Bump version: 0.24.0-beta.0 → 0.24.0

lance [release
details](https://github.com/lancedb/lance/releases/tag/v0.30.0)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->

## Summary by CodeRabbit

- **Chores**
- Updated dependency specifications to use exact version numbers instead
of referencing a git repository and tag.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->

Signed-off-by: BubbleCal <bubble-cal@outlook.com>

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.20.1-beta.0"
+current_version = "0.20.1-beta.2"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

.github/workflows/npm-publish.yml

@@ -541,6 +541,8 @@ jobs:
         run: npm deprecate vectordb "Use @lancedb/lancedb instead."
       - name: Checkout
         uses: actions/checkout@v4
+        with:
+          ref: main
       - name: Update package-lock.json
         run: |
           git config user.name 'Lance Release'
@@ -548,6 +550,9 @@ jobs:
           bash ci/update_lockfiles.sh
       - name: Push new commit
         uses: ad-m/github-push-action@master
+        with:
+          github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
+          branch: main
       - name: Notify Slack Action
         uses: ravsamhq/notify-slack-action@2.3.0
         if: ${{ always() }}

Cargo.toml

@@ -21,14 +21,14 @@ categories = ["database-implementations"]
 rust-version = "1.78.0"
 
 [workspace.dependencies]
-lance = { "version" = "=0.29.1", "features" = ["dynamodb"], tag = "v0.29.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-io = { version = "=0.29.1", tag = "v0.29.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-index = { version = "=0.29.1", tag = "v0.29.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-linalg = { version = "=0.29.1", tag = "v0.29.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-table = { version = "=0.29.1", tag = "v0.29.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-testing = { version = "=0.29.1", tag = "v0.29.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-datafusion = { version = "=0.29.1", tag = "v0.29.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-encoding = { version = "=0.29.1", tag = "v0.29.1-beta.2", git="https://github.com/lancedb/lance.git" }
+lance = { "version" = "=0.30.0", "features" = ["dynamodb"] }
+lance-io = "=0.30.0"
+lance-index = "=0.30.0"
+lance-linalg = "=0.30.0"
+lance-table = "=0.30.0"
+lance-testing = "=0.30.0"
+lance-datafusion = "=0.30.0"
+lance-encoding = "=0.30.0"
 # Note that this one does not include pyarrow
 arrow = { version = "55.1", optional = false }
 arrow-array = "55.1"

java/core/pom.xml

@@ -8,7 +8,7 @@
     <parent>
         <groupId>com.lancedb</groupId>
         <artifactId>lancedb-parent</artifactId>
-        <version>0.20.1-beta.0</version>
+        <version>0.20.1-beta.2</version>
         <relativePath>../pom.xml</relativePath>
     </parent>
 

java/pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-parent</artifactId>
-    <version>0.20.1-beta.0</version>
+    <version>0.20.1-beta.2</version>
     <packaging>pom</packaging>
 
