scikit-hep
Fix: sort v1 parents
This is a very hacky fix that just runs std::stable_sort on the CPU. Given that we won't (?) improve CUDA support in v1, and this is a v1-only fix, I think it's reasonable to (a) assume we're on the CPU, and (b) do this inline instead of with a CPU kernel.
Codecov Report
Merging #1389 (9884cf8) into main (b2fd2be) will decrease coverage by
0.89%. The diff coverage is54.02%.
| Impacted Files | Coverage Δ | |
|---|---|---|
| src/awkward/_v2/_connect/cling.py | 0.00% <0.00%> (ø) |
|
| src/awkward/_v2/_connect/cuda/__init__.py | 0.00% <0.00%> (ø) |
|
| src/awkward/_v2/_lookup.py | 97.50% <0.00%> (ø) |
|
| src/awkward/_v2/_prettyprint.py | 66.09% <0.00%> (+2.29%) |
:arrow_up: |
| src/awkward/_v2/identifier.py | 55.69% <0.00%> (ø) |
|
| src/awkward/_v2/operations/convert/ak_from_jax.py | 75.00% <0.00%> (ø) |
|
| src/awkward/_v2/operations/convert/ak_to_jax.py | 75.00% <0.00%> (ø) |
|
| src/awkward/_v2/operations/io/ak_from_parquet.py | 75.00% <0.00%> (ø) |
|
| src/awkward/_v2/operations/io/ak_to_parquet.py | 75.00% <0.00%> (ø) |
|
| src/awkward/_v2/operations/structure/ak_firsts.py | 75.00% <0.00%> (ø) |
|
| ... and 152 more |
![codecov[bot] avatar](https://avatars.githubusercontent.com/in/254?v=4 "Published by a pub.dev verified publisher"){kind=small}
That's true: we won't be supporting the GPU in v1 (the experimental support has been removed), so a direct call to std::stable_sort is a perfectly acceptable solution (and a good idea!).
I naively expected this to fix the MacOS tests. It seems like it maybe fixes 3.6, but not 3.7+. Perhaps this is a fluke, or maybe it reveals that there is more than one contributing bug here.
First, I'll re-run the 3.6 test.
We talked about this at today's meeting; it's old, addresses v1 only, and there haven't been any complaints about it not being done. I'm closing it. If need be, it can be reopened as a bug-fix in the 1.10.x series (v1 bug-fixes).