fix(qdrant): cover AsyncQdrantClient.query_points and query_batch_points

[NishchayMahor]

Summary

Follow-up to #3492 / #3500 picking up the async-parity gap @kngyeol flagged in the issue thread.

The original crash from upload_records is gone since #3500, but the query_points / query_batch_points mismatch between the sync and async method configs is still present on main. Both methods exist on AsyncQdrantClient in current qdrant-client releases, the sync JSON already lists them, the async JSON does not, so async users get no spans for the modern query APIs.

Changes

• async_qdrant_client_methods.json: add AsyncQdrantClient.query_points and AsyncQdrantClient.query_batch_points next to the existing query / query_batch entries, mirroring the sync layout.
• tests/test_qdrant_instrumentation.py: add two async tests that drive query_points and query_batch_points through the in-memory AsyncQdrantClient, run via asyncio.run (no new test deps), and assert the spans show up with the expected attributes. Used upsert for setup since AsyncQdrantClient.upload_collection is sync internally and returns None.

Test plan

• uv run pytest tests/ reports 6 passed (4 existing sync tests + 2 new async ones).
• uv run ruff check . reports All checks passed!.
• Verified locally that both methods exist on AsyncQdrantClient in current qdrant-client so the wrap will not skip them at instrumentation time.

Refs #3492.

Summary by CodeRabbit

• New Features

  • Added OpenTelemetry spans for two asynchronous Qdrant query operations: qdrant.query_points and qdrant.query_batch_points, providing clearer tracing and query-specific attributes.

• Tests

  • Added tests that run async queries and verify emitted span names and key attributes (collection, top-k, batch count) to ensure telemetry correctness.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: e689b993-970e-44b0-971c-6f50bea7187e

📥 Commits

Reviewing files that changed from the base of the PR and between ef41cfc and ace73d4.

📒 Files selected for processing (2)

• packages/opentelemetry-instrumentation-qdrant/opentelemetry/instrumentation/qdrant/async_qdrant_client_methods.json
• packages/opentelemetry-instrumentation-qdrant/tests/test_qdrant_instrumentation.py

🚧 Files skipped from review as they are similar to previous changes (2)

• packages/opentelemetry-instrumentation-qdrant/opentelemetry/instrumentation/qdrant/async_qdrant_client_methods.json
• packages/opentelemetry-instrumentation-qdrant/tests/test_qdrant_instrumentation.py

---

📝 Walkthrough

Walkthrough

The PR adds JSON span mappings for AsyncQdrantClient.query_points and query_batch_points, and introduces async tests that run those methods via AsyncQdrantClient and assert the emitted span names and attributes.

Changes

Qdrant async query span coverage

Layer / File(s)	Summary
Span mapping configuration packages/opentelemetry-instrumentation-qdrant/.../async_qdrant_client_methods.json	JSON configuration maps AsyncQdrantClient.query_points and query_batch_points to span names qdrant.query_points and qdrant.query_batch_points.
Async test setup and validation packages/opentelemetry-instrumentation-qdrant/tests/test_qdrant_instrumentation.py	Tests import asyncio and AsyncQdrantClient, add an async helper to create a collection and upsert points, and validate that query_points and query_batch_points emit the correct span names and expected attributes (collection name, top-k, batch request count).

Sequence Diagram(s)

sequenceDiagram
  participant Test as Test case
  participant AsyncClient as AsyncQdrantClient
  participant Instrumentation as OpenTelemetryInstrumentation
  participant Exporter as InMemoryExporter
  Test->>AsyncClient: call query_points / query_batch_points
  AsyncClient->>Instrumentation: instrumented call (span created)
  Instrumentation->>Exporter: export span (qdrant.query_points / qdrant.query_batch_points)

Loading

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Poem

A rabbit hops through traces bright,
Async calls made in moonlit night,
Spans now bloom for queries run,
Points and batches seen as one,
🐰✨

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main change: adding async span coverage for two specific AsyncQdrantClient query methods that were previously missing from instrumentation.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

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

🤖 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
`@packages/opentelemetry-instrumentation-qdrant/tests/test_qdrant_instrumentation.py`:
- Around line 167-174: The exporter fixture accumulates spans across the session
so async tests can pick up earlier sync spans; fix by clearing the in-memory
exporter at the start of each async test (e.g., in
test_qdrant_async_query_points and test_qdrant_async_query_batch_points) by
calling exporter.clear() (before invoking the async instrumented call and before
using exporter.get_finished_spans()) so next(...) will select the async span you
just produced.

🪄 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: defaults

Review profile: CHILL

Plan: Pro

Run ID: 46de7e04-2b4f-40f0-bc6d-394b71a54d2e

📥 Commits

Reviewing files that changed from the base of the PR and between 375a4d5 and bf662a3.

📒 Files selected for processing (2)

• packages/opentelemetry-instrumentation-qdrant/opentelemetry/instrumentation/qdrant/async_qdrant_client_methods.json
• packages/opentelemetry-instrumentation-qdrant/tests/test_qdrant_instrumentation.py

[NishchayMahor]

Thanks @coderabbitai :), fair callout. The clear_exporter autouse fixture in conftest.py already wipes the in-memory exporter between tests, so the failure mode you're describing does not actually trigger with the current setup, but you're right that the explicit clear inside each async test makes the contract obvious from the test body alone and survives any future fixture change.

Pushed in ef41cfc:

• exporter.clear() at the top of both test_qdrant_async_query_points and test_qdrant_async_query_batch_points, right before the asyncio.run(...) block, so the next(span for span ...) lookup only sees spans produced by this test.

Re-ran locally, still 6/6 passing. Appreciate the second pair of eyes :)

[coderabbitai]

Tip

For best results, initiate chat on the files or code changes.

@NishchayMahor: Great explanation — the belt-and-suspenders approach (conftest.py fixture + explicit exporter.clear()) is exactly the right call. The test intent is now self-documenting regardless of fixture ordering, and it's robust against future test infrastructure changes. Thanks for the clean follow-up! 🎉

(ﾉ◕ヮ◕)ﾉ*:･ﾟ✧ 🐇

---

If you found this review helpful, would you consider giving us a shout-out on X?

Thank you for using CodeRabbit!