@lancedb/lancedb 0.5.2 → 0.7.1

package/examples/basic.ts

@@ -0,0 +1,149 @@
+// --8<--  [start:imports]
+import * as lancedb from "@lancedb/lancedb";
+import * as arrow from "apache-arrow";
+import { Field, FixedSizeList, Float16, Int32, Schema } from "apache-arrow";
+
+// --8<-- [end:imports]
+
+// --8<-- [start:connect]
+const uri = "/tmp/lancedb/";
+const db = await lancedb.connect(uri);
+// --8<-- [end:connect]
+{
+  // --8<-- [start:create_table]
+  const data = [
+    { vector: [3.1, 4.1], item: "foo", price: 10.0 },
+    { vector: [5.9, 26.5], item: "bar", price: 20.0 },
+  ];
+  const _tbl = await db.createTable("myTable", data);
+  // --8<-- [end:create_table]
+  {
+    // --8<-- [start:create_table_exists_ok]
+    const _tbl = await db.createTable("myTable", data, {
+      existsOk: true,
+    });
+    // --8<-- [end:create_table_exists_ok]
+  }
+  {
+    // --8<-- [start:create_table_overwrite]
+    const _tbl = await db.createTable("myTable", data, {
+      mode: "overwrite",
+    });
+    // --8<-- [end:create_table_overwrite]
+  }
+}
+
+{
+  // --8<-- [start:create_table_with_schema]
+  const schema = new arrow.Schema([
+    new arrow.Field(
+      "vector",
+      new arrow.FixedSizeList(
+        2,
+        new arrow.Field("item", new arrow.Float32(), true),
+      ),
+    ),
+    new arrow.Field("item", new arrow.Utf8(), true),
+    new arrow.Field("price", new arrow.Float32(), true),
+  ]);
+  const data = [
+    { vector: [3.1, 4.1], item: "foo", price: 10.0 },
+    { vector: [5.9, 26.5], item: "bar", price: 20.0 },
+  ];
+  const _tbl = await db.createTable("myTable", data, {
+    schema,
+  });
+  // --8<-- [end:create_table_with_schema]
+}
+
+{
+  // --8<-- [start:create_empty_table]
+  const schema = new arrow.Schema([
+    new arrow.Field(
+      "vector",
+      new arrow.FixedSizeList(
+        2,
+        new arrow.Field("item", new arrow.Float32(), true),
+      ),
+    ),
+  ]);
+  const _tbl = await db.createEmptyTable("empty_table", schema);
+  // --8<-- [end:create_empty_table]
+}
+{
+  // --8<-- [start:open_table]
+  const _tbl = await db.openTable("myTable");
+  // --8<-- [end:open_table]
+}
+
+{
+  // --8<-- [start:table_names]
+  const tableNames = await db.tableNames();
+  console.log(tableNames);
+  // --8<-- [end:table_names]
+}
+
+const tbl = await db.openTable("myTable");
+{
+  // --8<-- [start:add_data]
+  const data = [
+    { vector: [1.3, 1.4], item: "fizz", price: 100.0 },
+    { vector: [9.5, 56.2], item: "buzz", price: 200.0 },
+  ];
+  await tbl.add(data);
+  // --8<-- [end:add_data]
+}
+{
+  // --8<-- [start:vector_search]
+  const _res = tbl.search([100, 100]).limit(2).toArray();
+  // --8<-- [end:vector_search]
+}
+{
+  const data = Array.from({ length: 1000 })
+    .fill(null)
+    .map(() => ({
+      vector: [Math.random(), Math.random()],
+      item: "autogen",
+      price: Math.round(Math.random() * 100),
+    }));
+
+  await tbl.add(data);
+}
+
+// --8<-- [start:create_index]
+await tbl.createIndex("vector");
+// --8<-- [end:create_index]
+
+// --8<-- [start:delete_rows]
+await tbl.delete('item = "fizz"');
+// --8<-- [end:delete_rows]
+
+// --8<-- [start:drop_table]
+await db.dropTable("myTable");
+// --8<-- [end:drop_table]
+await db.dropTable("empty_table");
+
+{
+  // --8<-- [start:create_f16_table]
+  const db = await lancedb.connect("/tmp/lancedb");
+  const dim = 16;
+  const total = 10;
+  const f16Schema = new Schema([
+    new Field("id", new Int32()),
+    new Field(
+      "vector",
+      new FixedSizeList(dim, new Field("item", new Float16(), true)),
+      false,
+    ),
+  ]);
+  const data = lancedb.makeArrowTable(
+    Array.from(Array(total), (_, i) => ({
+      id: i,
+      vector: Array.from(Array(dim), Math.random),
+    })),
+    { schema: f16Schema },
+  );
+  const _table = await db.createTable("f16_tbl", data);
+  // --8<-- [end:create_f16_table]
+  await db.dropTable("f16_tbl");
+}

