feat(nodejs): surface skip_auto_cleanup on add and merge insert

### Summary 

Adds a `connectNamespace(implName, properties, options?)` to the NodeJS
SDK`. Closes #3380.

### Testing
- pnpm test
- Ran smoke test

```
import { connectNamespace } from "lancedb"
import { tmpdir } from "os";
import { mkdtempSync } from "fs";
import { join } from "path";

const dir = mkdtempSync(join(tmpdir(), "lancedb-connect-namespace-smoke-"));
console.log(`Using temp dir: ${dir}\n`);

// 1. Happy path: connect via the "dir" namespace impl, create + list a table.
console.log('Connecting via connectNamespace("dir", { root })...');
const db = await connectNamespace("dir", { root: dir });
console.log("  ✓ connected:", db.display());

console.log("Creating a table and listing it...");
await db.createTable("users", [
  { id: 1, name: "alice" },
  { id: 2, name: "bob" },
]);
console.log("  ✓ tableNames ->", await db.tableNames());

const table = await db.openTable("users");
console.log("  ✓ users.countRows ->", await table.countRows());

// 2. Storage options pass-through.
console.log("\nReconnecting with storageOptions (plumbing check)...");
const dbWithOpts = await connectNamespace(
  "dir",
  { root: dir },
  { storageOptions: { newTableDataStorageVersion: "stable" } },
);
console.log("  ✓ connected with storageOptions:", dbWithOpts.display());
await dbWithOpts.close();

// 3. Empty implName -> clear error.
console.log("\nCalling connectNamespace('', {}) (expect error)...");
try {
  await connectNamespace("", {});
  console.error("  UNEXPECTED: empty implName did not throw");
} catch (err) {
  console.log(`  ✓ Got expected error: ${err.message.split("\n")[0]}`);
}

// 4. Unknown impl -> error.
console.log("\nCalling connectNamespace('not-a-real-impl', {}) (expect error)...");
try {
  await connectNamespace("not-a-real-impl", {});
  console.error("  UNEXPECTED: unknown impl did not throw");
} catch (err) {
  console.log(`  ✓ Got expected error: ${err.message.split("\n")[0]}`);
}

// 5. Create a table inside a child namespace, then reconnect with a fresh
//    connectNamespace call and confirm the table is reachable via that
//    namespace path. (The dir+manifest impl keeps the namespace hierarchy in
//    a root manifest, so "scoping" happens via namespacePath args, not by
//    pointing root at a subdir.)
console.log("\nCreating a table inside a child namespace...");
const dir2 = mkdtempSync(join(tmpdir(), "lancedb-connect-namespace-smoke-"));
const writer = await connectNamespace("dir", {
  root: dir2,
  manifest_enabled: "true",
});
await writer.createNamespace(["analytics"]);
await writer.createTable(
  "orders",
  [
    { id: 1, total: 10 },
    { id: 2, total: 20 },
  ],
  ["analytics"],
);
console.log(
  "  ✓ writer sees tables under [analytics] ->",
  await writer.tableNames(["analytics"]),
);
await writer.close();

console.log("Reconnecting and reading the table via its namespace path...");
const reader = await connectNamespace("dir", {
  root: dir2,
  manifest_enabled: "true",
});
console.log(
  "  ✓ reader tableNames(['analytics']) ->",
  await reader.tableNames(["analytics"]),
);
const orders = await reader.openTable("orders", ["analytics"]);
console.log("  ✓ orders.countRows via reader ->", await orders.countRows());
await reader.close();

await db.close();
console.log("\nAll checks passed.");
```

```
Using temp dir: /var/folders/bj/hn6jv9c50y301d1nx0y8xmn00000gn/T/lancedb-connect-namespace-smoke-WByF1P

Connecting via connectNamespace("dir", { root })...
  ✓ connected: LanceNamespaceDatabase
Creating a table and listing it...
  ✓ tableNames -> [ 'users' ]
  ✓ users.countRows -> 2

Reconnecting with storageOptions (plumbing check)...
  ✓ connected with storageOptions: LanceNamespaceDatabase

Calling connectNamespace('', {}) (expect error)...
  ✓ Got expected error: implName must be a non-empty string

Calling connectNamespace('not-a-real-impl', {}) (expect error)...
  ✓ Got expected error: Invalid input, Failed to connect to namespace: Namespace { source: Unsupported { message: "Implementation 'not-a-real-impl' is not available. Supported: dir, rest" }, location: Location { file: "/Users/brendan/.cargo/git/checkouts/lance-8ddea23c38163eda/f693245/rust/lance-namespace-impls/src/connect.rs", line: 216, column: 14 } }

Creating a table inside a child namespace...
  ✓ writer sees tables under [analytics] -> [ 'orders' ]
Reconnecting and reading the table via its namespace path...
  ✓ reader tableNames(['analytics']) -> [ 'orders' ]
  ✓ orders.countRows via reader -> 2

All checks passed.
```

### Docs
- regenerated docs

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.27.0-beta.5"
+current_version = "0.28.0-beta.11"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

.github/ISSUE_TEMPLATE/documentation.yml

@@ -18,6 +18,6 @@ body:
       label: Link
       description: >
         Provide a link to the existing documentation, if applicable.
-      placeholder: ex. https://lancedb.com/docs/tables/...
+      placeholder: ex. https://docs.lancedb.com/tables/...
     validations:
       required: false

