Bump version: 0.22.0-beta.12 → 0.22.0

Upstream changelog:
https://github.com/lancedb/lance/releases/tag/v0.26.0

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->

## Summary by CodeRabbit

- **Chores**
- Updated dependency management to use published crate versions for
improved reliability and maintainability.
- Added a temporary workaround for build issues by pinning a specific
version of a dependency.
- **Refactor**
- Improved resource management and concurrency by updating internal
ownership models for object storage components.

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

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.19.0-beta.1"
+current_version = "0.19.0-beta.11"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

.github/workflows/docs.yml

@@ -18,17 +18,24 @@ concurrency:
   group: "pages"
   cancel-in-progress: true
 
+env:
+  # This reduces the disk space needed for the build
+  RUSTFLAGS: "-C debuginfo=0"
+  # according to: https://matklad.github.io/2021/09/04/fast-rust-builds.html
+  # CI builds are faster with incremental disabled.
+  CARGO_INCREMENTAL: "0"
+
 jobs:
   # Single deploy job since we're just deploying
   build:
     environment:
       name: github-pages
       url: ${{ steps.deployment.outputs.page_url }}
-    runs-on: buildjet-8vcpu-ubuntu-2204
+    runs-on: ubuntu-24.04
     steps:
       - name: Checkout
         uses: actions/checkout@v4
-      - name: Install dependecies needed for ubuntu
+      - name: Install dependencies needed for ubuntu
         run: |
           sudo apt install -y protobuf-compiler libssl-dev
           rustup update && rustup default
@@ -38,6 +45,7 @@ jobs:
           python-version: "3.10"
           cache: "pip"
           cache-dependency-path: "docs/requirements.txt"
+      - uses: Swatinem/rust-cache@v2
       - name: Build Python
         working-directory: python
         run: |
@@ -49,7 +57,6 @@ jobs:
           node-version: 20
           cache: 'npm'
           cache-dependency-path: node/package-lock.json
-      - uses: Swatinem/rust-cache@v2
       - name: Install node dependencies
         working-directory: node
         run: |

.github/workflows/npm-publish.yml

@@ -18,6 +18,7 @@ on:
     # This should trigger a dry run (we skip the final publish step)
     paths:
       - .github/workflows/npm-publish.yml
+      - Cargo.toml # Change in dependency frequently breaks builds
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
@@ -130,29 +131,24 @@ jobs:
               set -e &&
               apt-get update &&
               apt-get install -y protobuf-compiler pkg-config
-
-          # TODO: re-enable x64 musl builds. I could not figure out why, but it
-          # consistently made GHA runners non-responsive at the end of build. Example:
-          # https://github.com/lancedb/lancedb/actions/runs/13980431071/job/39144319470?pr=2250
-
-          # - target: x86_64-unknown-linux-musl
-          #   # This one seems to need some extra memory
-          #   host: ubuntu-2404-8x-x64
-          #   # https://github.com/napi-rs/napi-rs/blob/main/alpine.Dockerfile
-          #   docker: ghcr.io/napi-rs/napi-rs/nodejs-rust:lts-alpine
-          #   features: ","
-          #   pre_build: |-
-          #     set -e &&
-          #     apk add protobuf-dev curl &&
-          #     ln -s /usr/lib/gcc/x86_64-alpine-linux-musl/14.2.0/crtbeginS.o /usr/lib/crtbeginS.o &&
-          #     ln -s /usr/lib/libgcc_s.so /usr/lib/libgcc.so
-
+          - target: x86_64-unknown-linux-musl
+            # This one seems to need some extra memory
+            host: ubuntu-2404-8x-x64
+            # https://github.com/napi-rs/napi-rs/blob/main/alpine.Dockerfile
+            docker: ghcr.io/napi-rs/napi-rs/nodejs-rust:lts-alpine
+            features: fp16kernels
+            pre_build: |-
+              set -e &&
+              apk add protobuf-dev curl &&
+              ln -s /usr/lib/gcc/x86_64-alpine-linux-musl/14.2.0/crtbeginS.o /usr/lib/crtbeginS.o &&
+              ln -s /usr/lib/libgcc_s.so /usr/lib/libgcc.so &&
+              CC=gcc &&
+              CXX=g++
           - target: aarch64-unknown-linux-gnu
             host: ubuntu-2404-8x-x64
             # https://github.com/napi-rs/napi-rs/blob/main/debian-aarch64.Dockerfile
             docker: ghcr.io/napi-rs/napi-rs/nodejs-rust:lts-debian-aarch64
-            # TODO: enable fp16kernels after https://github.com/lancedb/lance/pull/3559
-            features: ","
+            features: "fp16kernels"
             pre_build: |-
               set -e &&
               apt-get update &&
@@ -170,8 +166,8 @@ jobs:
               set -e &&
               apk add protobuf-dev &&
               rustup target add aarch64-unknown-linux-musl &&
-              export CC="/aarch64-linux-musl-cross/bin/aarch64-linux-musl-gcc" &&
-              export CXX="/aarch64-linux-musl-cross/bin/aarch64-linux-musl-g++"
+              export CC_aarch64_unknown_linux_musl=aarch64-linux-musl-gcc &&
+              export CXX_aarch64_unknown_linux_musl=aarch64-linux-musl-g++
     name: build - ${{ matrix.settings.target }}
     runs-on: ${{ matrix.settings.host }}
     defaults:
@@ -536,6 +532,8 @@ jobs:
             npm publish $PUBLISH_ARGS $filename
           done
       - name: Deprecate
+        env:
+          NODE_AUTH_TOKEN: ${{ secrets.LANCEDB_NPM_REGISTRY_TOKEN }}
         # We need to deprecate the old package to avoid confusion.
         # Each time we publish a new version, it gets undeprecated.
         run: npm deprecate vectordb "Use @lancedb/lancedb instead."

.github/workflows/pypi-publish.yml

@@ -8,6 +8,7 @@ on:
     # This should trigger a dry run (we skip the final publish step)
     paths:
       - .github/workflows/pypi-publish.yml
+      - Cargo.toml # Change in dependency frequently breaks builds
 
 jobs:
   linux:

.github/workflows/python.yml

@@ -136,9 +136,9 @@ jobs:
       - uses: ./.github/workflows/run_tests
         with:
           integration: true
-      - name: Test without pylance
+      - name: Test without pylance or pandas
         run: |
-          pip uninstall -y pylance
+          pip uninstall -y pylance pandas
           pytest -vv python/tests/test_table.py
       # Make sure wheels are not included in the Rust cache
       - name: Delete wheels

Cargo.toml

@@ -21,16 +21,14 @@ categories = ["database-implementations"]
 rust-version = "1.78.0"
 
 [workspace.dependencies]
-lance = { "version" = "=0.25.3", "features" = [
-    "dynamodb",
-], tag = "v0.25.3-beta.1", git = "https://github.com/lancedb/lance" }
-lance-io = { version = "=0.25.3", tag = "v0.25.3-beta.1", git = "https://github.com/lancedb/lance" }
-lance-index = { version = "=0.25.3", tag = "v0.25.3-beta.1", git = "https://github.com/lancedb/lance" }
-lance-linalg = { version = "=0.25.3", tag = "v0.25.3-beta.1", git = "https://github.com/lancedb/lance" }
-lance-table = { version = "=0.25.3", tag = "v0.25.3-beta.1", git = "https://github.com/lancedb/lance" }
-lance-testing = { version = "=0.25.3", tag = "v0.25.3-beta.1", git = "https://github.com/lancedb/lance" }
-lance-datafusion = { version = "=0.25.3", tag = "v0.25.3-beta.1", git = "https://github.com/lancedb/lance" }
-lance-encoding = { version = "=0.25.3", tag = "v0.25.3-beta.1", git = "https://github.com/lancedb/lance" }
+lance = { "version" = "=0.26.0", "features" = ["dynamodb"] }
+lance-io = "=0.26.0"
+lance-index = "=0.26.0"
+lance-linalg = "=0.26.0"
+lance-table = "=0.26.0"
+lance-testing = "=0.26.0"
+lance-datafusion = "=0.26.0"
+lance-encoding = "=0.26.0"
 # Note that this one does not include pyarrow
 arrow = { version = "54.1", optional = false }
 arrow-array = "54.1"

docs/README.md

@@ -2,7 +2,7 @@
 
 LanceDB docs are deployed to https://lancedb.github.io/lancedb/.
 
-Docs is built and deployed automatically by [Github Actions](.github/workflows/docs.yml)
+Docs is built and deployed automatically by [Github Actions](../.github/workflows/docs.yml)
 whenever a commit is pushed to the `main` branch. So it is possible for the docs to show
 unreleased features.
 

docs/src/guides/storage.md

@@ -342,7 +342,7 @@ For **read and write access**, LanceDB will need a policy such as:
             "Action": [
               "s3:PutObject",
               "s3:GetObject",
-              "s3:DeleteObject",
+              "s3:DeleteObject"
             ],
             "Resource": "arn:aws:s3:::<bucket>/<prefix>/*"
         },
@@ -374,7 +374,7 @@ For **read-only access**, LanceDB will need a policy such as:
         {
             "Effect": "Allow",
             "Action": [
-              "s3:GetObject",
+              "s3:GetObject"
             ],
             "Resource": "arn:aws:s3:::<bucket>/<prefix>/*"
         },

docs/src/guides/tables.md

@@ -765,7 +765,10 @@ This can be used to update zero to all rows depending on how many rows match the
         ];
         const tbl = await db.createTable("my_table", data)
 
-        await tbl.update({vector: [10, 10]}, { where: "x = 2"})
+        await tbl.update({ 
+            values: { vector: [10, 10] },
+            where: "x = 2"
+        });
         ```
 
     === "vectordb (deprecated)"
@@ -784,7 +787,10 @@ This can be used to update zero to all rows depending on how many rows match the
         ];
         const tbl = await db.createTable("my_table", data)
 
-        await tbl.update({ where: "x = 2", values: {vector: [10, 10]} })
+        await tbl.update({ 
+            where: "x = 2", 
+            values: { vector: [10, 10] } 
+        });
         ```
 
 #### Updating using a sql query
@@ -1001,11 +1007,9 @@ In LanceDB OSS, users can set the `read_consistency_interval` parameter on conne
 
 There are three possible settings for `read_consistency_interval`:
 
-1. **Unset**: The database does not check for updates to tables made by other processes. This setting is suitable for applications where the data does not change during the lifetime of the table reference.
-2. **Zero seconds (Strong consistency)**: The database checks for updates on every read. This provides the strongest consistency guarantees, ensuring that all clients see the latest committed data. However, it has the most overhead. This setting is suitable when consistency matters more than having high QPS. For best performance, combine this setting with the storage option `new_table_enable_v2_manifest_paths` set to `true`.
-3. **Custom interval (Eventual consistency, the default)**: The database checks for updates at a custom interval. By default, this is every 5 seconds. This provides eventual consistency, allowing for some lag between write and read operations. Performance wise, this is a middle ground between strong consistency and no consistency check. This setting is suitable for applications where immediate consistency is not critical, but clients should see updated data eventually.
-
-You can always force a synchronization by calling `checkout_latest()` / `checkoutLatest()` on a table.
+1. **Unset (default)**: The database does not check for updates to tables made by other processes. This provides the best query performance, but means that clients may not see the most up-to-date data. This setting is suitable for applications where the data does not change during the lifetime of the table reference.
+2. **Zero seconds (Strong consistency)**: The database checks for updates on every read. This provides the strongest consistency guarantees, ensuring that all clients see the latest committed data. However, it has the most overhead. This setting is suitable when consistency matters more than having high QPS.
+3. **Custom interval (Eventual consistency)**: The database checks for updates at a custom interval, such as every 5 seconds. This provides eventual consistency, allowing for some lag between write and read operations. Performance wise, this is a middle ground between strong consistency and no consistency check. This setting is suitable for applications where immediate consistency is not critical, but clients should see updated data eventually.
 
 !!! tip "Consistency in LanceDB Cloud"
 
@@ -1043,21 +1047,7 @@ You can always force a synchronization by calling `checkout_latest()` / `checkou
         --8<-- "python/python/tests/docs/test_guide_tables.py:table_async_eventual_consistency"
         ```
 
-    For no consistency, use `None`:
-
-    === "Sync API"
-
-        ```python
-        --8<-- "python/python/tests/docs/test_guide_tables.py:table_no_consistency"
-        ```
-
-    === "Async API"
-
-        ```python
-        --8<-- "python/python/tests/docs/test_guide_tables.py:table_async_no_consistency"
-        ```
-
-    To manually check for updates you can use `checkout_latest`:
+    By default, a `Table` will never check for updates from other writers. To manually check for updates you can use `checkout_latest`:
 
     === "Sync API"
 
@@ -1075,25 +1065,15 @@ You can always force a synchronization by calling `checkout_latest()` / `checkou
     To set strong consistency, use `0`:
 
     ```ts
-    --8<-- "nodejs/examples/basic.test.ts:table_strong_consistency"
+    const db = await lancedb.connect({ uri: "./.lancedb", readConsistencyInterval: 0 });
+    const tbl = await db.openTable("my_table");
     ```
 
     For eventual consistency, specify the update interval as seconds:
 
     ```ts
-    --8<-- "nodejs/examples/basic.test.ts:table_eventual_consistency"
-    ```
-
-    For no consistency, use `null`:
-
-    ```ts
-    --8<-- "nodejs/examples/basic.test.ts:table_no_consistency"
-    ```
-
-    To manually check for updates you can use `checkoutLatest`:
-
-    ```ts
-    --8<-- "nodejs/examples/basic.test.ts:table_checkout_latest"
+    const db = await lancedb.connect({ uri: "./.lancedb", readConsistencyInterval: 5 });
+    const tbl = await db.openTable("my_table");
     ```
 
 <!-- Node doesn't yet support the version time travel: https://github.com/lancedb/lancedb/issues/1007

docs/src/js/classes/BoostQuery.md

@@ -22,10 +22,13 @@ including methods to retrieve the query type and convert the query to a dictiona
 new BoostQuery(
    positive,
    negative,
-   negativeBoost): BoostQuery
+   options?): BoostQuery
 ```
 
 Creates an instance of BoostQuery.
+The boost returns documents that match the positive query,
+but penalizes those that match the negative query.
+the penalty is controlled by the `negativeBoost` parameter.
 
 #### Parameters
 
@@ -35,8 +38,11 @@ Creates an instance of BoostQuery.
 * **negative**: [`FullTextQuery`](../interfaces/FullTextQuery.md)
     The negative query that reduces the relevance score.
 
-* **negativeBoost**: `number`
-    The factor by which the negative query reduces the score.
+* **options?**
+    Optional parameters for the boost query.
+    - `negativeBoost`: The boost factor for the negative query (default is 0.0).
+
+* **options.negativeBoost?**: `number`
 
 #### Returns
 
@@ -50,6 +56,8 @@ Creates an instance of BoostQuery.
 queryType(): FullTextQueryType
 ```
 
+The type of the full-text query.
+
 #### Returns
 
 [`FullTextQueryType`](../enumerations/FullTextQueryType.md)
@@ -57,19 +65,3 @@ queryType(): FullTextQueryType
 #### Implementation of
 
 [`FullTextQuery`](../interfaces/FullTextQuery.md).[`queryType`](../interfaces/FullTextQuery.md#querytype)
-
-***
-
-### toDict()
-
-```ts
-toDict(): Record<string, unknown>
-```
-
-#### Returns
-
-`Record`&lt;`string`, `unknown`&gt;
-
-#### Implementation of
-
-[`FullTextQuery`](../interfaces/FullTextQuery.md).[`toDict`](../interfaces/FullTextQuery.md#todict)

docs/src/js/classes/MatchQuery.md

@@ -22,9 +22,7 @@ including methods to retrieve the query type and convert the query to a dictiona
 new MatchQuery(
    query,
    column,
-   boost,
-   fuzziness,
-   maxExpansions): MatchQuery
+   options?): MatchQuery
 ```
 
 Creates an instance of MatchQuery.
@@ -37,14 +35,17 @@ Creates an instance of MatchQuery.
 * **column**: `string`
     The name of the column to search within.
 
-* **boost**: `number` = `1.0`
-    (Optional) The boost factor to influence the relevance score of this query. Default is `1.0`.
+* **options?**
+    Optional parameters for the match query.
+    - `boost`: The boost factor for the query (default is 1.0).
+    - `fuzziness`: The fuzziness level for the query (default is 0).
+    - `maxExpansions`: The maximum number of terms to consider for fuzzy matching (default is 50).
 
-* **fuzziness**: `number` = `0`
-    (Optional) The allowed edit distance for fuzzy matching. Default is `0`.
+* **options.boost?**: `number`
 
-* **maxExpansions**: `number` = `50`
-    (Optional) The maximum number of terms to consider for fuzzy matching. Default is `50`.
+* **options.fuzziness?**: `number`
+
+* **options.maxExpansions?**: `number`
 
 #### Returns
 
@@ -58,6 +59,8 @@ Creates an instance of MatchQuery.
 queryType(): FullTextQueryType
 ```
 
+The type of the full-text query.
+
 #### Returns
 
 [`FullTextQueryType`](../enumerations/FullTextQueryType.md)
@@ -65,19 +68,3 @@ queryType(): FullTextQueryType
 #### Implementation of
 
 [`FullTextQuery`](../interfaces/FullTextQuery.md).[`queryType`](../interfaces/FullTextQuery.md#querytype)
-
-***
-
-### toDict()
-
-```ts
-toDict(): Record<string, unknown>
-```
-
-#### Returns
-
-`Record`&lt;`string`, `unknown`&gt;
-
-#### Implementation of
-
-[`FullTextQuery`](../interfaces/FullTextQuery.md).[`toDict`](../interfaces/FullTextQuery.md#todict)

docs/src/js/classes/MultiMatchQuery.md

@@ -22,7 +22,7 @@ including methods to retrieve the query type and convert the query to a dictiona
 new MultiMatchQuery(
    query,
    columns,
-   boosts): MultiMatchQuery
+   options?): MultiMatchQuery
 ```
 
 Creates an instance of MultiMatchQuery.
@@ -35,10 +35,11 @@ Creates an instance of MultiMatchQuery.
 * **columns**: `string`[]
     An array of column names to search within.
 
-* **boosts**: `number`[] = `...`
-    (Optional) An array of boost factors corresponding to each column. Default is an array of 1.0 for each column.
-    The `boosts` array should have the same length as `columns`. If not provided, all columns will have a default boost of 1.0.
-    If the length of `boosts` is less than `columns`, it will be padded with 1.0s.
+* **options?**
+    Optional parameters for the multi-match query.
+    - `boosts`: An array of boost factors for each column (default is 1.0 for all).
+
+* **options.boosts?**: `number`[]
 
 #### Returns
 
@@ -52,6 +53,8 @@ Creates an instance of MultiMatchQuery.
 queryType(): FullTextQueryType
 ```
 
+The type of the full-text query.
+
 #### Returns
 
 [`FullTextQueryType`](../enumerations/FullTextQueryType.md)
@@ -59,19 +62,3 @@ queryType(): FullTextQueryType
 #### Implementation of
 
 [`FullTextQuery`](../interfaces/FullTextQuery.md).[`queryType`](../interfaces/FullTextQuery.md#querytype)
-
-***
-
-### toDict()
-
-```ts
-toDict(): Record<string, unknown>
-```
-
-#### Returns
-
-`Record`&lt;`string`, `unknown`&gt;
-
-#### Implementation of
-
-[`FullTextQuery`](../interfaces/FullTextQuery.md).[`toDict`](../interfaces/FullTextQuery.md#todict)

docs/src/js/classes/PhraseQuery.md

@@ -44,6 +44,8 @@ Creates an instance of `PhraseQuery`.
 queryType(): FullTextQueryType
 ```
 
+The type of the full-text query.
+
 #### Returns
 
 [`FullTextQueryType`](../enumerations/FullTextQueryType.md)
@@ -51,19 +53,3 @@ queryType(): FullTextQueryType
 #### Implementation of
 
 [`FullTextQuery`](../interfaces/FullTextQuery.md).[`queryType`](../interfaces/FullTextQuery.md#querytype)
-
-***
-
-### toDict()
-
-```ts
-toDict(): Record<string, unknown>
-```
-
-#### Returns
-
-`Record`&lt;`string`, `unknown`&gt;
-
-#### Implementation of
-
-[`FullTextQuery`](../interfaces/FullTextQuery.md).[`toDict`](../interfaces/FullTextQuery.md#todict)

docs/src/js/classes/Table.md

@@ -454,6 +454,28 @@ Modeled after ``VACUUM`` in PostgreSQL.
 
 ***
 
+### prewarmIndex()
+
+```ts
+abstract prewarmIndex(name): Promise<void>
+```
+
+Prewarm an index in the table.
+
+#### Parameters
+
+* **name**: `string`
+    The name of the index.
+    This will load the index into memory.  This may reduce the cold-start time for
+    future queries.  If the index does not fit in the cache then this call may be
+    wasteful.
+
+#### Returns
+
+`Promise`&lt;`void`&gt;
+
+***
+
 ### query()
 
 ```ts
@@ -575,7 +597,7 @@ of the given query
 
 #### Parameters
 
-* **query**: `string` \| [`IntoVector`](../type-aliases/IntoVector.md)
+* **query**: `string` \| [`IntoVector`](../type-aliases/IntoVector.md) \| [`FullTextQuery`](../interfaces/FullTextQuery.md)
     the query, a vector or string
 
 * **queryType?**: `string`
@@ -731,3 +753,26 @@ Retrieve the version of the table
 #### Returns
 
 `Promise`&lt;`number`&gt;
+
+***
+
+### waitForIndex()
+
+```ts
+abstract waitForIndex(indexNames, timeoutSeconds): Promise<void>
+```
+
+Waits for asynchronous indexing to complete on the table.
+
+#### Parameters
+
+* **indexNames**: `string`[]
+    The name of the indices to wait for
+
+* **timeoutSeconds**: `number`
+    The number of seconds to wait before timing out
+    This will raise an error if the indices are not created and fully indexed within the timeout.
+
+#### Returns
+
+`Promise`&lt;`void`&gt;

docs/src/js/interfaces/ConnectionOptions.md

@@ -44,7 +44,7 @@ for testing purposes.
 ### readConsistencyInterval?
 
 ```ts
-optional readConsistencyInterval: null | number;
+optional readConsistencyInterval: number;
 ```
 