package/examples/embedding.ts

@@ -0,0 +1,83 @@
+// --8<-- [start:imports]
+import * as lancedb from "@lancedb/lancedb";
+import { LanceSchema, getRegistry, register } from "@lancedb/lancedb/embedding";
+import { EmbeddingFunction } from "@lancedb/lancedb/embedding";
+import { type Float, Float32, Utf8 } from "apache-arrow";
+// --8<-- [end:imports]
+
+{
+  // --8<-- [start:openai_embeddings]
+
+  const db = await lancedb.connect("/tmp/db");
+  const func = getRegistry()
+    .get("openai")
+    ?.create({ model: "text-embedding-ada-002" }) as EmbeddingFunction;
+
+  const wordsSchema = LanceSchema({
+    text: func.sourceField(new Utf8()),
+    vector: func.vectorField(),
+  });
+  const tbl = await db.createEmptyTable("words", wordsSchema, {
+    mode: "overwrite",
+  });
+  await tbl.add([{ text: "hello world" }, { text: "goodbye world" }]);
+
+  const query = "greetings";
+  const actual = (await (await tbl.search(query)).limit(1).toArray())[0];
+
+  // --8<-- [end:openai_embeddings]
+  console.log("result = ", actual.text);
+}
+
+{
+  // --8<-- [start:embedding_function]
+  const db = await lancedb.connect("/tmp/db");
+
+  @register("my_embedding")
+  class MyEmbeddingFunction extends EmbeddingFunction<string> {
+    toJSON(): object {
+      return {};
+    }
+    ndims() {
+      return 3;
+    }
+    embeddingDataType(): Float {
+      return new Float32();
+    }
+    async computeQueryEmbeddings(_data: string) {
+      // This is a placeholder for a real embedding function
+      return [1, 2, 3];
+    }
+    async computeSourceEmbeddings(data: string[]) {
+      // This is a placeholder for a real embedding function
+      return Array.from({ length: data.length }).fill([1, 2, 3]) as number[][];
+    }
+  }
+
+  const func = new MyEmbeddingFunction();
+
+  const data = [{ text: "pepperoni" }, { text: "pineapple" }];
+
+  // Option 1: manually specify the embedding function
+  const table = await db.createTable("vectors", data, {
+    embeddingFunction: {
+      function: func,
+      sourceColumn: "text",
+      vectorColumn: "vector",
+    },
+    mode: "overwrite",
+  });
+
+  // Option 2: provide the embedding function through a schema
+
+  const schema = LanceSchema({
+    text: func.sourceField(new Utf8()),
+    vector: func.vectorField(),
+  });
+
+  const table2 = await db.createTable("vectors2", data, {
+    schema,
+    mode: "overwrite",
+  });
+  // --8<-- [end:embedding_function]
+}

package/examples/filtering.ts

@@ -0,0 +1,34 @@
+import * as lancedb from "@lancedb/lancedb";
+
+const db = await lancedb.connect("data/sample-lancedb");
+
+const data = Array.from({ length: 10_000 }, (_, i) => ({
+  vector: Array(1536).fill(i),
+  id: i,
+  item: `item ${i}`,
+  strId: `${i}`,
+}));
+
+const tbl = await db.createTable("myVectors", data, { mode: "overwrite" });
+
+// --8<-- [start:search]
+const _result = await tbl
+  .search(Array(1536).fill(0.5))
+  .limit(1)
+  .where("id = 10")
+  .toArray();
+// --8<-- [end:search]
+
+// --8<-- [start:vec_search]
+await tbl
+  .search(Array(1536).fill(0))
+  .where("(item IN ('item 0', 'item 2')) AND (id > 10)")
+  .postfilter()
+  .toArray();
+// --8<-- [end:vec_search]
+
+// --8<-- [start:sql_search]
+await tbl.query().where("id = 10").limit(10).toArray();
+// --8<-- [end:sql_search]
+
+console.log("SQL search: done");

