Add a sort to tables.subset()

[hyanwong]

Fixes #2487, see extra discussion in #2473

PR Checklist:

• Tests that fully cover new/changed functionality.
• Documentation including tutorial content if appropriate.
• Changelogs, if there are API changes.

[hyanwong]

@petrelharp: I seem to have opened up a can of worms. This is failing on edge ordering when comparing the tsutil python version of subset() (here) to the new one, even though I added a tables.sort() to both versions. Can you figure out why?

[hyanwong]

@petrelharp: I seem to have opened up a can of worms. This is failing on edge ordering when comparing the tsutil python version of subset() (here) to the new one, even though I added a tables.sort() to both versions. Can you figure out why?

OK - fixed now. In verify_subset we now can't test that the edges are identical and in the same order as a subset of the full table, because sorting might change the edge order. So we have to make a set and test that instead. Here I test each element against the set of expected edges, but it might be nicer to make another set from subset.edges and simply test set equality.

Ready for review.

[codecov]

Codecov Report

Merging #2489 (5ddba5f) into main (1d8bc7e) will decrease coverage by 0.00%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main    #2489      +/-   ##
==========================================
- Coverage   93.91%   93.90%   -0.01%     
==========================================
  Files          28       27       -1     
  Lines       27216    27191      -25     
  Branches     1263     1263              
==========================================
- Hits        25559    25535      -24     
+ Misses       1623     1622       -1     
  Partials       34       34              

Flag	Coverage Δ
c-tests	92.24% <ø> (ø)
lwt-tests	89.05% <ø> (ø)
python-c-tests	73.19% <100.00%> (+1.49%)	⬆️
python-tests	98.95% <100.00%> (-0.01%)	⬇️

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

Impacted Files	Coverage Δ
python/tskit/trees.py	98.74% <ø> (-0.01%)	⬇️
python/tskit/tables.py	98.95% <100.00%> (+<0.01%)	⬆️
python/tskit/__init__.py
python/_tskitmodule.c	92.91% <0.00%> (+0.01%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 1d8bc7e...5ddba5f. Read the comment docs.

[benjeffery]

LGTM, one small paranoia point about the testing.

[hyanwong]

Thanks @benjeffery - changed

[jeromekelleher]

LGTM, but I think we need to make it clear that there'll be some extra sorting in the tables.subset case, as this could potentially break some code.