mirror of
https://github.com/lancedb/lancedb.git
synced 2026-10-17 09:12:28 +00:00
Brendan Clement
758f738b8c
This PR is part of the documentation revamp. Once the full set of PRs is merged, we’ll publish the combined site to staging for review before updating docs.lancedb.com. Ingestion, search and integration examples are incomplete or inconsistent across languages, and some guidance misstates SDK behavior. This brings the guides and their example sources together. It includes complete Python, TypeScript and Rust workflows, isolated provider and reranker examples, the Superlinked corrections, and guidance on retries that can duplicate writes. Stacked on #4491. Each MDX file has one PR owner. Rebase onto main after #4491 is squash-merged. Local assembly, Mint validation and link checks pass. Example code is unchanged from the previously validated package. Publication stays held through integration.
33 lines
11 KiB
Plaintext
33 lines
11 KiB
Plaintext
{/* Auto-generated by scripts/mdx_snippets_gen.py. Do not edit manually. */}
|
|
|
|
export const PyFilterSearch = "from tempfile import TemporaryDirectory\n\nimport lancedb\n\nwith TemporaryDirectory(prefix=\"lancedb-search-\") as uri:\n db = lancedb.connect(uri)\n table = db.create_table(\n \"documents\",\n [\n {\n \"id\": 1,\n \"text\": \"A puppy playing in the park\",\n \"category\": \"animals\",\n \"vector\": [1.0, 0.0],\n },\n {\n \"id\": 2,\n \"text\": \"A dog resting at home\",\n \"category\": \"animals\",\n \"vector\": [0.9, 0.1],\n },\n {\n \"id\": 3,\n \"text\": \"A guide to growing tomatoes\",\n \"category\": \"gardening\",\n \"vector\": [0.0, 1.0],\n },\n ],\n )\n rows = (\n table.search([0.0, 1.0])\n .where(\"category = 'animals'\", prefilter=True)\n .select([\"id\", \"text\", \"category\", \"_distance\"])\n .limit(2)\n .to_list()\n )\n print(rows)\n";
|
|
|
|
export const PyHybridSearch = "from tempfile import TemporaryDirectory\n\nimport lancedb\nfrom lancedb.index import FTS\n\nwith TemporaryDirectory(prefix=\"lancedb-search-\") as uri:\n db = lancedb.connect(uri)\n table = db.create_table(\n \"documents\",\n [\n {\n \"id\": 1,\n \"text\": \"A puppy playing in the park\",\n \"category\": \"animals\",\n \"vector\": [1.0, 0.0],\n },\n {\n \"id\": 2,\n \"text\": \"A dog resting at home\",\n \"category\": \"animals\",\n \"vector\": [0.9, 0.1],\n },\n {\n \"id\": 3,\n \"text\": \"A guide to growing tomatoes\",\n \"category\": \"gardening\",\n \"vector\": [0.0, 1.0],\n },\n ],\n )\n from lancedb.rerankers import RRFReranker\n\n table.create_index(\"text\", config=FTS())\n rows = (\n table.search(\n query_type=\"hybrid\", vector_column_name=\"vector\", fts_columns=\"text\"\n )\n .vector([1.0, 0.0])\n .text(\"puppy\")\n .rerank(RRFReranker())\n .limit(2)\n .to_list()\n )\n print(rows)\n";
|
|
|
|
export const PyQueryPlan = "from tempfile import TemporaryDirectory\n\nimport lancedb\n\nwith TemporaryDirectory(prefix=\"lancedb-search-\") as uri:\n db = lancedb.connect(uri)\n table = db.create_table(\n \"documents\",\n [\n {\n \"id\": 1,\n \"text\": \"A puppy playing in the park\",\n \"category\": \"animals\",\n \"vector\": [1.0, 0.0],\n },\n {\n \"id\": 2,\n \"text\": \"A dog resting at home\",\n \"category\": \"animals\",\n \"vector\": [0.9, 0.1],\n },\n {\n \"id\": 3,\n \"text\": \"A guide to growing tomatoes\",\n \"category\": \"gardening\",\n \"vector\": [0.0, 1.0],\n },\n ],\n )\n query = table.search([1.0, 0.0]).select([\"id\", \"text\", \"_distance\"]).limit(2)\n print(\"Plan:\", query.explain_plan())\n print(\"Measured execution:\", query.analyze_plan())\n rows = query.to_list()\n print(rows)\n";
