|
2 | 2 |
|
3 | 3 | [](https://www.npmjs.com/package/@photostructure/sqlite-diskann) |
4 | 4 | [](https://github.com/photostructure/sqlite-diskann/actions/workflows/ci.yml) |
| 5 | +[](https://photostructure.github.io/sqlite-diskann/) |
5 | 6 |
|
6 | 7 | SQLite extension for DiskANN approximate nearest neighbor vector search. |
7 | 8 |
|
| 9 | +📘 **[Full TypeScript API Documentation](https://photostructure.github.io/sqlite-diskann/)** |
| 10 | + |
8 | 11 | ## What is this? |
9 | 12 |
|
10 | 13 | A standalone SQLite extension implementing the [DiskANN algorithm](https://github.com/microsoft/DiskANN) for efficient vector similarity search at scale. Extracted from [libSQL's implementation](https://github.com/tursodatabase/libsql) and optimized for use as a standard SQLite extension. |
@@ -456,6 +459,21 @@ See [`_research/sqlite-vector-options.md`](./_research/sqlite-vector-options.md) |
456 | 459 |
|
457 | 460 | ## API Reference |
458 | 461 |
|
| 462 | +### TypeScript API |
| 463 | +
|
| 464 | +**📘 [Complete API Documentation](https://photostructure.github.io/sqlite-diskann/)** - Auto-generated from source with TypeDoc |
| 465 | +
|
| 466 | +The TypeScript package exports the following functions: |
| 467 | +
|
| 468 | +- **`loadDiskAnnExtension(db)`** - Load extension into database |
| 469 | +- **`getExtensionPath()`** - Get platform-specific extension path |
| 470 | +- **`createDiskAnnIndex(db, tableName, options)`** - Create virtual table with configuration |
| 471 | +- **`searchNearest(db, tableName, queryVector, k)`** - Search for k nearest neighbors |
| 472 | +- **`insertVector(db, tableName, rowid, vector)`** - Insert a vector |
| 473 | +- **`deleteVector(db, tableName, rowid)`** - Delete a vector |
| 474 | +
|
| 475 | +See the [full API documentation](https://photostructure.github.io/sqlite-diskann/) for detailed usage, parameters, and examples. |
| 476 | +
|
459 | 477 | ### Virtual Table SQL |
460 | 478 |
|
461 | 479 | ```sql |
|
0 commit comments