fix(documents): count documents instead of holdings in org facet

[jma]

The organisation -> library -> location facet is built on the
nested_holdings nested field, so terms sub-aggregations count holdings,
not documents: counts were inflated whenever a document had several
holdings in the same bucket.

Add a reverse_nested sub-aggregation (record_count) at each level to
count distinct parent documents, and expose it as doc_count in the
serializer. Buckets are also re-sorted by document count, since
Elasticsearch orders terms by holdings count and the displayed number is
now the document count.

• Closes The organisation/library facet counts are incorrect #4003.

Co-Authored-by: Johnny Mariéthoz johnny.mariethoz@rero.ch
Co-Authored-by: Ana Gomes anagoncalves7@icloud.com

[coderabbitai]

Walkthrough

The organisation facet aggregation now adds reverse-nested document counts at organisation and location levels. Serializer post-processing rewrites nested buckets to use those counts, removes empty buckets, sorts results, and applies local-view filtering. REST expectations and new unit coverage validate document-based counts, bucket pruning, and ordering.

Estimated code review effort: 3 (Moderate) | ~20 minutes

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly summarizes the main change: document counts in the organisation facet instead of holdings.
Description check	✅ Passed	The description matches the code changes and explains the reverse_nested counting and bucket reordering.
Linked Issues check	✅ Passed	The PR addresses issue #4003 by fixing facet counts to use distinct document counts rather than holdings counts.
Out of Scope Changes check	✅ Passed	The changes stay focused on the document facet counting fix and the related serializer and test updates.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

---

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.}

[coderabbitai]

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@rero_ils/config.py`:
- Around line 2129-2147: Update the nested terms aggregations in the
record-count query so both library and location buckets order by their
reverse-nested record_count in descending document-count order before applying
size limits. Add a regression case with more than 50 libraries and more than 100
locations that verifies the highest-record-count buckets are retained.

In `@rero_ils/modules/documents/serializers/json.py`:
- Around line 178-193: Update the organisation, library, and location bucket
rewriting flow around _rewrite_nested_buckets so returned aggregations no longer
retain stale sum_other_doc_count or doc_count_error_upper_bound metadata from
the nested terms response. Remove those fields or recompute them consistently
for the rewritten buckets, and add coverage using a non-zero metadata case.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 7e7f301e-64af-432e-80fb-e1c88fa84bb6

📥 Commits

Reviewing files that changed from the base of the PR and between 32459b2 and 73f7081.

📒 Files selected for processing (4)

• rero_ils/config.py
• rero_ils/modules/documents/serializers/json.py
• tests/api/documents/test_documents_rest.py
• tests/unit/documents/test_documents_serializers.py

[coveralls]

Coverage is 91.294% — jma:maj-fix-4003 into rero:staging. No base build found for rero:staging.