C++ refactoring: ak.parameters

[ioanaif]

No description provided.

[codecov]

Codecov Report

Merging #1242 (d0898d5) into main (ab65e20) will increase coverage by 0.20%.
The diff coverage is 61.11%.

Impacted Files	Coverage Δ
src/awkward/_v2/contents/content.py	81.89% <ø> (+0.31%)	⬆️
src/awkward/_v2/contents/indexedarray.py	60.64% <ø> (-0.10%)	⬇️
src/awkward/_v2/contents/indexedoptionarray.py	85.40% <ø> (+0.15%)	⬆️
src/awkward/_v2/contents/listarray.py	87.95% <ø> (+0.21%)	⬆️
src/awkward/_v2/contents/regulararray.py	82.85% <ø> (+0.18%)	⬆️
...c/awkward/_v2/operations/describe/ak_parameters.py	41.66% <22.22%> (-38.34%)	⬇️
src/awkward/_v2/_util.py	83.79% <100.00%> (+1.55%)	⬆️
src/awkward/_v2/contents/numpyarray.py	89.74% <100.00%> (+0.02%)	⬆️
src/awkward/_v2/contents/recordarray.py	80.86% <100.00%> (+1.12%)	⬆️
src/awkward/_v2/contents/unionarray.py	84.08% <100.00%> (+0.07%)	⬆️
... and 144 more

[jpivarski]

Looks great!

We talked about these things on Slack, but for posterity:

• The new merge_equal flag for merge_parameters is good, as it bring the use of this function in line with one of the same name in v1. The other use of this function is in Arrow conversions, which has the different merging behavior, and that's why you need a flag.
• A lot of the changes in the tests are replacing XYZ.parameters with ak._v2.operations.describe.parameters(XYZ), which is fine. The behavior defined by the tests has not changed.