.github/dependabot.yml

@@ -0,0 +1,18 @@
+version: 2
+
+# Scope: the root Cargo workspace, which produces the Rust binaries we
+# ship to users (the Node.js and Python native extensions). The
+# `rust/lancedb` library crate shares the same lockfile; its consumers
+# pick their own dependency versions, but bumping transitive deps here
+# keeps the binaries we ship current.
+updates:
+  - package-ecosystem: cargo
+    directory: /
+    schedule:
+      interval: weekly
+    open-pull-requests-limit: 10
+    groups:
+      rust-minor-patch:
+        update-types:
+          - minor
+          - patch

.github/workflows/build_linux_wheel/action.yml

@@ -23,8 +23,10 @@ runs:
   steps:
     - name: CONFIRM ARM BUILD
       shell: bash
+      env:
+        ARM_BUILD: ${{ inputs.arm-build }}
       run: |
-        echo "ARM BUILD: ${{ inputs.arm-build }}"
+        echo "ARM BUILD: $ARM_BUILD"
     - name: Build x86_64 Manylinux wheel
       if: ${{ inputs.arm-build == 'false' }}
       uses: PyO3/maturin-action@v1

.github/workflows/codex-fix-ci.yml

@@ -45,7 +45,9 @@ jobs:
       - name: Set up Node.js
         uses: actions/setup-node@v4
         with:
-          node-version: 20
+          # pnpm 11 (used by the nodejs install step below) requires
+          # Node >= 22.13; use 24 since 22 hits EOL in October.
+          node-version: 24
 
       - name: Install Codex CLI
         run: npm install -g @openai/codex
@@ -79,10 +81,14 @@ jobs:
           java-version: '11'
           cache: maven
 
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v4
+        with:
+          version: 11.1.1
       - name: Install Node.js dependencies for TypeScript bindings
         run: |
           cd nodejs
-          npm ci
+          pnpm install --frozen-lockfile
 
       - name: Configure git user
         run: |
@@ -137,7 +143,7 @@ jobs:
                - For Rust test failures: Run the specific test with "cargo test -p <crate> <test_name>"
                - For Python test failures: Build with "cd python && maturin develop" then run "pytest <specific_test_file>::<test_name>"
                - For Java test failures: Run "cd java && mvn test -Dtest=<TestClass>#<testMethod>"
-               - For TypeScript test failures: Run "cd nodejs && npm run build && npm test -- --testNamePattern='<test_name>'"
+               - For TypeScript test failures: Run "cd nodejs && pnpm build && pnpm test -- --testNamePattern='<test_name>'"
                - Do NOT run the full test suite - only run the tests that were failing
 
           7. If the additional guidelines are provided, follow them as well.

.github/workflows/dev.yml

@@ -8,6 +8,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   labeler:
     permissions:

.github/workflows/java-publish.yml

@@ -19,6 +19,9 @@ on:
     paths:
       - .github/workflows/java-publish.yml
 
+permissions:
+  contents: read
+
 jobs:
   publish:
     name: Build and Publish
@@ -40,7 +43,7 @@ jobs:
           server-username: SONATYPE_USER
           server-password: SONATYPE_TOKEN
           gpg-private-key: ${{ secrets.GPG_PRIVATE_KEY }}
-          gpg-passphrase: ${{ secrets.GPG_PASSPHRASE }}
+          gpg-passphrase: MAVEN_GPG_PASSPHRASE
       - name: Set git config
         run: |
           git config --global user.email "dev+gha@lancedb.com"
@@ -55,10 +58,11 @@ jobs:
           echo "use-agent" >> ~/.gnupg/gpg.conf
           echo "pinentry-mode loopback" >> ~/.gnupg/gpg.conf
           export GPG_TTY=$(tty)
-          ./mvnw --batch-mode -DskipTests -DpushChanges=false -Dgpg.passphrase=${{ secrets.GPG_PASSPHRASE }} deploy -pl lancedb-core -am -P deploy-to-ossrh
+          ./mvnw --batch-mode -DskipTests -DpushChanges=false deploy -pl lancedb-core -am -P deploy-to-ossrh
         env:
           SONATYPE_USER: ${{ secrets.SONATYPE_USER }}
           SONATYPE_TOKEN: ${{ secrets.SONATYPE_TOKEN }}
+          MAVEN_GPG_PASSPHRASE: ${{ secrets.GPG_PASSPHRASE }}
 
   report-failure:
     name: Report Workflow Failure

.github/workflows/java.yml

@@ -24,6 +24,9 @@ on:
       - java/**
       - .github/workflows/java.yml
 
+permissions:
+  contents: read
+
 jobs:
   build-java:
     runs-on: ubuntu-24.04

.github/workflows/license-header-check.yml

@@ -10,6 +10,10 @@ on:
       - nodejs/**
       - java/**
       - .github/workflows/license-header-check.yml
+
+permissions:
+  contents: read
+
 jobs:
   check-licenses:
     runs-on: ubuntu-latest

.github/workflows/nodejs.yml

@@ -7,12 +7,17 @@ on:
   pull_request:
     paths:
       - Cargo.toml
+      - Cargo.lock
+      - rust-toolchain.toml
       - nodejs/**
       - rust/**
       - docs/src/js/**
       - .github/workflows/nodejs.yml
       - docker-compose.yml
 
