarks copy of the examples, or something else?
(maybe test_benchmarks should run test_examples?)

[jdegenstein]

@fischman yes I agree that having two versions is suboptimal, I was just trying to get something up and running for benchmarking purposes with some changes that were happening on the OCCT/OCP and build123d side. Yes, I would like TTT 23-T-24 to be part of the benchmarks.

Can we simply import the models from docs/ into test_benchmarks.py ? That would probably be the "cleanest" solution requiring less work to maintain long-term.

RE: PPP0110 -- I am fine with your solution, I am not married to what I created. Hopefully OCCT 7.9 will make creating this model easier.

[fischman]

@jdegenstein dropped the copies in the test_benchmarks.py file and now pulling from the docs copy. PTAL.

[jdegenstein]

@fischman looking at the way that [benchmarks] category was eliminated resulted in benchmarks trying to run then failing because of multiple threads in the tests workflow.

/opt/hostedtoolcache/Python/3.13.2/x64/lib/python3.13/site-packages/pytest_benchmark/logger.py:39: PytestBenchmarkWarning: Benchmarks are automatically disabled because xdist plugin is active.Benchmarks cannot be performed reliably in a parallelized environment.

If this is the way it is setup, I would prefer that an appropriate flag is passed to pytest to explicitly disable benchmarking for the tests workflow. The current behavior seems fragile to me.

[fischman]

@jdegenstein good catch! Fixed to exclude benchmarks from the tests workflow.

[fischman]

@gumyr @jdegenstein any idea about the CI failure? Seems unrelated to this PR.

[jdegenstein]

@fischman I am going to take one more quick look, I don't believe the CI failure is related to anything you did -- just a random problem with the runner I suspect.

[jdegenstein]

OK I am merging despite what appears to be a different random CI failure.

EDIT: thanks for your contribution @fischman