     <name>LanceDB Parent</name>

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.20.1-beta.0",
+  "version": "0.20.1-beta.2",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.20.1-beta.0",
+      "version": "0.20.1-beta.2",
       "cpu": [
         "x64",
         "arm64"
@@ -52,11 +52,11 @@
         "uuid": "^9.0.0"
       },
       "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.20.1-beta.0",
-        "@lancedb/vectordb-darwin-x64": "0.20.1-beta.0",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.20.1-beta.0",
-        "@lancedb/vectordb-linux-x64-gnu": "0.20.1-beta.0",
-        "@lancedb/vectordb-win32-x64-msvc": "0.20.1-beta.0"
+        "@lancedb/vectordb-darwin-arm64": "0.20.1-beta.2",
+        "@lancedb/vectordb-darwin-x64": "0.20.1-beta.2",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.20.1-beta.2",
+        "@lancedb/vectordb-linux-x64-gnu": "0.20.1-beta.2",
+        "@lancedb/vectordb-win32-x64-msvc": "0.20.1-beta.2"
       },
       "peerDependencies": {
         "@apache-arrow/ts": "^14.0.2",
@@ -327,9 +327,9 @@
       }
     },
     "node_modules/@lancedb/vectordb-darwin-arm64": {
-      "version": "0.20.1-beta.0",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.20.1-beta.0.tgz",
-      "integrity": "sha512-EZl1nvF/2MbLkB8DkNPg+9SpYWpqnNR9kY5a1JWtNWQWw735oT2VPnH3B2htDKU42gJ/9DJGBdEvIJwzeHT85w==",
+      "version": "0.20.1-beta.2",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.20.1-beta.2.tgz",
+      "integrity": "sha512-mqi0yI+ZwBTydaDy1FRHAUZwrWS28u6tbHTe1s4uSrmERbVI6PfmoPR+NZWWAp6ZhlseSdl/+yeI4imk11rQSw==",
       "cpu": [
         "arm64"
       ],
@@ -339,9 +339,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-darwin-x64": {
-      "version": "0.20.1-beta.0",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.20.1-beta.0.tgz",
-      "integrity": "sha512-1ZkMcsXsysLRohAeHGpbytVHUp4yEU89A34rrh48vcQUNvYtqxbAw+TLjAbN0vvNvOZOI4DRllxSL1O+Dbybbg==",
+      "version": "0.20.1-beta.2",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.20.1-beta.2.tgz",
+      "integrity": "sha512-m8EYYA8JZIeNsJqQsBDUMu6r31/u7FzpjonJ4Y+CjapVl6UdvI65KUkeL2dYrFao++RuIoaiqcm3e7gRgFZpXQ==",
       "cpu": [
         "x64"
       ],
@@ -351,9 +351,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.20.1-beta.0",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.20.1-beta.0.tgz",
-      "integrity": "sha512-CxjSGaLJNRYxljdrC8MSirnHu73jctv3S3Q90CbsWMsij9za87zvnrjoiRIn7kv7UNS4ArwS9yyH6gNorCBf6Q==",
+      "version": "0.20.1-beta.2",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.20.1-beta.2.tgz",
+      "integrity": "sha512-3Og2+bk4GlWmMO1Yg2HBfeb5zrOMLaIHD7bEqQ4+6yw4IckAaV+ke05H0tyyqmOVrOQ0LpvtXgD7pPztjm9r9A==",
       "cpu": [
         "arm64"
       ],
@@ -363,9 +363,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.20.1-beta.0",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.20.1-beta.0.tgz",
-      "integrity": "sha512-WI2XWYYO5ygL0Az7SlX98VpNqrz8hKuTK/xC/PoM99s1xnfcCukM28DaDGZJpXOGnLbVnexcO2RW4daJ2xDPaQ==",
+      "version": "0.20.1-beta.2",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.20.1-beta.2.tgz",
+      "integrity": "sha512-mwTQyA/FBoU/FkPuvCNBZG3y83gBN+iYoejehBH2HBkLUIcmlsDgSRZ1OQ+f9ijj12EMBCA11tBUPA9zhHzyrw==",
       "cpu": [
         "x64"
       ],
@@ -375,9 +375,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.20.1-beta.0",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.20.1-beta.0.tgz",
-      "integrity": "sha512-Mxd7V3Y8whEBoQFQZhZGFQi0avq8ujHRI2c0LhjhYTdwGylrBS3bfGD+/nbDGhAjp7dp5U8P4kiBi30QNwoedA==",
+      "version": "0.20.1-beta.2",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.20.1-beta.2.tgz",
+      "integrity": "sha512-VkjNpqhK3l3uHLLPmox+HrmKPMaZgV+qsGQWx0nfseGnSOEmXAWZWQFe0APVCQ9y0xTypQB0oH7eSOPZv2t4WQ==",
       "cpu": [
         "x64"
       ],

node/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vectordb",
-  "version": "0.20.1-beta.0",
+  "version": "0.20.1-beta.2",
   "description": " Serverless, low-latency vector database for AI applications",
   "private": false,
   "main": "dist/index.js",
@@ -89,10 +89,10 @@
     }
   },
   "optionalDependencies": {
-    "@lancedb/vectordb-darwin-x64": "0.20.1-beta.0",
-    "@lancedb/vectordb-darwin-arm64": "0.20.1-beta.0",
-    "@lancedb/vectordb-linux-x64-gnu": "0.20.1-beta.0",
-    "@lancedb/vectordb-linux-arm64-gnu": "0.20.1-beta.0",
-    "@lancedb/vectordb-win32-x64-msvc": "0.20.1-beta.0"
+    "@lancedb/vectordb-darwin-x64": "0.20.1-beta.2",
+    "@lancedb/vectordb-darwin-arm64": "0.20.1-beta.2",
+    "@lancedb/vectordb-linux-x64-gnu": "0.20.1-beta.2",
+    "@lancedb/vectordb-linux-arm64-gnu": "0.20.1-beta.2",
+    "@lancedb/vectordb-win32-x64-msvc": "0.20.1-beta.2"
   }
 }

nodejs/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.20.1-beta.0"
+version = "0.20.1-beta.2"
 license.workspace = true
 description.workspace = true
 repository.workspace = true

nodejs/__test__/arrow.test.ts

@@ -592,14 +592,14 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
         ).rejects.toThrow("column vector was missing");
       });
 