+permissions:
+  contents: read
+
 concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
   cancel-in-progress: true
@@ -37,11 +42,17 @@ jobs:
       with:
         fetch-depth: 0
         lfs: true
+    - uses: pnpm/action-setup@v4
+      with:
+        version: 11.1.1
     - uses: actions/setup-node@v4
       with:
-        node-version: 20
-        cache: 'npm'
-        cache-dependency-path: nodejs/package-lock.json
+        # pnpm 11 requires Node >= 22.13; use 24 since 22 hits EOL
+        # in October. The library itself still supports Node >= 18
+        # (see test matrix below).
+        node-version: 24
+        cache: 'pnpm'
+        cache-dependency-path: nodejs/pnpm-lock.yaml
     - uses: actions-rust-lang/setup-rust-toolchain@v1
       with:
         components: rustfmt, clippy
@@ -56,11 +67,13 @@ jobs:
       run: cargo clippy --profile ci --all --all-features -- -D warnings
     - name: Lint Typescript
       run: |
-        npm ci
-        npm run lint-ci
+        pnpm install --frozen-lockfile
+        pnpm lint-ci
     - name: Lint examples
       working-directory: nodejs/examples
-      run: npm ci && npm run lint-ci
+      # The `@lancedb/lancedb` dep points at file:../dist; pnpm errors if
+      # that dir is missing, so create an empty one for lint-only runs.
+      run: mkdir -p ../dist && pnpm install --frozen-lockfile && pnpm lint-ci
   linux:
     name: Linux (NodeJS ${{ matrix.node-version }})
     timeout-minutes: 30
@@ -77,14 +90,18 @@ jobs:
       with:
         fetch-depth: 0
         lfs: true
-    - uses: actions/setup-node@v4
-      name: Setup Node.js 20 for build
+    - uses: pnpm/action-setup@v4
       with:
-        # @napi-rs/cli v3 requires Node >= 20.12 (via @inquirer/prompts@8).
-        # Build always on Node 20; tests run on the matrix version below.
-        node-version: 20
-        cache: 'npm'
-        cache-dependency-path: nodejs/package-lock.json
+        version: 11.1.1
+    - uses: actions/setup-node@v4
+      name: Setup Node.js 24 for build
+      with:
+        # pnpm 11 requires Node >= 22.13; use 24 since 22 hits EOL
+        # in October. Build/install runs on Node 24; tests run on the
+        # matrix version below using direct jest invocation.
+        node-version: 24
+        cache: 'pnpm'
+        cache-dependency-path: nodejs/pnpm-lock.yaml
     - uses: Swatinem/rust-cache@v2
     - name: Install dependencies
       run: |
@@ -92,45 +109,52 @@ jobs:
         sudo apt install -y protobuf-compiler libssl-dev
     - name: Build
       run: |
-        npm ci --include=optional
-        npm run build:debug -- --profile ci
+        pnpm install --frozen-lockfile
+        # No `--` separator: pnpm forwards it literally, which would
+        # make napi-rs treat `--profile ci` as a cargo passthrough arg.
+        pnpm build:debug --profile ci
+        pnpm tsc
+    - name: Setup examples
+      working-directory: nodejs/examples
+      run: pnpm install --frozen-lockfile
+    - name: Check docs
+      run: |
+        # We run this as part of the job because the binary needs to be built
+        # first to export the types of the native code.
+        set -e
+        # `pnpm docs` would invoke pnpm's built-in `docs` command, not
+        # the script — use `pnpm run docs`.
+        pnpm run docs
+        if ! git diff --exit-code -- ../ ':(exclude)Cargo.lock'; then
+          echo "Docs need to be updated"
+          echo "Run 'pnpm run docs', fix any warnings, and commit the changes."
+          exit 1
+        fi
     - uses: actions/setup-node@v4
       name: Setup Node.js ${{ matrix.node-version }} for test
       with:
         node-version: ${{ matrix.node-version }}
-    - name: Compile TypeScript
-      run: npm run tsc
     - name: Setup localstack
       working-directory: .
       run: docker compose up --detach --wait
     - name: Test
       env:
         S3_TEST: "1"
-      run: npm run test
-    - name: Setup examples
-      working-directory: nodejs/examples
-      run: npm ci
+        # Newer @smithy/core uses dynamic ESM imports.
+        NODE_OPTIONS: "--experimental-vm-modules"
+      # Invoke jest directly because pnpm 11 itself requires Node 22+
+      # while the matrix tests on older Node versions.
+      run: npx jest --verbose
     - name: Test examples
       working-directory: ./
       env:
         OPENAI_API_KEY: test
         OPENAI_BASE_URL: http://0.0.0.0:8000
+        NODE_OPTIONS: "--experimental-vm-modules"
       run: |
         python ci/mock_openai.py &
         cd nodejs/examples
-        npm test
-    - name: Check docs
-      run: |
-        # We run this as part of the job because the binary needs to be built
-        # first to export the types of the native code.
-        set -e
-        npm ci
-        npm run docs
-        if ! git diff --exit-code -- ../ ':(exclude)Cargo.lock'; then
-          echo "Docs need to be updated"
-          echo "Run 'npm run docs', fix any warnings, and commit the changes."
-          exit 1
-        fi
+        npx jest --testEnvironment jest-environment-node-single-context --verbose
   macos:
     timeout-minutes: 30
     runs-on: "macos-14"
@@ -143,20 +167,28 @@ jobs:
       with:
         fetch-depth: 0
         lfs: true
+    - uses: pnpm/action-setup@v4
+      with:
+        version: 11.1.1
     - uses: actions/setup-node@v4
       with:
-        node-version: 20
-        cache: 'npm'
-        cache-dependency-path: nodejs/package-lock.json
+        # pnpm 11 requires Node >= 22.13; use 24 since 22 hits EOL
+        # in October.
+        node-version: 24
+        cache: 'pnpm'
+        cache-dependency-path: nodejs/pnpm-lock.yaml
+    - uses: dtolnay/rust-toolchain@stable
     - uses: Swatinem/rust-cache@v2
     - name: Install dependencies
       run: |
         brew install protobuf
     - name: Build
       run: |
-        npm ci --include=optional
-        npm run build:debug -- --profile ci
-        npm run tsc
+        pnpm install --frozen-lockfile
+        # No `--` separator: pnpm forwards it literally, which would
+        # make napi-rs treat `--profile ci` as a cargo passthrough arg.
+        pnpm build:debug --profile ci
+        pnpm tsc
     - name: Test
       run: |
-        npm run test
+        pnpm test