 (For LanceDB OSS only): The interval, in seconds, at which to check for

docs/src/js/interfaces/FullTextQuery.md

@@ -18,18 +18,8 @@ including methods to retrieve the query type and convert the query to a dictiona
 queryType(): FullTextQueryType
 ```
 
+The type of the full-text query.
+
 #### Returns
 
 [`FullTextQueryType`](../enumerations/FullTextQueryType.md)
-
-***
-
-### toDict()
-
-```ts
-toDict(): Record<string, unknown>
-```
-
-#### Returns
-
-`Record`&lt;`string`, `unknown`&gt;

docs/src/js/interfaces/IndexOptions.md

@@ -39,3 +39,11 @@ and the same name, then an error will be returned.  This is true even if
 that index is out of date.
 
 The default is true
+
+***
+
+### waitTimeoutSeconds?
+
+```ts
+optional waitTimeoutSeconds: number;
+```

docs/src/js/interfaces/QueryExecutionOptions.md

@@ -20,3 +20,13 @@ The maximum number of rows to return in a single batch
 
 Batches may have fewer rows if the underlying data is stored
 in smaller chunks.
+
+***
+
+### timeoutMs?
+
+```ts
+optional timeoutMs: number;
+```
+
+Timeout for query execution in milliseconds

docs/src/troubleshooting.md

@@ -11,7 +11,6 @@ likely that someone who knows the answer will see your question.
 ## Common issues
 
 * Multiprocessing with `fork` is not supported. You should use `spawn` instead.
-* Data returned by queries may not reflect the most recent writes, depending on configuration. LanceDB uses eventual consistency by default. See [consistency](/docs/src/guides/tables.md#consistency) for more information.
 
 ## Enabling logging
 

java/core/pom.xml

@@ -8,7 +8,7 @@
     <parent>
         <groupId>com.lancedb</groupId>
         <artifactId>lancedb-parent</artifactId>
-        <version>0.19.0-beta.1</version>
+        <version>0.19.0-beta.11</version>
         <relativePath>../pom.xml</relativePath>
     </parent>
 

java/pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-parent</artifactId>
-    <version>0.19.0-beta.1</version>
+    <version>0.19.0-beta.11</version>
     <packaging>pom</packaging>
 
     <name>LanceDB Parent</name>

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.19.0-beta.1",
+  "version": "0.19.0-beta.11",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.19.0-beta.1",
+      "version": "0.19.0-beta.11",
       "cpu": [
         "x64",
         "arm64"
@@ -52,11 +52,11 @@
         "uuid": "^9.0.0"
       },
       "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.19.0-beta.1",
-        "@lancedb/vectordb-darwin-x64": "0.19.0-beta.1",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.19.0-beta.1",
-        "@lancedb/vectordb-linux-x64-gnu": "0.19.0-beta.1",
-        "@lancedb/vectordb-win32-x64-msvc": "0.19.0-beta.1"
+        "@lancedb/vectordb-darwin-arm64": "0.19.0-beta.11",
+        "@lancedb/vectordb-darwin-x64": "0.19.0-beta.11",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.19.0-beta.11",
+        "@lancedb/vectordb-linux-x64-gnu": "0.19.0-beta.11",
+        "@lancedb/vectordb-win32-x64-msvc": "0.19.0-beta.11"
       },
       "peerDependencies": {
         "@apache-arrow/ts": "^14.0.2",
@@ -326,6 +326,71 @@
         "@jridgewell/sourcemap-codec": "^1.4.10"
       }
     },
+    "node_modules/@lancedb/vectordb-darwin-arm64": {
+      "version": "0.19.0-beta.11",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.19.0-beta.11.tgz",
+      "integrity": "sha512-fLefGJYdlIRIjrJj8MU1r5Zix5LpKktpCYilA7tZrfvBWkubGceJ+U6RPsWz7VGBfWcETo3g5CBooUPhbtSMlQ==",
+      "cpu": [
+        "arm64"
+      ],
+      "license": "Apache-2.0",
+      "optional": true,
+      "os": [
+        "darwin"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-darwin-x64": {
+      "version": "0.19.0-beta.11",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.19.0-beta.11.tgz",
+      "integrity": "sha512-FkCa1TbPLDXAGhlRI4tafcltzApCsyvgi+I+kX07u5DKPNQVALpQ3R6X6GLlbiFsAFBdyv9t2fqQ9DlgjJIZpA==",
+      "cpu": [
+        "x64"
+      ],
+      "license": "Apache-2.0",
+      "optional": true,
+      "os": [
+        "darwin"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-linux-arm64-gnu": {
+      "version": "0.19.0-beta.11",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.19.0-beta.11.tgz",
+      "integrity": "sha512-iZkL/01HNUNQ8pGK0+hoNyrM7P1YtShsyIQVzJMfo41SAofCBf9qvi9YaYyd49sDb+dQXeRn1+cfaJ9siz1OHw==",
+      "cpu": [
+        "arm64"
+      ],
+      "license": "Apache-2.0",
+      "optional": true,
+      "os": [
+        "linux"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-linux-x64-gnu": {
+      "version": "0.19.0-beta.11",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.19.0-beta.11.tgz",
+      "integrity": "sha512-MdKRHxe2tRQqmExNLv3f6Wvx1mEi98eFtD0ysm4tNrQdaS1MJbTp+DUehrRKkfDWsooalHkIi9d02BVw5qseUQ==",
+      "cpu": [
+        "x64"
+      ],
+      "license": "Apache-2.0",
+      "optional": true,
+      "os": [
+        "linux"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-win32-x64-msvc": {
+      "version": "0.19.0-beta.11",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.19.0-beta.11.tgz",
+      "integrity": "sha512-KWy+t9jr0feJAW9NkmM/w9kfdpp78+7mkeh9lb0g3xI3OdYU1yizNqFjbIQqJf7/L4sou4wmOjAC+FcP8qCtzg==",
+      "cpu": [
+        "x64"
+      ],
+      "license": "Apache-2.0",
+      "optional": true,
+      "os": [
+        "win32"
+      ]
+    },
     "node_modules/@neon-rs/cli": {
       "version": "0.0.160",
       "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.160.tgz",
@@ -1119,9 +1184,10 @@
       }
     },
     "node_modules/axios": {
-      "version": "1.7.7",
-      "resolved": "https://registry.npmjs.org/axios/-/axios-1.7.7.tgz",
-      "integrity": "sha512-S4kL7XrjgBmvdGut0sN3yJxqYzrDOnivkBiN0OFs6hLiUam3UPvswUo0kqGyhqUZGEOytHyumEdXsAkgCOUf3Q==",
+      "version": "1.8.4",
+      "resolved": "https://registry.npmjs.org/axios/-/axios-1.8.4.tgz",
+      "integrity": "sha512-eBSYY4Y68NNlHbHBMdeDmKNtDgXWhQsJcGqzO3iLUM0GraQFSS9cVgPX5I9b3lbdFKyYoAEGAZF1DwhTaljNAw==",
+      "license": "MIT",
       "dependencies": {
         "follow-redirects": "^1.15.6",
         "form-data": "^4.0.0",

node/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vectordb",
-  "version": "0.19.0-beta.1",
+  "version": "0.19.0-beta.11",
   "description": " Serverless, low-latency vector database for AI applications",
   "private": false,
   "main": "dist/index.js",
@@ -89,10 +89,10 @@
     }
   },
   "optionalDependencies": {
-    "@lancedb/vectordb-darwin-x64": "0.19.0-beta.1",
-    "@lancedb/vectordb-darwin-arm64": "0.19.0-beta.1",
-    "@lancedb/vectordb-linux-x64-gnu": "0.19.0-beta.1",
-    "@lancedb/vectordb-linux-arm64-gnu": "0.19.0-beta.1",
-    "@lancedb/vectordb-win32-x64-msvc": "0.19.0-beta.1"
+    "@lancedb/vectordb-darwin-x64": "0.19.0-beta.11",
+    "@lancedb/vectordb-darwin-arm64": "0.19.0-beta.11",
+    "@lancedb/vectordb-linux-x64-gnu": "0.19.0-beta.11",
+    "@lancedb/vectordb-linux-arm64-gnu": "0.19.0-beta.11",
+    "@lancedb/vectordb-win32-x64-msvc": "0.19.0-beta.11"
   }
 }

node/src/integration_test/test.ts

@@ -110,7 +110,7 @@ describe('LanceDB Mirrored Store Integration test', function () {
 
       fs.readdir(path.join(mirroredPath, 'data'), { withFileTypes: true }, (err, files) => {
         if (err != null) throw err
-        assert.equal(files.length, 1, `Found files: ${files.map(f => f.name)}`)
+        assert.equal(files.length, 1)
         assert.isTrue(files[0].name.endsWith('.lance'))
       })
 

nodejs/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.19.0-beta.1"
+version = "0.19.0-beta.11"
 license.workspace = true
 description.workspace = true
 repository.workspace = true
@@ -28,6 +28,9 @@ napi-derive = "2.16.4"
 lzma-sys = { version = "*", features = ["static"] }
 log.workspace = true
 
+# Workaround for build failure until we can fix it.
+aws-lc-sys = "=0.28.0"
+
 [build-dependencies]
 napi-build = "2.1"
 

nodejs/__test__/connection.test.ts

@@ -17,7 +17,7 @@ describe("when connecting", () => {
   it("should connect", async () => {
     const db = await connect(tmpDir.name);
     expect(db.display()).toBe(
-      `ListingDatabase(uri=${tmpDir.name}, read_consistency_interval=5s)`,
+      `ListingDatabase(uri=${tmpDir.name}, read_consistency_interval=None)`,
     );
   });
 

nodejs/__test__/table.test.ts

@@ -10,7 +10,7 @@ import * as arrow16 from "apache-arrow-16";
 import * as arrow17 from "apache-arrow-17";
 import * as arrow18 from "apache-arrow-18";
 
-import { Table, connect } from "../lancedb";
+import { MatchQuery, PhraseQuery, Table, connect } from "../lancedb";
 import {
   Table as ArrowTable,
   Field,
@@ -33,6 +33,7 @@ import {
   register,
 } from "../lancedb/embedding";
 import { Index } from "../lancedb/indices";
+import { instanceOfFullTextQuery } from "../lancedb/query";
 
 describe.each([arrow15, arrow16, arrow17, arrow18])(
   "Given a table",
@@ -58,7 +59,7 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
 
     it("be displayable", async () => {
       expect(table.display()).toMatch(
-        /NativeTable\(some_table, uri=.*, read_consistency_interval=5s\)/,
+        /NativeTable\(some_table, uri=.*, read_consistency_interval=None\)/,
       );
       table.close();
       expect(table.display()).toBe("ClosedTable(some_table)");
@@ -506,6 +507,15 @@ describe("When creating an index", () => {
     expect(indices2.length).toBe(0);
   });
 
+  it("should wait for index readiness", async () => {
+    // Create an index and then wait for it to be ready
+    await tbl.createIndex("vec");
+    const indices = await tbl.listIndices();
+    expect(indices.length).toBeGreaterThan(0);
+    const idxName = indices[0].name;
+    await expect(tbl.waitForIndex([idxName], 5)).resolves.toBeUndefined();
+  });
+
   it("should search with distance range", async () => {
     await tbl.createIndex("vec");
 
@@ -823,6 +833,7 @@ describe("When creating an index", () => {
     // Only build index over v1
     await tbl.createIndex("vec", {
       config: Index.ivfPq({ numPartitions: 2, numSubVectors: 2 }),
+      waitTimeoutSeconds: 30,
     });
 
     const rst = await tbl
@@ -867,6 +878,44 @@ describe("When creating an index", () => {
   });
 });
 
+describe("When querying a table", () => {
+  let tmpDir: tmp.DirResult;
+  beforeEach(() => {
+    tmpDir = tmp.dirSync({ unsafeCleanup: true });
+  });
+  afterEach(() => tmpDir.removeCallback());
+
+  it("should throw an error when timeout is reached", async () => {
+    const db = await connect(tmpDir.name);
+    const data = makeArrowTable([
+      { text: "a", vector: [0.1, 0.2] },
+      { text: "b", vector: [0.3, 0.4] },
+    ]);
+    const table = await db.createTable("test", data);
+    await table.createIndex("text", { config: Index.fts() });
+
+    await expect(
+      table.query().where("text != 'a'").toArray({ timeoutMs: 0 }),
+    ).rejects.toThrow("Query timeout");
+
+    await expect(
+      table.query().nearestTo([0.0, 0.0]).toArrow({ timeoutMs: 0 }),
+    ).rejects.toThrow("Query timeout");
+
+    await expect(
+      table.search("a", "fts").toArray({ timeoutMs: 0 }),
+    ).rejects.toThrow("Query timeout");
+
+    await expect(
+      table
+        .query()
+        .nearestToText("a")
+        .nearestTo([0.0, 0.0])
+        .toArrow({ timeoutMs: 0 }),
+    ).rejects.toThrow("Query timeout");
+  });
+});
+
 describe("Read consistency interval", () => {
   let tmpDir: tmp.DirResult;
   beforeEach(() => {
@@ -1264,6 +1313,56 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
 
       const results = await table.search("hello").toArray();
       expect(results[0].text).toBe(data[0].text);
+
+      const query = new MatchQuery("goodbye", "text");
+      expect(instanceOfFullTextQuery(query)).toBe(true);
+      const results2 = await table
+        .search(new MatchQuery("goodbye", "text"))
+        .toArray();
+      expect(results2[0].text).toBe(data[1].text);
+    });
+
+    test("prewarm full text search index", async () => {
+      const db = await connect(tmpDir.name);
+      const data = [
+        { text: ["lance database", "the", "search"], vector: [0.1, 0.2, 0.3] },
+        { text: ["lance database"], vector: [0.4, 0.5, 0.6] },
+        { text: ["lance", "search"], vector: [0.7, 0.8, 0.9] },
+        { text: ["database", "search"], vector: [1.0, 1.1, 1.2] },
+        { text: ["unrelated", "doc"], vector: [1.3, 1.4, 1.5] },
+      ];
+      const table = await db.createTable("test", data);
+      await table.createIndex("text", {
+        config: Index.fts(),
+      });
+
+      // For the moment, we just confirm we can call prewarmIndex without error
+      // and still search it afterwards
+      await table.prewarmIndex("text_idx");
+
+      const results = await table.search("lance").toArray();
+      expect(results.length).toBe(3);
+    });
+
+    test("full text index on list", async () => {
+      const db = await connect(tmpDir.name);
+      const data = [
+        { text: ["lance database", "the", "search"], vector: [0.1, 0.2, 0.3] },
+        { text: ["lance database"], vector: [0.4, 0.5, 0.6] },
+        { text: ["lance", "search"], vector: [0.7, 0.8, 0.9] },
+        { text: ["database", "search"], vector: [1.0, 1.1, 1.2] },
+        { text: ["unrelated", "doc"], vector: [1.3, 1.4, 1.5] },
+      ];
+      const table = await db.createTable("test", data);
+      await table.createIndex("text", {
+        config: Index.fts(),
+      });
+
+      const results = await table.search("lance").toArray();
+      expect(results.length).toBe(3);
+
+      const results2 = await table.search('"lance database"').toArray();
+      expect(results2.length).toBe(2);
     });
 
     test("full text search without positions", async () => {
@@ -1316,6 +1415,43 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       expect(results.length).toBe(2);
       const phraseResults = await table.search('"hello world"').toArray();
       expect(phraseResults.length).toBe(1);
+      const phraseResults2 = await table
+        .search(new PhraseQuery("hello world", "text"))
+        .toArray();
+      expect(phraseResults2.length).toBe(1);
+    });
+
+    test("full text search fuzzy query", async () => {
+      const db = await connect(tmpDir.name);
+      const data = [
+        { text: "fa", vector: [0.1, 0.2, 0.3] },
+        { text: "fo", vector: [0.4, 0.5, 0.6] },
+        { text: "fob", vector: [0.4, 0.5, 0.6] },
+        { text: "focus", vector: [0.4, 0.5, 0.6] },
+        { text: "foo", vector: [0.4, 0.5, 0.6] },
+        { text: "food", vector: [0.4, 0.5, 0.6] },
+        { text: "foul", vector: [0.4, 0.5, 0.6] },
+      ];
+      const table = await db.createTable("test", data);
+      await table.createIndex("text", {
+        config: Index.fts(),
+      });
+
+      const results = await table
+        .search(new MatchQuery("foo", "text"))
+        .toArray();
+      expect(results.length).toBe(1);
+      expect(results[0].text).toBe("foo");
+
+      const fuzzyResults = await table
+        .search(new MatchQuery("foo", "text", { fuzziness: 1 }))
+        .toArray();
+      expect(fuzzyResults.length).toBe(4);
+      const resultSet = new Set(fuzzyResults.map((r) => r.text));
+      expect(resultSet.has("foo")).toBe(true);
+      expect(resultSet.has("fob")).toBe(true);
+      expect(resultSet.has("fo")).toBe(true);
+      expect(resultSet.has("food")).toBe(true);
     });
 
     test.each([

nodejs/examples/basic.test.ts

@@ -202,35 +202,5 @@ test("basic table examples", async () => {
       // --8<-- [end:create_f16_table]
       await db.dropTable("f16_tbl");
     }
-    const uri = databaseDir;
-    await db.createTable("my_table", [{ id: 1 }, { id: 2 }]);
-    {
-      // --8<-- [start:table_strong_consistency]
-      const db = await lancedb.connect({ uri, readConsistencyInterval: 0 });
-      const tbl = await db.openTable("my_table");
-      // --8<-- [end:table_strong_consistency]
-    }
-    {
-      // --8<-- [start:table_eventual_consistency]
-      const db = await lancedb.connect({ uri, readConsistencyInterval: 5 });
-      const tbl = await db.openTable("my_table");
-      // --8<-- [end:table_eventual_consistency]
-    }
-    {
-      // --8<-- [start:table_no_consistency]
-      const db = await lancedb.connect({ uri, readConsistencyInterval: null });
-      const tbl = await db.openTable("my_table");
-      // --8<-- [end:table_no_consistency]
-    }
-    {
-      // --8<-- [start:table_checkout_latest]
-      const tbl = await db.openTable("my_table");
-
-      // (Other writes happen to test_table_async from another process)
-
-      // Check for updates
-      tbl.checkoutLatest();
-      // --8<-- [end:table_checkout_latest]
-    }
   });
 });

nodejs/lancedb/indices.ts

@@ -681,4 +681,6 @@ export interface IndexOptions {
    * The default is true
    */
   replace?: boolean;
+
+  waitTimeoutSeconds?: number;
 }

nodejs/lancedb/query.ts

@@ -11,6 +11,7 @@ import {
 } from "./arrow";
 import { type IvfPqOptions } from "./indices";
 import {
+  JsFullTextQuery,
   RecordBatchIterator as NativeBatchIterator,
   Query as NativeQuery,
   Table as NativeTable,
@@ -63,7 +64,7 @@ class RecordBatchIterable<
   // biome-ignore lint/suspicious/noExplicitAny: skip
   [Symbol.asyncIterator](): AsyncIterator<RecordBatch<any>, any, undefined> {
     return new RecordBatchIterator(
-      this.inner.execute(this.options?.maxBatchLength),
+      this.inner.execute(this.options?.maxBatchLength, this.options?.timeoutMs),
     );
   }
 }
@@ -79,6 +80,11 @@ export interface QueryExecutionOptions {
    * in smaller chunks.
    */
   maxBatchLength?: number;
+
+  /**
+   * Timeout for query execution in milliseconds
+   */
+  timeoutMs?: number;
 }
 
 /**
@@ -172,9 +178,7 @@ export class QueryBase<NativeQueryType extends NativeQuery | NativeVectorQuery>
           columns: columns,
         });
       } else {
-        // If query is a FullTextQuery object, convert it to a dict
-        const queryObj = query.toDict();
-        inner.fullTextSearch(queryObj);
+        inner.fullTextSearch({ query: query.inner });
       }
     });
     return this;
@@ -283,9 +287,11 @@ export class QueryBase<NativeQueryType extends NativeQuery | NativeVectorQuery>
     options?: Partial<QueryExecutionOptions>,
   ): Promise<NativeBatchIterator> {
     if (this.inner instanceof Promise) {
-      return this.inner.then((inner) => inner.execute(options?.maxBatchLength));
+      return this.inner.then((inner) =>
+        inner.execute(options?.maxBatchLength, options?.timeoutMs),
+      );
     } else {
-      return this.inner.execute(options?.maxBatchLength);
+      return this.inner.execute(options?.maxBatchLength, options?.timeoutMs);
     }
   }
 
@@ -736,8 +742,7 @@ export class Query extends QueryBase<NativeQuery> {
           columns: columns,
         });
       } else {
-        const queryObj = query.toDict();
-        inner.fullTextSearch(queryObj);
+        inner.fullTextSearch({ query: query.inner });
       }
     });
     return this;
@@ -765,130 +770,141 @@ export enum FullTextQueryType {
  * including methods to retrieve the query type and convert the query to a dictionary format.
  */
 export interface FullTextQuery {
+  /**
+   * Returns the inner query object.
+   * This is the underlying query object used by the database engine.
+   * @ignore
+   */
+  inner: JsFullTextQuery;
+
+  /**
+   * The type of the full-text query.
+   */
   queryType(): FullTextQueryType;
-  toDict(): Record<string, unknown>;
+}
+
+// biome-ignore lint/suspicious/noExplicitAny: we want any here
+export function instanceOfFullTextQuery(obj: any): obj is FullTextQuery {
+  return obj != null && obj.inner instanceof JsFullTextQuery;
 }
 
 export class MatchQuery implements FullTextQuery {
+  /** @ignore */
+  public readonly inner: JsFullTextQuery;
   /**
    * Creates an instance of MatchQuery.
    *
    * @param query - The text query to search for.
    * @param column - The name of the column to search within.
-   * @param boost - (Optional) The boost factor to influence the relevance score of this query. Default is `1.0`.
-   * @param fuzziness - (Optional) The allowed edit distance for fuzzy matching. Default is `0`.
-   * @param maxExpansions - (Optional) The maximum number of terms to consider for fuzzy matching. Default is `50`.
+   * @param options - Optional parameters for the match query.
+   *   - `boost`: The boost factor for the query (default is 1.0).
+   *   - `fuzziness`: The fuzziness level for the query (default is 0).
+   *   - `maxExpansions`: The maximum number of terms to consider for fuzzy matching (default is 50).
    */
   constructor(
-    private query: string,
-    private column: string,
-    private boost: number = 1.0,
-    private fuzziness: number = 0,
-    private maxExpansions: number = 50,
-  ) {}
+    query: string,
+    column: string,
+    options?: {
+      boost?: number;
+      fuzziness?: number;
+      maxExpansions?: number;
+    },
+  ) {
+    let fuzziness = options?.fuzziness;
+    if (fuzziness === undefined) {
+      fuzziness = 0;
+    }
+    this.inner = JsFullTextQuery.matchQuery(
+      query,
+      column,
+      options?.boost ?? 1.0,
+      fuzziness,
+      options?.maxExpansions ?? 50,
+    );
+  }
 
   queryType(): FullTextQueryType {
     return FullTextQueryType.Match;
   }
-
-  toDict(): Record<string, unknown> {
-    return {
-      [this.queryType()]: {
-        [this.column]: {
-          query: this.query,
-          boost: this.boost,
-          fuzziness: this.fuzziness,
-          // biome-ignore lint/style/useNamingConvention: use underscore for consistency with the other APIs
-          max_expansions: this.maxExpansions,
-        },
-      },
-    };
-  }
 }
 
 export class PhraseQuery implements FullTextQuery {
+  /** @ignore */
+  public readonly inner: JsFullTextQuery;
   /**
    * Creates an instance of `PhraseQuery`.
    *
    * @param query - The phrase to search for in the specified column.
    * @param column - The name of the column to search within.
    */
-  constructor(
-    private query: string,
-    private column: string,
-  ) {}
+  constructor(query: string, column: string) {
+    this.inner = JsFullTextQuery.phraseQuery(query, column);
+  }
 
   queryType(): FullTextQueryType {
     return FullTextQueryType.MatchPhrase;
   }
-
-  toDict(): Record<string, unknown> {
-    return {
-      [this.queryType()]: {
-        [this.column]: this.query,
-      },
-    };
-  }
 }
 
 export class BoostQuery implements FullTextQuery {
+  /** @ignore */
+  public readonly inner: JsFullTextQuery;
   /**
    * Creates an instance of BoostQuery.
+   * The boost returns documents that match the positive query,
+   * but penalizes those that match the negative query.
+   * the penalty is controlled by the `negativeBoost` parameter.
    *
    * @param positive - The positive query that boosts the relevance score.
    * @param negative - The negative query that reduces the relevance score.
-   * @param negativeBoost - The factor by which the negative query reduces the score.
+   * @param options - Optional parameters for the boost query.
+   *  - `negativeBoost`: The boost factor for the negative query (default is 0.0).
    */
   constructor(
-    private positive: FullTextQuery,
-    private negative: FullTextQuery,
-    private negativeBoost: number,
-  ) {}
+    positive: FullTextQuery,
+    negative: FullTextQuery,
+    options?: {
+      negativeBoost?: number;
+    },
+  ) {
+    this.inner = JsFullTextQuery.boostQuery(
+      positive.inner,
+      negative.inner,
+      options?.negativeBoost,
+    );
+  }
 
   queryType(): FullTextQueryType {
     return FullTextQueryType.Boost;
   }
-
-  toDict(): Record<string, unknown> {
-    return {
-      [this.queryType()]: {
-        positive: this.positive.toDict(),
-        negative: this.negative.toDict(),
-        // biome-ignore lint/style/useNamingConvention: use underscore for consistency with the other APIs
-        negative_boost: this.negativeBoost,
-      },
-    };
-  }
 }
 
 export class MultiMatchQuery implements FullTextQuery {
+  /** @ignore */
+  public readonly inner: JsFullTextQuery;
   /**
    * Creates an instance of MultiMatchQuery.
    *
    * @param query - The text query to search for across multiple columns.
    * @param columns - An array of column names to search within.
-   * @param boosts - (Optional) An array of boost factors corresponding to each column. Default is an array of 1.0 for each column.
-   *
-   * The `boosts` array should have the same length as `columns`. If not provided, all columns will have a default boost of 1.0.
-   * If the length of `boosts` is less than `columns`, it will be padded with 1.0s.
+   * @param options - Optional parameters for the multi-match query.
+   *  - `boosts`: An array of boost factors for each column (default is 1.0 for all).
    */
   constructor(
-    private query: string,
-    private columns: string[],
-    private boosts: number[] = columns.map(() => 1.0),
-  ) {}
+    query: string,
+    columns: string[],
+    options?: {
+      boosts?: number[];
+    },
+  ) {
+    this.inner = JsFullTextQuery.multiMatchQuery(
+      query,
+      columns,
+      options?.boosts,
+    );
+  }
 
   queryType(): FullTextQueryType {
     return FullTextQueryType.MultiMatch;
   }
-
-  toDict(): Record<string, unknown> {
-    return {
-      [this.queryType()]: {
-        query: this.query,
-        columns: this.columns,
-        boost: this.boosts,
-      },
-    };
-  }
 }

nodejs/lancedb/table.ts

@@ -22,7 +22,12 @@ import {
   OptimizeStats,
   Table as _NativeTable,
 } from "./native";
-import { Query, VectorQuery } from "./query";
+import {
+  FullTextQuery,
+  Query,
+  VectorQuery,
+  instanceOfFullTextQuery,
+} from "./query";
 import { sanitizeType } from "./sanitize";
 import { IntoSql, toSQL } from "./util";
 export { IndexConfig } from "./native";
@@ -230,6 +235,30 @@ export abstract class Table {
    */
   abstract dropIndex(name: string): Promise<void>;
 
+  /**
+   * Prewarm an index in the table.
+   *
+   * @param name The name of the index.
+   *
+   * This will load the index into memory.  This may reduce the cold-start time for
+   * future queries.  If the index does not fit in the cache then this call may be
+   * wasteful.
+   */
+  abstract prewarmIndex(name: string): Promise<void>;
+
+  /**
+   * Waits for asynchronous indexing to complete on the table.
+   *
+   * @param indexNames The name of the indices to wait for
+   * @param timeoutSeconds The number of seconds to wait before timing out
+   *
+   * This will raise an error if the indices are not created and fully indexed within the timeout.
+   */
+  abstract waitForIndex(
+    indexNames: string[],
+    timeoutSeconds: number,
+  ): Promise<void>;
+
   /**
    * Create a {@link Query} Builder.
    *
@@ -294,7 +323,7 @@ export abstract class Table {
    * if the query is a string and no embedding function is defined, it will be treated as a full text search query
    */
   abstract search(
-    query: string | IntoVector,
+    query: string | IntoVector | FullTextQuery,
     queryType?: string,
     ftsColumns?: string | string[],
   ): VectorQuery | Query;
@@ -553,23 +582,39 @@ export class LocalTable extends Table {
     // Bit of a hack to get around the fact that TS has no package-scope.
     // biome-ignore lint/suspicious/noExplicitAny: skip
     const nativeIndex = (options?.config as any)?.inner;
-    await this.inner.createIndex(nativeIndex, column, options?.replace);
+    await this.inner.createIndex(
+      nativeIndex,
+      column,
+      options?.replace,
+      options?.waitTimeoutSeconds,
+    );
   }
 
   async dropIndex(name: string): Promise<void> {
     await this.inner.dropIndex(name);
   }
 
+  async prewarmIndex(name: string): Promise<void> {
+    await this.inner.prewarmIndex(name);
+  }
+
+  async waitForIndex(
+    indexNames: string[],
+    timeoutSeconds: number,
+  ): Promise<void> {
+    await this.inner.waitForIndex(indexNames, timeoutSeconds);
+  }
+
   query(): Query {
     return new Query(this.inner);
   }
 
   search(
-    query: string | IntoVector,
+    query: string | IntoVector | FullTextQuery,
     queryType: string = "auto",
     ftsColumns?: string | string[],
   ): VectorQuery | Query {
-    if (typeof query !== "string") {
+    if (typeof query !== "string" && !instanceOfFullTextQuery(query)) {
       if (queryType === "fts") {
         throw new Error("Cannot perform full text search on a vector query");
       }
@@ -585,7 +630,10 @@ export class LocalTable extends Table {
 
     // The query type is auto or vector
     // fall back to full text search if no embedding functions are defined and the query is a string
-    if (queryType === "auto" && getRegistry().length() === 0) {
+    if (
+      queryType === "auto" &&
+      (getRegistry().length() === 0 || instanceOfFullTextQuery(query))
+    ) {
       return this.query().fullTextSearch(query, {
         columns: ftsColumns,
       });

nodejs/npm/darwin-arm64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-arm64",
-	"version": "0.19.0-beta.1",
+	"version": "0.19.0-beta.11",
 	"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": "0.19.0-beta.1",
+	"version": "0.19.0-beta.11",
 	"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": "0.19.0-beta.1",
+	"version": "0.19.0-beta.11",
 	"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": "0.19.0-beta.1",
+	"version": "0.19.0-beta.11",
 	"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": "0.19.0-beta.1",
+	"version": "0.19.0-beta.11",
 	"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": "0.19.0-beta.1",
+	"version": "0.19.0-beta.11",
 	"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": "0.19.0-beta.1",
+  "version": "0.19.0-beta.11",
   "os": [
     "win32"
   ],

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-win32-x64-msvc",
-	"version": "0.19.0-beta.1",
+	"version": "0.19.0-beta.11",
 	"os": ["win32"],
 	"cpu": ["x64"],
 	"main": "lancedb.win32-x64-msvc.node",

nodejs/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "@lancedb/lancedb",
-  "version": "0.19.0-beta.1",
+  "version": "0.19.0-beta.11",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "@lancedb/lancedb",
-      "version": "0.19.0-beta.1",
+      "version": "0.19.0-beta.11",
       "cpu": [
         "x64",
         "arm64"
@@ -2304,89 +2304,20 @@
       }
     },
     "node_modules/@babel/code-frame": {
-      "version": "7.23.5",
-      "resolved": "https://registry.npmjs.org/@babel/code-frame/-/code-frame-7.23.5.tgz",
-      "integrity": "sha512-CgH3s1a96LipHCmSUmYFPwY7MNx8C3avkq7i4Wl3cfa662ldtUe4VM1TPXX70pfmrlWTb6jLqTYrZyT2ZTJBgA==",
+      "version": "7.26.2",
+      "resolved": "https://registry.npmjs.org/@babel/code-frame/-/code-frame-7.26.2.tgz",
+      "integrity": "sha512-RJlIHRueQgwWitWgF8OdFYGZX328Ax5BCemNGlqHfplnRT9ESi8JkFlvaVYbS+UubVY6dpv87Fs2u5M29iNFVQ==",
       "dev": true,
+      "license": "MIT",
       "dependencies": {
-        "@babel/highlight": "^7.23.4",
-        "chalk": "^2.4.2"
+        "@babel/helper-validator-identifier": "^7.25.9",
+        "js-tokens": "^4.0.0",
+        "picocolors": "^1.0.0"
       },
       "engines": {
         "node": ">=6.9.0"
       }
     },
-    "node_modules/@babel/code-frame/node_modules/ansi-styles": {
-      "version": "3.2.1",
-      "resolved": "https://registry.npmjs.org/ansi-styles/-/ansi-styles-3.2.1.tgz",