package/examples/jsconfig.json

@@ -0,0 +1,27 @@
+{
+  "compilerOptions": {
+    // Enable latest features
+    "lib": ["ESNext", "DOM"],
+    "target": "ESNext",
+    "module": "ESNext",
+    "moduleDetection": "force",
+    "jsx": "react-jsx",
+    "allowJs": true,
+
+    // Bundler mode
+    "moduleResolution": "bundler",
+    "allowImportingTsExtensions": true,
+    "verbatimModuleSyntax": true,
+    "noEmit": true,
+
+    // Best practices
+    "strict": true,
+    "skipLibCheck": true,
+    "noFallthroughCasesInSwitch": true,
+
+    // Some stricter flags (disabled by default)
+    "noUnusedLocals": false,
+    "noUnusedParameters": false,
+    "noPropertyAccessFromIndexSignature": false
+  }
+}

package/examples/package-lock.json

@@ -0,0 +1,79 @@
+{
+  "name": "examples",
+  "version": "1.0.0",
+  "lockfileVersion": 3,
+  "requires": true,
+  "packages": {
+    "": {
+      "name": "examples",
+      "version": "1.0.0",
+      "license": "Apache-2.0",
+      "dependencies": {
+        "@lancedb/lancedb": "file:../"
+      },
+      "peerDependencies": {
+        "typescript": "^5.0.0"
+      }
+    },
+    "..": {
+      "name": "@lancedb/lancedb",
+      "version": "0.6.0",
+      "cpu": [
+        "x64",
+        "arm64"
+      ],
+      "license": "Apache 2.0",
+      "os": [
+        "darwin",
+        "linux",
+        "win32"
+      ],
+      "dependencies": {
+        "apache-arrow": "^15.0.0",
+        "axios": "^1.7.2",
+        "openai": "^4.29.2",
+        "reflect-metadata": "^0.2.2"
+      },
+      "devDependencies": {
+        "@aws-sdk/client-kms": "^3.33.0",
+        "@aws-sdk/client-s3": "^3.33.0",
+        "@biomejs/biome": "^1.7.3",
+        "@jest/globals": "^29.7.0",
+        "@napi-rs/cli": "^2.18.0",
+        "@types/axios": "^0.14.0",
+        "@types/jest": "^29.1.2",
+        "@types/tmp": "^0.2.6",
+        "apache-arrow-old": "npm:apache-arrow@13.0.0",
+        "eslint": "^8.57.0",
+        "jest": "^29.7.0",
+        "shx": "^0.3.4",
+        "tmp": "^0.2.3",
+        "ts-jest": "^29.1.2",
+        "typedoc": "^0.25.7",
+        "typedoc-plugin-markdown": "^3.17.1",
+        "typescript": "^5.3.3",
+        "typescript-eslint": "^7.1.0"
+      },
+      "engines": {
+        "node": ">= 18"
+      }
+    },
+    "node_modules/@lancedb/lancedb": {
+      "resolved": "..",
+      "link": true
+    },
+    "node_modules/typescript": {
+      "version": "5.5.2",
+      "resolved": "https://registry.npmjs.org/typescript/-/typescript-5.5.2.tgz",
+      "integrity": "sha512-NcRtPEOsPFFWjobJEtfihkLCZCXZt/os3zf8nTxjVH3RvTSxjrCamJpbExGvYOF+tFHc3pA65qpdwPbzjohhew==",
+      "peer": true,
+      "bin": {
+        "tsc": "bin/tsc",
+        "tsserver": "bin/tsserver"
+      },
+      "engines": {
+        "node": ">=14.17"
+      }
+    }
+  }
+}

package/examples/package.json

@@ -0,0 +1,18 @@
+{
+  "name": "examples",
+  "version": "1.0.0",
+  "description": "Examples for LanceDB",
+  "main": "index.js",
+  "type": "module",
+  "scripts": {
+    "test": "echo \"Error: no test specified\" && exit 1"
+  },
+  "author": "Lance Devs",
+  "license": "Apache-2.0",
+  "dependencies": {
+    "@lancedb/lancedb": "file:../"
+  },
+  "peerDependencies": {
+    "typescript": "^5.0.0"
+  }
+}