-      it("will provide a nice error if run twice", async function () {
+      it("will skip embedding application if already applied", async function () {
         const records = sampleRecords();
         const table = await convertToTable(records, dummyEmbeddingConfig);
 
         // fromTableToBuffer will try and apply the embeddings again
-        await expect(
-          fromTableToBuffer(table, dummyEmbeddingConfig),
-        ).rejects.toThrow("already existed");
+        // but should skip since the column already has non-null values
+        const result = await fromTableToBuffer(table, dummyEmbeddingConfig);
+        expect(result.byteLength).toBeGreaterThan(0);
       });
     });
 

nodejs/__test__/table.test.ts

@@ -1650,13 +1650,25 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       expect(resultSet.has("fob")).toBe(true);
       expect(resultSet.has("fo")).toBe(true);
       expect(resultSet.has("food")).toBe(true);
+
+      const prefixResults = await table
+        .search(
+          new MatchQuery("foo", "text", { fuzziness: 3, prefixLength: 3 }),
+        )
+        .toArray();
+      expect(prefixResults.length).toBe(2);
+      const resultSet2 = new Set(prefixResults.map((r) => r.text));
+      expect(resultSet2.has("foo")).toBe(true);
+      expect(resultSet2.has("food")).toBe(true);
     });
 
     test("full text search boolean query", async () => {
       const db = await connect(tmpDir.name);
       const data = [
-        { text: "hello world", vector: [0.1, 0.2, 0.3] },
-        { text: "goodbye world", vector: [0.4, 0.5, 0.6] },
+        { text: "The cat and dog are playing" },
+        { text: "The cat is sleeping" },
+        { text: "The dog is barking" },
+        { text: "The dog chases the cat" },
       ];
       const table = await db.createTable("test", data);
       await table.createIndex("text", {
@@ -1666,22 +1678,32 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       const shouldResults = await table
         .search(
           new BooleanQuery([
-            [Occur.Should, new MatchQuery("hello", "text")],
-            [Occur.Should, new MatchQuery("goodbye", "text")],
+            [Occur.Should, new MatchQuery("cat", "text")],
+            [Occur.Should, new MatchQuery("dog", "text")],
           ]),
         )
         .toArray();
-      expect(shouldResults.length).toBe(2);
+      expect(shouldResults.length).toBe(4);
 
       const mustResults = await table
         .search(
           new BooleanQuery([
-            [Occur.Must, new MatchQuery("hello", "text")],
-            [Occur.Must, new MatchQuery("world", "text")],
+            [Occur.Must, new MatchQuery("cat", "text")],
+            [Occur.Must, new MatchQuery("dog", "text")],
           ]),
         )
         .toArray();
-      expect(mustResults.length).toBe(1);
+      expect(mustResults.length).toBe(2);
+
+      const mustNotResults = await table
+        .search(
+          new BooleanQuery([
+            [Occur.Must, new MatchQuery("cat", "text")],
+            [Occur.MustNot, new MatchQuery("dog", "text")],
+          ]),
+        )
+        .toArray();
+      expect(mustNotResults.length).toBe(1);
     });
 