-      "integrity": "sha512-VT0ZI6kZRdTh8YyJw3SMbYm/u+NqfsAxEpWO0Pf9sq8/e94WxxOpPKx9FR1FlyCtOVDNOQ+8ntlqFxiRc+r5qA==",
-      "dev": true,
-      "dependencies": {
-        "color-convert": "^1.9.0"
-      },
-      "engines": {
-        "node": ">=4"
-      }
-    },
-    "node_modules/@babel/code-frame/node_modules/chalk": {
-      "version": "2.4.2",
-      "resolved": "https://registry.npmjs.org/chalk/-/chalk-2.4.2.tgz",
-      "integrity": "sha512-Mti+f9lpJNcwF4tWV8/OrTTtF1gZi+f8FqlyAdouralcFWFQWF2+NgCHShjkCb+IFBLq9buZwE1xckQU4peSuQ==",
-      "dev": true,
-      "dependencies": {
-        "ansi-styles": "^3.2.1",
-        "escape-string-regexp": "^1.0.5",
-        "supports-color": "^5.3.0"
-      },
-      "engines": {
-        "node": ">=4"
-      }
-    },
-    "node_modules/@babel/code-frame/node_modules/color-convert": {
-      "version": "1.9.3",
-      "resolved": "https://registry.npmjs.org/color-convert/-/color-convert-1.9.3.tgz",
-      "integrity": "sha512-QfAUtd+vFdAtFQcC8CCyYt1fYWxSqAiK2cSD6zDB8N3cpsEBAvRxp9zOGg6G/SHHJYAT88/az/IuDGALsNVbGg==",
-      "dev": true,
-      "dependencies": {
-        "color-name": "1.1.3"
-      }
-    },
-    "node_modules/@babel/code-frame/node_modules/color-name": {
-      "version": "1.1.3",
-      "resolved": "https://registry.npmjs.org/color-name/-/color-name-1.1.3.tgz",
-      "integrity": "sha512-72fSenhMw2HZMTVHeCA9KCmpEIbzWiQsjN+BHcBbS9vr1mtt+vJjPdksIBNUmKAW8TFUDPJK5SUU3QhE9NEXDw==",
-      "dev": true
-    },
-    "node_modules/@babel/code-frame/node_modules/escape-string-regexp": {
-      "version": "1.0.5",
-      "resolved": "https://registry.npmjs.org/escape-string-regexp/-/escape-string-regexp-1.0.5.tgz",
-      "integrity": "sha512-vbRorB5FUQWvla16U8R/qgaFIya2qGzwDrNmCZuYKrbdSUMG6I1ZCGQRefkRVhuOkIGVne7BQ35DSfo1qvJqFg==",
-      "dev": true,
-      "engines": {
-        "node": ">=0.8.0"
-      }
-    },
-    "node_modules/@babel/code-frame/node_modules/has-flag": {
-      "version": "3.0.0",
-      "resolved": "https://registry.npmjs.org/has-flag/-/has-flag-3.0.0.tgz",
-      "integrity": "sha512-sKJf1+ceQBr4SMkvQnBDNDtf4TXpVhVGateu0t918bl30FnbE2m4vNLX+VWe/dpjlb+HugGYzW7uQXH98HPEYw==",
-      "dev": true,
-      "engines": {
-        "node": ">=4"
-      }
-    },
-    "node_modules/@babel/code-frame/node_modules/supports-color": {
-      "version": "5.5.0",
-      "resolved": "https://registry.npmjs.org/supports-color/-/supports-color-5.5.0.tgz",
-      "integrity": "sha512-QjVjwdXIt408MIiAqCX4oUKsgU2EqAGzs2Ppkm4aQYbjm+ZEWEcW4SfFNTr4uMNZma0ey4f5lgLrkB0aX0QMow==",
-      "dev": true,
-      "dependencies": {
-        "has-flag": "^3.0.0"
-      },
-      "engines": {
-        "node": ">=4"
-      }
-    },
     "node_modules/@babel/compat-data": {
       "version": "7.23.5",
       "resolved": "https://registry.npmjs.org/@babel/compat-data/-/compat-data-7.23.5.tgz",
@@ -2589,19 +2520,21 @@
       }
     },
     "node_modules/@babel/helper-string-parser": {
-      "version": "7.23.4",
-      "resolved": "https://registry.npmjs.org/@babel/helper-string-parser/-/helper-string-parser-7.23.4.tgz",
-      "integrity": "sha512-803gmbQdqwdf4olxrX4AJyFBV/RTr3rSmOj0rKwesmzlfhYNDEs+/iOcznzpNWlJlIlTJC2QfPFcHB6DlzdVLQ==",
+      "version": "7.25.9",
+      "resolved": "https://registry.npmjs.org/@babel/helper-string-parser/-/helper-string-parser-7.25.9.tgz",
+      "integrity": "sha512-4A/SCr/2KLd5jrtOMFzaKjVtAei3+2r/NChoBNoZ3EyP/+GlhoaEGoWOZUmFmoITP7zOJyHIMm+DYRd8o3PvHA==",
       "dev": true,
+      "license": "MIT",
       "engines": {
         "node": ">=6.9.0"
       }
     },
     "node_modules/@babel/helper-validator-identifier": {
-      "version": "7.22.20",
-      "resolved": "https://registry.npmjs.org/@babel/helper-validator-identifier/-/helper-validator-identifier-7.22.20.tgz",
-      "integrity": "sha512-Y4OZ+ytlatR8AI+8KZfKuL5urKp7qey08ha31L8b3BwewJAoJamTzyvxPR/5D+KkdJCGPq/+8TukHBlY10FX9A==",
+      "version": "7.25.9",
+      "resolved": "https://registry.npmjs.org/@babel/helper-validator-identifier/-/helper-validator-identifier-7.25.9.tgz",
+      "integrity": "sha512-Ed61U6XJc3CVRfkERJWDz4dJwKe7iLmmJsbOGu9wSloNSFttHV0I8g6UAgb7qnK5ly5bGLPd4oXZlxCdANBOWQ==",
       "dev": true,
+      "license": "MIT",
       "engines": {
         "node": ">=6.9.0"
       }
@@ -2616,109 +2549,28 @@
       }
     },
     "node_modules/@babel/helpers": {
-      "version": "7.23.8",
-      "resolved": "https://registry.npmjs.org/@babel/helpers/-/helpers-7.23.8.tgz",
-      "integrity": "sha512-KDqYz4PiOWvDFrdHLPhKtCThtIcKVy6avWD2oG4GEvyQ+XDZwHD4YQd+H2vNMnq2rkdxsDkU82T+Vk8U/WXHRQ==",
+      "version": "7.27.0",
+      "resolved": "https://registry.npmjs.org/@babel/helpers/-/helpers-7.27.0.tgz",
+      "integrity": "sha512-U5eyP/CTFPuNE3qk+WZMxFkp/4zUzdceQlfzf7DdGdhp+Fezd7HD+i8Y24ZuTMKX3wQBld449jijbGq6OdGNQg==",
       "dev": true,
+      "license": "MIT",
       "dependencies": {
-        "@babel/template": "^7.22.15",
-        "@babel/traverse": "^7.23.7",
-        "@babel/types": "^7.23.6"
+        "@babel/template": "^7.27.0",
+        "@babel/types": "^7.27.0"
       },
       "engines": {
         "node": ">=6.9.0"
       }
     },
-    "node_modules/@babel/highlight": {
-      "version": "7.23.4",
-      "resolved": "https://registry.npmjs.org/@babel/highlight/-/highlight-7.23.4.tgz",
-      "integrity": "sha512-acGdbYSfp2WheJoJm/EBBBLh/ID8KDc64ISZ9DYtBmC8/Q204PZJLHyzeB5qMzJ5trcOkybd78M4x2KWsUq++A==",
-      "dev": true,
-      "dependencies": {
-        "@babel/helper-validator-identifier": "^7.22.20",
-        "chalk": "^2.4.2",
-        "js-tokens": "^4.0.0"
-      },
-      "engines": {
-        "node": ">=6.9.0"
-      }
-    },
-    "node_modules/@babel/highlight/node_modules/ansi-styles": {
-      "version": "3.2.1",
-      "resolved": "https://registry.npmjs.org/ansi-styles/-/ansi-styles-3.2.1.tgz",
-      "integrity": "sha512-VT0ZI6kZRdTh8YyJw3SMbYm/u+NqfsAxEpWO0Pf9sq8/e94WxxOpPKx9FR1FlyCtOVDNOQ+8ntlqFxiRc+r5qA==",
-      "dev": true,
-      "dependencies": {
-        "color-convert": "^1.9.0"
-      },
-      "engines": {
-        "node": ">=4"
-      }
-    },
-    "node_modules/@babel/highlight/node_modules/chalk": {
-      "version": "2.4.2",
-      "resolved": "https://registry.npmjs.org/chalk/-/chalk-2.4.2.tgz",
-      "integrity": "sha512-Mti+f9lpJNcwF4tWV8/OrTTtF1gZi+f8FqlyAdouralcFWFQWF2+NgCHShjkCb+IFBLq9buZwE1xckQU4peSuQ==",
-      "dev": true,
-      "dependencies": {
-        "ansi-styles": "^3.2.1",
-        "escape-string-regexp": "^1.0.5",
-        "supports-color": "^5.3.0"
-      },
-      "engines": {
-        "node": ">=4"
-      }
-    },
-    "node_modules/@babel/highlight/node_modules/color-convert": {
-      "version": "1.9.3",
-      "resolved": "https://registry.npmjs.org/color-convert/-/color-convert-1.9.3.tgz",
-      "integrity": "sha512-QfAUtd+vFdAtFQcC8CCyYt1fYWxSqAiK2cSD6zDB8N3cpsEBAvRxp9zOGg6G/SHHJYAT88/az/IuDGALsNVbGg==",
-      "dev": true,
-      "dependencies": {
-        "color-name": "1.1.3"
-      }
-    },
-    "node_modules/@babel/highlight/node_modules/color-name": {
-      "version": "1.1.3",
-      "resolved": "https://registry.npmjs.org/color-name/-/color-name-1.1.3.tgz",
-      "integrity": "sha512-72fSenhMw2HZMTVHeCA9KCmpEIbzWiQsjN+BHcBbS9vr1mtt+vJjPdksIBNUmKAW8TFUDPJK5SUU3QhE9NEXDw==",
-      "dev": true
-    },
-    "node_modules/@babel/highlight/node_modules/escape-string-regexp": {
-      "version": "1.0.5",
-      "resolved": "https://registry.npmjs.org/escape-string-regexp/-/escape-string-regexp-1.0.5.tgz",
-      "integrity": "sha512-vbRorB5FUQWvla16U8R/qgaFIya2qGzwDrNmCZuYKrbdSUMG6I1ZCGQRefkRVhuOkIGVne7BQ35DSfo1qvJqFg==",
-      "dev": true,
-      "engines": {
-        "node": ">=0.8.0"
-      }
-    },
-    "node_modules/@babel/highlight/node_modules/has-flag": {
-      "version": "3.0.0",
-      "resolved": "https://registry.npmjs.org/has-flag/-/has-flag-3.0.0.tgz",
-      "integrity": "sha512-sKJf1+ceQBr4SMkvQnBDNDtf4TXpVhVGateu0t918bl30FnbE2m4vNLX+VWe/dpjlb+HugGYzW7uQXH98HPEYw==",
-      "dev": true,
-      "engines": {
-        "node": ">=4"
-      }
-    },
-    "node_modules/@babel/highlight/node_modules/supports-color": {
-      "version": "5.5.0",
-      "resolved": "https://registry.npmjs.org/supports-color/-/supports-color-5.5.0.tgz",
-      "integrity": "sha512-QjVjwdXIt408MIiAqCX4oUKsgU2EqAGzs2Ppkm4aQYbjm+ZEWEcW4SfFNTr4uMNZma0ey4f5lgLrkB0aX0QMow==",
-      "dev": true,
-      "dependencies": {
-        "has-flag": "^3.0.0"
-      },
-      "engines": {
-        "node": ">=4"
-      }
-    },
     "node_modules/@babel/parser": {
-      "version": "7.23.6",
-      "resolved": "https://registry.npmjs.org/@babel/parser/-/parser-7.23.6.tgz",
-      "integrity": "sha512-Z2uID7YJ7oNvAI20O9X0bblw7Qqs8Q2hFy0R9tAfnfLkp5MW0UH9eUvnDSnFwKZ0AvgS1ucqR4KzvVHgnke1VQ==",
+      "version": "7.27.0",
+      "resolved": "https://registry.npmjs.org/@babel/parser/-/parser-7.27.0.tgz",
+      "integrity": "sha512-iaepho73/2Pz7w2eMS0Q5f83+0RKI7i4xmiYeBmDzfRVbQtTOG7Ts0S4HzJVsTMGI9keU8rNfuZr8DKfSt7Yyg==",
       "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "@babel/types": "^7.27.0"
+      },
       "bin": {
         "parser": "bin/babel-parser.js"
       },
@@ -2904,14 +2756,15 @@
       }
     },
     "node_modules/@babel/template": {
-      "version": "7.22.15",
-      "resolved": "https://registry.npmjs.org/@babel/template/-/template-7.22.15.tgz",
-      "integrity": "sha512-QPErUVm4uyJa60rkI73qneDacvdvzxshT3kksGqlGWYdOTIUOwJ7RDUL8sGqslY1uXWSL6xMFKEXDS3ox2uF0w==",
+      "version": "7.27.0",
+      "resolved": "https://registry.npmjs.org/@babel/template/-/template-7.27.0.tgz",
+      "integrity": "sha512-2ncevenBqXI6qRMukPlXwHKHchC7RyMuu4xv5JBXRfOGVcTy1mXCD12qrp7Jsoxll1EV3+9sE4GugBVRjT2jFA==",
       "dev": true,
+      "license": "MIT",
       "dependencies": {
-        "@babel/code-frame": "^7.22.13",
-        "@babel/parser": "^7.22.15",
-        "@babel/types": "^7.22.15"
+        "@babel/code-frame": "^7.26.2",
+        "@babel/parser": "^7.27.0",
+        "@babel/types": "^7.27.0"
       },
       "engines": {
         "node": ">=6.9.0"
@@ -2948,14 +2801,14 @@
       }
     },
     "node_modules/@babel/types": {
-      "version": "7.23.6",
-      "resolved": "https://registry.npmjs.org/@babel/types/-/types-7.23.6.tgz",
-      "integrity": "sha512-+uarb83brBzPKN38NX1MkB6vb6+mwvR6amUulqAE7ccQw1pEl+bCia9TbdG1lsnFP7lZySvUn37CHyXQdfTwzg==",
+      "version": "7.27.0",
+      "resolved": "https://registry.npmjs.org/@babel/types/-/types-7.27.0.tgz",
+      "integrity": "sha512-H45s8fVLYjbhFH62dIJ3WtmJ6RSPt/3DRO0ZcT2SUiYiQyz3BLVb9ADEnLl91m74aQPS3AzzeajZHYOalWe3bg==",
       "dev": true,
+      "license": "MIT",
       "dependencies": {
-        "@babel/helper-string-parser": "^7.23.4",
-        "@babel/helper-validator-identifier": "^7.22.20",
-        "to-fast-properties": "^2.0.0"
+        "@babel/helper-string-parser": "^7.25.9",
+        "@babel/helper-validator-identifier": "^7.25.9"
       },
       "engines": {
         "node": ">=6.9.0"
@@ -5550,10 +5403,11 @@
       "devOptional": true
     },
     "node_modules/axios": {
-      "version": "1.7.7",
-      "resolved": "https://registry.npmjs.org/axios/-/axios-1.7.7.tgz",
-      "integrity": "sha512-S4kL7XrjgBmvdGut0sN3yJxqYzrDOnivkBiN0OFs6hLiUam3UPvswUo0kqGyhqUZGEOytHyumEdXsAkgCOUf3Q==",
+      "version": "1.8.4",
+      "resolved": "https://registry.npmjs.org/axios/-/axios-1.8.4.tgz",
+      "integrity": "sha512-eBSYY4Y68NNlHbHBMdeDmKNtDgXWhQsJcGqzO3iLUM0GraQFSS9cVgPX5I9b3lbdFKyYoAEGAZF1DwhTaljNAw==",
       "dev": true,
+      "license": "MIT",
       "dependencies": {
         "follow-redirects": "^1.15.6",
         "form-data": "^4.0.0",
@@ -7869,7 +7723,8 @@
       "version": "4.0.0",
       "resolved": "https://registry.npmjs.org/js-tokens/-/js-tokens-4.0.0.tgz",
       "integrity": "sha512-RdJUflcE3cUzKiMqQgsCu06FPu9UdIJO0beYbPhHN4k6apgJtifcoCtT9bcxOpYBtpD2kCM6Sbzg4CausW/PKQ==",
-      "dev": true
+      "dev": true,
+      "license": "MIT"
     },
     "node_modules/js-yaml": {
       "version": "3.14.1",
@@ -9360,15 +9215,6 @@
       "integrity": "sha512-3f0uOEAQwIqGuWW2MVzYg8fV/QNnc/IpuJNG837rLuczAaLVHslWHZQj4IGiEl5Hs3kkbhwL9Ab7Hrsmuj+Smw==",
       "dev": true
     },
-    "node_modules/to-fast-properties": {
-      "version": "2.0.0",
-      "resolved": "https://registry.npmjs.org/to-fast-properties/-/to-fast-properties-2.0.0.tgz",
-      "integrity": "sha512-/OaKK0xYrs3DmxRYqL/yDc+FxFUVYhDlXMhRmv3z915w2HF1tnN1omB354j8VUGO/hbRzyD6Y3sA7v7GS/ceog==",
-      "dev": true,
-      "engines": {
-        "node": ">=4"
-      }
-    },
     "node_modules/to-regex-range": {
       "version": "5.0.1",
       "resolved": "https://registry.npmjs.org/to-regex-range/-/to-regex-range-5.0.1.tgz",

nodejs/package.json

@@ -11,7 +11,7 @@
     "ann"
   ],
   "private": false,
-  "version": "0.19.0-beta.1",
+  "version": "0.19.0-beta.11",
   "main": "dist/index.js",
   "exports": {
     ".": "./dist/index.js",
@@ -29,6 +29,7 @@
         "aarch64-apple-darwin",
         "x86_64-unknown-linux-gnu",
         "aarch64-unknown-linux-gnu",
+        "x86_64-unknown-linux-musl",
         "aarch64-unknown-linux-musl",
         "x86_64-pc-windows-msvc",
         "aarch64-pc-windows-msvc"

nodejs/src/connection.rs

@@ -48,16 +48,8 @@ impl Connection {
     pub async fn new(uri: String, options: ConnectionOptions) -> napi::Result<Self> {
         let mut builder = ConnectBuilder::new(&uri);
         if let Some(interval) = options.read_consistency_interval {
-            match interval {
-                Either::A(seconds) => {
-                    builder = builder.read_consistency_interval(Some(
-                        std::time::Duration::from_secs_f64(seconds),
-                    ));
-                }
-                Either::B(_) => {
-                    builder = builder.read_consistency_interval(None);
-                }
-            }
+            builder =
+                builder.read_consistency_interval(std::time::Duration::from_secs_f64(interval));
         }
         if let Some(storage_options) = options.storage_options {
             for (key, value) in storage_options {

nodejs/src/lib.rs

@@ -4,7 +4,6 @@
 use std::collections::HashMap;
 
 use env_logger::Env;
-use napi::{bindgen_prelude::Null, Either};
 use napi_derive::*;
 
 mod connection;
@@ -19,6 +18,7 @@ mod table;
 mod util;
 
 #[napi(object)]
+#[derive(Debug)]
 pub struct ConnectionOptions {
     /// (For LanceDB OSS only): The interval, in seconds, at which to check for
     /// updates to the table from other processes. If None, then consistency is not
@@ -29,7 +29,7 @@ pub struct ConnectionOptions {
     /// has passed since the last check, then the table will be checked for updates.
     /// Note: this consistency only applies to read operations. Write operations are
     /// always consistent.
-    pub read_consistency_interval: Option<Either<f64, Null>>,
+    pub read_consistency_interval: Option<f64>,
     /// (For LanceDB OSS only): configuration for object storage.
     ///
     /// The available options are described at https://lancedb.github.io/lancedb/guides/storage/

nodejs/src/query.rs

@@ -3,7 +3,9 @@
 
 use std::sync::Arc;
 
-use lancedb::index::scalar::{FtsQuery, FullTextSearchQuery, MatchQuery, PhraseQuery};
+use lancedb::index::scalar::{
+    BoostQuery, FtsQuery, FullTextSearchQuery, MatchQuery, MultiMatchQuery, PhraseQuery,
+};
 use lancedb::query::ExecutableQuery;
 use lancedb::query::Query as LanceDbQuery;
 use lancedb::query::QueryBase;
@@ -18,7 +20,7 @@ use crate::error::NapiErrorExt;
 use crate::iterator::RecordBatchIterator;
 use crate::rerankers::Reranker;
 use crate::rerankers::RerankerCallbacks;
-use crate::util::{parse_distance_type, parse_fts_query};
+use crate::util::parse_distance_type;
 
 #[napi]
 pub struct Query {
@@ -38,51 +40,8 @@ impl Query {
     }
 
     #[napi]
-    pub fn full_text_search(&mut self, query: napi::JsUnknown) -> napi::Result<()> {
-        let query = unsafe { query.cast::<napi::JsObject>() };
-        let query = if let Some(query_text) = query.get::<_, String>("query").transpose() {
-            let mut query_text = query_text?;
-            let columns = query.get::<_, Option<Vec<String>>>("columns")?.flatten();
-
-            let is_phrase =
-                query_text.len() >= 2 && query_text.starts_with('"') && query_text.ends_with('"');
-            let is_multi_match = columns.as_ref().map(|cols| cols.len() > 1).unwrap_or(false);
-
-            if is_phrase {
-                // Remove the surrounding quotes for phrase queries
-                query_text = query_text[1..query_text.len() - 1].to_string();
-            }
-
-            let query: FtsQuery = match (is_phrase, is_multi_match) {
-                (false, _) => MatchQuery::new(query_text).into(),
-                (true, false) => PhraseQuery::new(query_text).into(),
-                (true, true) => {
-                    return Err(napi::Error::from_reason(
-                        "Phrase queries cannot be used with multiple columns.",
-                    ));
-                }
-            };
-            let mut query = FullTextSearchQuery::new_query(query);
-            if let Some(cols) = columns {
-                if !cols.is_empty() {
-                    query = query.with_columns(&cols).map_err(|e| {
-                        napi::Error::from_reason(format!(
-                            "Failed to set full text search columns: {}",
-                            e
-                        ))
-                    })?;
-                }
-            }
-            query
-        } else if let Some(query) = query.get::<_, napi::JsObject>("query")? {
-            let query = parse_fts_query(&query)?;
-            FullTextSearchQuery::new_query(query)
-        } else {
-            return Err(napi::Error::from_reason(
-                "Invalid full text search query object".to_string(),
-            ));
-        };
-
+    pub fn full_text_search(&mut self, query: napi::JsObject) -> napi::Result<()> {
+        let query = parse_fts_query(query)?;
         self.inner = self.inner.clone().full_text_search(query);
         Ok(())
     }
@@ -131,11 +90,15 @@ impl Query {
     pub async fn execute(
         &self,
         max_batch_length: Option<u32>,
+        timeout_ms: Option<u32>,
     ) -> napi::Result<RecordBatchIterator> {
         let mut execution_opts = QueryExecutionOptions::default();
         if let Some(max_batch_length) = max_batch_length {
             execution_opts.max_batch_length = max_batch_length;
         }
+        if let Some(timeout_ms) = timeout_ms {
+            execution_opts.timeout = Some(std::time::Duration::from_millis(timeout_ms as u64))
+        }
         let inner_stream = self
             .inner
             .execute_with_options(execution_opts)
@@ -239,51 +202,8 @@ impl VectorQuery {
     }
 
     #[napi]
-    pub fn full_text_search(&mut self, query: napi::JsUnknown) -> napi::Result<()> {
-        let query = unsafe { query.cast::<napi::JsObject>() };
-        let query = if let Some(query_text) = query.get::<_, String>("query").transpose() {
-            let mut query_text = query_text?;
-            let columns = query.get::<_, Option<Vec<String>>>("columns")?.flatten();
-
-            let is_phrase =
-                query_text.len() >= 2 && query_text.starts_with('"') && query_text.ends_with('"');
-            let is_multi_match = columns.as_ref().map(|cols| cols.len() > 1).unwrap_or(false);
-
-            if is_phrase {
-                // Remove the surrounding quotes for phrase queries
-                query_text = query_text[1..query_text.len() - 1].to_string();
-            }
-
-            let query: FtsQuery = match (is_phrase, is_multi_match) {
-                (false, _) => MatchQuery::new(query_text).into(),
-                (true, false) => PhraseQuery::new(query_text).into(),
-                (true, true) => {
-                    return Err(napi::Error::from_reason(
-                        "Phrase queries cannot be used with multiple columns.",
-                    ));
-                }
-            };
-            let mut query = FullTextSearchQuery::new_query(query);
-            if let Some(cols) = columns {
-                if !cols.is_empty() {
-                    query = query.with_columns(&cols).map_err(|e| {
-                        napi::Error::from_reason(format!(
-                            "Failed to set full text search columns: {}",
-                            e
-                        ))
-                    })?;
-                }
-            }
-            query
-        } else if let Some(query) = query.get::<_, napi::JsObject>("query")? {
-            let query = parse_fts_query(&query)?;
-            FullTextSearchQuery::new_query(query)
-        } else {
-            return Err(napi::Error::from_reason(
-                "Invalid full text search query object".to_string(),
-            ));
-        };
-
+    pub fn full_text_search(&mut self, query: napi::JsObject) -> napi::Result<()> {
+        let query = parse_fts_query(query)?;
         self.inner = self.inner.clone().full_text_search(query);
         Ok(())
     }
@@ -330,11 +250,15 @@ impl VectorQuery {
     pub async fn execute(
         &self,
         max_batch_length: Option<u32>,
+        timeout_ms: Option<u32>,
     ) -> napi::Result<RecordBatchIterator> {
         let mut execution_opts = QueryExecutionOptions::default();
         if let Some(max_batch_length) = max_batch_length {
             execution_opts.max_batch_length = max_batch_length;
         }
+        if let Some(timeout_ms) = timeout_ms {
+            execution_opts.timeout = Some(std::time::Duration::from_millis(timeout_ms as u64))
+        }
         let inner_stream = self
             .inner
             .execute_with_options(execution_opts)
@@ -368,3 +292,116 @@ impl VectorQuery {
         })
     }
 }
+
+#[napi]
+#[derive(Debug, Clone)]
+pub struct JsFullTextQuery {
+    pub(crate) inner: FtsQuery,
+}
+
+#[napi]
+impl JsFullTextQuery {
+    #[napi(factory)]
+    pub fn match_query(
+        query: String,
+        column: String,
+        boost: f64,
+        fuzziness: Option<u32>,
+        max_expansions: u32,
+    ) -> napi::Result<Self> {
+        Ok(Self {
+            inner: MatchQuery::new(query)
+                .with_column(Some(column))
+                .with_boost(boost as f32)
+                .with_fuzziness(fuzziness)
+                .with_max_expansions(max_expansions as usize)
+                .into(),
+        })
+    }
+
+    #[napi(factory)]
+    pub fn phrase_query(query: String, column: String) -> napi::Result<Self> {
+        Ok(Self {
+            inner: PhraseQuery::new(query).with_column(Some(column)).into(),
+        })
+    }
+
+    #[napi(factory)]
+    #[allow(clippy::use_self)] // NAPI doesn't allow Self here but clippy reports it
+    pub fn boost_query(
+        positive: &JsFullTextQuery,
+        negative: &JsFullTextQuery,
+        negative_boost: Option<f64>,
+    ) -> napi::Result<Self> {
+        Ok(Self {
+            inner: BoostQuery::new(
+                positive.inner.clone(),
+                negative.inner.clone(),
+                negative_boost.map(|v| v as f32),
+            )
+            .into(),
+        })
+    }
+
+    #[napi(factory)]
+    pub fn multi_match_query(
+        query: String,
+        columns: Vec<String>,
+        boosts: Option<Vec<f64>>,
+    ) -> napi::Result<Self> {
+        let q = match boosts {
+            Some(boosts) => MultiMatchQuery::try_new(query, columns)
+                .and_then(|q| q.try_with_boosts(boosts.into_iter().map(|v| v as f32).collect())),
+            None => MultiMatchQuery::try_new(query, columns),
+        }
+        .map_err(|e| {
+            napi::Error::from_reason(format!("Failed to create multi match query: {}", e))
+        })?;
+
+        Ok(Self { inner: q.into() })
+    }
+}
+
+fn parse_fts_query(query: napi::JsObject) -> napi::Result<FullTextSearchQuery> {
+    if let Ok(Some(query)) = query.get::<_, &JsFullTextQuery>("query") {
+        Ok(FullTextSearchQuery::new_query(query.inner.clone()))
+    } else if let Ok(Some(query_text)) = query.get::<_, String>("query") {
+        let mut query_text = query_text;
+        let columns = query.get::<_, Option<Vec<String>>>("columns")?.flatten();
+
+        let is_phrase =
+            query_text.len() >= 2 && query_text.starts_with('"') && query_text.ends_with('"');
+        let is_multi_match = columns.as_ref().map(|cols| cols.len() > 1).unwrap_or(false);
+
+        if is_phrase {
+            // Remove the surrounding quotes for phrase queries
+            query_text = query_text[1..query_text.len() - 1].to_string();
+        }
+
+        let query: FtsQuery = match (is_phrase, is_multi_match) {
+            (false, _) => MatchQuery::new(query_text).into(),
+            (true, false) => PhraseQuery::new(query_text).into(),
+            (true, true) => {
+                return Err(napi::Error::from_reason(
+                    "Phrase queries cannot be used with multiple columns.",
+                ));
+            }
+        };
+        let mut query = FullTextSearchQuery::new_query(query);
+        if let Some(cols) = columns {
+            if !cols.is_empty() {
+                query = query.with_columns(&cols).map_err(|e| {
+                    napi::Error::from_reason(format!(
+                        "Failed to set full text search columns: {}",
+                        e
+                    ))
+                })?;
+            }
+        }
+        Ok(query)
+    } else {
+        Err(napi::Error::from_reason(
+            "Invalid full text search query object".to_string(),
+        ))
+    }
+}

nodejs/src/table.rs

@@ -111,6 +111,7 @@ impl Table {
         index: Option<&Index>,
         column: String,
         replace: Option<bool>,
+        wait_timeout_s: Option<i64>,
     ) -> napi::Result<()> {
         let lancedb_index = if let Some(index) = index {
             index.consume()?
@@ -121,6 +122,10 @@ impl Table {
         if let Some(replace) = replace {
             builder = builder.replace(replace);
         }
+        if let Some(timeout) = wait_timeout_s {
+            builder =
+                builder.wait_timeout(std::time::Duration::from_secs(timeout.try_into().unwrap()));
+        }
         builder.execute().await.default_error()
     }
 
@@ -132,6 +137,26 @@ impl Table {
             .default_error()
     }
 
+    #[napi(catch_unwind)]
+    pub async fn prewarm_index(&self, index_name: String) -> napi::Result<()> {
+        self.inner_ref()?
+            .prewarm_index(&index_name)
+            .await
+            .default_error()
+    }
+
+    #[napi(catch_unwind)]
+    pub async fn wait_for_index(&self, index_names: Vec<String>, timeout_s: i64) -> Result<()> {
+        let timeout = std::time::Duration::from_secs(timeout_s.try_into().unwrap());
+        let index_names: Vec<&str> = index_names.iter().map(|s| s.as_str()).collect();
+        let slice: &[&str] = &index_names;
+
+        self.inner_ref()?
+            .wait_for_index(slice, timeout)
+            .await
+            .default_error()
+    }
+
     #[napi(catch_unwind)]
     pub async fn update(
         &self,

nodejs/src/util.rs

@@ -1,7 +1,6 @@
 // SPDX-License-Identifier: Apache-2.0
 // SPDX-FileCopyrightText: Copyright The LanceDB Authors
 
-use lancedb::index::scalar::{BoostQuery, FtsQuery, MatchQuery, MultiMatchQuery, PhraseQuery};
 use lancedb::DistanceType;
 
 pub fn parse_distance_type(distance_type: impl AsRef<str>) -> napi::Result<DistanceType> {
@@ -16,144 +15,3 @@ pub fn parse_distance_type(distance_type: impl AsRef<str>) -> napi::Result<Dista
         ))),
     }
 }
-
-pub fn parse_fts_query(query: &napi::JsObject) -> napi::Result<FtsQuery> {
-    let query_type = query
-        .get_property_names()?
-        .get_element::<napi::JsString>(0)?;
-    let query_type = query_type.into_utf8()?.into_owned()?;
-    let query_value =
-        query
-            .get::<_, napi::JsObject>(&query_type)?
-            .ok_or(napi::Error::from_reason(format!(
-                "query value {} not found",
-                query_type
-            )))?;
-
-    match query_type.as_str() {
-        "match" => {
-            let column = query_value
-                .get_property_names()?
-                .get_element::<napi::JsString>(0)?
-                .into_utf8()?
-                .into_owned()?;
-            let params =
-                query_value
-                    .get::<_, napi::JsObject>(&column)?
-                    .ok_or(napi::Error::from_reason(format!(
-                        "column {} not found",
-                        column
-                    )))?;
-
-            let query = params
-                .get::<_, napi::JsString>("query")?
-                .ok_or(napi::Error::from_reason("query not found"))?
-                .into_utf8()?
-                .into_owned()?;
-            let boost = params
-                .get::<_, napi::JsNumber>("boost")?
-                .ok_or(napi::Error::from_reason("boost not found"))?
-                .get_double()? as f32;
-            let fuzziness = params
-                .get::<_, napi::JsNumber>("fuzziness")?
-                .map(|f| f.get_uint32())
-                .transpose()?;
-            let max_expansions = params
-                .get::<_, napi::JsNumber>("max_expansions")?
-                .ok_or(napi::Error::from_reason("max_expansions not found"))?
-                .get_uint32()? as usize;
-
-            let query = MatchQuery::new(query)
-                .with_column(Some(column))
-                .with_boost(boost)
-                .with_fuzziness(fuzziness)
-                .with_max_expansions(max_expansions);
-            Ok(query.into())
-        }
-
-        "match_phrase" => {
-            let column = query_value
-                .get_property_names()?
-                .get_element::<napi::JsString>(0)?
-                .into_utf8()?
-                .into_owned()?;
-            let query = query_value
-                .get::<_, napi::JsString>(&column)?
-                .ok_or(napi::Error::from_reason(format!(
-                    "column {} not found",
-                    column
-                )))?
-                .into_utf8()?
-                .into_owned()?;
-
-            let query = PhraseQuery::new(query).with_column(Some(column));
-            Ok(query.into())
-        }
-
-        "boost" => {
-            let positive = query_value
-                .get::<_, napi::JsObject>("positive")?
-                .ok_or(napi::Error::from_reason("positive not found"))?;
-
-            let negative = query_value
-                .get::<_, napi::JsObject>("negative")?
-                .ok_or(napi::Error::from_reason("negative not found"))?;
-            let negative_boost = query_value
-                .get::<_, napi::JsNumber>("negative_boost")?
-                .ok_or(napi::Error::from_reason("negative_boost not found"))?
-                .get_double()? as f32;
-
-            let positive = parse_fts_query(&positive)?;
-            let negative = parse_fts_query(&negative)?;
-            let query = BoostQuery::new(positive, negative, Some(negative_boost));
-            Ok(query.into())
-        }
-
-        "multi_match" => {
-            let query = query_value
-                .get::<_, napi::JsString>("query")?
