make sure the ref patch used in finest_part_data is not empty

[rochSmets]

When using finest_part_data of a particle hierarchy (for example in a dist_plot with finest=True) we need to iterate through the patches, hence know their names. Then, these names was taken from the first patch_data, of index 0. But this one could eventually be empty, hence not able to provide these names.

In this version, we hence iterate on all the patches of level 0, and then get the index of the first one that is not empty. This index is then used to get the keys of the patch_data dict, as it is then the first one to know them.

With this fix, we should then be able to make the dist_plot of a set of particles containing empty patch(es).

[coderabbitai]

📝 Walkthrough

Walkthrough

finest_part_data now scans level-0 patches to find the first patch with non-empty patch_datas, uses that patch (index i_ref) to initialize aPatch and build finalParticles, and raises ValueError("This particle hierarchy seems empty !") if none are found.

Changes

Cohort / File(s)	Summary
Patch initialization logic pyphare/pyphare/pharesee/hierarchy/hierarchy.py	Updated finest_part_data to locate the first non-empty level-0 patch and use its index (i_ref) for initialization instead of unconditionally using patches[0]. Adds explicit ValueError when no non-empty patch exists.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately and concisely describes the main change: ensuring the reference patch used in finest_part_data is not empty to prevent errors.
Description check	✅ Passed	The description is well-detailed and directly related to the changeset, explaining the problem, the solution, and the expected benefit.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

📜 Review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 8cd180f and 70141be.

📒 Files selected for processing (1)

• pyphare/pyphare/pharesee/hierarchy/hierarchy.py

🧰 Additional context used

🧠 Learnings (2)

📓 Common learnings

Learnt from: PhilipDeegan
Repo: PHAREHUB/PHARE PR: 910
File: pyphare/pyphare/pharesee/hierarchy/hierarchy_utils.py:7-7
Timestamp: 2024-10-18T13:23:32.074Z
Learning: In the `pyphare.pharesee.hierarchy` module, importing `PatchHierarchy` and `format_timestamp` from `hierarchy.py` into `hierarchy_utils.py` is acceptable as long as `hierarchy.py` does not import `hierarchy_utils.py`, thereby avoiding a cyclic import.

📚 Learning: 2024-10-18T13:23:32.074Z

Learnt from: PhilipDeegan
Repo: PHAREHUB/PHARE PR: 910
File: pyphare/pyphare/pharesee/hierarchy/hierarchy_utils.py:7-7
Timestamp: 2024-10-18T13:23:32.074Z
Learning: In the `pyphare.pharesee.hierarchy` module, importing `PatchHierarchy` and `format_timestamp` from `hierarchy.py` into `hierarchy_utils.py` is acceptable as long as `hierarchy.py` does not import `hierarchy_utils.py`, thereby avoiding a cyclic import.

Applied to files:

• pyphare/pyphare/pharesee/hierarchy/hierarchy.py

🪛 Ruff (0.14.10)

pyphare/pyphare/pharesee/hierarchy/hierarchy.py

628-628: Loop control variable i_ref not used within loop body

(B007)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (5)

• GitHub Check: build (macos-latest)
• GitHub Check: build (macos-14)
• GitHub Check: Analyze (cpp)
• GitHub Check: build (ubuntu-latest, gcc)
• GitHub Check: build (ubuntu-latest, clang)

[PhilipDeegan]

this to me indicates another issue, not sure when as can have patches with no data perse

[nicolasaunai]

for particles, some populations may exist only in some parts of the domain.
the question is should python patch datas be created when there is no particles in the patch..?
It probably exists in the cpp hierarchy with size 0 particle arrays, no?

[PhilipDeegan]

level ghost can have a bunch of "empty patches"

we could drop these patches here if there's no data

[nicolasaunai]

ok after thinking a bit I think we may have uses cases where we need to keep empty particle patches, for instance if we want to browse two hierarchies, one field, one particles, to calculate, say, the total energy, then we'd like the two hierarchies to have the same patch structure.
In general I think it's probably better to remove patches from a hierarchy only when based on user selections

[nicolasaunai]

any reason not to call len() directly?