.github/workflows/npm-publish.yml

@@ -19,6 +19,7 @@ on:
     paths:
       - .github/workflows/npm-publish.yml
       - Cargo.toml # Change in dependency frequently breaks builds
+      - Cargo.lock
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
@@ -124,7 +125,12 @@ jobs:
             pre_build: |-
               set -e &&
               apt-get update &&
-              apt-get install -y protobuf-compiler pkg-config
+              apt-get install -y protobuf-compiler pkg-config &&
+              # The base image (manylinux2014-cross) sets TARGET_CC to the old
+              # GCC 4.8 cross-compiler. aws-lc-sys checks TARGET_CC before CC,
+              # so it picks up GCC even though the napi-rs image sets CC=clang.
+              # Override to use the image's clang-18 which supports -fuse-ld=lld.
+              export TARGET_CC=clang TARGET_CXX=clang++
           - target: x86_64-unknown-linux-musl
             # This one seems to need some extra memory
             host: ubuntu-2404-8x-x64
@@ -144,9 +150,10 @@ jobs:
               set -e &&
               apt-get update &&
               apt-get install -y protobuf-compiler pkg-config &&
-              # https://github.com/aws/aws-lc-rs/issues/737#issuecomment-2725918627
-              ln -s /usr/aarch64-unknown-linux-gnu/lib/gcc/aarch64-unknown-linux-gnu/4.8.5/crtbeginS.o /usr/aarch64-unknown-linux-gnu/aarch64-unknown-linux-gnu/sysroot/usr/lib/crtbeginS.o &&
-              ln -s /usr/aarch64-unknown-linux-gnu/lib/gcc /usr/aarch64-unknown-linux-gnu/aarch64-unknown-linux-gnu/sysroot/usr/lib/gcc &&
+              export TARGET_CC=clang TARGET_CXX=clang++ &&
+              # The manylinux2014 sysroot has glibc 2.17 headers which lack
+              # AT_HWCAP2 (added in Linux 3.17). Define it for aws-lc-sys.
+              export CFLAGS="$CFLAGS -DAT_HWCAP2=26" &&
               rustup target add aarch64-unknown-linux-gnu
           - target: aarch64-unknown-linux-musl
             host: ubuntu-2404-8x-x64
@@ -164,13 +171,18 @@ jobs:
         working-directory: nodejs
     steps:
       - uses: actions/checkout@v4
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v4
+        with:
+          version: 11.1.1
       - name: Setup node
         uses: actions/setup-node@v4
-        if: ${{ !matrix.settings.docker }}
         with:
-          node-version: 20
-          cache: npm
-          cache-dependency-path: nodejs/package-lock.json
+          # pnpm 11 requires Node >= 22.13; use 24 since 22 hits EOL
+          # in October.
+          node-version: 24
+          cache: pnpm
+          cache-dependency-path: nodejs/pnpm-lock.yaml
       - name: Install
         uses: dtolnay/rust-toolchain@stable
         if: ${{ !matrix.settings.docker }}
@@ -188,7 +200,7 @@ jobs:
             target/
           key: nodejs-${{ matrix.settings.target }}-cargo-${{ matrix.settings.host }}
       - name: Install dependencies
-        run: npm ci
+        run: pnpm install --frozen-lockfile
       - name: Install Zig
         uses: mlugg/setup-zig@v2
         if: ${{ contains(matrix.settings.target, 'musl') }}
@@ -241,7 +253,7 @@ jobs:
       # one to do the upload.
       - name: Make generic artifacts
         if: ${{ matrix.settings.target == 'aarch64-apple-darwin' }}
-        run: npm run tsc
+        run: pnpm tsc
       - name: Upload Generic Artifacts
         if: ${{ matrix.settings.target == 'aarch64-apple-darwin' }}
         uses: actions/upload-artifact@v4
@@ -266,7 +278,7 @@ jobs:
           - target: x86_64-unknown-linux-gnu
             host: ubuntu-latest
           - target: aarch64-unknown-linux-gnu
-            host: buildjet-16vcpu-ubuntu-2204-arm
+            host: ubuntu-2404-8x-arm64
         node:
           - '20'
     runs-on: ${{ matrix.settings.host }}
@@ -276,14 +288,24 @@ jobs:
         working-directory: nodejs
     steps:
       - uses: actions/checkout@v4
-      - name: Setup node
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v4
+        with:
+          version: 11.1.1
+      - name: Setup Node.js 24 for install
+        uses: actions/setup-node@v4
+        with:
+          # pnpm 11 requires Node >= 22.13; use 24 since 22 hits EOL
+          # in October.
+          node-version: 24
+          cache: pnpm
+          cache-dependency-path: nodejs/pnpm-lock.yaml
+      - name: Install dependencies
+        run: pnpm install --frozen-lockfile
+      - name: Setup Node.js ${{ matrix.node }} for test
         uses: actions/setup-node@v4
         with:
           node-version: ${{ matrix.node }}
-          cache: npm
-          cache-dependency-path: nodejs/package-lock.json
-      - name: Install dependencies
-        run: npm ci
       - name: Download artifacts
         uses: actions/download-artifact@v4
         with:
@@ -304,7 +326,9 @@ jobs:
       - name: Move built files
         run: cp dist/native.d.ts dist/native.js dist/*.node lancedb/
       - name: Test bindings
-        run: npm test
+        # Invoke jest directly because pnpm 11 itself requires Node 22+
+        # while the matrix tests on older Node versions.
+        run: npx jest --verbose
   publish:
     name: Publish
     runs-on: ubuntu-latest
@@ -316,15 +340,19 @@ jobs:
       - test-lancedb
     steps:
       - uses: actions/checkout@v4
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v4
+        with:
+          version: 11.1.1
       - name: Setup node
         uses: actions/setup-node@v4
         with:
           node-version: 24
-          cache: npm
-          cache-dependency-path: nodejs/package-lock.json
+          cache: pnpm
+          cache-dependency-path: nodejs/pnpm-lock.yaml
           registry-url: "https://registry.npmjs.org"
       - name: Install dependencies
-        run: npm ci
+        run: pnpm install --frozen-lockfile
       - uses: actions/download-artifact@v4
         with:
           name: nodejs-dist
@@ -344,7 +372,7 @@ jobs:
       - name: Display structure of downloaded files
         run: find dist && find nodejs-artifacts
       - name: Move artifacts
-        run: npx napi artifacts -d nodejs-artifacts
+        run: pnpm exec napi artifacts -d nodejs-artifacts
       - name: List packages
         run: find npm
       - name: Publish