package/examples/search.ts

@@ -0,0 +1,37 @@
+// --8<-- [end:import]
+import * as fs from "node:fs";
+// --8<-- [start:import]
+import * as lancedb from "@lancedb/lancedb";
+
+async function setup() {
+  fs.rmSync("data/sample-lancedb", { recursive: true, force: true });
+  const db = await lancedb.connect("data/sample-lancedb");
+
+  const data = Array.from({ length: 10_000 }, (_, i) => ({
+    vector: Array(1536).fill(i),
+    id: `${i}`,
+    content: "",
+    longId: `${i}`,
+  }));
+
+  await db.createTable("my_vectors", data);
+}
+
+await setup();
+
+// --8<-- [start:search1]
+const db = await lancedb.connect("data/sample-lancedb");
+const tbl = await db.openTable("my_vectors");
+
+const _results1 = await tbl.search(Array(1536).fill(1.2)).limit(10).toArray();
+// --8<-- [end:search1]
+
+// --8<-- [start:search2]
+const _results2 = await tbl
+  .search(Array(1536).fill(1.2))
+  .distanceType("cosine")
+  .limit(10)
+  .toArray();
+// --8<-- [end:search2]
+
+console.log("search: done");

package/lancedb/arrow.ts

@@ -15,6 +15,7 @@
 import {
   Table as ArrowTable,
   Binary,
+  BufferType,
   DataType,
   Field,
   FixedSizeBinary,
@@ -37,14 +38,72 @@ import {
   type makeTable,
   vectorFromArray,
 } from "apache-arrow";
+import { Buffers } from "apache-arrow/data";
 import { type EmbeddingFunction } from "./embedding/embedding_function";
 import { EmbeddingFunctionConfig, getRegistry } from "./embedding/registry";
-import { sanitizeField, sanitizeSchema, sanitizeType } from "./sanitize";
+import {
+  sanitizeField,
+  sanitizeSchema,
+  sanitizeTable,
+  sanitizeType,
+} from "./sanitize";
 export * from "apache-arrow";
-
-export type IntoVector = Float32Array | Float64Array | number[];
-
-export function isArrowTable(value: object): value is ArrowTable {
+export type SchemaLike =
+  | Schema
+  | {
+      fields: FieldLike[];
+      metadata: Map<string, string>;
+      get names(): unknown[];
+    };
+export type FieldLike =
+  | Field
+  | {
+      type: string;
+      name: string;
+      nullable?: boolean;
+      metadata?: Map<string, string>;
+    };
+
+export type DataLike =
+  // biome-ignore lint/suspicious/noExplicitAny: <explanation>
+  | import("apache-arrow").Data<Struct<any>>
+  | {
+      // biome-ignore lint/suspicious/noExplicitAny: <explanation>
+      type: any;
+      length: number;
+      offset: number;
+      stride: number;
+      nullable: boolean;
+      children: DataLike[];
+      get nullCount(): number;
+      // biome-ignore lint/suspicious/noExplicitAny: <explanation>
+      values: Buffers<any>[BufferType.DATA];
+      // biome-ignore lint/suspicious/noExplicitAny: <explanation>
+      typeIds: Buffers<any>[BufferType.TYPE];
+      // biome-ignore lint/suspicious/noExplicitAny: <explanation>
+      nullBitmap: Buffers<any>[BufferType.VALIDITY];
+      // biome-ignore lint/suspicious/noExplicitAny: <explanation>
+      valueOffsets: Buffers<any>[BufferType.OFFSET];
+    };
+
+export type RecordBatchLike =
+  | RecordBatch
+  | {
+      schema: SchemaLike;
+      data: DataLike;
+    };
+
+export type TableLike =
+  | ArrowTable
+  | { schema: SchemaLike; batches: RecordBatchLike[] };
+
+export type IntoVector =
+  | Float32Array
+  | Float64Array
+  | number[]
+  | Promise<Float32Array | Float64Array | number[]>;
+
+export function isArrowTable(value: object): value is TableLike {
   if (value instanceof ArrowTable) return true;
   return "schema" in value && "batches" in value;
 }
@@ -135,7 +194,7 @@ export function isFixedSizeList(value: unknown): value is FixedSizeList {
 }
 
 /** Data type accepted by NodeJS SDK */
-export type Data = Record<string, unknown>[] | ArrowTable;
+export type Data = Record<string, unknown>[] | TableLike;
 
 /*
  * Options to control how a column should be converted to a vector array
@@ -162,7 +221,7 @@ export class MakeArrowTableOptions {
    * The schema must be specified if there are no records (e.g. to make
    * an empty table)
    */
-  schema?: Schema;
+  schema?: SchemaLike;
 