     test.each([

nodejs/lancedb/arrow.ts

@@ -417,7 +417,9 @@ function inferSchema(
         } else {
           const inferredType = inferType(value, path, opts);
           if (inferredType === undefined) {
-            throw new Error(`Failed to infer data type for field ${path.join(".")} at row ${rowI}. \
+            throw new Error(`Failed to infer data type for field ${path.join(
+              ".",
+            )} at row ${rowI}. \
                              Consider providing an explicit schema.`);
           }
           pathTree.set(path, inferredType);
@@ -799,11 +801,17 @@ async function applyEmbeddingsFromMetadata(
         `Cannot apply embedding function because the source column '${functionEntry.sourceColumn}' was not present in the data`,
       );
     }
+
+    // Check if destination column exists and handle accordingly
     if (columns[destColumn] !== undefined) {
-      throw new Error(
-        `Attempt to apply embeddings to table failed because column ${destColumn} already existed`,
-      );
+      const existingColumn = columns[destColumn];
+      // If the column exists but is all null, we can fill it with embeddings
+      if (existingColumn.nullCount !== existingColumn.length) {
+        // Column has non-null values, skip embedding application
+        continue;
+      }
     }
+
     if (table.batches.length > 1) {
       throw new Error(
         "Internal error: `makeArrowTable` unexpectedly created a table with more than one batch",
@@ -903,11 +911,23 @@ async function applyEmbeddings<T>(
       );
     }
   } else {
+    // Check if destination column exists and handle accordingly
     if (Object.prototype.hasOwnProperty.call(newColumns, destColumn)) {
-      throw new Error(
-        `Attempt to apply embeddings to table failed because column ${destColumn} already existed`,
-      );
+      const existingColumn = newColumns[destColumn];
+      // If the column exists but is all null, we can fill it with embeddings
+      if (existingColumn.nullCount !== existingColumn.length) {
+        // Column has non-null values, skip embedding application and return table as-is
+        let newTable = new ArrowTable(newColumns);
+        if (schema != null) {
+          newTable = alignTable(newTable, schema as Schema);
+        }
+        return new ArrowTable(
+          new Schema(newTable.schema.fields, schemaMetadata),
+          newTable.batches,
+        );
+      }
     }
+
     if (table.batches.length > 1) {
       throw new Error(
         "Internal error: `makeArrowTable` unexpectedly created a table with more than one batch",

nodejs/lancedb/query.ts

@@ -812,10 +812,12 @@ export enum Operator {
  *
  * - `Must`: The term must be present in the document.
  * - `Should`: The term should contribute to the document score, but is not required.
+ * - `MustNot`: The term must not be present in the document.
  */
 export enum Occur {
-  Must = "MUST",
   Should = "SHOULD",
+  Must = "MUST",
+  MustNot = "MUST_NOT",
 }
 
 /**
@@ -856,6 +858,7 @@ export class MatchQuery implements FullTextQuery {
    *   - `fuzziness`: The fuzziness level for the query (default is 0).
    *   - `maxExpansions`: The maximum number of terms to consider for fuzzy matching (default is 50).
    *   - `operator`: The logical operator to use for combining terms in the query (default is "OR").
+   *   - `prefixLength`: The number of beginning characters being unchanged for fuzzy matching.
    */
   constructor(
     query: string,
@@ -865,6 +868,7 @@ export class MatchQuery implements FullTextQuery {
       fuzziness?: number;
       maxExpansions?: number;
       operator?: Operator;
+      prefixLength?: number;
     },
   ) {
     let fuzziness = options?.fuzziness;
@@ -878,6 +882,7 @@ export class MatchQuery implements FullTextQuery {
       fuzziness,
       options?.maxExpansions ?? 50,
       options?.operator ?? Operator.Or,
+      options?.prefixLength ?? 0,
     );
   }
 

nodejs/npm/darwin-arm64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-arm64",
-	"version": "0.20.1-beta.0",
+	"version": "0.20.1-beta.2",
 	"os": ["darwin"],
 	"cpu": ["arm64"],
 	"main": "lancedb.darwin-arm64.node",

nodejs/npm/darwin-x64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-x64",
-	"version": "0.20.1-beta.0",
+	"version": "0.20.1-beta.2",
 	"os": ["darwin"],
 	"cpu": ["x64"],
 	"main": "lancedb.darwin-x64.node",

nodejs/npm/linux-arm64-gnu/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-gnu",
-	"version": "0.20.1-beta.0",
+	"version": "0.20.1-beta.2",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-gnu.node",

nodejs/npm/linux-arm64-musl/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-musl",
-	"version": "0.20.1-beta.0",
+	"version": "0.20.1-beta.2",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-musl.node",

nodejs/npm/linux-x64-gnu/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-gnu",
-	"version": "0.20.1-beta.0",
+	"version": "0.20.1-beta.2",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-gnu.node",