.github/workflows/pypi-publish.yml

@@ -9,14 +9,21 @@ on:
     paths:
       - .github/workflows/pypi-publish.yml
       - Cargo.toml # Change in dependency frequently breaks builds
+      - Cargo.lock
 
 env:
   PIP_EXTRA_INDEX_URL: "https://pypi.fury.io/lance-format/ https://pypi.fury.io/lancedb/"
 
+permissions:
+  contents: read
+
 jobs:
   linux:
     name: Python ${{ matrix.config.platform }} manylinux${{ matrix.config.manylinux }}
     timeout-minutes: 60
+    permissions:
+      id-token: write
+      contents: read
     strategy:
       matrix:
         config:
@@ -56,10 +63,12 @@ jobs:
       - uses: ./.github/workflows/upload_wheel
         if: startsWith(github.ref, 'refs/tags/python-v')
         with:
-          pypi_token: ${{ secrets.LANCEDB_PYPI_API_TOKEN }}
           fury_token: ${{ secrets.FURY_TOKEN }}
   mac:
     timeout-minutes: 90
+    permissions:
+      id-token: write
+      contents: read
     runs-on: ${{ matrix.config.runner }}
     strategy:
       matrix:
@@ -84,10 +93,12 @@ jobs:
       - uses: ./.github/workflows/upload_wheel
         if: startsWith(github.ref, 'refs/tags/python-v')
         with:
-          pypi_token: ${{ secrets.LANCEDB_PYPI_API_TOKEN }}
           fury_token: ${{ secrets.FURY_TOKEN }}
   windows:
     timeout-minutes: 60
+    permissions:
+      id-token: write
+      contents: read
     runs-on: windows-latest
     steps:
       - uses: actions/checkout@v4
@@ -106,7 +117,6 @@ jobs:
       - uses: ./.github/workflows/upload_wheel
         if: startsWith(github.ref, 'refs/tags/python-v')
         with:
-          pypi_token: ${{ secrets.LANCEDB_PYPI_API_TOKEN }}
           fury_token: ${{ secrets.FURY_TOKEN }}
   gh-release:
     if: startsWith(github.ref, 'refs/tags/python-v')

.github/workflows/python.yml

@@ -7,6 +7,8 @@ on:
   pull_request:
     paths:
       - Cargo.toml
+      - Cargo.lock
+      - rust-toolchain.toml
       - python/**
       - rust/**
       - .github/workflows/python.yml
@@ -15,6 +17,9 @@ on:
       - .github/workflows/build_windows_wheel/**
       - .github/workflows/run_tests/**
 