-                .ok_or(napi::Error::from_reason("query not found"))?
-                .into_utf8()?
-                .into_owned()?;
-            let columns_array = query_value
-                .get::<_, napi::JsTypedArray>("columns")?
-                .ok_or(napi::Error::from_reason("columns not found"))?;
-            let columns_num = columns_array.get_array_length()?;
-            let mut columns = Vec::with_capacity(columns_num as usize);
-            for i in 0..columns_num {
-                let column = columns_array
-                    .get_element::<napi::JsString>(i)?
-                    .into_utf8()?
-                    .into_owned()?;
-                columns.push(column);
-            }
-            let boost_array = query_value
-                .get::<_, napi::JsTypedArray>("boost")?
-                .ok_or(napi::Error::from_reason("boost not found"))?;
-            if boost_array.get_array_length()? != columns_num {
-                return Err(napi::Error::from_reason(format!(
-                    "boost array length ({}) does not match columns length ({})",
-                    boost_array.get_array_length()?,
-                    columns_num
-                )));
-            }
-            let mut boost = Vec::with_capacity(columns_num as usize);
-            for i in 0..columns_num {
-                let b = boost_array.get_element::<napi::JsNumber>(i)?.get_double()? as f32;
-                boost.push(b);
-            }
-
-            let query =
-                MultiMatchQuery::try_new_with_boosts(query, columns, boost).map_err(|e| {
-                    napi::Error::from_reason(format!("Error creating MultiMatchQuery: {}", e))
-                })?;
-
-            Ok(query.into())
-        }
-
-        _ => Err(napi::Error::from_reason(format!(
-            "Unsupported query type: {}",
-            query_type
-        ))),
-    }
-}

python/.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.22.0-beta.2"
+current_version = "0.22.0"
 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 = "0.22.0-beta.2"
+version = "0.22.0"
 edition.workspace = true
 description = "Python bindings for LanceDB"
 license.workspace = true

python/pyproject.toml

@@ -4,11 +4,12 @@ name = "lancedb"
 dynamic = ["version"]
 dependencies = [
     "deprecation",
-    "tqdm>=4.27.0",
+    "numpy",
+    "overrides>=0.7",
+    "packaging",
     "pyarrow>=14",
     "pydantic>=1.10",
-    "packaging",
-    "overrides>=0.7",
+    "tqdm>=4.27.0",
 ]
 description = "lancedb"
 authors = [{ name = "LanceDB Devs", email = "dev@lancedb.com" }]
@@ -42,6 +43,9 @@ classifiers = [
 repository = "https://github.com/lancedb/lancedb"
 
 [project.optional-dependencies]
+pylance = [
+    "pylance>=0.25",
+]
 tests = [
     "aiohttp",
     "boto3",
@@ -54,7 +58,8 @@ tests = [
     "polars>=0.19, <=1.3.0",
     "tantivy",
     "pyarrow-stubs",
-    "pylance>=0.23.2",
+    "pylance>=0.25",
+    "requests",
 ]
 dev = [
     "ruff",
@@ -72,6 +77,7 @@ embeddings = [
     "pillow",
     "open-clip-torch",
     "cohere",
+    "colpali-engine>=0.3.10",
     "huggingface_hub",
     "InstructorEmbedding",
     "google.generativeai",

python/python/lancedb/__init__.py

@@ -26,7 +26,7 @@ def connect(
     api_key: Optional[str] = None,
     region: str = "us-east-1",
     host_override: Optional[str] = None,
-    read_consistency_interval: Optional[timedelta] = timedelta(seconds=5),
+    read_consistency_interval: Optional[timedelta] = None,
     request_thread_pool: Optional[Union[int, ThreadPoolExecutor]] = None,
     client_config: Union[ClientConfig, Dict[str, Any], None] = None,
     storage_options: Optional[Dict[str, str]] = None,
@@ -49,8 +49,9 @@ def connect(
     read_consistency_interval: timedelta, default None
         (For LanceDB OSS only)
         The interval at which to check for updates to the table from other
-        processes. If None, then consistency is not checked. For strong consistency,
-        set this to zero seconds. Then every read will check for updates from other
+        processes. If None, then consistency is not checked. For performance
+        reasons, this is the default. For strong consistency, set this to
+        zero seconds. Then every read will check for updates from other
         processes. As a compromise, you can set this to a non-zero timedelta
         for eventual consistency. If more than that interval has passed since
         the last check, then the table will be checked for updates. Note: this
@@ -121,7 +122,7 @@ async def connect_async(
     api_key: Optional[str] = None,
     region: str = "us-east-1",
     host_override: Optional[str] = None,
-    read_consistency_interval: Optional[timedelta] = timedelta(seconds=5),
+    read_consistency_interval: Optional[timedelta] = None,
     client_config: Optional[Union[ClientConfig, Dict[str, Any]]] = None,
     storage_options: Optional[Dict[str, str]] = None,
 ) -> AsyncConnection:
@@ -142,8 +143,9 @@ async def connect_async(
     read_consistency_interval: timedelta, default None
         (For LanceDB OSS only)
         The interval at which to check for updates to the table from other
-        processes. If None, then consistency is not checked. For strong consistency,
-        set this to zero seconds. Then every read will check for updates from other
+        processes. If None, then consistency is not checked. For performance
+        reasons, this is the default. For strong consistency, set this to
+        zero seconds. Then every read will check for updates from other
         processes. As a compromise, you can set this to a non-zero timedelta
         for eventual consistency. If more than that interval has passed since
         the last check, then the table will be checked for updates. Note: this

python/python/lancedb/_lancedb.pyi

@@ -1,3 +1,4 @@
+from datetime import timedelta
 from typing import Dict, List, Optional, Tuple, Any, Union, Literal
 
 import pyarrow as pa
@@ -94,7 +95,9 @@ class Query:
     def postfilter(self): ...
     def nearest_to(self, query_vec: pa.Array) -> VectorQuery: ...
     def nearest_to_text(self, query: dict) -> FTSQuery: ...
-    async def execute(self, max_batch_length: Optional[int]) -> RecordBatchStream: ...
+    async def execute(
+        self, max_batch_length: Optional[int], timeout: Optional[timedelta]
+    ) -> RecordBatchStream: ...
     async def explain_plan(self, verbose: Optional[bool]) -> str: ...
     async def analyze_plan(self) -> str: ...
     def to_query_request(self) -> PyQueryRequest: ...
@@ -110,7 +113,9 @@ class FTSQuery:
     def get_query(self) -> str: ...
     def add_query_vector(self, query_vec: pa.Array) -> None: ...
     def nearest_to(self, query_vec: pa.Array) -> HybridQuery: ...
-    async def execute(self, max_batch_length: Optional[int]) -> RecordBatchStream: ...
+    async def execute(
+        self, max_batch_length: Optional[int], timeout: Optional[timedelta]
+    ) -> RecordBatchStream: ...
     def to_query_request(self) -> PyQueryRequest: ...
 
 class VectorQuery:

python/python/lancedb/common.py

@@ -9,7 +9,7 @@ import numpy as np
 import pyarrow as pa
 import pyarrow.dataset
 
-from .dependencies import pandas as pd
+from .dependencies import _check_for_pandas, pandas as pd
 
 DATA = Union[List[dict], "pd.DataFrame", pa.Table, Iterable[pa.RecordBatch]]
 VEC = Union[list, np.ndarray, pa.Array, pa.ChunkedArray]
@@ -63,7 +63,7 @@ def data_to_reader(
     data: DATA, schema: Optional[pa.Schema] = None
 ) -> pa.RecordBatchReader:
     """Convert various types of input into a RecordBatchReader"""
-    if pd is not None and isinstance(data, pd.DataFrame):
+    if _check_for_pandas(data) and isinstance(data, pd.DataFrame):
         return pa.Table.from_pandas(data, schema=schema).to_reader()
     elif isinstance(data, pa.Table):
         return data.to_reader()

python/python/lancedb/db.py

@@ -6,7 +6,6 @@ from __future__ import annotations
 
 from abc import abstractmethod
 from pathlib import Path
-from datetime import timedelta
 from typing import TYPE_CHECKING, Dict, Iterable, List, Literal, Optional, Union
 
 from lancedb.embeddings.registry import EmbeddingFunctionRegistry
@@ -33,6 +32,7 @@ import deprecation
 if TYPE_CHECKING:
     import pyarrow as pa
     from .pydantic import LanceModel
+    from datetime import timedelta
 
     from ._lancedb import Connection as LanceDbConnection
     from .common import DATA, URI
@@ -318,8 +318,9 @@ class LanceDBConnection(DBConnection):
         The root uri of the database.
     read_consistency_interval: timedelta, default None
         The interval at which to check for updates to the table from other
-        processes. If None, then consistency is not checked. For strong consistency,
-        set this to zero seconds. Then every read will check for updates from other
+        processes. If None, then consistency is not checked. For performance
+        reasons, this is the default. For strong consistency, set this to
+        zero seconds. Then every read will check for updates from other
         processes. As a compromise, you can set this to a non-zero timedelta
         for eventual consistency. If more than that interval has passed since
         the last check, then the table will be checked for updates. Note: this
@@ -351,7 +352,7 @@ class LanceDBConnection(DBConnection):
         self,
         uri: URI,
         *,
-        read_consistency_interval: Optional[timedelta] = timedelta(seconds=5),
+        read_consistency_interval: Optional[timedelta] = None,
         storage_options: Optional[Dict[str, str]] = None,
     ):
         if not isinstance(uri, Path):

python/python/lancedb/embeddings/__init__.py

@@ -19,3 +19,4 @@ from .imagebind import ImageBindEmbeddings
 from .jinaai import JinaEmbeddings
 from .watsonx import WatsonxEmbeddings
 from .voyageai import VoyageAIEmbeddingFunction
+from .colpali import ColPaliEmbeddings

python/python/lancedb/embeddings/colpali.py

@@ -0,0 +1,255 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+
+from functools import lru_cache
+from typing import List, Union, Optional, Any
+import numpy as np
+import io
+
+from ..util import attempt_import_or_raise
+from .base import EmbeddingFunction
+from .registry import register
+from .utils import TEXT, IMAGES, is_flash_attn_2_available
+
+
+@register("colpali")
+class ColPaliEmbeddings(EmbeddingFunction):
+    """
+    An embedding function that uses the ColPali engine for
+    multimodal multi-vector embeddings.
+
+    This embedding function supports ColQwen2.5 models, producing multivector outputs
+    for both text and image inputs. The output embeddings are lists of vectors, each
+    vector being 128-dimensional by default, represented as List[List[float]].
+
+    Parameters
+    ----------
+    model_name : str
+        The name of the model to use (e.g., "Metric-AI/ColQwen2.5-3b-multilingual-v1.0")
+    device : str
+        The device for inference (default "cuda:0").
+    dtype : str
+        Data type for model weights (default "bfloat16").
+    use_token_pooling : bool
+        Whether to use token pooling to reduce embedding size (default True).
+    pool_factor : int
+        Factor to reduce sequence length if token pooling is enabled (default 2).
+    quantization_config : Optional[BitsAndBytesConfig]
+        Quantization configuration for the model. (default None, bitsandbytes needed)
+    batch_size : int
+        Batch size for processing inputs (default 2).
+    """
+
+    model_name: str = "Metric-AI/ColQwen2.5-3b-multilingual-v1.0"
+    device: str = "auto"
+    dtype: str = "bfloat16"
+    use_token_pooling: bool = True
+    pool_factor: int = 2
+    quantization_config: Optional[Any] = None
+    batch_size: int = 2
+
+    _model = None
+    _processor = None
+    _token_pooler = None
+    _vector_dim = None
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        (
+            self._model,
+            self._processor,
+            self._token_pooler,
+        ) = self._load_model(
+            self.model_name,
+            self.dtype,
+            self.device,
+            self.use_token_pooling,
+            self.quantization_config,
+        )
+
+    @staticmethod
+    @lru_cache(maxsize=1)
+    def _load_model(
+        model_name: str,
+        dtype: str,
+        device: str,
+        use_token_pooling: bool,
+        quantization_config: Optional[Any],
+    ):
+        """
+        Initialize and cache the ColPali model, processor, and token pooler.
+        """
+        torch = attempt_import_or_raise("torch", "torch")
+        transformers = attempt_import_or_raise("transformers", "transformers")
+        colpali_engine = attempt_import_or_raise("colpali_engine", "colpali_engine")
+        from colpali_engine.compression.token_pooling import HierarchicalTokenPooler
+
+        if quantization_config is not None:
+            if not isinstance(quantization_config, transformers.BitsAndBytesConfig):
+                raise ValueError("quantization_config must be a BitsAndBytesConfig")
+
+        if dtype == "bfloat16":
+            torch_dtype = torch.bfloat16
+        elif dtype == "float16":
+            torch_dtype = torch.float16
+        elif dtype == "float64":
+            torch_dtype = torch.float64
+        else:
+            torch_dtype = torch.float32
+
+        model = colpali_engine.models.ColQwen2_5.from_pretrained(
+            model_name,
+            torch_dtype=torch_dtype,
+            device_map=device,
+            quantization_config=quantization_config
+            if quantization_config is not None
+            else None,
+            attn_implementation="flash_attention_2"
+            if is_flash_attn_2_available()
+            else None,
+        ).eval()
+        processor = colpali_engine.models.ColQwen2_5_Processor.from_pretrained(
+            model_name
+        )
+        token_pooler = HierarchicalTokenPooler() if use_token_pooling else None
+        return model, processor, token_pooler
+
+    def ndims(self):
+        """
+        Return the dimension of a vector in the multivector output (e.g., 128).
+        """
+        torch = attempt_import_or_raise("torch", "torch")
+        if self._vector_dim is None:
+            dummy_query = "test"
+            batch_queries = self._processor.process_queries([dummy_query]).to(
+                self._model.device
+            )
+            with torch.no_grad():
+                query_embeddings = self._model(**batch_queries)
+
+            if self.use_token_pooling and self._token_pooler is not None:
+                query_embeddings = self._token_pooler.pool_embeddings(
+                    query_embeddings,
+                    pool_factor=self.pool_factor,
+                    padding=True,
+                    padding_side=self._processor.tokenizer.padding_side,
+                )
+
+            self._vector_dim = query_embeddings[0].shape[-1]
+        return self._vector_dim
+
+    def _process_embeddings(self, embeddings):
+        """
+        Format model embeddings into List[List[float]].
+        Use token pooling if enabled.
+        """
+        torch = attempt_import_or_raise("torch", "torch")
+        if self.use_token_pooling and self._token_pooler is not None:
+            embeddings = self._token_pooler.pool_embeddings(
+                embeddings,
+                pool_factor=self.pool_factor,
+                padding=True,
+                padding_side=self._processor.tokenizer.padding_side,
+            )
+
+        if isinstance(embeddings, torch.Tensor):
+            tensors = embeddings.detach().cpu()
+            if tensors.dtype == torch.bfloat16:
+                tensors = tensors.to(torch.float32)
+            return (
+                tensors.numpy()
+                .astype(np.float64 if self.dtype == "float64" else np.float32)
+                .tolist()
+            )
+        return []
+
+    def generate_text_embeddings(self, text: TEXT) -> List[List[List[float]]]:
+        """
+        Generate embeddings for text input.
+        """
+        torch = attempt_import_or_raise("torch", "torch")
+        text = self.sanitize_input(text)
+        all_embeddings = []
+
+        for i in range(0, len(text), self.batch_size):
+            batch_text = text[i : i + self.batch_size]
+            batch_queries = self._processor.process_queries(batch_text).to(
+                self._model.device
+            )
+            with torch.no_grad():
+                query_embeddings = self._model(**batch_queries)
+            all_embeddings.extend(self._process_embeddings(query_embeddings))
+        return all_embeddings
+
+    def _prepare_images(self, images: IMAGES) -> List:
+        """
+        Convert image inputs to PIL Images.
+        """
+        PIL = attempt_import_or_raise("PIL", "pillow")
+        requests = attempt_import_or_raise("requests", "requests")
+        images = self.sanitize_input(images)
+        pil_images = []
+        try:
+            for image in images:
+                if isinstance(image, str):
+                    if image.startswith(("http://", "https://")):
+                        response = requests.get(image, timeout=10)
+                        response.raise_for_status()
+                        pil_images.append(PIL.Image.open(io.BytesIO(response.content)))
+                    else:
+                        with PIL.Image.open(image) as im:
+                            pil_images.append(im.copy())
+                elif isinstance(image, bytes):
+                    pil_images.append(PIL.Image.open(io.BytesIO(image)))
+                else:
+                    # Assume it's a PIL Image; will raise if invalid
+                    pil_images.append(image)
+        except Exception as e:
+            raise ValueError(f"Failed to process image: {e}")
+
+        return pil_images
+
+    def generate_image_embeddings(self, images: IMAGES) -> List[List[List[float]]]:
+        """
+        Generate embeddings for a batch of images.
+        """
+        torch = attempt_import_or_raise("torch", "torch")
+        pil_images = self._prepare_images(images)
+        all_embeddings = []
+
+        for i in range(0, len(pil_images), self.batch_size):
+            batch_images = pil_images[i : i + self.batch_size]
+            batch_images = self._processor.process_images(batch_images).to(
+                self._model.device
+            )
+            with torch.no_grad():
+                image_embeddings = self._model(**batch_images)
+            all_embeddings.extend(self._process_embeddings(image_embeddings))
+        return all_embeddings
+
+    def compute_query_embeddings(
+        self, query: Union[str, IMAGES], *args, **kwargs
+    ) -> List[List[List[float]]]:
+        """
+        Compute embeddings for a single user query (text only).
+        """
+        if not isinstance(query, str):
+            raise ValueError(
+                "Query must be a string, image to image search is not supported"
+            )
+        return self.generate_text_embeddings([query])
+
+    def compute_source_embeddings(
+        self, images: IMAGES, *args, **kwargs
+    ) -> List[List[List[float]]]:
+        """
+        Compute embeddings for a batch of source images.
+
+        Parameters
+        ----------
+        images : Union[str, bytes, List, pa.Array, pa.ChunkedArray, np.ndarray]
+            Batch of images (paths, URLs, bytes, or PIL Images).
+        """
+        images = self.sanitize_input(images)
+        return self.generate_image_embeddings(images)

python/python/lancedb/embeddings/utils.py

@@ -18,6 +18,7 @@ import numpy as np
 import pyarrow as pa
 
 from ..dependencies import pandas as pd
+from ..util import attempt_import_or_raise
 
 
 # ruff: noqa: PERF203
@@ -275,3 +276,12 @@ def url_retrieve(url: str):
 def api_key_not_found_help(provider):
     logging.error("Could not find API key for %s", provider)
     raise ValueError(f"Please set the {provider.upper()}_API_KEY environment variable.")
+
+
+def is_flash_attn_2_available():
+    try:
+        attempt_import_or_raise("flash_attn", "flash_attn")
+
+        return True
+    except ImportError:
+        return False

python/python/lancedb/embeddings/voyageai.py

@@ -1,9 +1,12 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-
+import base64
 import os
-from typing import ClassVar, TYPE_CHECKING, List, Union
+from typing import ClassVar, TYPE_CHECKING, List, Union, Any
+
+from pathlib import Path
+from urllib.parse import urlparse
+from io import BytesIO
 
 import numpy as np
 import pyarrow as pa
@@ -11,12 +14,100 @@ import pyarrow as pa
 from ..util import attempt_import_or_raise
 from .base import EmbeddingFunction
 from .registry import register
-from .utils import api_key_not_found_help, IMAGES
+from .utils import api_key_not_found_help, IMAGES, TEXT
 
 if TYPE_CHECKING:
     import PIL
 
 
+def is_valid_url(text):
+    try:
+        parsed = urlparse(text)
+        return bool(parsed.scheme) and bool(parsed.netloc)
+    except Exception:
+        return False
+
+
+def transform_input(input_data: Union[str, bytes, Path]):
+    PIL = attempt_import_or_raise("PIL", "pillow")
+    if isinstance(input_data, str):
+        if is_valid_url(input_data):
+            content = {"type": "image_url", "image_url": input_data}
+        else:
+            content = {"type": "text", "text": input_data}
+    elif isinstance(input_data, PIL.Image.Image):
+        buffered = BytesIO()
+        input_data.save(buffered, format="JPEG")
+        img_str = base64.b64encode(buffered.getvalue()).decode("utf-8")
+        content = {
+            "type": "image_base64",
+            "image_base64": "data:image/jpeg;base64," + img_str,
+        }
+    elif isinstance(input_data, bytes):
+        img = PIL.Image.open(BytesIO(input_data))
+        buffered = BytesIO()
+        img.save(buffered, format="JPEG")
+        img_str = base64.b64encode(buffered.getvalue()).decode("utf-8")
+        content = {
+            "type": "image_base64",
+            "image_base64": "data:image/jpeg;base64," + img_str,
+        }
+    elif isinstance(input_data, Path):
+        img = PIL.Image.open(input_data)
+        buffered = BytesIO()
+        img.save(buffered, format="JPEG")
+        img_str = base64.b64encode(buffered.getvalue()).decode("utf-8")
+        content = {
+            "type": "image_base64",
+            "image_base64": "data:image/jpeg;base64," + img_str,
+        }
+    else:
+        raise ValueError("Each input should be either str, bytes, Path or Image.")
+
+    return {"content": [content]}
+
+
+def sanitize_multimodal_input(inputs: Union[TEXT, IMAGES]) -> List[Any]:
+    """
+    Sanitize the input to the embedding function.
+    """
+    PIL = attempt_import_or_raise("PIL", "pillow")
+    if isinstance(inputs, (str, bytes, Path, PIL.Image.Image)):
+        inputs = [inputs]
+    elif isinstance(inputs, pa.Array):
+        inputs = inputs.to_pylist()
+    elif isinstance(inputs, pa.ChunkedArray):
+        inputs = inputs.combine_chunks().to_pylist()
+    else:
+        raise ValueError(
+            f"Input type {type(inputs)} not allowed with multimodal model."
+        )
+
+    if not all(isinstance(x, (str, bytes, Path, PIL.Image.Image)) for x in inputs):
+        raise ValueError("Each input should be either str, bytes, Path or Image.")
+
+    return [transform_input(i) for i in inputs]
+
+
+def sanitize_text_input(inputs: TEXT) -> List[str]:
+    """
+    Sanitize the input to the embedding function.
+    """
+    if isinstance(inputs, str):
+        inputs = [inputs]
+    elif isinstance(inputs, pa.Array):
+        inputs = inputs.to_pylist()
+    elif isinstance(inputs, pa.ChunkedArray):
+        inputs = inputs.combine_chunks().to_pylist()
+    else:
+        raise ValueError(f"Input type {type(inputs)} not allowed with text model.")
+
+    if not all(isinstance(x, str) for x in inputs):
+        raise ValueError("Each input should be str.")
+
+    return inputs
+
+
 @register("voyageai")
 class VoyageAIEmbeddingFunction(EmbeddingFunction):
     """
@@ -74,6 +165,11 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
     ]
     multimodal_embedding_models: list = ["voyage-multimodal-3"]
 
+    def _is_multimodal_model(self, model_name: str):
+        return (
+            model_name in self.multimodal_embedding_models or "multimodal" in model_name
+        )
+
     def ndims(self):
         if self.name == "voyage-3-lite":
             return 512
@@ -85,55 +181,12 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
             "voyage-finance-2",
             "voyage-multilingual-2",
             "voyage-law-2",
+            "voyage-multimodal-3",
         ]:
             return 1024
         else:
             raise ValueError(f"Model {self.name} not supported")
 
-    def sanitize_input(self, images: IMAGES) -> Union[List[bytes], np.ndarray]:
-        """
-        Sanitize the input to the embedding function.
-        """
-        if isinstance(images, (str, bytes)):
-            images = [images]
-        elif isinstance(images, pa.Array):
-            images = images.to_pylist()
-        elif isinstance(images, pa.ChunkedArray):
-            images = images.combine_chunks().to_pylist()
-        return images
-
-    def generate_text_embeddings(self, text: str, **kwargs) -> np.ndarray:
-        """
-        Get the embeddings for the given texts
-
-        Parameters
-        ----------
-        texts: list[str] or np.ndarray (of str)
-            The texts to embed
-        input_type: Optional[str]
-
-        truncation: Optional[bool]
-        """
-        client = VoyageAIEmbeddingFunction._get_client()
-        if self.name in self.text_embedding_models:
-            rs = client.embed(texts=[text], model=self.name, **kwargs)
-        elif self.name in self.multimodal_embedding_models:
-            rs = client.multimodal_embed(inputs=[[text]], model=self.name, **kwargs)
-        else:
-            raise ValueError(
-                f"Model {self.name} not supported to generate text embeddings"
-            )
-
-        return rs.embeddings[0]
-
-    def generate_image_embedding(
-        self, image: "PIL.Image.Image", **kwargs
-    ) -> np.ndarray:
-        rs = VoyageAIEmbeddingFunction._get_client().multimodal_embed(
-            inputs=[[image]], model=self.name, **kwargs
-        )
-        return rs.embeddings[0]
-
     def compute_query_embeddings(
         self, query: Union[str, "PIL.Image.Image"], *args, **kwargs
     ) -> List[np.ndarray]:
@@ -144,23 +197,52 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
         ----------
         query : Union[str, PIL.Image.Image]
             The query to embed. A query can be either text or an image.
+
+        Returns
+        -------
+            List[np.array]: the list of embeddings
         """
-        if isinstance(query, str):
-            return [self.generate_text_embeddings(query, input_type="query")]
+        client = VoyageAIEmbeddingFunction._get_client()
+        if self._is_multimodal_model(self.name):
+            result = client.multimodal_embed(
+                inputs=[[query]], model=self.name, input_type="query", **kwargs
+            )
         else:
-            PIL = attempt_import_or_raise("PIL", "pillow")
-            if isinstance(query, PIL.Image.Image):
-                return [self.generate_image_embedding(query, input_type="query")]
-            else:
-                raise TypeError("Only text PIL images supported as query")
+            result = client.embed(
+                texts=[query], model=self.name, input_type="query", **kwargs
+            )
+
+        return [result.embeddings[0]]
 
     def compute_source_embeddings(
-        self, images: IMAGES, *args, **kwargs
+        self, inputs: Union[TEXT, IMAGES], *args, **kwargs
     ) -> List[np.array]:
-        images = self.sanitize_input(images)
-        return [
-            self.generate_image_embedding(img, input_type="document") for img in images
-        ]
+        """
+        Compute the embeddings for the inputs
+
+        Parameters
+        ----------
+        inputs : Union[TEXT, IMAGES]
+            The inputs to embed. The input can be either str, bytes, Path (to an image),
+            PIL.Image or list of these.
+
+        Returns
+        -------
+            List[np.array]: the list of embeddings
+        """
+        client = VoyageAIEmbeddingFunction._get_client()
+        if self._is_multimodal_model(self.name):
+            inputs = sanitize_multimodal_input(inputs)
+            result = client.multimodal_embed(
+                inputs=inputs, model=self.name, input_type="document", **kwargs
+            )
+        else:
+            inputs = sanitize_text_input(inputs)
+            result = client.embed(
+                texts=inputs, model=self.name, input_type="document", **kwargs
+            )
+
+        return result.embeddings
 
     @staticmethod
     def _get_client():

python/python/lancedb/pydantic.py

@@ -152,6 +152,104 @@ def Vector(
     return FixedSizeList
 
 
+def MultiVector(
+    dim: int, value_type: pa.DataType = pa.float32(), nullable: bool = True
+) -> Type:
+    """Pydantic MultiVector Type for multi-vector embeddings.
+
+    This type represents a list of vectors, each with the same dimension.
+    Useful for models that produce multiple embeddings per input, like ColPali.
+
+    Parameters
+    ----------
+    dim : int
+        The dimension of each vector in the multi-vector.
+    value_type : pyarrow.DataType, optional
+        The value type of the vectors, by default pa.float32()
+    nullable : bool, optional
+        Whether the multi-vector is nullable, by default it is True.
+
+    Examples
+    --------
+
+    >>> import pydantic
+    >>> from lancedb.pydantic import MultiVector
+    ...
+    >>> class MyModel(pydantic.BaseModel):
+    ...     id: int
+    ...     text: str
+    ...     embeddings: MultiVector(128)  # List of 128-dimensional vectors
+    >>> schema = pydantic_to_schema(MyModel)
+    >>> assert schema == pa.schema([
+    ...     pa.field("id", pa.int64(), False),
+    ...     pa.field("text", pa.utf8(), False),
+    ...     pa.field("embeddings", pa.list_(pa.list_(pa.float32(), 128)))
+    ... ])
+    """
+
+    class MultiVectorList(list, FixedSizeListMixin):
+        def __repr__(self):
+            return f"MultiVector(dim={dim})"
+
+        @staticmethod
+        def nullable() -> bool:
+            return nullable
+
+        @staticmethod
+        def dim() -> int:
+            return dim
+
+        @staticmethod
+        def value_arrow_type() -> pa.DataType:
+            return value_type
+
+        @staticmethod
+        def is_multi_vector() -> bool:
+            return True
+
+        @classmethod
+        def __get_pydantic_core_schema__(
+            cls, _source_type: Any, _handler: pydantic.GetCoreSchemaHandler
+        ) -> CoreSchema:
+            return core_schema.no_info_after_validator_function(
+                cls,
+                core_schema.list_schema(
+                    items_schema=core_schema.list_schema(
+                        min_length=dim,
+                        max_length=dim,
+                        items_schema=core_schema.float_schema(),
+                    ),
+                ),
+            )
+
+        @classmethod
+        def __get_validators__(cls) -> Generator[Callable, None, None]:
+            yield cls.validate