nodejs/npm/linux-x64-musl/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-musl",
-	"version": "0.20.1-beta.0",
+	"version": "0.20.1-beta.2",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-musl.node",

nodejs/npm/win32-arm64-msvc/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lancedb/lancedb-win32-arm64-msvc",
-  "version": "0.20.1-beta.0",
+  "version": "0.20.1-beta.2",
   "os": [
     "win32"
   ],

nodejs/npm/win32-x64-msvc/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-win32-x64-msvc",
-	"version": "0.20.1-beta.0",
+	"version": "0.20.1-beta.2",
 	"os": ["win32"],
 	"cpu": ["x64"],
 	"main": "lancedb.win32-x64-msvc.node",

nodejs/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "@lancedb/lancedb",
-  "version": "0.20.1-beta.0",
+  "version": "0.20.1-beta.2",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "@lancedb/lancedb",
-      "version": "0.20.1-beta.0",
+      "version": "0.20.1-beta.2",
       "cpu": [
         "x64",
         "arm64"

nodejs/package.json

@@ -11,7 +11,7 @@
     "ann"
   ],
   "private": false,
-  "version": "0.20.1-beta.0",
+  "version": "0.20.1-beta.2",
   "main": "dist/index.js",
   "exports": {
     ".": "./dist/index.js",

nodejs/src/query.rs

@@ -335,6 +335,7 @@ impl JsFullTextQuery {
         fuzziness: Option<u32>,
         max_expansions: u32,
         operator: String,
+        prefix_length: u32,
     ) -> napi::Result<Self> {
         Ok(Self {
             inner: MatchQuery::new(query)
@@ -347,6 +348,7 @@ impl JsFullTextQuery {
                         napi::Error::from_reason(format!("Invalid operator: {}", e))
                     })?,
                 )
+                .with_prefix_length(prefix_length)
                 .into(),
         })
     }

python/.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.23.1-beta.1"
+current_version = "0.24.0"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

python/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-python"
-version = "0.23.1-beta.1"
+version = "0.24.0"
 edition.workspace = true
 description = "Python bindings for LanceDB"
 license.workspace = true

python/python/lancedb/query.py

@@ -101,8 +101,9 @@ class FullTextOperator(str, Enum):
 
 
 class Occur(str, Enum):
-    MUST = "MUST"
     SHOULD = "SHOULD"
+    MUST = "MUST"
+    MUST_NOT = "MUST_NOT"
 
 
 @pydantic.dataclasses.dataclass
@@ -181,6 +182,9 @@ class MatchQuery(FullTextQuery):
         Can be either `AND` or `OR`.
         If `AND`, all terms in the query must match.
         If `OR`, at least one term in the query must match.
+    prefix_length : int, optional
+        The number of beginning characters being unchanged for fuzzy matching.
+        This is useful to achieve prefix matching.
     """
 
     query: str
@@ -189,6 +193,7 @@ class MatchQuery(FullTextQuery):
     fuzziness: int = pydantic.Field(0, kw_only=True)
     max_expansions: int = pydantic.Field(50, kw_only=True)
     operator: FullTextOperator = pydantic.Field(FullTextOperator.OR, kw_only=True)
+    prefix_length: int = pydantic.Field(0, kw_only=True)
 
     def query_type(self) -> FullTextQueryType:
         return FullTextQueryType.MATCH
@@ -1446,10 +1451,13 @@ class LanceFtsQueryBuilder(LanceQueryBuilder):
 
         query = self._query
         if self._phrase_query:
-            raise NotImplementedError(
-                "Phrase query is not yet supported in Lance FTS. "
-                "Use tantivy-based index instead for now."
-            )
+            if isinstance(query, str):
+                if not query.startswith('"') or not query.endswith('"'):
+                    query = f'"{query}"'
+            elif isinstance(query, FullTextQuery) and not isinstance(
+                query, PhraseQuery
+            ):
+                raise TypeError("Please use PhraseQuery for phrase queries.")
         query = self.to_query_object()
         results = self._table._execute_query(query, timeout=timeout)
         results = results.read_all()