|
|
|
|
export const PyTextSearch = "from tempfile import TemporaryDirectory\n\nimport lancedb\nfrom lancedb.index import FTS\n\nwith TemporaryDirectory(prefix=\"lancedb-search-\") as uri:\n db = lancedb.connect(uri)\n table = db.create_table(\n \"documents\",\n [\n {\n \"id\": 1,\n \"text\": \"A puppy playing in the park\",\n \"category\": \"animals\",\n },\n {\n \"id\": 2,\n \"text\": \"A dog resting at home\",\n \"category\": \"animals\",\n },\n {\n \"id\": 3,\n \"text\": \"A guide to growing tomatoes\",\n \"category\": \"gardening\",\n },\n ],\n )\n table.create_index(\"text\", config=FTS())\n rows = (\n table.search(\"puppy\", query_type=\"fts\", fts_columns=\"text\")\n .select([\"id\", \"text\", \"_score\"])\n .limit(2)\n .to_list()\n )\n print(rows)\n";
|
|
|
|
export const PyVectorSearch = "from tempfile import TemporaryDirectory\n\nimport lancedb\n\nwith TemporaryDirectory(prefix=\"lancedb-search-\") as uri:\n db = lancedb.connect(uri)\n table = db.create_table(\n \"documents\",\n [\n {\n \"id\": 1,\n \"text\": \"A puppy playing in the park\",\n \"category\": \"animals\",\n \"vector\": [1.0, 0.0],\n },\n {\n \"id\": 2,\n \"text\": \"A dog resting at home\",\n \"category\": \"animals\",\n \"vector\": [0.9, 0.1],\n },\n {\n \"id\": 3,\n \"text\": \"A guide to growing tomatoes\",\n \"category\": \"gardening\",\n \"vector\": [0.0, 1.0],\n },\n ],\n )\n rows = (\n table.search([1.0, 0.0])\n .distance_type(\"l2\")\n .select([\"id\", \"text\", \"_distance\"])\n .limit(2)\n .to_list()\n )\n print(rows)\n";
|
|
|
|
export const TsFilterSearchBody = "const lancedb = await import(\"@lancedb/lancedb\");\nconst { mkdtemp, rm } = await import(\"node:fs/promises\");\nconst { tmpdir } = await import(\"node:os\");\nconst { join } = await import(\"node:path\");\nconst uri = await mkdtemp(join(tmpdir(), \"lancedb-search-\"));\ntry {\n const db = await lancedb.connect(uri);\n const table = await db.createTable(\"documents\", [\n {\n id: 1,\n text: \"A puppy playing in the park\",\n category: \"animals\",\n vector: [1.0, 0.0],\n },\n {\n id: 2,\n text: \"A dog resting at home\",\n category: \"animals\",\n vector: [0.9, 0.1],\n },\n {\n id: 3,\n text: \"A guide to growing tomatoes\",\n category: \"gardening\",\n vector: [0.0, 1.0],\n },\n ]);\n const rows = await table\n .vectorSearch([0.0, 1.0])\n .where(\"category = 'animals'\")\n .select([\"id\", \"text\", \"category\", \"_distance\"])\n .limit(2)\n .toArray();\n console.log(rows);\n";
|
|
|
|
export const TsFilterSearchTail = "} finally {\n await rm(uri, { recursive: true, force: true });\n}\n";
|
|
|
|
export const TsHybridSearchBody = "const lancedb = await import(\"@lancedb/lancedb\");\nconst { mkdtemp, rm } = await import(\"node:fs/promises\");\nconst { tmpdir } = await import(\"node:os\");\nconst { join } = await import(\"node:path\");\nconst uri = await mkdtemp(join(tmpdir(), \"lancedb-search-\"));\ntry {\n const db = await lancedb.connect(uri);\n const table = await db.createTable(\"documents\", [\n {\n id: 1,\n text: \"A puppy playing in the park\",\n category: \"animals\",\n vector: [1.0, 0.0],\n },\n {\n id: 2,\n text: \"A dog resting at home\",\n category: \"animals\",\n vector: [0.9, 0.1],\n },\n {\n id: 3,\n text: \"A guide to growing tomatoes\",\n category: \"gardening\",\n vector: [0.0, 1.0],\n },\n ]);\n await table.createIndex(\"text\", { config: lancedb.Index.fts() });\n const reranker = await lancedb.rerankers.RRFReranker.create();\n const rows = await table\n .vectorSearch([1.0, 0.0])\n .fullTextSearch(\"puppy\", { columns: [\"text\"] })\n .rerank(reranker)\n .limit(2)\n .toArray();\n console.log(rows);\n";