   /*
    * Mapping from vector column name to expected type
@@ -310,7 +369,7 @@ export function makeArrowTable(
   if (opt.schema !== undefined && opt.schema !== null) {
     opt.schema = sanitizeSchema(opt.schema);
     opt.schema = validateSchemaEmbeddings(
-      opt.schema,
+      opt.schema as Schema,
       data,
       options?.embeddingFunction,
     );
@@ -394,7 +453,7 @@ export function makeArrowTable(
     // `new ArrowTable(schema, batches)` which does not do any schema inference
     const firstTable = new ArrowTable(columns);
     const batchesFixed = firstTable.batches.map(
-      (batch) => new RecordBatch(opt.schema!, batch.data),
+      (batch) => new RecordBatch(opt.schema as Schema, batch.data),
     );
     let schema: Schema;
     if (metadata !== undefined) {
@@ -407,9 +466,9 @@ export function makeArrowTable(
         }
       }
 
-      schema = new Schema(opt.schema.fields, schemaMetadata);
+      schema = new Schema(opt.schema.fields as Field[], schemaMetadata);
     } else {
-      schema = opt.schema;
+      schema = opt.schema as Schema;
     }
     return new ArrowTable(schema, batchesFixed);
   }
@@ -425,7 +484,7 @@ export function makeArrowTable(
  * Create an empty Arrow table with the provided schema
  */
 export function makeEmptyTable(
-  schema: Schema,
+  schema: SchemaLike,
   metadata?: Map<string, string>,
 ): ArrowTable {
   return makeArrowTable([], { schema }, metadata);
@@ -563,18 +622,17 @@ async function applyEmbeddingsFromMetadata(
 async function applyEmbeddings<T>(
   table: ArrowTable,
   embeddings?: EmbeddingFunctionConfig,
-  schema?: Schema,
+  schema?: SchemaLike,
 ): Promise<ArrowTable> {
+  if (schema !== undefined && schema !== null) {
+    schema = sanitizeSchema(schema);
+  }
   if (schema?.metadata.has("embedding_functions")) {
-    return applyEmbeddingsFromMetadata(table, schema!);
+    return applyEmbeddingsFromMetadata(table, schema! as Schema);
   } else if (embeddings == null || embeddings === undefined) {
     return table;
   }
 
-  if (schema !== undefined && schema !== null) {
-    schema = sanitizeSchema(schema);
-  }
-
   // Convert from ArrowTable to Record<String, Vector>
   const colEntries = [...Array(table.numCols).keys()].map((_, idx) => {
     const name = table.schema.fields[idx].name;
@@ -650,7 +708,7 @@ async function applyEmbeddings<T>(
         `When using embedding functions and specifying a schema the schema should include the embedding column but the column ${destColumn} was missing`,
       );
     }
-    return alignTable(newTable, schema);
+    return alignTable(newTable, schema as Schema);
   }
   return newTable;
 }
@@ -744,7 +802,7 @@ export async function fromRecordsToStreamBuffer(
 export async function fromTableToBuffer(
   table: ArrowTable,
   embeddings?: EmbeddingFunctionConfig,
-  schema?: Schema,
+  schema?: SchemaLike,
 ): Promise<Buffer> {
   if (schema !== undefined && schema !== null) {
     schema = sanitizeSchema(schema);
@@ -771,7 +829,7 @@ export async function fromDataToBuffer(
     schema = sanitizeSchema(schema);
   }
   if (isArrowTable(data)) {
-    return fromTableToBuffer(data, embeddings, schema);
+    return fromTableToBuffer(sanitizeTable(data), embeddings, schema);
   } else {
     const table = await convertToTable(data, embeddings, { schema });
     return fromTableToBuffer(table);
@@ -789,7 +847,7 @@ export async function fromDataToBuffer(
 export async function fromTableToStreamBuffer(
   table: ArrowTable,
   embeddings?: EmbeddingFunctionConfig,
-  schema?: Schema,
+  schema?: SchemaLike,
 ): Promise<Buffer> {
   const tableWithEmbeddings = await applyEmbeddings(table, embeddings, schema);
   const writer = RecordBatchStreamWriter.writeAll(tableWithEmbeddings);
@@ -854,7 +912,6 @@ function validateSchemaEmbeddings(
   for (let field of schema.fields) {
     if (isFixedSizeList(field.type)) {
       field = sanitizeField(field);
-
       if (data.length !== 0 && data?.[0]?.[field.name] === undefined) {
         if (schema.metadata.has("embedding_functions")) {
           const embeddings = JSON.parse(