+permissions:
+  contents: read
+
 concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
   cancel-in-progress: true
@@ -106,7 +111,6 @@ jobs:
       - name: Install
         run: |
           pip install --extra-index-url https://pypi.fury.io/lance-format/ --extra-index-url https://pypi.fury.io/lancedb/ -e .[tests,dev,embeddings]
-          pip install tantivy
           pip install mlx
       - name: Doctest
         run: pytest --doctest-modules python/lancedb
@@ -225,6 +229,5 @@ jobs:
           pip install "pydantic<2"
           pip install pyarrow==16
           pip install --extra-index-url https://pypi.fury.io/lance-format/ --extra-index-url https://pypi.fury.io/lancedb/ -e .[tests]
-          pip install tantivy
       - name: Run tests
         run: pytest -m "not slow and not s3_test" -x -v --durations=30 python/tests

.github/workflows/rust.yml

@@ -7,9 +7,17 @@ on:
   pull_request:
     paths:
       - Cargo.toml
+      - Cargo.lock
+      - rust-toolchain.toml
+      - deny.toml
       - rust/**
+      - nodejs/Cargo.toml
+      - python/Cargo.toml
       - .github/workflows/rust.yml
 
+permissions:
+  contents: read
+
 concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
   cancel-in-progress: true
@@ -51,6 +59,17 @@ jobs:
       - name: Run clippy (without remote feature)
         run: cargo clippy --profile ci --workspace --tests -- -D warnings
 
+  deny:
+    # Supply-chain checks: advisories, licenses, banned crates, and source
+    # restrictions. Configuration lives in `deny.toml` at the workspace root.
+    timeout-minutes: 10
+    runs-on: ubuntu-24.04
+    steps:
+      - uses: actions/checkout@v4
+      - uses: EmbarkStudios/cargo-deny-action@v2
+        with:
+          command: check advisories bans licenses sources
+
   build-no-lock:
     runs-on: ubuntu-24.04
     timeout-minutes: 30
@@ -206,14 +225,14 @@ jobs:
       - name: Downgrade  dependencies
         # These packages have newer requirements for MSRV
         run: |
-          cargo update -p aws-sdk-bedrockruntime --precise 1.64.0
-          cargo update -p aws-sdk-dynamodb --precise 1.55.0
-          cargo update -p aws-config --precise 1.5.10
-          cargo update -p aws-sdk-kms --precise 1.51.0
-          cargo update -p aws-sdk-s3 --precise 1.65.0
-          cargo update -p aws-sdk-sso --precise 1.50.0
-          cargo update -p aws-sdk-ssooidc --precise 1.51.0
-          cargo update -p aws-sdk-sts --precise 1.51.0
+          cargo update -p aws-sdk-bedrockruntime --precise 1.77.0
+          cargo update -p aws-sdk-dynamodb --precise 1.68.0
+          cargo update -p aws-config --precise 1.6.0
+          cargo update -p aws-sdk-kms --precise 1.63.0
+          cargo update -p aws-sdk-s3 --precise 1.79.0
+          cargo update -p aws-sdk-sso --precise 1.62.0
+          cargo update -p aws-sdk-ssooidc --precise 1.63.0
+          cargo update -p aws-sdk-sts --precise 1.63.0
           cargo update -p home --precise 0.5.9
       - name: cargo +${{ matrix.msrv }} check
         env:

.github/workflows/update_package_lock_run.yml

@@ -3,6 +3,9 @@ name: Update package-lock.json
 on:
   workflow_dispatch:
 
+permissions:
+  contents: read
+
 jobs:
   publish:
     runs-on: ubuntu-latest

.github/workflows/update_package_lock_run_nodejs.yml

@@ -3,6 +3,9 @@ name: Update NodeJs package-lock.json
 on:
   workflow_dispatch:
 
+permissions:
+  contents: read
+
 jobs:
   publish:
     runs-on: ubuntu-latest

.github/workflows/upload_wheel/action.yml

@@ -2,9 +2,6 @@ name: upload-wheel
 
 description: "Upload wheels to Pypi"
 inputs:
-  pypi_token:
-    required: true
-    description: "release token for the repo"
   fury_token:
     required: true
     description: "release token for the fury repo"
@@ -12,12 +9,6 @@ inputs:
 runs:
   using: "composite"
   steps:
-  - name: Install dependencies
-    shell: bash
-    run: |
-      python -m pip install --upgrade pip
-      pip install twine
-      python3 -m pip install --upgrade pkginfo
   - name: Choose repo
     shell: bash
     id: choose_repo
@@ -27,19 +18,17 @@ runs:
       else
         echo "repo=pypi" >> $GITHUB_OUTPUT
       fi
-  - name: Publish to PyPI
+  - name: Publish to Fury
+    if: steps.choose_repo.outputs.repo == 'fury'
     shell: bash
     env:
       FURY_TOKEN: ${{ inputs.fury_token }}
-      PYPI_TOKEN: ${{ inputs.pypi_token }}
     run: |
-      if [[ ${{ steps.choose_repo.outputs.repo }} == fury ]]; then
-        WHEEL=$(ls target/wheels/lancedb-*.whl 2> /dev/null | head -n 1)
-        echo "Uploading $WHEEL to Fury"
-        curl -f -F package=@$WHEEL https://$FURY_TOKEN@push.fury.io/lancedb/
-      else
-        twine upload --repository ${{ steps.choose_repo.outputs.repo }} \
-          --username __token__ \
-          --password $PYPI_TOKEN \
-          target/wheels/lancedb-*.whl
-      fi
+      WHEEL=$(ls target/wheels/lancedb-*.whl 2> /dev/null | head -n 1)
+      echo "Uploading $WHEEL to Fury"
+      curl -f -F package=@$WHEEL https://$FURY_TOKEN@push.fury.io/lancedb/
+  - name: Publish to PyPI
+    if: steps.choose_repo.outputs.repo == 'pypi'
+    uses: pypa/gh-action-pypi-publish@release/v1
+    with:
+      packages-dir: target/wheels/

