feature/v2.2 features - #12
Conversation
CoderDayton
commented
Jan 17, 2026
CoderDayton
commented
- feat: add cross-collection search API
- fix: address clustering assignment and dependency messages
- docs: fix clustering anchor links
- chore: bump version to 2.2.0
Add list_collections() and search_collections() methods to VectorDB for searching across multiple collections with unified, ranked results. - Score normalization: 1/(1+distance) for fair comparison across metrics - Parallel search with ThreadPoolExecutor for >1 collections - Dimension validation across searched collections - Async versions in AsyncVectorDB - 20 comprehensive unit tests - Documentation and examples
There was a problem hiding this comment.
Pull request overview
This PR introduces v2.2.0 with major new features: vector clustering with auto-tagging and cross-collection search. The implementation adds K-means, MiniBatch K-means, and HDBSCAN clustering algorithms with cluster persistence capabilities, plus the ability to search across multiple collections with unified, ranked results.
Changes:
- Added comprehensive clustering engine with three algorithms (kmeans, minibatch_kmeans, hdbscan) and auto-tagging via TF-IDF
- Implemented cross-collection search API with score normalization and parallel execution
- Moved encryption and clustering dependencies from optional to standard install
- Added 29 new tests (26 clustering + 3 async clustering tests)
Reviewed changes
Copilot reviewed 20 out of 21 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| uv.lock | Added dependencies: black, hdbscan, pytokens, sqlite-vec; updated simplevecdb to 2.2.0 |
| pyproject.toml | Bumped version to 2.2.0, moved encryption/clustering to core dependencies |
| src/simplevecdb/init.py | Updated version to 2.2.0 |
| src/simplevecdb/types.py | Added ClusterResult dataclass with quality metrics |
| src/simplevecdb/engine/clustering.py | New clustering engine with K-means variants, HDBSCAN, and TF-IDF tagging |
| src/simplevecdb/engine/usearch_index.py | Added keys() and get() methods for cluster vector retrieval |
| src/simplevecdb/engine/catalog.py | Added cluster persistence (save/load/list/delete) and batch metadata updates |
| src/simplevecdb/encryption.py | Updated dependency messages for standard install |
| src/simplevecdb/core.py | Added clustering methods and cross-collection search to VectorCollection and VectorDB |
| src/simplevecdb/async_core.py | Added async wrappers for all clustering methods |
| tests/unit/test_clustering.py | Comprehensive clustering tests (26 tests) |
| tests/unit/test_cross_collection_search.py | Cross-collection search tests (new file) |
| tests/unit/test_async.py | Added 3 async clustering tests |
| docs/guides/clustering.md | New comprehensive clustering guide |
| docs/api/types.md | New types reference documentation |
| docs/api/core.md | Updated with clustering and cross-collection search APIs |
| docs/CHANGELOG.md | Detailed v2.2.0 changelog entry |
| docs/index.md | Updated with clustering examples |
| docs/examples.md | Added cross-collection search examples |
| README.md | Updated feature matrix and examples |
| mkdocs.yml | Added clustering guide and types API pages |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 75f0c2fb1e
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
|
@copilot open a new pull request to apply changes based on the comments in this thread |
|
@CoderDayton I've opened a new pull request, #13, to work on those changes. Once the pull request is ready, I'll request review from you. |
[WIP] Add cross-collection search API and clustering fixes