+
+        # For pydantic v1
+        @classmethod
+        def validate(cls, v):
+            if not isinstance(v, (list, range)):
+                raise TypeError("A list of vectors is needed")
+            for vec in v:
+                if not isinstance(vec, (list, range, np.ndarray)) or len(vec) != dim:
+                    raise TypeError(f"Each vector must be a list of {dim} numbers")
+            return cls(v)
+
+        if PYDANTIC_VERSION.major < 2:
+
+            @classmethod
+            def __modify_schema__(cls, field_schema: Dict[str, Any]):
+                field_schema["items"] = {
+                    "type": "array",
+                    "items": {"type": "number"},
+                    "minItems": dim,
+                    "maxItems": dim,
+                }
+
+    return MultiVectorList
+
+
 def _py_type_to_arrow_type(py_type: Type[Any], field: FieldInfo) -> pa.DataType:
     """Convert a field with native Python type to Arrow data type.
 
@@ -206,6 +304,9 @@ def _pydantic_type_to_arrow_type(tp: Any, field: FieldInfo) -> pa.DataType:
             fields = _pydantic_model_to_fields(tp)
             return pa.struct(fields)
         if issubclass(tp, FixedSizeListMixin):
+            if getattr(tp, "is_multi_vector", lambda: False)():
+                return pa.list_(pa.list_(tp.value_arrow_type(), tp.dim()))
+            # For regular Vector
             return pa.list_(tp.value_arrow_type(), tp.dim())
     return _py_type_to_arrow_type(tp, field)
 

python/python/lancedb/query.py

@@ -7,6 +7,7 @@ from abc import ABC, abstractmethod
 import abc
 from concurrent.futures import ThreadPoolExecutor
 from enum import Enum
+from datetime import timedelta
 from typing import (
     TYPE_CHECKING,
     Dict,
@@ -27,6 +28,8 @@ import pyarrow.compute as pc
 import pyarrow.fs as pa_fs
 import pydantic
 
+from lancedb.pydantic import PYDANTIC_VERSION
+
 from . import __version__
 from .arrow import AsyncRecordBatchReader
 from .dependencies import pandas as pd
@@ -117,6 +120,12 @@ class FullTextQuery(abc.ABC, pydantic.BaseModel):
 
 
 class MatchQuery(FullTextQuery):
+    query: str
+    column: str
+    boost: float = 1.0
+    fuzziness: int = 0
+    max_expansions: int = 50
+
     def __init__(
         self,
         query: str,
@@ -149,11 +158,13 @@ class MatchQuery(FullTextQuery):
             The maximum number of terms to consider for fuzzy matching.
             Defaults to 50.
         """
-        self.column = column
-        self.query = query
-        self.boost = boost
-        self.fuzziness = fuzziness
-        self.max_expansions = max_expansions
+        super().__init__(
+            query=query,
+            column=column,
+            boost=boost,
+            fuzziness=fuzziness,
+            max_expansions=max_expansions,
+        )
 
     def query_type(self) -> FullTextQueryType:
         return FullTextQueryType.MATCH
@@ -172,6 +183,9 @@ class MatchQuery(FullTextQuery):
 
 
 class PhraseQuery(FullTextQuery):
+    query: str
+    column: str
+
     def __init__(self, query: str, column: str):
         """
         Phrase query for full-text search.
@@ -183,8 +197,7 @@ class PhraseQuery(FullTextQuery):
         column : str
             The name of the column to match against.
         """
-        self.column = column
-        self.query = query
+        super().__init__(query=query, column=column)
 
     def query_type(self) -> FullTextQueryType:
         return FullTextQueryType.MATCH_PHRASE
@@ -198,11 +211,16 @@ class PhraseQuery(FullTextQuery):
 
 
 class BoostQuery(FullTextQuery):
+    positive: FullTextQuery
+    negative: FullTextQuery
+    negative_boost: float = 0.5
+
     def __init__(
         self,
         positive: FullTextQuery,
         negative: FullTextQuery,
-        negative_boost: float,
+        *,
+        negative_boost: float = 0.5,
     ):
         """
         Boost query for full-text search.
@@ -216,9 +234,9 @@ class BoostQuery(FullTextQuery):
         negative_boost : float
             The boost factor for the negative query.
         """
-        self.positive = positive
-        self.negative = negative
-        self.negative_boost = negative_boost
+        super().__init__(
+            positive=positive, negative=negative, negative_boost=negative_boost
+        )
 
     def query_type(self) -> FullTextQueryType:
         return FullTextQueryType.BOOST
@@ -234,6 +252,10 @@ class BoostQuery(FullTextQuery):
 
 
 class MultiMatchQuery(FullTextQuery):
+    query: str
+    columns: list[str]
+    boosts: list[float]
+
     def __init__(
         self,
         query: str,
@@ -246,8 +268,8 @@ class MultiMatchQuery(FullTextQuery):
 
         Parameters
         ----------
-        query : str | list[Query]
-            If a string, the query string to match against.
+        query : str
+            The query string to match against.
 
         columns : list[str]
             The list of columns to match against.
@@ -256,11 +278,9 @@ class MultiMatchQuery(FullTextQuery):
             The list of boost factors for each column. If not provided,
             all columns will have the same boost factor.
         """
-        self.query = query
-        self.columns = columns
         if boosts is None:
             boosts = [1.0] * len(columns)
-        self.boosts = boosts
+        super().__init__(query=query, columns=columns, boosts=boosts)
 
     def query_type(self) -> FullTextQueryType:
         return FullTextQueryType.MULTI_MATCH
@@ -480,10 +500,14 @@ class Query(pydantic.BaseModel):
             )
         return query
 
-    class Config:
-        # This tells pydantic to allow custom types (needed for the `vector` query since
-        # pa.Array wouln't be allowed otherwise)
-        arbitrary_types_allowed = True
+    # This tells pydantic to allow custom types (needed for the `vector` query since
+    # pa.Array wouln't be allowed otherwise)
+    if PYDANTIC_VERSION.major < 2:  # Pydantic 1.x compat
+
+        class Config:
+            arbitrary_types_allowed = True
+    else:
+        model_config = {"arbitrary_types_allowed": True}
 
 
 class LanceQueryBuilder(ABC):
@@ -544,7 +568,7 @@ class LanceQueryBuilder(ABC):
                 table, query, vector_column_name, fts_columns=fts_columns
             )
 
-        if isinstance(query, str):
+        if isinstance(query, (str, FullTextQuery)):
             # fts
             return LanceFtsQueryBuilder(
                 table,
@@ -569,8 +593,10 @@ class LanceQueryBuilder(ABC):
         # If query_type is fts, then query must be a string.
         # otherwise raise TypeError
         if query_type == "fts":
-            if not isinstance(query, str):
-                raise TypeError(f"'fts' queries must be a string: {type(query)}")
+            if not isinstance(query, (str, FullTextQuery)):
+                raise TypeError(
+                    f"'fts' query must be a string or FullTextQuery: {type(query)}"
+                )
             return query, query_type
         elif query_type == "vector":
             query = cls._query_to_vector(table, query, vector_column_name)
@@ -631,7 +657,12 @@ class LanceQueryBuilder(ABC):
         """
         return self.to_pandas()
 
-    def to_pandas(self, flatten: Optional[Union[int, bool]] = None) -> "pd.DataFrame":
+    def to_pandas(
+        self,
+        flatten: Optional[Union[int, bool]] = None,
+        *,
+        timeout: Optional[timedelta] = None,
+    ) -> "pd.DataFrame":
         """
         Execute the query and return the results as a pandas DataFrame.
         In addition to the selected columns, LanceDB also returns a vector
@@ -645,12 +676,15 @@ class LanceQueryBuilder(ABC):
             If flatten is an integer, flatten the nested columns up to the
             specified depth.
             If unspecified, do not flatten the nested columns.
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If None, wait indefinitely.
         """
-        tbl = flatten_columns(self.to_arrow(), flatten)
+        tbl = flatten_columns(self.to_arrow(timeout=timeout), flatten)
         return tbl.to_pandas()
 
     @abstractmethod
-    def to_arrow(self) -> pa.Table:
+    def to_arrow(self, *, timeout: Optional[timedelta] = None) -> pa.Table:
         """
         Execute the query and return the results as an
         [Apache Arrow Table](https://arrow.apache.org/docs/python/generated/pyarrow.Table.html#pyarrow.Table).
@@ -658,34 +692,65 @@ class LanceQueryBuilder(ABC):
         In addition to the selected columns, LanceDB also returns a vector
         and also the "_distance" column which is the distance between the query
         vector and the returned vectors.
+
+        Parameters
+        ----------
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If None, wait indefinitely.
         """
         raise NotImplementedError
 
     @abstractmethod
-    def to_batches(self, /, batch_size: Optional[int] = None) -> pa.RecordBatchReader:
+    def to_batches(
+        self,
+        /,
+        batch_size: Optional[int] = None,
+        *,
+        timeout: Optional[timedelta] = None,
+    ) -> pa.RecordBatchReader:
         """
         Execute the query and return the results as a pyarrow
         [RecordBatchReader](https://arrow.apache.org/docs/python/generated/pyarrow.RecordBatchReader.html)
+
+        Parameters
+        ----------
+        batch_size: int
+            The maximum number of selected records in a RecordBatch object.
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If None, wait indefinitely.
         """
         raise NotImplementedError
 
-    def to_list(self) -> List[dict]:
+    def to_list(self, *, timeout: Optional[timedelta] = None) -> List[dict]:
         """
         Execute the query and return the results as a list of dictionaries.
 
         Each list entry is a dictionary with the selected column names as keys,
         or all table columns if `select` is not called. The vector and the "_distance"
         fields are returned whether or not they're explicitly selected.
-        """
-        return self.to_arrow().to_pylist()
 
-    def to_pydantic(self, model: Type[LanceModel]) -> List[LanceModel]:
+        Parameters
+        ----------
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If None, wait indefinitely.
+        """
+        return self.to_arrow(timeout=timeout).to_pylist()
+
+    def to_pydantic(
+        self, model: Type[LanceModel], *, timeout: Optional[timedelta] = None
+    ) -> List[LanceModel]:
         """Return the table as a list of pydantic models.
 
         Parameters
         ----------
         model: Type[LanceModel]
             The pydantic model to use.
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If None, wait indefinitely.
 
         Returns
         -------
@@ -693,19 +758,25 @@ class LanceQueryBuilder(ABC):
         """
         return [
             model(**{k: v for k, v in row.items() if k in model.field_names()})
-            for row in self.to_arrow().to_pylist()
+            for row in self.to_arrow(timeout=timeout).to_pylist()
         ]
 
-    def to_polars(self) -> "pl.DataFrame":
+    def to_polars(self, *, timeout: Optional[timedelta] = None) -> "pl.DataFrame":
         """
         Execute the query and return the results as a Polars DataFrame.
         In addition to the selected columns, LanceDB also returns a vector
         and also the "_distance" column which is the distance between the query
         vector and the returned vector.
+
+        Parameters
+        ----------
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If None, wait indefinitely.
         """
         import polars as pl
 
-        return pl.from_arrow(self.to_arrow())
+        return pl.from_arrow(self.to_arrow(timeout=timeout))
 
     def limit(self, limit: Union[int, None]) -> Self:
         """Set the maximum number of results to return.
@@ -1120,7 +1191,7 @@ class LanceVectorQueryBuilder(LanceQueryBuilder):
         self._refine_factor = refine_factor
         return self
 
-    def to_arrow(self) -> pa.Table:
+    def to_arrow(self, *, timeout: Optional[timedelta] = None) -> pa.Table:
         """
         Execute the query and return the results as an
         [Apache Arrow Table](https://arrow.apache.org/docs/python/generated/pyarrow.Table.html#pyarrow.Table).
@@ -1128,8 +1199,14 @@ class LanceVectorQueryBuilder(LanceQueryBuilder):
         In addition to the selected columns, LanceDB also returns a vector
         and also the "_distance" column which is the distance between the query
         vector and the returned vectors.
+
+        Parameters
+        ----------
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If None, wait indefinitely.
         """
-        return self.to_batches().read_all()
+        return self.to_batches(timeout=timeout).read_all()
 
     def to_query_object(self) -> Query:
         """
@@ -1159,7 +1236,13 @@ class LanceVectorQueryBuilder(LanceQueryBuilder):
             bypass_vector_index=self._bypass_vector_index,
         )
 
-    def to_batches(self, /, batch_size: Optional[int] = None) -> pa.RecordBatchReader:
+    def to_batches(
+        self,
+        /,
+        batch_size: Optional[int] = None,
+        *,
+        timeout: Optional[timedelta] = None,
+    ) -> pa.RecordBatchReader:
         """
         Execute the query and return the result as a RecordBatchReader object.
 
@@ -1167,6 +1250,9 @@ class LanceVectorQueryBuilder(LanceQueryBuilder):
         ----------
         batch_size: int
             The maximum number of selected records in a RecordBatch object.
+        timeout: timedelta, default None
+            The maximum time to wait for the query to complete.
+            If None, wait indefinitely.
 
         Returns
         -------
@@ -1176,7 +1262,9 @@ class LanceVectorQueryBuilder(LanceQueryBuilder):
         if isinstance(vector[0], np.ndarray):
             vector = [v.tolist() for v in vector]
         query = self.to_query_object()
-        result_set = self._table._execute_query(query, batch_size)
+        result_set = self._table._execute_query(
+            query, batch_size=batch_size, timeout=timeout
+        )
         if self._reranker is not None:
             rs_table = result_set.read_all()
             result_set = self._reranker.rerank_vector(self._str_query, rs_table)
@@ -1315,7 +1403,7 @@ class LanceFtsQueryBuilder(LanceQueryBuilder):
             offset=self._offset,
         )
 
-    def to_arrow(self) -> pa.Table:
+    def to_arrow(self, *, timeout: Optional[timedelta] = None) -> pa.Table:
         path, fs, exist = self._table._get_fts_index_path()
         if exist:
             return self.tantivy_to_arrow()
@@ -1327,14 +1415,16 @@ class LanceFtsQueryBuilder(LanceQueryBuilder):
                 "Use tantivy-based index instead for now."
             )
         query = self.to_query_object()
-        results = self._table._execute_query(query)
+        results = self._table._execute_query(query, timeout=timeout)
         results = results.read_all()
         if self._reranker is not None:
             results = self._reranker.rerank_fts(self._query, results)
             check_reranker_result(results)
         return results
 
-    def to_batches(self, /, batch_size: Optional[int] = None):
+    def to_batches(
+        self, /, batch_size: Optional[int] = None, timeout: Optional[timedelta] = None
+    ):
         raise NotImplementedError("to_batches on an FTS query")
 
     def tantivy_to_arrow(self) -> pa.Table:
@@ -1439,8 +1529,8 @@ class LanceFtsQueryBuilder(LanceQueryBuilder):
 
 
 class LanceEmptyQueryBuilder(LanceQueryBuilder):
-    def to_arrow(self) -> pa.Table:
-        return self.to_batches().read_all()
+    def to_arrow(self, *, timeout: Optional[timedelta] = None) -> pa.Table:
+        return self.to_batches(timeout=timeout).read_all()
 
     def to_query_object(self) -> Query:
         return Query(
@@ -1451,9 +1541,11 @@ class LanceEmptyQueryBuilder(LanceQueryBuilder):
             offset=self._offset,
         )
 
-    def to_batches(self, /, batch_size: Optional[int] = None) -> pa.RecordBatchReader:
+    def to_batches(
+        self, /, batch_size: Optional[int] = None, timeout: Optional[timedelta] = None
+    ) -> pa.RecordBatchReader:
         query = self.to_query_object()
-        return self._table._execute_query(query, batch_size)
+        return self._table._execute_query(query, batch_size=batch_size, timeout=timeout)
 
     def rerank(self, reranker: Reranker) -> LanceEmptyQueryBuilder:
         """Rerank the results using the specified reranker.
@@ -1486,7 +1578,7 @@ class LanceHybridQueryBuilder(LanceQueryBuilder):
     def __init__(
         self,
         table: "Table",
-        query: Optional[str] = None,
+        query: Optional[Union[str, FullTextQuery]] = None,
         vector_column: Optional[str] = None,
         fts_columns: Optional[Union[str, List[str]]] = None,
     ):
@@ -1500,6 +1592,8 @@ class LanceHybridQueryBuilder(LanceQueryBuilder):
         self._refine_factor = None
         self._distance_type = None
         self._phrase_query = None
+        self._lower_bound = None
+        self._upper_bound = None
 
     def _validate_query(self, query, vector=None, text=None):
         if query is not None and (vector is not None or text is not None):
@@ -1516,8 +1610,8 @@ class LanceHybridQueryBuilder(LanceQueryBuilder):
         text_query = text or query
         if text_query is None:
             raise ValueError("Text query must be provided for hybrid search.")
-        if not isinstance(text_query, str):
-            raise ValueError("Text query must be a string")
+        if not isinstance(text_query, (str, FullTextQuery)):
+            raise ValueError("Text query must be a string or FullTextQuery")
 
         return vector_query, text_query
 
@@ -1541,7 +1635,7 @@ class LanceHybridQueryBuilder(LanceQueryBuilder):
     def to_query_object(self) -> Query:
         raise NotImplementedError("to_query_object not yet supported on a hybrid query")
 
-    def to_arrow(self) -> pa.Table:
+    def to_arrow(self, *, timeout: Optional[timedelta] = None) -> pa.Table:
         vector_query, fts_query = self._validate_query(
             self._query, self._vector, self._text
         )
@@ -1579,14 +1673,20 @@ class LanceHybridQueryBuilder(LanceQueryBuilder):
             self._vector_query.ef(self._ef)
         if self._bypass_vector_index:
             self._vector_query.bypass_vector_index()
+        if self._lower_bound or self._upper_bound:
+            self._vector_query.distance_range(
+                lower_bound=self._lower_bound, upper_bound=self._upper_bound
+            )
 
         if self._reranker is None:
             self._reranker = RRFReranker()
 
         with ThreadPoolExecutor() as executor:
-            fts_future = executor.submit(self._fts_query.with_row_id(True).to_arrow)
+            fts_future = executor.submit(
+                self._fts_query.with_row_id(True).to_arrow, timeout=timeout
+            )
             vector_future = executor.submit(
-                self._vector_query.with_row_id(True).to_arrow
+                self._vector_query.with_row_id(True).to_arrow, timeout=timeout
             )
             fts_results = fts_future.result()
             vector_results = vector_future.result()
@@ -1673,7 +1773,9 @@ class LanceHybridQueryBuilder(LanceQueryBuilder):
 
         return results
 
-    def to_batches(self):
+    def to_batches(
+        self, /, batch_size: Optional[int] = None, timeout: Optional[timedelta] = None
+    ):
         raise NotImplementedError("to_batches not yet supported on a hybrid query")
 
     @staticmethod
@@ -2037,7 +2139,10 @@ class AsyncQueryBase(object):
         return self
 
     async def to_batches(
-        self, *, max_batch_length: Optional[int] = None
+        self,
+        *,
+        max_batch_length: Optional[int] = None,
+        timeout: Optional[timedelta] = None,
     ) -> AsyncRecordBatchReader:
         """
         Execute the query and return the results as an Apache Arrow RecordBatchReader.
@@ -2050,34 +2155,56 @@ class AsyncQueryBase(object):
             If not specified, a default batch length is used.
             It is possible for batches to be smaller than the provided length if the
             underlying data is stored in smaller chunks.
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If not specified, no timeout is applied. If the query does not
+            complete within the specified time, an error will be raised.
         """
-        return AsyncRecordBatchReader(await self._inner.execute(max_batch_length))
+        return AsyncRecordBatchReader(
+            await self._inner.execute(max_batch_length, timeout)
+        )
 
-    async def to_arrow(self) -> pa.Table:
+    async def to_arrow(self, timeout: Optional[timedelta] = None) -> pa.Table:
         """
         Execute the query and collect the results into an Apache Arrow Table.
 
         This method will collect all results into memory before returning.  If
         you expect a large number of results, you may want to use
         [to_batches][lancedb.query.AsyncQueryBase.to_batches]
+
+        Parameters
+        ----------
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If not specified, no timeout is applied. If the query does not
+            complete within the specified time, an error will be raised.
         """
-        batch_iter = await self.to_batches()
+        batch_iter = await self.to_batches(timeout=timeout)
         return pa.Table.from_batches(
             await batch_iter.read_all(), schema=batch_iter.schema
         )
 
-    async def to_list(self) -> List[dict]:
+    async def to_list(self, timeout: Optional[timedelta] = None) -> List[dict]:
         """
         Execute the query and return the results as a list of dictionaries.
 
         Each list entry is a dictionary with the selected column names as keys,
         or all table columns if `select` is not called. The vector and the "_distance"
         fields are returned whether or not they're explicitly selected.
+
+        Parameters
+        ----------
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If not specified, no timeout is applied. If the query does not
+            complete within the specified time, an error will be raised.
         """
-        return (await self.to_arrow()).to_pylist()
+        return (await self.to_arrow(timeout=timeout)).to_pylist()
 
     async def to_pandas(
-        self, flatten: Optional[Union[int, bool]] = None
+        self,
+        flatten: Optional[Union[int, bool]] = None,
+        timeout: Optional[timedelta] = None,
     ) -> "pd.DataFrame":
         """
         Execute the query and collect the results into a pandas DataFrame.
@@ -2106,10 +2233,19 @@ class AsyncQueryBase(object):
             If flatten is an integer, flatten the nested columns up to the
             specified depth.
             If unspecified, do not flatten the nested columns.
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If not specified, no timeout is applied. If the query does not
+            complete within the specified time, an error will be raised.
         """
-        return (flatten_columns(await self.to_arrow(), flatten)).to_pandas()
+        return (
+            flatten_columns(await self.to_arrow(timeout=timeout), flatten)
+        ).to_pandas()
 
-    async def to_polars(self) -> "pl.DataFrame":
+    async def to_polars(
+        self,
+        timeout: Optional[timedelta] = None,
+    ) -> "pl.DataFrame":
         """
         Execute the query and collect the results into a Polars DataFrame.
 
@@ -2118,6 +2254,13 @@ class AsyncQueryBase(object):
         [to_batches][lancedb.query.AsyncQueryBase.to_batches] and convert each batch to
         polars separately.
 
+        Parameters
+        ----------
+        timeout: Optional[timedelta]
+            The maximum time to wait for the query to complete.
+            If not specified, no timeout is applied. If the query does not
+            complete within the specified time, an error will be raised.
+
         Examples
         --------
 
@@ -2133,7 +2276,7 @@ class AsyncQueryBase(object):
         """
         import polars as pl
 
-        return pl.from_arrow(await self.to_arrow())
+        return pl.from_arrow(await self.to_arrow(timeout=timeout))
 
     async def explain_plan(self, verbose: Optional[bool] = False):
         """Return the execution plan for this query.
@@ -2308,7 +2451,7 @@ class AsyncQuery(AsyncQueryBase):
                 self._inner.nearest_to_text({"query": query, "columns": columns})
             )
         # FullTextQuery object
-        return AsyncFTSQuery(self._inner.nearest_to_text(query.to_dict()))
+        return AsyncFTSQuery(self._inner.nearest_to_text({"query": query.to_dict()}))
 
 
 class AsyncFTSQuery(AsyncQueryBase):
@@ -2404,9 +2547,12 @@ class AsyncFTSQuery(AsyncQueryBase):
             )
 
     async def to_batches(
-        self, *, max_batch_length: Optional[int] = None
+        self,
+        *,
+        max_batch_length: Optional[int] = None,
+        timeout: Optional[timedelta] = None,
     ) -> AsyncRecordBatchReader:
-        reader = await super().to_batches()
+        reader = await super().to_batches(timeout=timeout)
         results = pa.Table.from_batches(await reader.read_all(), reader.schema)
         if self._reranker:
             results = self._reranker.rerank_fts(self.get_query(), results)
@@ -2627,12 +2773,15 @@ class AsyncVectorQuery(AsyncQueryBase, AsyncVectorQueryBase):
                 self._inner.nearest_to_text({"query": query, "columns": columns})
             )
         # FullTextQuery object
-        return AsyncHybridQuery(self._inner.nearest_to_text(query.to_dict()))
+        return AsyncHybridQuery(self._inner.nearest_to_text({"query": query.to_dict()}))
 
     async def to_batches(
-        self, *, max_batch_length: Optional[int] = None
+        self,
+        *,
+        max_batch_length: Optional[int] = None,
+        timeout: Optional[timedelta] = None,
     ) -> AsyncRecordBatchReader:
-        reader = await super().to_batches()
+        reader = await super().to_batches(timeout=timeout)
         results = pa.Table.from_batches(await reader.read_all(), reader.schema)
         if self._reranker:
             results = self._reranker.rerank_vector(self._query_string, results)
@@ -2688,7 +2837,10 @@ class AsyncHybridQuery(AsyncQueryBase, AsyncVectorQueryBase):
         return self
 
     async def to_batches(
-        self, *, max_batch_length: Optional[int] = None
+        self,
+        *,
+        max_batch_length: Optional[int] = None,
+        timeout: Optional[timedelta] = None,
     ) -> AsyncRecordBatchReader:
         fts_query = AsyncFTSQuery(self._inner.to_fts_query())
         vec_query = AsyncVectorQuery(self._inner.to_vector_query())
@@ -2700,8 +2852,8 @@ class AsyncHybridQuery(AsyncQueryBase, AsyncVectorQueryBase):
         vec_query.with_row_id()
 
         fts_results, vector_results = await asyncio.gather(
-            fts_query.to_arrow(),
-            vec_query.to_arrow(),
+            fts_query.to_arrow(timeout=timeout),
+            vec_query.to_arrow(timeout=timeout),
         )
 
         result = LanceHybridQueryBuilder._combine_hybrid_results(

python/python/lancedb/remote/table.py

@@ -104,6 +104,7 @@ class RemoteTable(Table):
         index_type: Literal["BTREE", "BITMAP", "LABEL_LIST", "scalar"] = "scalar",
         *,
         replace: bool = False,
+        wait_timeout: timedelta = None,
     ):
         """Creates a scalar index
         Parameters
@@ -126,13 +127,18 @@ class RemoteTable(Table):
         else:
             raise ValueError(f"Unknown index type: {index_type}")
 
-        LOOP.run(self._table.create_index(column, config=config, replace=replace))
+        LOOP.run(
+            self._table.create_index(
+                column, config=config, replace=replace, wait_timeout=wait_timeout
+            )
+        )
 
     def create_fts_index(
         self,
         column: str,
         *,
         replace: bool = False,
+        wait_timeout: timedelta = None,
         with_position: bool = True,
         # tokenizer configs:
         base_tokenizer: str = "simple",
@@ -153,7 +159,11 @@ class RemoteTable(Table):
             remove_stop_words=remove_stop_words,
             ascii_folding=ascii_folding,
         )
-        LOOP.run(self._table.create_index(column, config=config, replace=replace))
+        LOOP.run(
+            self._table.create_index(
+                column, config=config, replace=replace, wait_timeout=wait_timeout
+            )
+        )
 
     def create_index(
         self,
@@ -165,6 +175,7 @@ class RemoteTable(Table):
         replace: Optional[bool] = None,
         accelerator: Optional[str] = None,
         index_type="vector",
+        wait_timeout: Optional[timedelta] = None,
     ):
         """Create an index on the table.
         Currently, the only parameters that matter are
@@ -236,7 +247,11 @@ class RemoteTable(Table):
                 " 'IVF_FLAT', 'IVF_PQ', 'IVF_HNSW_PQ', 'IVF_HNSW_SQ'"
             )
 
-        LOOP.run(self._table.create_index(vector_column_name, config=config))
+        LOOP.run(
+            self._table.create_index(
+                vector_column_name, config=config, wait_timeout=wait_timeout
+            )
+        )
 
     def add(
         self,
@@ -355,9 +370,15 @@ class RemoteTable(Table):
         )
 
     def _execute_query(
-        self, query: Query, batch_size: Optional[int] = None
+        self,
+        query: Query,
+        *,
+        batch_size: Optional[int] = None,
+        timeout: Optional[timedelta] = None,
     ) -> pa.RecordBatchReader:
-        async_iter = LOOP.run(self._table._execute_query(query, batch_size=batch_size))
+        async_iter = LOOP.run(
+            self._table._execute_query(query, batch_size=batch_size, timeout=timeout)
+        )
 
         def iter_sync():
             try:
@@ -548,6 +569,11 @@ class RemoteTable(Table):
     def drop_index(self, index_name: str):
         return LOOP.run(self._table.drop_index(index_name))
 
+    def wait_for_index(
+        self, index_names: Iterable[str], timeout: timedelta = timedelta(seconds=300)
+    ):
+        return LOOP.run(self._table.wait_for_index(index_names, timeout))
+
     def uses_v2_manifest_paths(self) -> bool:
         raise NotImplementedError(
             "uses_v2_manifest_paths() is not supported on the LanceDB Cloud"

python/python/lancedb/rerankers/answerdotai.py

@@ -47,6 +47,9 @@ class AnswerdotaiRerankers(Reranker):
         )
 
     def _rerank(self, result_set: pa.Table, query: str):
+        result_set = self._handle_empty_results(result_set)
+        if len(result_set) == 0:
+            return result_set
         docs = result_set[self.column].to_pylist()
         doc_ids = list(range(len(docs)))
         result = self.reranker.rank(query, docs, doc_ids=doc_ids)
@@ -83,7 +86,6 @@ class AnswerdotaiRerankers(Reranker):
         vector_results = self._rerank(vector_results, query)
         if self.score == "relevance":
             vector_results = vector_results.drop_columns(["_distance"])
-
         vector_results = vector_results.sort_by([("_relevance_score", "descending")])
         return vector_results
 
@@ -91,7 +93,5 @@ class AnswerdotaiRerankers(Reranker):