Cargo.toml

@@ -1,7 +1,5 @@
 [workspace]
 members = ["rust/lancedb", "nodejs", "python"]
-# Python package needs to be built by maturin.
-exclude = ["python"]
 resolver = "2"
 
 [workspace.package]
@@ -15,40 +13,40 @@ categories = ["database-implementations"]
 rust-version = "1.91.0"
 
 [workspace.dependencies]
-lance = { "version" = "=4.0.0-beta.8", default-features = false, "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-core = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-datagen = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-file = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-io = { "version" = "=4.0.0-beta.8", default-features = false, "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-index = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-linalg = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-namespace = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-namespace-impls = { "version" = "=4.0.0-beta.8", default-features = false, "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-table = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-testing = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-datafusion = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-encoding = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
-lance-arrow = { "version" = "=4.0.0-beta.8", "tag" = "v4.0.0-beta.8", "git" = "https://github.com/lance-format/lance.git" }
+lance = { "version" = "=7.0.0-beta.7", default-features = false, "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-core = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-datagen = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-file = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-io = { "version" = "=7.0.0-beta.7", default-features = false, "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-index = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-linalg = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-namespace = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-namespace-impls = { "version" = "=7.0.0-beta.7", default-features = false, "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-table = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-testing = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-datafusion = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-encoding = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
+lance-arrow = { "version" = "=7.0.0-beta.7", "tag" = "v7.0.0-beta.7", "git" = "https://github.com/lance-format/lance.git" }
 ahash = "0.8"
 # Note that this one does not include pyarrow
-arrow = { version = "57.2", optional = false }
-arrow-array = "57.2"
-arrow-data = "57.2"
-arrow-ipc = "57.2"
-arrow-ord = "57.2"
-arrow-schema = "57.2"
-arrow-select = "57.2"
-arrow-cast = "57.2"
+arrow = { version = "58.0.0", optional = false }
+arrow-array = "58.0.0"
+arrow-data = "58.0.0"
+arrow-ipc = "58.0.0"
+arrow-ord = "58.0.0"
+arrow-schema = "58.0.0"
+arrow-select = "58.0.0"
+arrow-cast = "58.0.0"
 async-trait = "0"
-datafusion = { version = "52.1", default-features = false }
-datafusion-catalog = "52.1"
-datafusion-common = { version = "52.1", default-features = false }
-datafusion-execution = "52.1"
-datafusion-expr = "52.1"
-datafusion-functions = "52.1"
-datafusion-physical-plan = "52.1"
-datafusion-physical-expr = "52.1"
-datafusion-sql = "52.1"
+datafusion = { version = "53.0.0", default-features = false }
+datafusion-catalog = "53.0.0"
+datafusion-common = { version = "53.0.0", default-features = false }
+datafusion-execution = "53.0.0"
+datafusion-expr = "53.0.0"
+datafusion-functions = "53.0.0"
+datafusion-physical-plan = "53.0.0"
+datafusion-physical-expr = "53.0.0"
+datafusion-sql = "53.0.0"
 env_logger = "0.11"
 half = { "version" = "2.7.1", default-features = false, features = [
     "num-traits",
@@ -56,7 +54,7 @@ half = { "version" = "2.7.1", default-features = false, features = [
 futures = "0"
 log = "0.4"
 moka = { version = "0.12", features = ["future"] }
-object_store = "0.12.0"
+object_store = "0.13.2"
 pin-project = "1.0.7"
 rand = "0.9"
 snafu = "0.8"

README.md

@@ -15,7 +15,7 @@
 
 # **The Multimodal AI Lakehouse**
 
-[**How to Install** ](#how-to-install) ✦ [**Detailed Documentation**](https://lancedb.com/docs) ✦ [**Tutorials and Recipes**](https://github.com/lancedb/vectordb-recipes/tree/main) ✦  [**Contributors**](#contributors) 
+[**How to Install** ](#how-to-install) ✦ [**Detailed Documentation**](https://docs.lancedb.com) ✦ [**Tutorials and Recipes**](https://github.com/lancedb/vectordb-recipes/tree/main) ✦  [**Contributors**](#contributors) 
 
 **The ultimate multimodal data platform for AI/ML applications.** 
 
@@ -57,7 +57,7 @@ LanceDB is a central location where developers can build, train and analyze thei
 
 ## **How to Install**:
 
-Follow the [Quickstart](https://lancedb.com/docs/quickstart/) doc to set up LanceDB locally. 
+Follow the [Quickstart](https://docs.lancedb.com/quickstart) doc to set up LanceDB locally. 
 
