Fix for find_children() with stacks #2

[darbyjohnston]

Fixes #1652

This PR is a replacement for #1755 which has some trouble with the git history.

After some investigation, it looks like the find_children() function was not working with stacks when given a search range. To fix this, an override is added for the function children_in_range() that works with "stacks" of children.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 92.59259% with 2 lines in your changes missing coverage. Please review.

Project coverage is 81.69%. Comparing base (c0e97b0) to head (3a61e3c).
Report is 12 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1766      +/-   ##
==========================================
- Coverage   84.11%   81.69%   -2.43%     
==========================================
  Files         198      176      -22     
  Lines       22241    12342    -9899     
  Branches     4687     3027    -1660     
==========================================
- Hits        18709    10083    -8626     
+ Misses       2610     1723     -887     
+ Partials      922      536     -386     

Flag	Coverage Δ
py-unittests	81.69% <92.59%> (-2.43%)	⬇️

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

Files	Coverage Δ
src/opentimelineio/composition.h	65.51% <ø> (ø)
tests/test_track.py	96.77% <100.00%> (+1.42%)	⬆️
src/opentimelineio/stack.cpp	63.38% <75.00%> (-6.47%)	⬇️

... and 66 files with indirect coverage changes

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update e10fdd2...3a61e3c. Read the comment docs.

[darbyjohnston]

@jminor Sorry, I managed to mess up the git history on the previous PR while trying to update with the latest changes from main. I created this new PR as a replacement.

I believe your last comment was about using trimmed_range_in_parent() and testing with a source range on the track? I didn't realize that tracks could have source ranges, what would that test look like?

[jminor]

There's an example of a track with a source_range in Figure 3 here:
https://opentimelineio.readthedocs.io/en/stable/tutorials/otio-timeline-structure.html
In that diagram, Track-003 is offset by using a source_range with a negative start time. A simpler use case would be to use a track's source range to just trim a nested track to a sub-range of the whole track.

There's also an example of a stack with a source_range in Figure 4 on that same page. In that case the source_range is used to trim the nested stack. Although rare, the source_range could be set on the top level stack of a timeline also.

[darbyjohnston]

Thanks, those figures were very helpful; I guess I should have checked there first. :)

I changed the code to use trimmed_range_in_parent() and added extra C++ tests that put various source ranges on the track. If the C++ tests look OK I will update the Python tests to match.

[darbyjohnston]

@jminor Just checking if you have time to review the latest changes?

[jminor]

Looks great. The use of trimmed_range_in_parent looks correct to me :)

[darbyjohnston]

Thanks!