python/python/lancedb/table.py

@@ -827,7 +827,7 @@ class Table(ABC):
         ordering_field_names: Optional[Union[str, List[str]]] = None,
         replace: bool = False,
         writer_heap_size: Optional[int] = 1024 * 1024 * 1024,
-        use_tantivy: bool = True,
+        use_tantivy: bool = False,
         tokenizer_name: Optional[str] = None,
         with_position: bool = False,
         # tokenizer configs:
@@ -864,7 +864,7 @@ class Table(ABC):
             The tokenizer to use for the index. Can be "raw", "default" or the 2 letter
             language code followed by "_stem". So for english it would be "en_stem".
             For available languages see: https://docs.rs/tantivy/latest/tantivy/tokenizer/enum.Language.html
-        use_tantivy: bool, default True
+        use_tantivy: bool, default False
             If True, use the legacy full-text search implementation based on tantivy.
             If False, use the new full-text search implementation based on lance-index.
         with_position: bool, default False
@@ -1970,7 +1970,7 @@ class LanceTable(Table):
         ordering_field_names: Optional[Union[str, List[str]]] = None,
         replace: bool = False,
         writer_heap_size: Optional[int] = 1024 * 1024 * 1024,
-        use_tantivy: bool = True,
+        use_tantivy: bool = False,
         tokenizer_name: Optional[str] = None,
         with_position: bool = False,
         # tokenizer configs:

python/python/tests/docs/test_search.py

@@ -6,7 +6,7 @@ import lancedb
 
 # --8<-- [end:import-lancedb]
 # --8<-- [start:import-numpy]
-from lancedb.query import BoostQuery, MatchQuery
+from lancedb.query import BooleanQuery, BoostQuery, MatchQuery, Occur
 import numpy as np
 import pyarrow as pa
 
@@ -191,6 +191,15 @@ def test_fts_fuzzy_query():
         "food",  # 1 insertion
     }
 
+    results = table.search(
+        MatchQuery("foo", "text", fuzziness=1, prefix_length=3)
+    ).to_pandas()
+    assert len(results) == 2
+    assert set(results["text"].to_list()) == {
+        "foo",
+        "food",
+    }
+
 
 @pytest.mark.skipif(
     os.name == "nt", reason="Need to fix https://github.com/lancedb/lance/issues/3905"
@@ -240,6 +249,60 @@ def test_fts_boost_query():
     )
 
 
+@pytest.mark.skipif(
+    os.name == "nt", reason="Need to fix https://github.com/lancedb/lance/issues/3905"
+)
+def test_fts_boolean_query(tmp_path):
+    uri = tmp_path / "boolean-example"
+    db = lancedb.connect(uri)
+    table = db.create_table(
+        "my_table_fts_boolean",
+        data=[
+            {"text": "The cat and dog are playing"},
+            {"text": "The cat is sleeping"},
+            {"text": "The dog is barking"},
+            {"text": "The dog chases the cat"},
+        ],
+        mode="overwrite",
+    )
+    table.create_fts_index("text", use_tantivy=False, replace=True)
+
+    # SHOULD
+    results = table.search(
+        MatchQuery("cat", "text") | MatchQuery("dog", "text")
+    ).to_pandas()
+    assert len(results) == 4
+    assert set(results["text"].to_list()) == {
+        "The cat and dog are playing",
+        "The cat is sleeping",
+        "The dog is barking",
+        "The dog chases the cat",
+    }
+    # MUST
+    results = table.search(
+        MatchQuery("cat", "text") & MatchQuery("dog", "text")
+    ).to_pandas()
+    assert len(results) == 2
+    assert set(results["text"].to_list()) == {
+        "The cat and dog are playing",
+        "The dog chases the cat",
+    }
+
+    # MUST NOT
+    results = table.search(
+        BooleanQuery(
+            [
+                (Occur.MUST, MatchQuery("cat", "text")),
+                (Occur.MUST_NOT, MatchQuery("dog", "text")),
+            ]
+        )
+    ).to_pandas()
+    assert len(results) == 1
+    assert set(results["text"].to_list()) == {
+        "The cat is sleeping",
+    }
+
+
 @pytest.mark.skipif(
     os.name == "nt", reason="Need to fix https://github.com/lancedb/lance/issues/3905"
 )

python/python/tests/test_s3.py

@@ -245,7 +245,7 @@ def test_s3_dynamodb_sync(s3_bucket: str, commit_table: str, monkeypatch):
         NotImplementedError,
         match="Full-text search is only supported on the local filesystem",
     ):
-        table.create_fts_index("x")
+        table.create_fts_index("x", use_tantivy=True)
 