         fts_results = self._rerank(fts_results, query)
         if self.score == "relevance":
             fts_results = fts_results.drop_columns(["_score"])
-
         fts_results = fts_results.sort_by([("_relevance_score", "descending")])
-
         return fts_results

python/python/lancedb/rerankers/base.py

@@ -65,6 +65,16 @@ class Reranker(ABC):
             f"{self.__class__.__name__} does not implement rerank_vector"
         )
 
+    def _handle_empty_results(self, results: pa.Table):
+        """
+        Helper method to handle empty FTS results consistently
+        """
+        if len(results) > 0:
+            return results
+        return results.append_column(
+            "_relevance_score", pa.array([], type=pa.float32())
+        )
+
     def rerank_fts(
         self,
         query: str,

python/python/lancedb/rerankers/cohere.py

@@ -62,6 +62,9 @@ class CohereReranker(Reranker):
         return cohere.Client(os.environ.get("COHERE_API_KEY") or self.api_key)
 
     def _rerank(self, result_set: pa.Table, query: str):
+        result_set = self._handle_empty_results(result_set)
+        if len(result_set) == 0:
+            return result_set
         docs = result_set[self.column].to_pylist()
         response = self._client.rerank(
             query=query,
@@ -99,24 +102,14 @@ class CohereReranker(Reranker):
             )
         return combined_results
 
-    def rerank_vector(
-        self,
-        query: str,
-        vector_results: pa.Table,
-    ):
-        result_set = self._rerank(vector_results, query)
+    def rerank_vector(self, query: str, vector_results: pa.Table):
+        vector_results = self._rerank(vector_results, query)
         if self.score == "relevance":
-            result_set = result_set.drop_columns(["_distance"])
+            vector_results = vector_results.drop_columns(["_distance"])
+        return vector_results
 
-        return result_set
-
-    def rerank_fts(
-        self,
-        query: str,
-        fts_results: pa.Table,
-    ):
-        result_set = self._rerank(fts_results, query)
+    def rerank_fts(self, query: str, fts_results: pa.Table):
+        fts_results = self._rerank(fts_results, query)
         if self.score == "relevance":
-            result_set = result_set.drop_columns(["_score"])
-
-        return result_set
+            fts_results = fts_results.drop_columns(["_score"])
+        return fts_results

python/python/lancedb/rerankers/cross_encoder.py

@@ -63,6 +63,9 @@ class CrossEncoderReranker(Reranker):
         return cross_encoder
 
     def _rerank(self, result_set: pa.Table, query: str):
+        result_set = self._handle_empty_results(result_set)
+        if len(result_set) == 0:
+            return result_set
         passages = result_set[self.column].to_pylist()
         cross_inp = [[query, passage] for passage in passages]
         cross_scores = self.model.predict(cross_inp)
@@ -93,11 +96,7 @@ class CrossEncoderReranker(Reranker):
 
         return combined_results
 
-    def rerank_vector(
-        self,
-        query: str,
-        vector_results: pa.Table,
-    ):
+    def rerank_vector(self, query: str, vector_results: pa.Table):
         vector_results = self._rerank(vector_results, query)
         if self.score == "relevance":
             vector_results = vector_results.drop_columns(["_distance"])
@@ -105,11 +104,7 @@ class CrossEncoderReranker(Reranker):
         vector_results = vector_results.sort_by([("_relevance_score", "descending")])
         return vector_results
 
-    def rerank_fts(
-        self,
-        query: str,
-        fts_results: pa.Table,
-    ):
+    def rerank_fts(self, query: str, fts_results: pa.Table):
         fts_results = self._rerank(fts_results, query)
         if self.score == "relevance":
             fts_results = fts_results.drop_columns(["_score"])

python/python/lancedb/rerankers/jinaai.py

@@ -62,6 +62,9 @@ class JinaReranker(Reranker):
         return self._session
 
     def _rerank(self, result_set: pa.Table, query: str):
+        result_set = self._handle_empty_results(result_set)
+        if len(result_set) == 0:
+            return result_set
         docs = result_set[self.column].to_pylist()
         response = self._client.post(  # type: ignore
             API_URL,
@@ -104,24 +107,14 @@ class JinaReranker(Reranker):
             )
         return combined_results
 
-    def rerank_vector(
-        self,
-        query: str,
-        vector_results: pa.Table,
-    ):
-        result_set = self._rerank(vector_results, query)
+    def rerank_vector(self, query: str, vector_results: pa.Table):
+        vector_results = self._rerank(vector_results, query)
         if self.score == "relevance":
-            result_set = result_set.drop_columns(["_distance"])
+            vector_results = vector_results.drop_columns(["_distance"])
+        return vector_results
 
-        return result_set
-
-    def rerank_fts(
-        self,
-        query: str,
-        fts_results: pa.Table,
-    ):
-        result_set = self._rerank(fts_results, query)
+    def rerank_fts(self, query: str, fts_results: pa.Table):
+        fts_results = self._rerank(fts_results, query)
         if self.score == "relevance":
-            result_set = result_set.drop_columns(["_score"])
-
-        return result_set
+            fts_results = fts_results.drop_columns(["_score"])
+        return fts_results

python/python/lancedb/rerankers/openai.py

@@ -44,6 +44,9 @@ class OpenaiReranker(Reranker):
         self.api_key = api_key
 
     def _rerank(self, result_set: pa.Table, query: str):
+        result_set = self._handle_empty_results(result_set)
+        if len(result_set) == 0:
+            return result_set
         docs = result_set[self.column].to_pylist()
         response = self._client.chat.completions.create(
             model=self.model_name,
@@ -104,18 +107,14 @@ class OpenaiReranker(Reranker):
         vector_results = self._rerank(vector_results, query)
         if self.score == "relevance":
             vector_results = vector_results.drop_columns(["_distance"])
-
         vector_results = vector_results.sort_by([("_relevance_score", "descending")])
-
         return vector_results
 
     def rerank_fts(self, query: str, fts_results: pa.Table):
         fts_results = self._rerank(fts_results, query)
         if self.score == "relevance":
             fts_results = fts_results.drop_columns(["_score"])
-
         fts_results = fts_results.sort_by([("_relevance_score", "descending")])
-
         return fts_results
 
     @cached_property

python/python/lancedb/rerankers/voyageai.py

@@ -63,6 +63,9 @@ class VoyageAIReranker(Reranker):
         )
 
     def _rerank(self, result_set: pa.Table, query: str):
+        result_set = self._handle_empty_results(result_set)
+        if len(result_set) == 0:
+            return result_set
         docs = result_set[self.column].to_pylist()
         response = self._client.rerank(
             query=query,
@@ -101,24 +104,14 @@ class VoyageAIReranker(Reranker):
             )
         return combined_results
 
-    def rerank_vector(
-        self,
-        query: str,
-        vector_results: pa.Table,
-    ):
-        result_set = self._rerank(vector_results, query)
+    def rerank_vector(self, query: str, vector_results: pa.Table):
+        vector_results = self._rerank(vector_results, query)
         if self.score == "relevance":
-            result_set = result_set.drop_columns(["_distance"])
+            vector_results = vector_results.drop_columns(["_distance"])
+        return vector_results
 
-        return result_set
-
-    def rerank_fts(
-        self,
-        query: str,
-        fts_results: pa.Table,
-    ):
-        result_set = self._rerank(fts_results, query)
+    def rerank_fts(self, query: str, fts_results: pa.Table):
+        fts_results = self._rerank(fts_results, query)
         if self.score == "relevance":
-            result_set = result_set.drop_columns(["_score"])
-
-        return result_set
+            fts_results = fts_results.drop_columns(["_score"])
+        return fts_results

python/python/lancedb/table.py

@@ -52,6 +52,7 @@ from .query import (
     AsyncHybridQuery,
     AsyncQuery,
     AsyncVectorQuery,
+    FullTextQuery,
     LanceEmptyQueryBuilder,
     LanceFtsQueryBuilder,
     LanceHybridQueryBuilder,
@@ -630,6 +631,7 @@ class Table(ABC):
         index_cache_size: Optional[int] = None,
         *,
         index_type: VectorIndexType = "IVF_PQ",
+        wait_timeout: Optional[timedelta] = None,
         num_bits: int = 8,
         max_iterations: int = 50,
         sample_rate: int = 256,
@@ -665,6 +667,8 @@ class Table(ABC):
         num_bits: int
             The number of bits to encode sub-vectors. Only used with the IVF_PQ index.
             Only 4 and 8 are supported.
+        wait_timeout: timedelta, optional
+            The timeout to wait if indexing is asynchronous.
         """
         raise NotImplementedError
 
@@ -688,6 +692,23 @@ class Table(ABC):
         """
         raise NotImplementedError
 
+    def wait_for_index(
+        self, index_names: Iterable[str], timeout: timedelta = timedelta(seconds=300)
+    ) -> None:
+        """
+        Wait for indexing to complete for the given index names.
+        This will poll the table until all the indices are fully indexed,
+        or raise a timeout exception if the timeout is reached.
+
+        Parameters
+        ----------
+        index_names: str
+            The name of the indices to poll
+        timeout: timedelta
+            Timeout to wait for asynchronous indexing. The default is 5 minutes.
+        """
+        raise NotImplementedError
+
     @abstractmethod
     def create_scalar_index(
         self,
@@ -695,6 +716,7 @@ class Table(ABC):
         *,
         replace: bool = True,
         index_type: ScalarIndexType = "BTREE",
+        wait_timeout: Optional[timedelta] = None,
     ):
         """Create a scalar index on a column.
 
@@ -707,7 +729,8 @@ class Table(ABC):
             Replace the existing index if it exists.
         index_type: Literal["BTREE", "BITMAP", "LABEL_LIST"], default "BTREE"
             The type of index to create.
-
+        wait_timeout: timedelta, optional
+            The timeout to wait if indexing is asynchronous.
         Examples
         --------
 
@@ -766,6 +789,7 @@ class Table(ABC):
         stem: bool = False,
         remove_stop_words: bool = False,
         ascii_folding: bool = False,
+        wait_timeout: Optional[timedelta] = None,
     ):
         """Create a full-text search index on the table.
 
@@ -821,6 +845,8 @@ class Table(ABC):
         ascii_folding : bool, default False
             Whether to fold ASCII characters. This converts accented characters to
             their ASCII equivalent. For example, "café" would be converted to "cafe".
+        wait_timeout: timedelta, optional
+            The timeout to wait if indexing is asynchronous.
         """
         raise NotImplementedError
 
@@ -919,7 +945,9 @@ class Table(ABC):
     @abstractmethod
     def search(
         self,
-        query: Optional[Union[VEC, str, "PIL.Image.Image", Tuple]] = None,
+        query: Optional[
+            Union[VEC, str, "PIL.Image.Image", Tuple, FullTextQuery]
+        ] = None,
         vector_column_name: Optional[str] = None,
         query_type: QueryType = "auto",
         ordering_field_name: Optional[str] = None,
@@ -1004,7 +1032,11 @@ class Table(ABC):
 
     @abstractmethod
     def _execute_query(
-        self, query: Query, batch_size: Optional[int] = None
+        self,
+        query: Query,
+        *,
+        batch_size: Optional[int] = None,
+        timeout: Optional[timedelta] = None,
     ) -> pa.RecordBatchReader: ...
 
     @abstractmethod
@@ -1738,8 +1770,37 @@ class LanceTable(Table):
         )
 
     def drop_index(self, name: str) -> None:
+        """
+        Drops an index from the table
+
+        Parameters
+        ----------
+        name: str
+            The name of the index to drop
+        """
         return LOOP.run(self._table.drop_index(name))
 
+    def prewarm_index(self, name: str) -> None:
+        """
+        Prewarms an index in the table
+
+        This loads the entire index into memory
+
+        If the index does not fit into the available cache this call
+        may be wasteful
+
+        Parameters
+        ----------
+        name: str
+            The name of the index to prewarm
+        """
+        return LOOP.run(self._table.prewarm_index(name))
+
+    def wait_for_index(
+        self, index_names: Iterable[str], timeout: timedelta = timedelta(seconds=300)
+    ) -> None:
+        return LOOP.run(self._table.wait_for_index(index_names, timeout))
+
     def create_scalar_index(
         self,
         column: str,
@@ -2039,7 +2100,9 @@ class LanceTable(Table):
     @overload
     def search(
         self,
-        query: Optional[Union[VEC, str, "PIL.Image.Image", Tuple]] = None,
+        query: Optional[
+            Union[VEC, str, "PIL.Image.Image", Tuple, FullTextQuery]
+        ] = None,
         vector_column_name: Optional[str] = None,
         query_type: Literal["hybrid"] = "hybrid",
         ordering_field_name: Optional[str] = None,
@@ -2058,7 +2121,9 @@ class LanceTable(Table):
 
     def search(
         self,
-        query: Optional[Union[VEC, str, "PIL.Image.Image", Tuple]] = None,
+        query: Optional[
+            Union[VEC, str, "PIL.Image.Image", Tuple, FullTextQuery]
+        ] = None,
         vector_column_name: Optional[str] = None,
         query_type: QueryType = "auto",
         ordering_field_name: Optional[str] = None,
@@ -2130,6 +2195,8 @@ class LanceTable(Table):
             and also the "_distance" column which is the distance between the query
             vector and the returned vector.
         """
+        if isinstance(query, FullTextQuery):
+            query_type = "fts"
         vector_column_name = infer_vector_column_name(
             schema=self.schema,
             query_type=query_type,
@@ -2305,9 +2372,15 @@ class LanceTable(Table):
         LOOP.run(self._table.update(values, where=where, updates_sql=values_sql))
 
     def _execute_query(
-        self, query: Query, batch_size: Optional[int] = None
+        self,
+        query: Query,
+        *,
+        batch_size: Optional[int] = None,
+        timeout: Optional[timedelta] = None,
     ) -> pa.RecordBatchReader:
-        async_iter = LOOP.run(self._table._execute_query(query, batch_size))
+        async_iter = LOOP.run(
+            self._table._execute_query(query, batch_size=batch_size, timeout=timeout)
+        )
 
         def iter_sync():
             try:
@@ -2921,6 +2994,7 @@ class AsyncTable:
         config: Optional[
             Union[IvfFlat, IvfPq, HnswPq, HnswSq, BTree, Bitmap, LabelList, FTS]
         ] = None,
+        wait_timeout: Optional[timedelta] = None,
     ):
         """Create an index to speed up queries
 
@@ -2945,6 +3019,8 @@ class AsyncTable:
             For advanced configuration you can specify the type of index you would
             like to create.   You can also specify index-specific parameters when
             creating an index object.
+        wait_timeout: timedelta, optional
+            The timeout to wait if indexing is asynchronous.
         """
         if config is not None:
             if not isinstance(
@@ -2955,7 +3031,9 @@ class AsyncTable:
                     " Bitmap, LabelList, or FTS"
                 )
         try:
-            await self._inner.create_index(column, index=config, replace=replace)
+            await self._inner.create_index(
+                column, index=config, replace=replace, wait_timeout=wait_timeout
+            )
         except ValueError as e:
             if "not support the requested language" in str(e):
                 supported_langs = ", ".join(lang_mapping.values())
@@ -2983,6 +3061,40 @@ class AsyncTable:
         """
         await self._inner.drop_index(name)
 
+    async def prewarm_index(self, name: str) -> None:
+        """
+        Prewarm an index in the table.
+
+        Parameters
+        ----------
+        name: str
+            The name of the index to prewarm
+
+        Notes
+        -----
+        This will load the index into memory.  This may reduce the cold-start time for
+        future queries.  If the index does not fit in the cache then this call may be
+        wasteful.
+        """
+        await self._inner.prewarm_index(name)
+
+    async def wait_for_index(
+        self, index_names: Iterable[str], timeout: timedelta = timedelta(seconds=300)
+    ) -> None:
+        """
+        Wait for indexing to complete for the given index names.
+        This will poll the table until all the indices are fully indexed,
+        or raise a timeout exception if the timeout is reached.
+
+        Parameters
+        ----------
+        index_names: str
+            The name of the indices to poll
+        timeout: timedelta
+            Timeout to wait for asynchronous indexing. The default is 5 minutes.
+        """
+        await self._inner.wait_for_index(index_names, timeout)
+
     async def add(
         self,
         data: DATA,
@@ -3134,7 +3246,9 @@ class AsyncTable:
     @overload
     async def search(
         self,
-        query: Optional[Union[VEC, str, "PIL.Image.Image", Tuple]] = None,
+        query: Optional[
+            Union[VEC, str, "PIL.Image.Image", Tuple, FullTextQuery]
+        ] = None,
         vector_column_name: Optional[str] = None,
         query_type: Literal["vector"] = ...,
         ordering_field_name: Optional[str] = None,
@@ -3143,7 +3257,9 @@ class AsyncTable:
 
     async def search(
         self,
-        query: Optional[Union[VEC, str, "PIL.Image.Image", Tuple]] = None,
+        query: Optional[
+            Union[VEC, str, "PIL.Image.Image", Tuple, FullTextQuery]
+        ] = None,
         vector_column_name: Optional[str] = None,
         query_type: QueryType = "auto",
         ordering_field_name: Optional[str] = None,
@@ -3202,8 +3318,10 @@ class AsyncTable:
         async def get_embedding_func(
             vector_column_name: Optional[str],
             query_type: QueryType,
-            query: Optional[Union[VEC, str, "PIL.Image.Image", Tuple]],
+            query: Optional[Union[VEC, str, "PIL.Image.Image", Tuple, FullTextQuery]],
         ) -> Tuple[str, EmbeddingFunctionConfig]:
+            if isinstance(query, FullTextQuery):
+                query_type = "fts"
             schema = await self.schema()
             vector_column_name = infer_vector_column_name(
                 schema=schema,
@@ -3253,6 +3371,8 @@ class AsyncTable:
             if is_embedding(query):
                 vector_query = query
                 query_type = "vector"
+            elif isinstance(query, FullTextQuery):
+                query_type = "fts"
             elif isinstance(query, str):
                 try:
                     (
@@ -3377,7 +3497,11 @@ class AsyncTable:
         return async_query
 
     async def _execute_query(
-        self, query: Query, batch_size: Optional[int] = None
+        self,
+        query: Query,
+        *,
+        batch_size: Optional[int] = None,
+        timeout: Optional[timedelta] = None,
     ) -> pa.RecordBatchReader:
         # The sync table calls into this method, so we need to map the
         # query to the async version of the query and run that here. This is only
@@ -3385,7 +3509,9 @@ class AsyncTable:
 
         async_query = self._sync_query_to_async(query)
 
-        return await async_query.to_batches(max_batch_length=batch_size)
+        return await async_query.to_batches(
+            max_batch_length=batch_size, timeout=timeout
+        )
 
     async def _explain_plan(self, query: Query, verbose: Optional[bool]) -> str:
         # This method is used by the sync table

python/python/lancedb/util.py

@@ -253,9 +253,14 @@ def infer_vector_column_name(
     query: Optional[Any],  # inferred later in query builder
     vector_column_name: Optional[str],
 ):
-    if (vector_column_name is None and query is not None and query_type != "fts") or (
-        vector_column_name is None and query_type == "hybrid"
-    ):
+    if vector_column_name is not None:
+        return vector_column_name
+
+    if query_type == "fts":
+        # FTS queries do not require a vector column
+        return None
+
+    if query is not None or query_type == "hybrid":
         try:
             vector_column_name = inf_vector_column_query(schema)
         except Exception as e:

python/python/tests/docs/test_guide_tables.py

@@ -315,11 +315,6 @@ def test_table():
     db = lancedb.connect(uri, read_consistency_interval=timedelta(seconds=5))
     tbl = db.open_table("test_table")
     # --8<-- [end:table_eventual_consistency]
-    # --8<-- [start:table_no_consistency]
-    uri = "data/sample-lancedb"
-    db = lancedb.connect(uri, read_consistency_interval=None)
-    tbl = db.open_table("test_table")
-    # --8<-- [end:table_no_consistency]
     # --8<-- [start:table_checkout_latest]
     tbl = db.open_table("test_table")
 
@@ -574,12 +569,6 @@ async def test_table_async():
     )
     async_tbl = await async_db.open_table("test_table_async")
     # --8<-- [end:table_async_eventual_consistency]
-    # --8<-- [start:table_async_no_consistency]
-    uri = "data/sample-lancedb"
-    async_db = await lancedb.connect_async(uri, read_consistency_interval=None)
-    async_tbl = await async_db.open_table("test_table_async")
-    # --8<-- [end:table_async_no_consistency]
-
     # --8<-- [start:table_async_checkout_latest]
     async_tbl = await async_db.open_table("test_table_async")
 

python/python/tests/docs/test_search.py

@@ -6,7 +6,9 @@ import lancedb
 
 # --8<-- [end:import-lancedb]
 # --8<-- [start:import-numpy]
+from lancedb.query import BoostQuery, MatchQuery
 import numpy as np
+import pyarrow as pa
 
 # --8<-- [end:import-numpy]
 # --8<-- [start:import-datetime]
@@ -154,6 +156,84 @@ async def test_vector_search_async():
     # --8<-- [end:search_result_async_as_list]
 
 
+def test_fts_fuzzy_query():
+    uri = "data/fuzzy-example"
+    db = lancedb.connect(uri)
+
+    table = db.create_table(
+        "my_table_fts_fuzzy",
+        data=pa.table(
+            {
+                "text": [
+                    "fa",
+                    "fo",  # spellchecker:disable-line
+                    "fob",
+                    "focus",
+                    "foo",
+                    "food",
+                    "foul",
+                ]
+            }
+        ),
+        mode="overwrite",
+    )
+    table.create_fts_index("text", use_tantivy=False, replace=True)
+
+    results = table.search(MatchQuery("foo", "text", fuzziness=1)).to_pandas()
+    assert len(results) == 4
+    assert set(results["text"].to_list()) == {
+        "foo",
+        "fo",  # 1 deletion # spellchecker:disable-line
+        "fob",  # 1 substitution
+        "food",  # 1 insertion
+    }
+
+
+def test_fts_boost_query():
+    uri = "data/boost-example"
+    db = lancedb.connect(uri)
+
+    table = db.create_table(
+        "my_table_fts_boost",
+        data=pa.table(
+            {
+                "title": [
+                    "The Hidden Gems of Travel",
+                    "Exploring Nature's Wonders",
+                    "Cultural Treasures Unveiled",
+                    "The Nightlife Chronicles",
+                    "Scenic Escapes and Challenges",
+                ],
+                "desc": [
+                    "A vibrant city with occasional traffic jams.",
+                    "Beautiful landscapes but overpriced tourist spots.",
+                    "Rich cultural heritage but humid summers.",
+                    "Bustling nightlife but noisy streets.",
+                    "Scenic views but limited public transport options.",
+                ],
+            }
+        ),
+        mode="overwrite",
+    )
+    table.create_fts_index("desc", use_tantivy=False, replace=True)
+
+    results = table.search(
+        BoostQuery(
+            MatchQuery("beautiful, cultural, nightlife", "desc"),
+            MatchQuery("bad traffic jams, overpriced", "desc"),
+        ),
+    ).to_pandas()
+
+    # we will hit 3 results because the positive query has 3 hits
+    assert len(results) == 3
+    # the one containing "overpriced" will be negatively boosted,
+    # so it will be the last one
+    assert (
+        results["desc"].to_list()[2]
+        == "Beautiful landscapes but overpriced tourist spots."
+    )
+
+
 def test_fts_native():
     # --8<-- [start:basic_fts]
     uri = "data/sample-lancedb"

python/python/tests/test_db.py

@@ -3,6 +3,7 @@
 
 
 import re
+from datetime import timedelta
 import os
 
 import lancedb
@@ -298,11 +299,13 @@ def test_create_exist_ok(tmp_db: lancedb.DBConnection):
 @pytest.mark.asyncio
 async def test_connect(tmp_path):
     db = await lancedb.connect_async(tmp_path)
-    assert str(db) == f"ListingDatabase(uri={tmp_path}, read_consistency_interval=5s)"
-
-    db = await lancedb.connect_async(tmp_path, read_consistency_interval=None)
     assert str(db) == f"ListingDatabase(uri={tmp_path}, read_consistency_interval=None)"
 
+    db = await lancedb.connect_async(
+        tmp_path, read_consistency_interval=timedelta(seconds=5)
+    )
+    assert str(db) == f"ListingDatabase(uri={tmp_path}, read_consistency_interval=5s)"
+
 
 @pytest.mark.asyncio
 async def test_close(mem_db_async: lancedb.AsyncConnection):
@@ -450,7 +453,7 @@ async def test_open_table(tmp_path):
     assert tbl.name == "test"
     assert (
         re.search(
-            r"NativeTable\(test, uri=.*test\.lance, read_consistency_interval=5s\)",
+            r"NativeTable\(test, uri=.*test\.lance, read_consistency_interval=None\)",
             str(tbl),
         )
         is not None

python/python/tests/test_embeddings_slow.py

@@ -11,7 +11,8 @@ import pandas as pd
 import pyarrow as pa
 import pytest
 from lancedb.embeddings import get_registry
-from lancedb.pydantic import LanceModel, Vector
+from lancedb.pydantic import LanceModel, Vector, MultiVector
+import requests
 
 # These are integration tests for embedding functions.
 # They are slow because they require downloading models
@@ -516,3 +517,125 @@ def test_voyageai_embedding_function():
 
     tbl.add(df)
     assert len(tbl.to_pandas()["vector"][0]) == voyageai.ndims()
+
+
+@pytest.mark.slow
+@pytest.mark.skipif(
+    os.environ.get("VOYAGE_API_KEY") is None, reason="VOYAGE_API_KEY not set"
+)
+def test_voyageai_multimodal_embedding_function():
+    voyageai = (
+        get_registry().get("voyageai").create(name="voyage-multimodal-3", max_retries=0)
+    )
+
+    class Images(LanceModel):
+        label: str
+        image_uri: str = voyageai.SourceField()  # image uri as the source
+        image_bytes: bytes = voyageai.SourceField()  # image bytes as the source
+        vector: Vector(voyageai.ndims()) = voyageai.VectorField()  # vector column
+        vec_from_bytes: Vector(voyageai.ndims()) = (
+            voyageai.VectorField()
+        )  # Another vector column
+
+    db = lancedb.connect("~/lancedb")
+    table = db.create_table("test", schema=Images, mode="overwrite")
+    labels = ["cat", "cat", "dog", "dog", "horse", "horse"]
+    uris = [
+        "http://farm1.staticflickr.com/53/167798175_7c7845bbbd_z.jpg",
+        "http://farm1.staticflickr.com/134/332220238_da527d8140_z.jpg",
+        "http://farm9.staticflickr.com/8387/8602747737_2e5c2a45d4_z.jpg",
+        "http://farm5.staticflickr.com/4092/5017326486_1f46057f5f_z.jpg",
+        "http://farm9.staticflickr.com/8216/8434969557_d37882c42d_z.jpg",
+        "http://farm6.staticflickr.com/5142/5835678453_4f3a4edb45_z.jpg",
+    ]
+    # get each uri as bytes
+    image_bytes = [requests.get(uri).content for uri in uris]
+    table.add(
+        pd.DataFrame({"label": labels, "image_uri": uris, "image_bytes": image_bytes})
+    )
+    assert len(table.to_pandas()["vector"][0]) == voyageai.ndims()
+
+
+@pytest.mark.slow
+@pytest.mark.skipif(
+    os.environ.get("VOYAGE_API_KEY") is None, reason="VOYAGE_API_KEY not set"
+)
+def test_voyageai_multimodal_embedding_text_function():
+    voyageai = (
+        get_registry().get("voyageai").create(name="voyage-multimodal-3", max_retries=0)
+    )
+
+    class TextModel(LanceModel):
+        text: str = voyageai.SourceField()
+        vector: Vector(voyageai.ndims()) = voyageai.VectorField()
+
+    df = pd.DataFrame({"text": ["hello world", "goodbye world"]})
+    db = lancedb.connect("~/lancedb")
+    tbl = db.create_table("test", schema=TextModel, mode="overwrite")
+
+    tbl.add(df)
+    assert len(tbl.to_pandas()["vector"][0]) == voyageai.ndims()
+
+
+@pytest.mark.slow
+@pytest.mark.skipif(
+    importlib.util.find_spec("colpali_engine") is None,
+    reason="colpali_engine not installed",
+)
+def test_colpali(tmp_path):
+    import requests
+    from lancedb.pydantic import LanceModel
+
+    db = lancedb.connect(tmp_path)
+    registry = get_registry()
+    func = registry.get("colpali").create()
+
+    class MediaItems(LanceModel):
+        text: str
+        image_uri: str = func.SourceField()
+        image_bytes: bytes = func.SourceField()
+        image_vectors: MultiVector(func.ndims()) = (
+            func.VectorField()
+        )  # Multivector image embeddings
+
+    table = db.create_table("media", schema=MediaItems)
+
+    texts = [
+        "a cute cat playing with yarn",
+        "a puppy in a flower field",
+        "a red sports car on the highway",
+        "a vintage bicycle leaning against a wall",
+        "a plate of delicious pasta",
+        "fresh fruit salad in a bowl",
+    ]
+
+    uris = [
+        "http://farm1.staticflickr.com/53/167798175_7c7845bbbd_z.jpg",
+        "http://farm1.staticflickr.com/134/332220238_da527d8140_z.jpg",
+        "http://farm9.staticflickr.com/8387/8602747737_2e5c2a45d4_z.jpg",
+        "http://farm5.staticflickr.com/4092/5017326486_1f46057f5f_z.jpg",
+        "http://farm9.staticflickr.com/8216/8434969557_d37882c42d_z.jpg",
+        "http://farm6.staticflickr.com/5142/5835678453_4f3a4edb45_z.jpg",
+    ]
+
+    # Get images as bytes
+    image_bytes = [requests.get(uri).content for uri in uris]
+
+    table.add(
+        pd.DataFrame({"text": texts, "image_uri": uris, "image_bytes": image_bytes})
+    )
+
+    # Test text-to-image search
+    image_results = (
+        table.search("fluffy companion", vector_column_name="image_vectors")
+        .limit(1)
+        .to_pydantic(MediaItems)[0]
+    )
+    assert "cat" in image_results.text.lower() or "puppy" in image_results.text.lower()
+
+    # Verify multivector dimensions
+    first_row = table.to_arrow().to_pylist()[0]
+    assert len(first_row["image_vectors"]) > 1, "Should have multiple image vectors"
+    assert len(first_row["image_vectors"][0]) == func.ndims(), (
+        "Vector dimension mismatch"
+    )

python/python/tests/test_fts.py

@@ -20,7 +20,9 @@ from unittest import mock
 import lancedb as ldb
 from lancedb.db import DBConnection
 from lancedb.index import FTS
+from lancedb.query import BoostQuery, MatchQuery, MultiMatchQuery, PhraseQuery
 import numpy as np
+import pyarrow as pa
 import pandas as pd
 import pytest
 from utils import exception_output
@@ -178,11 +180,47 @@ def test_search_fts(table, use_tantivy):