|
|
|
|
export const TsHybridSearchTail = "} finally {\n await rm(uri, { recursive: true, force: true });\n}\n";
|
|
|
|
export const TsQueryPlanBody = "const lancedb = await import(\"@lancedb/lancedb\");\nconst { mkdtemp, rm } = await import(\"node:fs/promises\");\nconst { tmpdir } = await import(\"node:os\");\nconst { join } = await import(\"node:path\");\nconst uri = await mkdtemp(join(tmpdir(), \"lancedb-search-\"));\ntry {\n const db = await lancedb.connect(uri);\n const table = await db.createTable(\"documents\", [\n {\n id: 1,\n text: \"A puppy playing in the park\",\n category: \"animals\",\n vector: [1.0, 0.0],\n },\n {\n id: 2,\n text: \"A dog resting at home\",\n category: \"animals\",\n vector: [0.9, 0.1],\n },\n {\n id: 3,\n text: \"A guide to growing tomatoes\",\n category: \"gardening\",\n vector: [0.0, 1.0],\n },\n ]);\n const query = table\n .vectorSearch([1.0, 0.0])\n .select([\"id\", \"text\", \"_distance\"])\n .limit(2);\n console.log(\"Plan:\", await query.explainPlan());\n console.log(\"Measured execution:\", await query.analyzePlan());\n const rows = await query.toArray();\n console.log(rows);\n";
|
|
|
|
export const TsQueryPlanTail = "} finally {\n await rm(uri, { recursive: true, force: true });\n}\n";
|
|
|
|
export const TsTextSearchBody = "const lancedb = await import(\"@lancedb/lancedb\");\nconst { mkdtemp, rm } = await import(\"node:fs/promises\");\nconst { tmpdir } = await import(\"node:os\");\nconst { join } = await import(\"node:path\");\nconst uri = await mkdtemp(join(tmpdir(), \"lancedb-search-\"));\ntry {\n const db = await lancedb.connect(uri);\n const table = await db.createTable(\"documents\", [\n { id: 1, text: \"A puppy playing in the park\", category: \"animals\" },\n { id: 2, text: \"A dog resting at home\", category: \"animals\" },\n { id: 3, text: \"A guide to growing tomatoes\", category: \"gardening\" },\n ]);\n await table.createIndex(\"text\", { config: lancedb.Index.fts() });\n const rows = await table\n .query()\n .fullTextSearch(\"puppy\", { columns: [\"text\"] })\n .select([\"id\", \"text\", \"_score\"])\n .limit(2)\n .toArray();\n console.log(rows);\n";
|
|
|
|
export const TsTextSearchTail = "} finally {\n await rm(uri, { recursive: true, force: true });\n}\n";
|
|
|
|
export const TsVectorSearchBody = "const lancedb = await import(\"@lancedb/lancedb\");\nconst { mkdtemp, rm } = await import(\"node:fs/promises\");\nconst { tmpdir } = await import(\"node:os\");\nconst { join } = await import(\"node:path\");\nconst uri = await mkdtemp(join(tmpdir(), \"lancedb-search-\"));\ntry {\n const db = await lancedb.connect(uri);\n const table = await db.createTable(\"documents\", [\n {\n id: 1,\n text: \"A puppy playing in the park\",\n category: \"animals\",\n vector: [1.0, 0.0],\n },\n {\n id: 2,\n text: \"A dog resting at home\",\n category: \"animals\",\n vector: [0.9, 0.1],\n },\n {\n id: 3,\n text: \"A guide to growing tomatoes\",\n category: \"gardening\",\n vector: [0.0, 1.0],\n },\n ]);\n const rows = await table\n .vectorSearch([1.0, 0.0])\n .distanceType(\"l2\")\n .select([\"id\", \"text\", \"_distance\"])\n .limit(2)\n .toArray();\n console.log(rows);\n";
|
|
|
|
export const TsVectorSearchTail = "} finally {\n await rm(uri, { recursive: true, force: true });\n}\n";
|
|
|