     # make sure list tables still works
     assert db.table_names() == ["test_ddb_sync"]

python/src/query.rs

@@ -50,6 +50,7 @@ impl FromPyObject<'_> for PyLanceDB<FtsQuery> {
                 let fuzziness = ob.getattr("fuzziness")?.extract()?;
                 let max_expansions = ob.getattr("max_expansions")?.extract()?;
                 let operator = ob.getattr("operator")?.extract::<String>()?;
+                let prefix_length = ob.getattr("prefix_length")?.extract()?;
 
                 Ok(PyLanceDB(
                     MatchQuery::new(query)
@@ -60,6 +61,7 @@ impl FromPyObject<'_> for PyLanceDB<FtsQuery> {
                         .with_operator(Operator::try_from(operator.as_str()).map_err(|e| {
                             PyValueError::new_err(format!("Invalid operator: {}", e))
                         })?)
+                        .with_prefix_length(prefix_length)
                         .into(),
                 ))
             }
@@ -139,7 +141,8 @@ impl<'py> IntoPyObject<'py> for PyLanceDB<FtsQuery> {
                 kwargs.set_item("boost", query.boost)?;
                 kwargs.set_item("fuzziness", query.fuzziness)?;
                 kwargs.set_item("max_expansions", query.max_expansions)?;
-                kwargs.set_item("operator", operator_to_str(query.operator))?;
+                kwargs.set_item::<_, &str>("operator", query.operator.into())?;
+                kwargs.set_item("prefix_length", query.prefix_length)?;
                 namespace
                     .getattr(intern!(py, "MatchQuery"))?
                     .call((query.terms, query.column.unwrap()), Some(&kwargs))
@@ -169,19 +172,25 @@ impl<'py> IntoPyObject<'py> for PyLanceDB<FtsQuery> {
                     .unzip();
                 let kwargs = PyDict::new(py);
                 kwargs.set_item("boosts", boosts)?;
-                kwargs.set_item("operator", operator_to_str(first.operator))?;
+                kwargs.set_item::<_, &str>("operator", first.operator.into())?;
                 namespace
                     .getattr(intern!(py, "MultiMatchQuery"))?
                     .call((first.terms.clone(), columns), Some(&kwargs))
             }
             FtsQuery::Boolean(query) => {
-                let mut queries = Vec::with_capacity(query.must.len() + query.should.len());
-                for q in query.must {
-                    queries.push((occur_to_str(Occur::Must), PyLanceDB(q).into_pyobject(py)?));
-                }
+                let mut queries: Vec<(&str, Bound<'py, PyAny>)> = Vec::with_capacity(
+                    query.should.len() + query.must.len() + query.must_not.len(),
+                );
                 for q in query.should {
-                    queries.push((occur_to_str(Occur::Should), PyLanceDB(q).into_pyobject(py)?));
+                    queries.push((Occur::Should.into(), PyLanceDB(q).into_pyobject(py)?));
                 }
+                for q in query.must {
+                    queries.push((Occur::Must.into(), PyLanceDB(q).into_pyobject(py)?));
+                }
+                for q in query.must_not {
+                    queries.push((Occur::MustNot.into(), PyLanceDB(q).into_pyobject(py)?));
+                }
+
                 namespace
                     .getattr(intern!(py, "BooleanQuery"))?
                     .call1((queries,))
@@ -190,21 +199,6 @@ impl<'py> IntoPyObject<'py> for PyLanceDB<FtsQuery> {
     }
 }
 
-fn operator_to_str(op: Operator) -> &'static str {
-    match op {
-        Operator::And => "AND",
-        Operator::Or => "OR",
-    }
-}
-
-fn occur_to_str(occur: Occur) -> &'static str {
-    match occur {
-        Occur::Must => "MUST",
-        Occur::Should => "SHOULD",
-        Occur::MustNot => "MUST NOT",
-    }
-}
-
 // Python representation of query vector(s)
 #[derive(Clone)]
 pub struct PyQueryVectors(Vec<Arc<dyn Array>>);

rust/ffi/node/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-node"
-version = "0.20.1-beta.0"
+version = "0.20.1-beta.2"
 description = "Serverless, low-latency vector database for AI applications"
 license.workspace = true
 edition.workspace = true

rust/lancedb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb"
-version = "0.20.1-beta.0"
+version = "0.20.1-beta.2"
 edition.workspace = true
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license.workspace = true