     results = table.search("puppy").select(["id", "text"]).to_list()
     assert len(results) == 10
 
+    if not use_tantivy:
+        # Test with a query
+        results = (
+            table.search(MatchQuery("puppy", "text"))
+            .select(["id", "text"])
+            .limit(5)
+            .to_list()
+        )
+        assert len(results) == 5
+
+        # Test boost query
+        results = (
+            table.search(
+                BoostQuery(
+                    MatchQuery("puppy", "text"),
+                    MatchQuery("runs", "text"),
+                )
+            )
+            .select(["id", "text"])
+            .limit(5)
+            .to_list()
+        )
+        assert len(results) == 5
+
+        # Test multi match query
+        table.create_fts_index("text2", use_tantivy=use_tantivy)
+        results = (
+            table.search(MultiMatchQuery("puppy", ["text", "text2"]))
+            .select(["id", "text"])
+            .limit(5)
+            .to_list()
+        )
+        assert len(results) == 5
+        assert len(results[0]) == 3  # id, text, _score
+
 
 @pytest.mark.asyncio
 async def test_fts_select_async(async_table):
     tbl = await async_table
     await tbl.create_index("text", config=FTS())
+    await tbl.create_index("text2", config=FTS())
     results = (
         await tbl.query()
         .nearest_to_text("puppy")
@@ -193,6 +231,54 @@ async def test_fts_select_async(async_table):
     assert len(results) == 5
     assert len(results[0]) == 3  # id, text, _score
 
+    # Test with FullTextQuery
+    results = (
+        await tbl.query()
+        .nearest_to_text(MatchQuery("puppy", "text"))
+        .select(["id", "text"])
+        .limit(5)
+        .to_list()
+    )
+    assert len(results) == 5
+    assert len(results[0]) == 3  # id, text, _score
+
+    # Test with BoostQuery
+    results = (
+        await tbl.query()
+        .nearest_to_text(
+            BoostQuery(
+                MatchQuery("puppy", "text"),
+                MatchQuery("runs", "text"),
+            )
+        )
+        .select(["id", "text"])
+        .limit(5)
+        .to_list()
+    )
+    assert len(results) == 5
+    assert len(results[0]) == 3  # id, text, _score
+
+    # Test with MultiMatchQuery
+    results = (
+        await tbl.query()
+        .nearest_to_text(MultiMatchQuery("puppy", ["text", "text2"]))
+        .select(["id", "text"])
+        .limit(5)
+        .to_list()
+    )
+    assert len(results) == 5
+    assert len(results[0]) == 3  # id, text, _score
+
+    # Test with search() API
+    results = (
+        await (await tbl.search(MatchQuery("puppy", "text")))
+        .select(["id", "text"])
+        .limit(5)
+        .to_list()
+    )
+    assert len(results) == 5
+    assert len(results[0]) == 3  # id, text, _score
+
 
 def test_search_fts_phrase_query(table):
     table.create_fts_index("text", use_tantivy=False, with_position=False)
@@ -207,6 +293,13 @@ def test_search_fts_phrase_query(table):
     assert len(results) > len(phrase_results)
     assert len(phrase_results) > 0
 
+    # Test with a query
+    phrase_results = (
+        table.search(PhraseQuery("puppy runs", "text")).limit(100).to_list()
+    )
+    assert len(results) > len(phrase_results)
+    assert len(phrase_results) > 0
+
 
 @pytest.mark.asyncio
 async def test_search_fts_phrase_query_async(async_table):
@@ -227,6 +320,16 @@ async def test_search_fts_phrase_query_async(async_table):
     assert len(results) > len(phrase_results)
     assert len(phrase_results) > 0
 
+    # Test with a query
+    phrase_results = (
+        await async_table.query()
+        .nearest_to_text(PhraseQuery("puppy runs", "text"))
+        .limit(100)
+        .to_list()
+    )
+    assert len(results) > len(phrase_results)
+    assert len(phrase_results) > 0
+
 
 def test_search_fts_specify_column(table):
     table.create_fts_index("text", use_tantivy=False)
@@ -524,3 +627,32 @@ def test_language(mem_db: DBConnection):
     # Stop words -> no results
     results = table.search("la", query_type="fts").limit(5).to_list()
     assert len(results) == 0
+
+
+def test_fts_on_list(mem_db: DBConnection):
+    data = pa.table(
+        {
+            "text": [
+                ["lance database", "the", "search"],
+                ["lance database"],
+                ["lance", "search"],
+                ["database", "search"],
+                ["unrelated", "doc"],
+            ],
+            "vector": [
+                [1.0, 2.0, 3.0],
+                [4.0, 5.0, 6.0],
+                [7.0, 8.0, 9.0],
+                [10.0, 11.0, 12.0],
+                [13.0, 14.0, 15.0],
+            ],
+        }
+    )
+    table = mem_db.create_table("test", data=data)
+    table.create_fts_index("text", use_tantivy=False)
+
+    res = table.search("lance").limit(5).to_list()
+    assert len(res) == 3
+
+    res = table.search(PhraseQuery("lance database", "text")).limit(5).to_list()
+    assert len(res) == 2

python/python/tests/test_hybrid_query.py

@@ -4,13 +4,32 @@
 import lancedb
 
 from lancedb.query import LanceHybridQueryBuilder
+from lancedb.rerankers.rrf import RRFReranker
 import pyarrow as pa
 import pyarrow.compute as pc
 import pytest
 import pytest_asyncio
 
 from lancedb.index import FTS
-from lancedb.table import AsyncTable
+from lancedb.table import AsyncTable, Table
+
+
+@pytest.fixture
+def sync_table(tmpdir_factory) -> Table:
+    tmp_path = str(tmpdir_factory.mktemp("data"))
+    db = lancedb.connect(tmp_path)
+    data = pa.table(
+        {
+            "text": pa.array(["a", "b", "cat", "dog"]),
+            "vector": pa.array(
+                [[0.1, 0.1], [2, 2], [-0.1, -0.1], [0.5, -0.5]],
+                type=pa.list_(pa.float32(), list_size=2),
+            ),
+        }
+    )
+    table = db.create_table("test", data)
+    table.create_fts_index("text", with_position=False, use_tantivy=False)
+    return table
 
 
 @pytest_asyncio.fixture
@@ -102,6 +121,42 @@ async def test_async_hybrid_query_default_limit(table: AsyncTable):
     assert texts.count("a") == 1
 
 
+def test_hybrid_query_distance_range(sync_table: Table):
+    reranker = RRFReranker(return_score="all")
+    result = (
+        sync_table.search(query_type="hybrid")
+        .vector([0.0, 0.4])
+        .text("cat and dog")
+        .distance_range(lower_bound=0.2, upper_bound=0.5)
+        .rerank(reranker)
+        .limit(2)
+        .to_arrow()
+    )
+    assert len(result) == 2
+    print(result)
+    for dist in result["_distance"]:
+        if dist.is_valid:
+            assert 0.2 <= dist.as_py() <= 0.5
+
+
+@pytest.mark.asyncio
+async def test_hybrid_query_distance_range_async(table: AsyncTable):
+    reranker = RRFReranker(return_score="all")
+    result = await (
+        table.query()
+        .nearest_to([0.0, 0.4])
+        .nearest_to_text("cat and dog")
+        .distance_range(lower_bound=0.2, upper_bound=0.5)
+        .rerank(reranker)
+        .limit(2)
+        .to_arrow()
+    )
+    assert len(result) == 2
+    for dist in result["_distance"]:
+        if dist.is_valid:
+            assert 0.2 <= dist.as_py() <= 0.5
+
+
 @pytest.mark.asyncio
 async def test_explain_plan(table: AsyncTable):
     plan = await (

python/python/tests/test_index.py

@@ -8,7 +8,7 @@ import pyarrow as pa
 import pytest
 import pytest_asyncio
 from lancedb import AsyncConnection, AsyncTable, connect_async
-from lancedb.index import BTree, IvfFlat, IvfPq, Bitmap, LabelList, HnswPq, HnswSq
+from lancedb.index import BTree, IvfFlat, IvfPq, Bitmap, LabelList, HnswPq, HnswSq, FTS
 
 
 @pytest_asyncio.fixture
@@ -119,6 +119,18 @@ async def test_create_label_list_index(some_table: AsyncTable):
     assert str(indices) == '[Index(LabelList, columns=["tags"], name="tags_idx")]'
 
 
+@pytest.mark.asyncio
+async def test_full_text_search_index(some_table: AsyncTable):
+    await some_table.create_index("tags", config=FTS(with_position=False))
+    indices = await some_table.list_indices()
+    assert str(indices) == '[Index(FTS, columns=["tags"], name="tags_idx")]'
+
+    await some_table.prewarm_index("tags_idx")
+
+    res = await (await some_table.search("tag0")).to_arrow()
+    assert res.num_rows > 0
+
+
 @pytest.mark.asyncio
 async def test_create_vector_index(some_table: AsyncTable):
     # Can create

python/python/tests/test_pydantic.py

@@ -9,7 +9,13 @@ from typing import List, Optional, Tuple
 import pyarrow as pa
 import pydantic
 import pytest
-from lancedb.pydantic import PYDANTIC_VERSION, LanceModel, Vector, pydantic_to_schema
+from lancedb.pydantic import (
+    PYDANTIC_VERSION,
+    LanceModel,
+    Vector,
+    pydantic_to_schema,
+    MultiVector,
+)
 from pydantic import BaseModel
 from pydantic import Field
 
@@ -354,3 +360,55 @@ def test_optional_nested_model():
             ),
         ]
     )
+
+
+def test_multi_vector():
+    class TestModel(pydantic.BaseModel):
+        vec: MultiVector(8)
+
+    schema = pydantic_to_schema(TestModel)
+    assert schema == pa.schema(
+        [pa.field("vec", pa.list_(pa.list_(pa.float32(), 8)), True)]
+    )
+
+    with pytest.raises(pydantic.ValidationError):
+        TestModel(vec=[[1.0] * 7])
+
+    with pytest.raises(pydantic.ValidationError):
+        TestModel(vec=[[1.0] * 9])
+
+    TestModel(vec=[[1.0] * 8])
+    TestModel(vec=[[1.0] * 8, [2.0] * 8])
+
+    TestModel(vec=[])
+
+
+def test_multi_vector_nullable():
+    class NullableModel(pydantic.BaseModel):
+        vec: MultiVector(16, nullable=False)
+
+    schema = pydantic_to_schema(NullableModel)
+    assert schema == pa.schema(
+        [pa.field("vec", pa.list_(pa.list_(pa.float32(), 16)), False)]
+    )
+
+    class DefaultModel(pydantic.BaseModel):
+        vec: MultiVector(16)
+
+    schema = pydantic_to_schema(DefaultModel)
+    assert schema == pa.schema(
+        [pa.field("vec", pa.list_(pa.list_(pa.float32(), 16)), True)]
+    )
+
+
+def test_multi_vector_in_lance_model():
+    class TestModel(LanceModel):
+        id: int
+        vectors: MultiVector(16) = Field(default=[[0.0] * 16])
+
+    schema = pydantic_to_schema(TestModel)
+    assert schema == TestModel.to_arrow_schema()
+    assert TestModel.field_names() == ["id", "vectors"]
+
+    t = TestModel(id=1)
+    assert t.vectors == [[0.0] * 16]

python/python/tests/test_query.py

@@ -257,7 +257,9 @@ async def test_distance_range_with_new_rows_async():
         }
     )
     table = await conn.create_table("test", data)
-    table.create_index("vector", config=IvfPq(num_partitions=1, num_sub_vectors=2))
+    await table.create_index(
+        "vector", config=IvfPq(num_partitions=1, num_sub_vectors=2)
+    )
 
     q = [0, 0]
     rs = await table.query().nearest_to(q).to_arrow()
@@ -511,7 +513,8 @@ def test_query_builder_with_different_vector_column():
             columns=["b"],
             vector_column="foo_vector",
         ),
-        None,
+        batch_size=None,
+        timeout=None,
     )
 
 
@@ -1076,3 +1079,67 @@ async def test_query_serialization_async(table_async: AsyncTable):
         full_text_query=FullTextSearchQuery(columns=[], query="foo"),
         with_row_id=False,
     )
+
+
+def test_query_timeout(tmp_path):
+    # Use local directory instead of memory:// to add a bit of latency to
+    # operations so a timeout of zero will trigger exceptions.
+    db = lancedb.connect(tmp_path)
+    data = pa.table(
+        {
+            "text": ["a", "b"],
+            "vector": pa.FixedSizeListArray.from_arrays(
+                pc.random(4).cast(pa.float32()), 2
+            ),
+        }
+    )
+    table = db.create_table("test", data)
+    table.create_fts_index("text", use_tantivy=False)
+
+    with pytest.raises(Exception, match="Query timeout"):
+        table.search().where("text = 'a'").to_list(timeout=timedelta(0))
+
+    with pytest.raises(Exception, match="Query timeout"):
+        table.search([0.0, 0.0]).to_arrow(timeout=timedelta(0))
+
+    with pytest.raises(Exception, match="Query timeout"):
+        table.search("a", query_type="fts").to_pandas(timeout=timedelta(0))
+
+    with pytest.raises(Exception, match="Query timeout"):
+        table.search(query_type="hybrid").vector([0.0, 0.0]).text("a").to_arrow(
+            timeout=timedelta(0)
+        )
+
+
+@pytest.mark.asyncio
+async def test_query_timeout_async(tmp_path):
+    db = await lancedb.connect_async(tmp_path)
+    data = pa.table(
+        {
+            "text": ["a", "b"],
+            "vector": pa.FixedSizeListArray.from_arrays(
+                pc.random(4).cast(pa.float32()), 2
+            ),
+        }
+    )
+    table = await db.create_table("test", data)
+    await table.create_index("text", config=FTS())
+
+    with pytest.raises(Exception, match="Query timeout"):
+        await table.query().where("text != 'a'").to_list(timeout=timedelta(0))
+
+    with pytest.raises(Exception, match="Query timeout"):
+        await table.vector_search([0.0, 0.0]).to_arrow(timeout=timedelta(0))
+
+    with pytest.raises(Exception, match="Query timeout"):
+        await (await table.search("a", query_type="fts")).to_pandas(
+            timeout=timedelta(0)
+        )
+
+    with pytest.raises(Exception, match="Query timeout"):
+        await (
+            table.query()
+            .nearest_to_text("a")
+            .nearest_to([0.0, 0.0])
+            .to_list(timeout=timedelta(0))
+        )

python/python/tests/test_remote_db.py

@@ -1,6 +1,6 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
+import re
 from concurrent.futures import ThreadPoolExecutor
 import contextlib
 from datetime import timedelta
@@ -235,6 +235,10 @@ def test_table_add_in_threadpool():
 
 def test_table_create_indices():
     def handler(request):
+        index_stats = dict(
+            index_type="IVF_PQ", num_indexed_rows=1000, num_unindexed_rows=0
+        )
+
         if request.path == "/v1/table/test/create_index/":
             request.send_response(200)
             request.end_headers()
@@ -258,6 +262,47 @@ def test_table_create_indices():
                 )
             )
             request.wfile.write(payload.encode())
+        elif request.path == "/v1/table/test/index/list/":
+            request.send_response(200)
+            request.send_header("Content-Type", "application/json")
+            request.end_headers()
+            payload = json.dumps(
+                dict(
+                    indexes=[
+                        {
+                            "index_name": "id_idx",
+                            "columns": ["id"],
+                        },
+                        {
+                            "index_name": "text_idx",
+                            "columns": ["text"],
+                        },
+                        {
+                            "index_name": "vector_idx",
+                            "columns": ["vector"],
+                        },
+                    ]
+                )
+            )
+            request.wfile.write(payload.encode())
+        elif request.path == "/v1/table/test/index/id_idx/stats/":
+            request.send_response(200)
+            request.send_header("Content-Type", "application/json")
+            request.end_headers()
+            payload = json.dumps(index_stats)
+            request.wfile.write(payload.encode())
+        elif request.path == "/v1/table/test/index/text_idx/stats/":
+            request.send_response(200)
+            request.send_header("Content-Type", "application/json")
+            request.end_headers()
+            payload = json.dumps(index_stats)
+            request.wfile.write(payload.encode())
+        elif request.path == "/v1/table/test/index/vector_idx/stats/":
+            request.send_response(200)
+            request.send_header("Content-Type", "application/json")
+            request.end_headers()
+            payload = json.dumps(index_stats)
+            request.wfile.write(payload.encode())
         elif "/drop/" in request.path:
             request.send_response(200)
             request.end_headers()
@@ -269,14 +314,81 @@ def test_table_create_indices():
         # Parameters are well-tested through local and async tests.
         # This is a smoke-test.
         table = db.create_table("test", [{"id": 1}])
-        table.create_scalar_index("id")
-        table.create_fts_index("text")
-        table.create_scalar_index("vector")
+        table.create_scalar_index("id", wait_timeout=timedelta(seconds=2))
+        table.create_fts_index("text", wait_timeout=timedelta(seconds=2))
+        table.create_index(
+            vector_column_name="vector", wait_timeout=timedelta(seconds=10)
+        )
+        table.wait_for_index(["id_idx"], timedelta(seconds=2))
+        table.wait_for_index(["text_idx", "vector_idx"], timedelta(seconds=2))
         table.drop_index("vector_idx")
         table.drop_index("id_idx")
         table.drop_index("text_idx")
 
 
+def test_table_wait_for_index_timeout():
+    def handler(request):
+        index_stats = dict(
+            index_type="BTREE", num_indexed_rows=1000, num_unindexed_rows=1
+        )
+
+        if request.path == "/v1/table/test/create/?mode=create":
+            request.send_response(200)
+            request.send_header("Content-Type", "application/json")
+            request.end_headers()
+            request.wfile.write(b"{}")
+        elif request.path == "/v1/table/test/describe/":
+            request.send_response(200)
+            request.send_header("Content-Type", "application/json")
+            request.end_headers()
+            payload = json.dumps(
+                dict(
+                    version=1,
+                    schema=dict(
+                        fields=[
+                            dict(name="id", type={"type": "int64"}, nullable=False),
+                        ]
+                    ),
+                )
+            )
+            request.wfile.write(payload.encode())
+        elif request.path == "/v1/table/test/index/list/":
+            request.send_response(200)
+            request.send_header("Content-Type", "application/json")
+            request.end_headers()
+            payload = json.dumps(
+                dict(
+                    indexes=[
+                        {
+                            "index_name": "id_idx",
+                            "columns": ["id"],
+                        },
+                    ]
+                )
+            )
+            request.wfile.write(payload.encode())
+        elif request.path == "/v1/table/test/index/id_idx/stats/":
+            request.send_response(200)
+            request.send_header("Content-Type", "application/json")
+            request.end_headers()
+            payload = json.dumps(index_stats)
+            print(f"{index_stats=}")
+            request.wfile.write(payload.encode())
+        else:
+            request.send_response(404)
+            request.end_headers()
+
+    with mock_lancedb_connection(handler) as db:
+        table = db.create_table("test", [{"id": 1}])
+        with pytest.raises(
+            RuntimeError,
+            match=re.escape(
+                'Timeout error: timed out waiting for indices: ["id_idx"] after 1s'
+            ),
+        ):
+            table.wait_for_index(["id_idx"], timedelta(seconds=1))
+
+
 @contextlib.contextmanager
 def query_test_table(query_handler, *, server_version=Version("0.1.0")):
     def handler(request):

python/python/tests/test_rerankers.py

@@ -457,3 +457,45 @@ def test_voyageai_reranker(tmp_path, use_tantivy):
     reranker = VoyageAIReranker(model_name="rerank-2")
     table, schema = get_test_table(tmp_path, use_tantivy)
     _run_test_reranker(reranker, table, "single player experience", None, schema)
+
+
+def test_empty_result_reranker():
+    pytest.importorskip("sentence_transformers")
+    db = lancedb.connect("memory://")
+
+    # Define schema
+    schema = pa.schema(
+        [
+            ("id", pa.int64()),
+            ("text", pa.string()),
+            ("vector", pa.list_(pa.float32(), 128)),  # 128-dimensional vector
+        ]
+    )
+
+    # Create empty table with schema
+    empty_table = db.create_table("empty_table", schema=schema, mode="overwrite")
+    empty_table.create_fts_index("text", use_tantivy=False, replace=True)
+    for reranker in [
+        CrossEncoderReranker(),
+        # ColbertReranker(),
+        # AnswerdotaiRerankers(),
+        # OpenaiReranker(),
+        # JinaReranker(),
+        # VoyageAIReranker(model_name="rerank-2"),
+    ]:
+        results = (
+            empty_table.search(list(range(128)))
+            .limit(3)
+            .rerank(reranker, "query")
+            .to_arrow()
+        )
+        # check if empty set contains _relevance_score column
+        assert "_relevance_score" in results.column_names
+        assert len(results) == 0
+
+        results = (
+            empty_table.search("query", query_type="fts")
+            .limit(3)
+            .rerank(reranker)
+            .to_arrow()
+        )

python/python/tests/test_table.py

@@ -9,9 +9,9 @@ from typing import List
 from unittest.mock import patch
 
 import lancedb
+from lancedb.dependencies import _PANDAS_AVAILABLE
 from lancedb.index import HnswPq, HnswSq, IvfPq
 import numpy as np
-import pandas as pd
 import polars as pl
 import pyarrow as pa
 import pyarrow.dataset
@@ -32,11 +32,7 @@ def test_basic(mem_db: DBConnection):
     table = mem_db.create_table("test", data=data)
 
     assert table.name == "test"
-    assert (
-        "LanceTable(name='test', version=1, "
-        "read_consistency_interval=datetime.timedelta(seconds=5), "
-        "_conn=LanceDBConnection("
-    ) in repr(table)
+    assert "LanceTable(name='test', version=1, _conn=LanceDBConnection(" in repr(table)
     expected_schema = pa.schema(
         {
             "vector": pa.list_(pa.float32(), 2),
@@ -142,13 +138,16 @@ def test_create_table(mem_db: DBConnection):
         {"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
         {"vector": [5.9, 26.5], "item": "bar", "price": 20.0},
     ]
-    df = pd.DataFrame(rows)
-    pa_table = pa.Table.from_pandas(df, schema=schema)
+    pa_table = pa.Table.from_pylist(rows, schema=schema)
     data = [
         ("Rows", rows),
-        ("pd_DataFrame", df),
         ("pa_Table", pa_table),
     ]
+    if _PANDAS_AVAILABLE:
+        import pandas as pd
+
+        df = pd.DataFrame(rows)
+        data.append(("pd_DataFrame", df))
 
     for name, d in data:
         tbl = mem_db.create_table(name, data=d, schema=schema).to_arrow()
@@ -300,7 +299,7 @@ def test_add_subschema(mem_db: DBConnection):
 
     data = {"price": 10.0, "item": "foo"}
     table.add([data])
-    data = pd.DataFrame({"price": [2.0], "vector": [[3.1, 4.1]]})
+    data = pa.Table.from_pydict({"price": [2.0], "vector": [[3.1, 4.1]]})
     table.add(data)
     data = {"price": 3.0, "vector": [5.9, 26.5], "item": "bar"}
     table.add([data])
@@ -409,6 +408,7 @@ def test_add_nullability(mem_db: DBConnection):
 
 
 def test_add_pydantic_model(mem_db: DBConnection):
+    pytest.importorskip("pandas")
     # https://github.com/lancedb/lancedb/issues/562
 
     class Metadata(BaseModel):
@@ -477,10 +477,10 @@ def test_polars(mem_db: DBConnection):
     table = mem_db.create_table("test", data=pl.DataFrame(data))
     assert len(table) == 2
 
-    result = table.to_pandas()
-    assert np.allclose(result["vector"].tolist(), data["vector"])
-    assert result["item"].tolist() == data["item"]
-    assert np.allclose(result["price"].tolist(), data["price"])
+    result = table.to_arrow()
+    assert np.allclose(result["vector"].to_pylist(), data["vector"])
+    assert result["item"].to_pylist() == data["item"]
+    assert np.allclose(result["price"].to_pylist(), data["price"])
 
     schema = pa.schema(
         [
@@ -692,7 +692,7 @@ def test_delete(mem_db: DBConnection):
     assert len(table.list_versions()) == 2
     assert table.version == 2
     assert len(table) == 1
-    assert table.to_pandas()["id"].tolist() == [1]
+    assert table.to_arrow()["id"].to_pylist() == [1]
 
 
 def test_update(mem_db: DBConnection):
@@ -856,6 +856,7 @@ def test_merge_insert(mem_db: DBConnection):
     ids=["pa.Table", "pd.DataFrame", "rows"],
 )
 def test_merge_insert_subschema(mem_db: DBConnection, data_format):
+    pytest.importorskip("pandas")
     initial_data = pa.table(
         {"id": range(3), "a": [1.0, 2.0, 3.0], "c": ["x", "x", "x"]}
     )
@@ -952,7 +953,7 @@ def test_create_with_embedding_function(mem_db: DBConnection):
 
     func = MockTextEmbeddingFunction.create()
     texts = ["hello world", "goodbye world", "foo bar baz fizz buzz"]
-    df = pd.DataFrame({"text": texts, "vector": func.compute_source_embeddings(texts)})
+    df = pa.table({"text": texts, "vector": func.compute_source_embeddings(texts)})
 
     conf = EmbeddingFunctionConfig(
         source_column="text", vector_column="vector", function=func
@@ -977,7 +978,7 @@ def test_create_f16_table(mem_db: DBConnection):
         text: str
         vector: Vector(32, value_type=pa.float16())
 
-    df = pd.DataFrame(
+    df = pa.table(
         {
             "text": [f"s-{i}" for i in range(512)],
             "vector": [np.random.randn(32).astype(np.float16) for _ in range(512)],
@@ -990,7 +991,7 @@ def test_create_f16_table(mem_db: DBConnection):
     table.add(df)
     table.create_index(num_partitions=2, num_sub_vectors=2)
 
-    query = df.vector.iloc[2]
+    query = df["vector"][2].as_py()
     expected = table.search(query).limit(2).to_arrow()
 
     assert "s-2" in expected["text"].to_pylist()
@@ -1006,7 +1007,7 @@ def test_add_with_embedding_function(mem_db: DBConnection):
     table = mem_db.create_table("my_table", schema=MyTable)
 
     texts = ["hello world", "goodbye world", "foo bar baz fizz buzz"]
-    df = pd.DataFrame({"text": texts})
+    df = pa.table({"text": texts})
     table.add(df)
 
     texts = ["the quick brown fox", "jumped over the lazy dog"]
@@ -1037,14 +1038,14 @@ def test_multiple_vector_columns(mem_db: DBConnection):
         {"vector1": v1, "vector2": v2, "text": "foo"},
         {"vector1": v2, "vector2": v1, "text": "bar"},
     ]
-    df = pd.DataFrame(data)
+    df = pa.Table.from_pylist(data)
     table.add(df)
 
     q = np.random.randn(10)
-    result1 = table.search(q, vector_column_name="vector1").limit(1).to_pandas()
-    result2 = table.search(q, vector_column_name="vector2").limit(1).to_pandas()
+    result1 = table.search(q, vector_column_name="vector1").limit(1).to_arrow()
+    result2 = table.search(q, vector_column_name="vector2").limit(1).to_arrow()
 
-    assert result1["text"].iloc[0] != result2["text"].iloc[0]
+    assert result1["text"][0] != result2["text"][0]
 
 
 def test_create_scalar_index(mem_db: DBConnection):
@@ -1082,22 +1083,22 @@ def test_empty_query(mem_db: DBConnection):
         "my_table",
         data=[{"text": "foo", "id": 0}, {"text": "bar", "id": 1}],
     )
-    df = table.search().select(["id"]).where("text='bar'").limit(1).to_pandas()
-    val = df.id.iloc[0]
+    df = table.search().select(["id"]).where("text='bar'").limit(1).to_arrow()
+    val = df["id"][0].as_py()
     assert val == 1
 
     table = mem_db.create_table("my_table2", data=[{"id": i} for i in range(100)])
-    df = table.search().select(["id"]).to_pandas()
-    assert len(df) == 100
+    df = table.search().select(["id"]).to_arrow()
+    assert df.num_rows == 100
     # None is the same as default
-    df = table.search().select(["id"]).limit(None).to_pandas()
-    assert len(df) == 100
+    df = table.search().select(["id"]).limit(None).to_arrow()
+    assert df.num_rows == 100
     # invalid limist is the same as None, wihch is the same as default
-    df = table.search().select(["id"]).limit(-1).to_pandas()
-    assert len(df) == 100
+    df = table.search().select(["id"]).limit(-1).to_arrow()
+    assert df.num_rows == 100
     # valid limit should work
-    df = table.search().select(["id"]).limit(42).to_pandas()
-    assert len(df) == 42
+    df = table.search().select(["id"]).limit(42).to_arrow()
+    assert df.num_rows == 42
 
 
 def test_search_with_schema_inf_single_vector(mem_db: DBConnection):
@@ -1116,14 +1117,14 @@ def test_search_with_schema_inf_single_vector(mem_db: DBConnection):
         {"vector_col": v1, "text": "foo"},
         {"vector_col": v2, "text": "bar"},
     ]
-    df = pd.DataFrame(data)
+    df = pa.Table.from_pylist(data)
     table.add(df)
 
     q = np.random.randn(10)
-    result1 = table.search(q, vector_column_name="vector_col").limit(1).to_pandas()
-    result2 = table.search(q).limit(1).to_pandas()
+    result1 = table.search(q, vector_column_name="vector_col").limit(1).to_arrow()
+    result2 = table.search(q).limit(1).to_arrow()
 
-    assert result1["text"].iloc[0] == result2["text"].iloc[0]
+    assert result1["text"][0].as_py() == result2["text"][0].as_py()
 
 
 def test_search_with_schema_inf_multiple_vector(mem_db: DBConnection):
@@ -1143,12 +1144,12 @@ def test_search_with_schema_inf_multiple_vector(mem_db: DBConnection):
         {"vector1": v1, "vector2": v2, "text": "foo"},
         {"vector1": v2, "vector2": v1, "text": "bar"},
     ]
-    df = pd.DataFrame(data)
+    df = pa.Table.from_pylist(data)
     table.add(df)
 
     q = np.random.randn(10)
     with pytest.raises(ValueError):
-        table.search(q).limit(1).to_pandas()
+        table.search(q).limit(1).to_arrow()
 
 
 def test_compact_cleanup(tmp_db: DBConnection):

python/src/connection.rs

@@ -204,9 +204,7 @@ pub fn connect(
         }
         if let Some(read_consistency_interval) = read_consistency_interval {
             let read_consistency_interval = Duration::from_secs_f64(read_consistency_interval);
-            builder = builder.read_consistency_interval(Some(read_consistency_interval));
-        } else {
-            builder = builder.read_consistency_interval(None);
+            builder = builder.read_consistency_interval(read_consistency_interval);
         }
         if let Some(storage_options) = storage_options {
             builder = builder.storage_options(storage_options);

python/src/query.rs

@@ -2,6 +2,7 @@
 // SPDX-FileCopyrightText: Copyright The LanceDB Authors
 
 use std::sync::Arc;
+use std::time::Duration;
 
 use arrow::array::make_array;
 use arrow::array::Array;
@@ -45,7 +46,7 @@ pub struct PyFullTextSearchQuery {
 
 impl From<FullTextSearchQuery> for PyFullTextSearchQuery {
     fn from(query: FullTextSearchQuery) -> Self {
-        PyFullTextSearchQuery {
+        Self {
             columns: query.columns().into_iter().collect(),
             query: query.query.query().to_owned(),
             limit: query.limit,
@@ -99,7 +100,7 @@ pub struct PyQueryRequest {
 impl From<AnyQuery> for PyQueryRequest {
     fn from(query: AnyQuery) -> Self {
         match query {
-            AnyQuery::Query(query_request) => PyQueryRequest {
+            AnyQuery::Query(query_request) => Self {
                 limit: query_request.limit,
                 offset: query_request.offset,
                 filter: query_request.filter.map(PyQueryFilter),
@@ -121,7 +122,7 @@ impl From<AnyQuery> for PyQueryRequest {
                 postfilter: None,
                 norm: None,
             },
-            AnyQuery::VectorQuery(vector_query) => PyQueryRequest {
+            AnyQuery::VectorQuery(vector_query) => Self {
                 limit: vector_query.base.limit,
                 offset: vector_query.base.offset,
                 filter: vector_query.base.filter.map(PyQueryFilter),
@@ -279,7 +280,7 @@ impl Query {
                 }
             }
             query
-        } else if let Ok(query) = query.downcast::<PyDict>() {
+        } else if let Ok(query) = fts_query.downcast::<PyDict>() {
             let query = parse_fts_query(query)?;
             FullTextSearchQuery::new_query(query)
         } else {
@@ -294,10 +295,11 @@ impl Query {
         })
     }
 
-    #[pyo3(signature = (max_batch_length=None))]
+    #[pyo3(signature = (max_batch_length=None, timeout=None))]
     pub fn execute(
         self_: PyRef<'_, Self>,
         max_batch_length: Option<u32>,
+        timeout: Option<Duration>,
     ) -> PyResult<Bound<'_, PyAny>> {
         let inner = self_.inner.clone();
         future_into_py(self_.py(), async move {
@@ -305,6 +307,9 @@ impl Query {
             if let Some(max_batch_length) = max_batch_length {
                 opts.max_batch_length = max_batch_length;
             }
+            if let Some(timeout) = timeout {
+                opts.timeout = Some(timeout);
+            }
             let inner_stream = inner.execute_with_options(opts).await.infer_error()?;
             Ok(RecordBatchStream::new(inner_stream))
         })
@@ -376,10 +381,11 @@ impl FTSQuery {
         self.inner = self.inner.clone().postfilter();
     }
 
-    #[pyo3(signature = (max_batch_length=None))]
+    #[pyo3(signature = (max_batch_length=None, timeout=None))]
     pub fn execute(
         self_: PyRef<'_, Self>,
         max_batch_length: Option<u32>,
+        timeout: Option<Duration>,
     ) -> PyResult<Bound<'_, PyAny>> {
         let inner = self_
             .inner
@@ -391,6 +397,9 @@ impl FTSQuery {
             if let Some(max_batch_length) = max_batch_length {
                 opts.max_batch_length = max_batch_length;
             }
+            if let Some(timeout) = timeout {
+                opts.timeout = Some(timeout);
+            }
             let inner_stream = inner.execute_with_options(opts).await.infer_error()?;
             Ok(RecordBatchStream::new(inner_stream))
         })
@@ -513,10 +522,11 @@ impl VectorQuery {
         self.inner = self.inner.clone().bypass_vector_index()
     }
 
-    #[pyo3(signature = (max_batch_length=None))]
+    #[pyo3(signature = (max_batch_length=None, timeout=None))]
     pub fn execute(
         self_: PyRef<'_, Self>,
         max_batch_length: Option<u32>,
+        timeout: Option<Duration>,
     ) -> PyResult<Bound<'_, PyAny>> {
         let inner = self_.inner.clone();
         future_into_py(self_.py(), async move {
@@ -524,6 +534,9 @@ impl VectorQuery {
             if let Some(max_batch_length) = max_batch_length {
                 opts.max_batch_length = max_batch_length;
             }
+            if let Some(timeout) = timeout {
+                opts.timeout = Some(timeout);
+            }
             let inner_stream = inner.execute_with_options(opts).await.infer_error()?;
             Ok(RecordBatchStream::new(inner_stream))
         })
@@ -639,6 +652,11 @@ impl HybridQuery {
         self.inner_vec.bypass_vector_index();
     }
 
+    #[pyo3(signature = (lower_bound=None, upper_bound=None))]
+    pub fn distance_range(&mut self, lower_bound: Option<f32>, upper_bound: Option<f32>) {
+        self.inner_vec.distance_range(lower_bound, upper_bound);
+    }
+
     pub fn to_vector_query(&mut self) -> PyResult<VectorQuery> {
         Ok(VectorQuery {
             inner: self.inner_vec.inner.clone(),

python/src/table.rs

@@ -177,15 +177,19 @@ impl Table {
         })
     }
 
-    #[pyo3(signature = (column, index=None, replace=None))]
+    #[pyo3(signature = (column, index=None, replace=None, wait_timeout=None))]
     pub fn create_index<'a>(
         self_: PyRef<'a, Self>,
         column: String,
         index: Option<Bound<'_, PyAny>>,
         replace: Option<bool>,
+        wait_timeout: Option<Bound<'_, PyAny>>,
     ) -> PyResult<Bound<'a, PyAny>> {
         let index = extract_index_params(&index)?;
-        let mut op = self_.inner_ref()?.create_index(&[column], index);
+        let timeout = wait_timeout.map(|t| t.extract::<std::time::Duration>().unwrap());
+        let mut op = self_
+            .inner_ref()?
+            .create_index_with_timeout(&[column], index, timeout);
         if let Some(replace) = replace {
             op = op.replace(replace);
         }
@@ -204,6 +208,34 @@ impl Table {
         })
     }
 
+    pub fn wait_for_index<'a>(
+        self_: PyRef<'a, Self>,
+        index_names: Vec<String>,
+        timeout: Bound<'_, PyAny>,
+    ) -> PyResult<Bound<'a, PyAny>> {
+        let inner = self_.inner_ref()?.clone();
+        let timeout = timeout.extract::<std::time::Duration>()?;
+        future_into_py(self_.py(), async move {
+            let index_refs = index_names
+                .iter()
+                .map(String::as_str)
+                .collect::<Vec<&str>>();
+            inner
+                .wait_for_index(&index_refs, timeout)
+                .await
+                .infer_error()?;
+            Ok(())
+        })
+    }
+
+    pub fn prewarm_index(self_: PyRef<'_, Self>, index_name: String) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.inner_ref()?.clone();
+        future_into_py(self_.py(), async move {
+            inner.prewarm_index(&index_name).await.infer_error()?;
+            Ok(())
+        })
+    }
+
     pub fn list_indices(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
         let inner = self_.inner_ref()?.clone();
         future_into_py(self_.py(), async move {

python/src/util.rs

@@ -163,8 +163,9 @@ pub fn parse_fts_query(query: &Bound<'_, PyDict>) -> PyResult<FtsQuery> {
                 .ok_or(PyValueError::new_err("boost not found"))?
                 .extract::<Vec<f32>>()?;
 
-            let query =
-                MultiMatchQuery::try_new_with_boosts(query, columns, boost).map_err(|e| {
+            let query = MultiMatchQuery::try_new(query, columns)
+                .and_then(|q| q.try_with_boosts(boost))
+                .map_err(|e| {
                     PyValueError::new_err(format!("Error creating MultiMatchQuery: {}", e))
                 })?;
             Ok(query.into())

rust/ffi/node/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-node"
-version = "0.19.0-beta.1"
+version = "0.19.0-beta.11"
 description = "Serverless, low-latency vector database for AI applications"
 license.workspace = true
 edition.workspace = true

rust/ffi/node/src/lib.rs

@@ -60,7 +60,7 @@ fn database_new(mut cx: FunctionContext) -> JsResult<JsPromise> {
     let mut conn_builder = connect(&path).storage_options(storage_options);
 
     if let Some(interval) = read_consistency_interval {
-        conn_builder = conn_builder.read_consistency_interval(Some(interval));
+        conn_builder = conn_builder.read_consistency_interval(interval);
     }
     rt.spawn(async move {
         let database = conn_builder.execute().await;

rust/lancedb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb"
-version = "0.19.0-beta.1"
+version = "0.19.0-beta.11"
 edition.workspace = true
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license.workspace = true

rust/lancedb/src/catalog/listing.rs

@@ -12,7 +12,7 @@ use super::{
     Catalog, CatalogOptions, CreateDatabaseMode, CreateDatabaseRequest, DatabaseNamesRequest,
     OpenDatabaseRequest,
 };
-use crate::connection::{ConnectRequest, DEFAULT_READ_CONSISTENCY_INTERVAL};
+use crate::connection::ConnectRequest;
 use crate::database::listing::{ListingDatabase, ListingDatabaseOptions};
 use crate::database::{Database, DatabaseOptions};
 use crate::error::{CreateDirSnafu, Error, Result};
@@ -81,7 +81,7 @@ impl ListingCatalogOptionsBuilder {
 /// [`crate::database::listing::ListingDatabase`]
 #[derive(Debug)]
 pub struct ListingCatalog {
-    object_store: ObjectStore,
+    object_store: Arc<ObjectStore>,
 
     uri: String,
 
@@ -105,7 +105,7 @@ impl ListingCatalog {
     }
 
     async fn open_path(path: &str) -> Result<Self> {
-        let (object_store, base_path) = ObjectStore::from_path(path).unwrap();
+        let (object_store, base_path) = ObjectStore::from_uri(path).await.unwrap();
         if object_store.is_local() {
             Self::try_create_dir(path).context(CreateDirSnafu { path })?;
         }
@@ -214,7 +214,7 @@ impl Catalog for ListingCatalog {
             uri: db_uri,
             #[cfg(feature = "remote")]
             client_config: Default::default(),
-            read_consistency_interval: DEFAULT_READ_CONSISTENCY_INTERVAL,
+            read_consistency_interval: None,
             options: Default::default(),
         };
 
@@ -241,7 +241,7 @@ impl Catalog for ListingCatalog {
             uri: db_path.to_string(),
             #[cfg(feature = "remote")]
             client_config: Default::default(),
-            read_consistency_interval: DEFAULT_READ_CONSISTENCY_INTERVAL,
+            read_consistency_interval: None,
             options: Default::default(),
         };
 
@@ -311,7 +311,7 @@ mod tests {
             #[cfg(feature = "remote")]
             client_config: Default::default(),
             options: Default::default(),
-            read_consistency_interval: DEFAULT_READ_CONSISTENCY_INTERVAL,
+            read_consistency_interval: None,
         };
 
         let catalog = ListingCatalog::connect(&request).await.unwrap();

rust/lancedb/src/connection.rs

@@ -36,9 +36,6 @@ pub use lance_encoding::version::LanceFileVersion;
 #[cfg(feature = "remote")]
 use lance_io::object_store::StorageOptions;
 
-pub(crate) const DEFAULT_READ_CONSISTENCY_INTERVAL: Option<std::time::Duration> =
-    Some(std::time::Duration::from_secs(5));
-
 /// A builder for configuring a [`Connection::table_names`] operation
 pub struct TableNamesBuilder {
     parent: Arc<dyn Database>,
@@ -142,12 +139,6 @@ impl CreateTableBuilder<true> {
         }
     }
 
-    /// Apply the given write options when writing the initial data
-    pub fn write_options(mut self, write_options: WriteOptions) -> Self {
-        self.request.write_options = write_options;
-        self
-    }
-
     /// Execute the create table operation
     pub async fn execute(self) -> Result<Table> {
         let embedding_registry = self.embedding_registry.clone();
@@ -229,6 +220,12 @@ impl<const HAS_DATA: bool> CreateTableBuilder<HAS_DATA> {
         self
     }
 
+    /// Apply the given write options when writing the initial data
+    pub fn write_options(mut self, write_options: WriteOptions) -> Self {
+        self.request.write_options = write_options;
+        self
+    }
+
     /// Set an option for the storage layer.
     ///
     /// Options already set on the connection will be inherited by the table,
@@ -621,15 +618,14 @@ pub struct ConnectRequest {
 
     /// The interval at which to check for updates from other processes.
     ///
-    /// If None, then consistency is not checked. For strong consistency, set this to
+    /// If None, then consistency is not checked. For performance
+    /// reasons, this is the default. For strong consistency, set this to
     /// zero seconds. Then every read will check for updates from other
     /// processes. As a compromise, you can set this to a non-zero timedelta
     /// for eventual consistency. If more than that interval has passed since
     /// the last check, then the table will be checked for updates. Note: this
     /// consistency only applies to read operations. Write operations are
     /// always consistent.
-    ///
-    /// The default is 5 seconds.
     pub read_consistency_interval: Option<std::time::Duration>,
 }
 
@@ -647,7 +643,7 @@ impl ConnectBuilder {
                 uri: uri.to_string(),
                 #[cfg(feature = "remote")]
                 client_config: Default::default(),
-                read_consistency_interval: DEFAULT_READ_CONSISTENCY_INTERVAL,
+                read_consistency_interval: None,
                 options: HashMap::new(),
             },
             embedding_registry: None,
@@ -786,7 +782,8 @@ impl ConnectBuilder {
     /// The interval at which to check for updates from other processes. This
     /// only affects LanceDB OSS.
     ///
-    /// If left unset, consistency is not checked. For strong consistency, set this to
+    /// If left unset, consistency is not checked. For maximum read
+    /// performance, this is the default. For strong consistency, set this to
     /// zero seconds. Then every read will check for updates from other processes.
     /// As a compromise, set this to a non-zero duration for eventual consistency.
     /// If more than that duration has passed since the last read, the read will
@@ -795,15 +792,13 @@ impl ConnectBuilder {
     /// This only affects read operations. Write operations are always
     /// consistent.
     ///
-    /// The default is 5 seconds.
-    ///
     /// LanceDB Cloud uses eventual consistency under the hood, and is not
     /// currently configurable.
     pub fn read_consistency_interval(
         mut self,
-        read_consistency_interval: Option<std::time::Duration>,
+        read_consistency_interval: std::time::Duration,
     ) -> Self {
-        self.request.read_consistency_interval = read_consistency_interval;
+        self.request.read_consistency_interval = Some(read_consistency_interval);
         self
     }
 
@@ -868,7 +863,7 @@ impl ConnectBuilder {
 /// # Arguments
 ///
 /// * `uri` - URI where the database is located, can be a local directory, supported remote cloud storage,
-///           or a LanceDB Cloud database.  See [ConnectOptions::uri] for a list of accepted formats
+///   or a LanceDB Cloud database.  See [ConnectOptions::uri] for a list of accepted formats
 pub fn connect(uri: &str) -> ConnectBuilder {
     ConnectBuilder::new(uri)
 }
@@ -887,7 +882,7 @@ impl CatalogConnectBuilder {
                 uri: uri.to_string(),
                 #[cfg(feature = "remote")]
                 client_config: Default::default(),
-                read_consistency_interval: DEFAULT_READ_CONSISTENCY_INTERVAL,
+                read_consistency_interval: None,
                 options: HashMap::new(),
             },
         }

rust/lancedb/src/data/inspect.rs

@@ -41,7 +41,7 @@ where
 /// ----------
 /// - reader: RecordBatchReader
 /// - strict: if set true, only `fixed_size_list<float>` is considered as vector column. If set to false,
-///           a `list<float>` column with same length is also considered as vector column.
+///   a `list<float>` column with same length is also considered as vector column.
 pub fn infer_vector_columns(
     reader: impl RecordBatchReader + Send,
     strict: bool,

rust/lancedb/src/database/listing.rs

@@ -201,7 +201,7 @@ impl ListingDatabaseOptionsBuilder {
 /// We will have two tables named `table1` and `table2`.
 #[derive(Debug)]
 pub struct ListingDatabase {
-    object_store: ObjectStore,
+    object_store: Arc<ObjectStore>,
     query_string: Option<String>,
 
     pub(crate) uri: String,

rust/lancedb/src/error.rs

@@ -35,6 +35,8 @@ pub enum Error {
     Schema { message: String },
     #[snafu(display("Runtime error: {message}"))]
     Runtime { message: String },
+    #[snafu(display("Timeout error: {message}"))]
+    Timeout { message: String },
 
     // 3rd party / external errors
     #[snafu(display("object_store error: {source}"))]

rust/lancedb/src/index.rs

@@ -1,11 +1,11 @@
 // SPDX-License-Identifier: Apache-2.0
 // SPDX-FileCopyrightText: Copyright The LanceDB Authors
 
-use std::sync::Arc;
-
 use scalar::FtsIndexBuilder;
 use serde::Deserialize;
 use serde_with::skip_serializing_none;
+use std::sync::Arc;
+use std::time::Duration;
 use vector::IvfFlatIndexBuilder;
 
 use crate::{table::BaseTable, DistanceType, Error, Result};
@@ -17,6 +17,7 @@ use self::{
 
 pub mod scalar;
 pub mod vector;
+pub mod waiter;
 
 /// Supported index types.
 #[derive(Debug, Clone)]
@@ -69,6 +70,7 @@ pub struct IndexBuilder {
     pub(crate) index: Index,
     pub(crate) columns: Vec<String>,
     pub(crate) replace: bool,
+    pub(crate) wait_timeout: Option<Duration>,
 }
 
 impl IndexBuilder {
@@ -78,6 +80,7 @@ impl IndexBuilder {
             index,
             columns,
             replace: true,
+            wait_timeout: None,
         }
     }
 
@@ -91,6 +94,15 @@ impl IndexBuilder {
         self
     }
 
+    /// Duration of time to wait for asynchronous indexing to complete. If not set,
+    /// `create_index()` will not wait.
+    ///
+    /// This is not supported for `NativeTable` since indexing is synchronous.
+    pub fn wait_timeout(mut self, d: Duration) -> Self {
+        self.wait_timeout = Some(d);
+        self
+    }
+
     pub async fn execute(self) -> Result<()> {
         self.parent.clone().create_index(self).await
     }

rust/lancedb/src/index/waiter.rs

@@ -0,0 +1,89 @@
+// SPDX-License-Identifier: Apache-2.0
+// SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+use crate::error::Result;
+use crate::table::BaseTable;
+use crate::Error;
+use log::debug;
+use std::time::{Duration, Instant};
+use tokio::time::sleep;
+
+const DEFAULT_SLEEP_MS: u64 = 1000;
+const MAX_WAIT: Duration = Duration::from_secs(2 * 60 * 60);
+
+/// Poll the table using list_indices() and index_stats() until all of the indices have 0 un-indexed rows.
+/// Will return Error::Timeout if the columns are not fully indexed within the timeout.
+pub async fn wait_for_index(
+    table: &dyn BaseTable,
+    index_names: &[&str],
+    timeout: Duration,
+) -> Result<()> {
+    if timeout > MAX_WAIT {
+        return Err(Error::InvalidInput {
+            message: format!("timeout must be less than {:?}", MAX_WAIT),
+        });
+    }
+    let start = Instant::now();
+    let mut remaining = index_names.to_vec();
+
+    // poll via list_indices() and index_stats() until all indices are created and fully indexed
+    while start.elapsed() < timeout {
+        let mut completed = vec![];
+        let indices = table.list_indices().await?;
+
+        for &idx in &remaining {
+            if !indices.iter().any(|i| i.name == *idx) {
+                debug!("still waiting for new index '{}'", idx);
+                continue;
+            }
+
+            let stats = table.index_stats(idx.as_ref()).await?;
+            match stats {
+                None => {
+                    debug!("still waiting for new index '{}'", idx);
+                    continue;
+                }
+                Some(s) => {
+                    if s.num_unindexed_rows == 0 {
+                        // note: this may never stabilize under constant writes.
+                        // we should later replace this with a status/job model
+                        completed.push(idx);
+                        debug!(
+                            "fully indexed '{}'. indexed rows: {}",
+                            idx, s.num_indexed_rows
+                        );
+                    } else {
+                        debug!(
+                            "still waiting for index '{}'. unindexed rows: {}",
+                            idx, s.num_unindexed_rows
+                        );
+                    }
+                }
+            }
+        }
+        remaining.retain(|idx| !completed.contains(idx));
+        if remaining.is_empty() {
+            return Ok(());
+        }
+        sleep(Duration::from_millis(DEFAULT_SLEEP_MS)).await;
+    }
+
+    // debug log index diagnostics
+    for &r in &remaining {
+        let stats = table.index_stats(r.as_ref()).await?;
+        match stats {
+            Some(s) => debug!(
+                "index '{}' not fully indexed after {:?}. stats: {:?}",
+                r, timeout, s
+            ),
+            None => debug!("index '{}' not found after {:?}", r, timeout),
+        }
+    }
+
+    Err(Error::Timeout {
+        message: format!(
+            "timed out waiting for indices: {:?} after {:?}",
+            remaining, timeout
+        ),
+    })
+}

rust/lancedb/src/io/object_store.rs

@@ -14,6 +14,9 @@ use object_store::{
 
 use async_trait::async_trait;
 
+#[cfg(test)]
+pub mod io_tracking;
+
 #[derive(Debug)]
 struct MirroringObjectStore {
     primary: Arc<dyn ObjectStore>,

rust/lancedb/src/io/object_store/io_tracking.rs

@@ -0,0 +1,237 @@
+// SPDX-License-Identifier: Apache-2.0
+// SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+use std::{
+    fmt::{Display, Formatter},
+    sync::{Arc, Mutex},
+};
+
+use bytes::Bytes;
+use futures::stream::BoxStream;
+use lance::io::WrappingObjectStore;
+use object_store::{
+    path::Path, GetOptions, GetResult, ListResult, MultipartUpload, ObjectMeta, ObjectStore,
+    PutMultipartOpts, PutOptions, PutPayload, PutResult, Result as OSResult, UploadPart,
+};
+
+#[derive(Debug, Default)]
+pub struct IoStats {
+    pub read_iops: u64,
+    pub read_bytes: u64,
+    pub write_iops: u64,
+    pub write_bytes: u64,
+}
+
+impl Display for IoStats {
+    fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
+        write!(f, "{:#?}", self)
+    }
+}
+
+#[derive(Debug, Clone)]
+pub struct IoTrackingStore {
+    target: Arc<dyn ObjectStore>,
+    stats: Arc<Mutex<IoStats>>,
+}
+
+impl Display for IoTrackingStore {
+    fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
+        write!(f, "{:#?}", self)
+    }
+}
+
+#[derive(Debug, Default, Clone)]
+pub struct IoStatsHolder(Arc<Mutex<IoStats>>);
+
+impl IoStatsHolder {
+    pub fn incremental_stats(&self) -> IoStats {
+        std::mem::take(&mut self.0.lock().expect("failed to lock IoStats"))
+    }
+}
+
+impl WrappingObjectStore for IoStatsHolder {
+    fn wrap(&self, target: Arc<dyn ObjectStore>) -> Arc<dyn ObjectStore> {
+        Arc::new(IoTrackingStore {
+            target,
+            stats: self.0.clone(),
+        })
+    }
+}
+
+impl IoTrackingStore {
+    pub fn new_wrapper() -> (Arc<dyn WrappingObjectStore>, Arc<Mutex<IoStats>>) {
+        let stats = Arc::new(Mutex::new(IoStats::default()));
+        (Arc::new(IoStatsHolder(stats.clone())), stats)
+    }
+
+    fn record_read(&self, num_bytes: u64) {
+        let mut stats = self.stats.lock().unwrap();
+        stats.read_iops += 1;
+        stats.read_bytes += num_bytes;
+    }
+
+    fn record_write(&self, num_bytes: u64) {
+        let mut stats = self.stats.lock().unwrap();
+        stats.write_iops += 1;
+        stats.write_bytes += num_bytes;
+    }
+}
+
+#[async_trait::async_trait]
+#[deny(clippy::missing_trait_methods)]
+impl ObjectStore for IoTrackingStore {
+    async fn put(&self, location: &Path, bytes: PutPayload) -> OSResult<PutResult> {
+        self.record_write(bytes.content_length() as u64);
+        self.target.put(location, bytes).await
+    }
+
+    async fn put_opts(
+        &self,
+        location: &Path,
+        bytes: PutPayload,
+        opts: PutOptions,
+    ) -> OSResult<PutResult> {
+        self.record_write(bytes.content_length() as u64);
+        self.target.put_opts(location, bytes, opts).await
+    }
+
+    async fn put_multipart(&self, location: &Path) -> OSResult<Box<dyn MultipartUpload>> {
+        let target = self.target.put_multipart(location).await?;
+        Ok(Box::new(IoTrackingMultipartUpload {
+            target,
+            stats: self.stats.clone(),
+        }))
+    }
+
+    async fn put_multipart_opts(
+        &self,
+        location: &Path,
+        opts: PutMultipartOpts,
+    ) -> OSResult<Box<dyn MultipartUpload>> {
+        let target = self.target.put_multipart_opts(location, opts).await?;
+        Ok(Box::new(IoTrackingMultipartUpload {
+            target,
+            stats: self.stats.clone(),
+        }))
+    }
+
+    async fn get(&self, location: &Path) -> OSResult<GetResult> {
+        let result = self.target.get(location).await;
+        if let Ok(result) = &result {
+            let num_bytes = result.range.end - result.range.start;
+            self.record_read(num_bytes as u64);
+        }
+        result
+    }
+
+    async fn get_opts(&self, location: &Path, options: GetOptions) -> OSResult<GetResult> {
+        let result = self.target.get_opts(location, options).await;
+        if let Ok(result) = &result {
+            let num_bytes = result.range.end - result.range.start;
+            self.record_read(num_bytes as u64);
+        }
+        result
+    }
+
+    async fn get_range(&self, location: &Path, range: std::ops::Range<usize>) -> OSResult<Bytes> {
+        let result = self.target.get_range(location, range).await;
+        if let Ok(result) = &result {
+            self.record_read(result.len() as u64);
+        }
+        result
+    }
+
+    async fn get_ranges(
+        &self,
+        location: &Path,
+        ranges: &[std::ops::Range<usize>],
+    ) -> OSResult<Vec<Bytes>> {
+        let result = self.target.get_ranges(location, ranges).await;
+        if let Ok(result) = &result {
+            self.record_read(result.iter().map(|b| b.len() as u64).sum());
+        }
+        result
+    }
+
+    async fn head(&self, location: &Path) -> OSResult<ObjectMeta> {
+        self.record_read(0);
+        self.target.head(location).await
+    }
+
+    async fn delete(&self, location: &Path) -> OSResult<()> {
+        self.record_write(0);
+        self.target.delete(location).await
+    }
+
+    fn delete_stream<'a>(
+        &'a self,
+        locations: BoxStream<'a, OSResult<Path>>,
+    ) -> BoxStream<'a, OSResult<Path>> {
+        self.target.delete_stream(locations)
+    }
+
+    fn list(&self, prefix: Option<&Path>) -> BoxStream<'_, OSResult<ObjectMeta>> {
+        self.record_read(0);
+        self.target.list(prefix)
+    }
+
+    fn list_with_offset(
+        &self,
+        prefix: Option<&Path>,
+        offset: &Path,
+    ) -> BoxStream<'_, OSResult<ObjectMeta>> {
+        self.record_read(0);
+        self.target.list_with_offset(prefix, offset)
+    }
+
+    async fn list_with_delimiter(&self, prefix: Option<&Path>) -> OSResult<ListResult> {
+        self.record_read(0);
+        self.target.list_with_delimiter(prefix).await
+    }
+
+    async fn copy(&self, from: &Path, to: &Path) -> OSResult<()> {
+        self.record_write(0);
+        self.target.copy(from, to).await
+    }
+
+    async fn rename(&self, from: &Path, to: &Path) -> OSResult<()> {
+        self.record_write(0);
+        self.target.rename(from, to).await
+    }
+
+    async fn rename_if_not_exists(&self, from: &Path, to: &Path) -> OSResult<()> {
+        self.record_write(0);
+        self.target.rename_if_not_exists(from, to).await
+    }
+
+    async fn copy_if_not_exists(&self, from: &Path, to: &Path) -> OSResult<()> {
+        self.record_write(0);
+        self.target.copy_if_not_exists(from, to).await
+    }
+}
+
+#[derive(Debug)]
+struct IoTrackingMultipartUpload {
+    target: Box<dyn MultipartUpload>,
+    stats: Arc<Mutex<IoStats>>,
+}
+
+#[async_trait::async_trait]
+impl MultipartUpload for IoTrackingMultipartUpload {
+    async fn abort(&mut self) -> OSResult<()> {
+        self.target.abort().await
+    }
+
+    async fn complete(&mut self) -> OSResult<PutResult> {
+        self.target.complete().await
+    }
+
+    fn put_part(&mut self, payload: PutPayload) -> UploadPart {
+        {
+            let mut stats = self.stats.lock().unwrap();
+            stats.write_iops += 1;
+            stats.write_bytes += payload.content_length() as u64;
+        }
+        self.target.put_part(payload)
+    }
+}

rust/lancedb/src/lib.rs

@@ -31,7 +31,7 @@
 //! are not yet ready to be released.
 //!
 //! - `remote` - Enable remote client to connect to LanceDB cloud.  This is not yet fully implemented
-//!              and should not be enabled.
+//!   and should not be enabled.
 //!
 //! ### Quick Start
 //!

rust/lancedb/src/query.rs

@@ -1,8 +1,8 @@
 // SPDX-License-Identifier: Apache-2.0
 // SPDX-FileCopyrightText: Copyright The LanceDB Authors
 
-use std::future::Future;
 use std::sync::Arc;
+use std::{future::Future, time::Duration};
 
 use arrow::compute::concat_batches;
 use arrow_array::{make_array, Array, Float16Array, Float32Array, Float64Array};
@@ -25,6 +25,7 @@ use crate::error::{Error, Result};
 use crate::rerankers::rrf::RRFReranker;
 use crate::rerankers::{check_reranker_result, NormalizeMethod, Reranker};
 use crate::table::BaseTable;
+use crate::utils::TimeoutStream;
 use crate::DistanceType;
 use crate::{arrow::SendableRecordBatchStream, table::AnyQuery};
 
@@ -525,12 +526,15 @@ pub struct QueryExecutionOptions {
     ///
     /// By default, this is 1024
     pub max_batch_length: u32,
+    /// Max duration to wait for the query to execute before timing out.
+    pub timeout: Option<Duration>,
 }
 
 impl Default for QueryExecutionOptions {
     fn default() -> Self {
         Self {
             max_batch_length: 1024,
+            timeout: None,
         }
     }
 }
@@ -1007,7 +1011,10 @@ impl VectorQuery {
         self
     }
 
-    pub async fn execute_hybrid(&self) -> Result<SendableRecordBatchStream> {
+    pub async fn execute_hybrid(
+        &self,
+        options: QueryExecutionOptions,
+    ) -> Result<SendableRecordBatchStream> {
         // clone query and specify we want to include row IDs, which can be needed for reranking
         let mut fts_query = Query::new(self.parent.clone());
         fts_query.request = self.request.base.clone();
@@ -1016,7 +1023,10 @@ impl VectorQuery {
         let mut vector_query = self.clone().with_row_id();
 
         vector_query.request.base.full_text_search = None;
-        let (fts_results, vec_results) = try_join!(fts_query.execute(), vector_query.execute())?;
+        let (fts_results, vec_results) = try_join!(
+            fts_query.execute_with_options(options.clone()),
+            vector_query.inner_execute_with_options(options)
+        )?;
 
         let (fts_results, vec_results) = try_join!(
             fts_results.try_collect::<Vec<_>>(),
@@ -1074,6 +1084,20 @@ impl VectorQuery {
             RecordBatchStreamAdapter::new(results.schema(), stream::iter([Ok(results)])),
         ))
     }
+
+    async fn inner_execute_with_options(
+        &self,
+        options: QueryExecutionOptions,
+    ) -> Result<SendableRecordBatchStream> {
+        let plan = self.create_plan(options.clone()).await?;
+        let inner = execute_plan(plan, Default::default())?;
+        let inner = if let Some(timeout) = options.timeout {
+            TimeoutStream::new_boxed(inner, timeout)
+        } else {
+            inner
+        };
+        Ok(DatasetRecordBatchStream::new(inner).into())
+    }
 }
 
 impl ExecutableQuery for VectorQuery {
@@ -1087,16 +1111,13 @@ impl ExecutableQuery for VectorQuery {
         options: QueryExecutionOptions,
     ) -> Result<SendableRecordBatchStream> {
         if self.request.base.full_text_search.is_some() {
-            let hybrid_result = async move { self.execute_hybrid().await }.boxed().await?;
+            let hybrid_result = async move { self.execute_hybrid(options).await }
+                .boxed()
+                .await?;
             return Ok(hybrid_result);
         }
 
-        Ok(SendableRecordBatchStream::from(
-            DatasetRecordBatchStream::new(execute_plan(
-                self.create_plan(options).await?,
-                Default::default(),
-            )?),
-        ))
+        self.inner_execute_with_options(options).await
     }
 
     async fn explain_plan(&self, verbose: bool) -> Result<String> {