 **API & SDK:** We also support Python, Typescript and Rust SDKs
 

ci/check_lance_release.py

@@ -3,6 +3,7 @@
 from __future__ import annotations
 
 import argparse
+import functools
 import json
 import os
 import re
@@ -26,6 +27,7 @@ SEMVER_RE = re.compile(
 )
 
 
+@functools.total_ordering
 @dataclass(frozen=True)
 class SemVer:
     major: int
@@ -156,7 +158,9 @@ def read_current_version(repo_root: Path) -> str:
 
 
 def determine_latest_tag(tags: Iterable[TagInfo]) -> TagInfo:
-    return max(tags, key=lambda tag: tag.semver)
+    # Stable releases (no prerelease) are always preferred over pre-releases.
+    # Within each group, standard semver ordering applies.
+    return max(tags, key=lambda tag: (not tag.semver.prerelease, tag.semver))
 
 
 def write_outputs(args: argparse.Namespace, payload: dict) -> None:

deny.toml

@@ -0,0 +1,196 @@
+# cargo-deny configuration for LanceDB.
+#
+# Run locally with `cargo deny check`. See
+# https://embarkstudios.github.io/cargo-deny/ for the full reference.
+
+# The set of target triples we care about. cargo-deny will only consider
+# dependencies that are used on at least one of these targets. Keeping this
+# explicit avoids noise from platform-specific crates (e.g. wasm, android,
+# ios) that we never actually ship.
+[graph]
+targets = [
+    "x86_64-unknown-linux-gnu",
+    "aarch64-unknown-linux-gnu",
+    "x86_64-apple-darwin",
+    "aarch64-apple-darwin",
+    "x86_64-pc-windows-msvc",
+    "aarch64-pc-windows-msvc",
+]
+all-features = true
+
+[output]
+feature-depth = 1
+
+# ---------------------------------------------------------------------------
+# Advisories: security vulnerabilities and yanked crates.
+# ---------------------------------------------------------------------------
+[advisories]
+version = 2
+# Fail the check if any crate in the lockfile has been yanked from crates.io.
+# Yanked crates are a signal the author retracted the release (often due to
+# bugs or security issues) and should not be depended on.
+yanked = "deny"
+# Advisory IDs we have explicitly reviewed and chosen to accept. Every
+# entry must include a rationale and, where possible, an upstream issue
+# pointing to a fix. Revisit this list whenever dependencies are updated.
+ignore = [
+    # rsa: Marvin Attack timing side-channel in PKCS#1 v1.5 decryption.
+    # Reached only through opendal → reqsign → rsa. We do not use RSA
+    # decryption in LanceDB ourselves; this is dormant in the signing path.
+    # No fixed release exists upstream as of this writing.
+    # https://rustsec.org/advisories/RUSTSEC-2023-0071
+    { id = "RUSTSEC-2023-0071", reason = "rsa crate via opendal/reqsign; no fixed upstream release" },
+
+    # instant: unmaintained. Pulled in via backoff → instant. Upstream
+    # recommends switching to `web-time`; fix has to come from backoff.
+    # https://rustsec.org/advisories/RUSTSEC-2024-0384
+    { id = "RUSTSEC-2024-0384", reason = "transitive via backoff; waiting on backoff replacement" },
+
+    # paste: unmaintained (author archived the repo). Used transitively by
+    # datafusion and the arrow ecosystem; widespread, no drop-in replacement.
+    # https://rustsec.org/advisories/RUSTSEC-2024-0436
+    { id = "RUSTSEC-2024-0436", reason = "transitive via datafusion; awaiting ecosystem migration" },
+
+    # encoding: unmaintained. Reached through lindera-dictionary, which is
+    # required by the native Lindera tokenizer path. Lindera has not migrated
+    # off this crate yet.
+    # https://rustsec.org/advisories/RUSTSEC-2021-0153
+    { id = "RUSTSEC-2021-0153", reason = "transitive via lindera-dictionary for native Lindera tokenizer" },
+
+    # fast-float: unsound and unmaintained. Reached only through polars-arrow
+    # from the optional Polars integration; replacement requires a Polars
+    # dependency upgrade.
+    # https://rustsec.org/advisories/RUSTSEC-2024-0379
+    { id = "RUSTSEC-2024-0379", reason = "transitive via polars-arrow; waiting on Polars migration" },
+
+    # tantivy: segfault on malformed input due to missing bounds check.
+    # Pulled in via lance for full-text search. We only feed tantivy
+    # documents we construct ourselves, not attacker-controlled bytes.
+    # Tracked for a lance dependency bump.
+    # https://rustsec.org/advisories/RUSTSEC-2025-0003
+    { id = "RUSTSEC-2025-0003", reason = "tantivy via lance; inputs are internally produced, not user-supplied bytes" },
+
+    # backoff: unmaintained. Reached only via async-openai. Replacement
+    # requires async-openai to migrate (or us to drop async-openai).
+    # https://rustsec.org/advisories/RUSTSEC-2025-0012
+    { id = "RUSTSEC-2025-0012", reason = "transitive via async-openai; waiting on upstream migration" },
+
+    # number_prefix: unmaintained. Transitive via indicatif → hf-hub.
+    # No security impact, just maintenance status.
+    # https://rustsec.org/advisories/RUSTSEC-2025-0119
+    { id = "RUSTSEC-2025-0119", reason = "transitive via hf-hub/indicatif; cosmetic formatting crate" },
+
+    # bincode: unmaintained. Reached through lindera and lindera-dictionary,
+    # which are required by the native Lindera tokenizer path. Lindera has not
+    # migrated to another serialization format yet.
+    # https://rustsec.org/advisories/RUSTSEC-2025-0141
+    { id = "RUSTSEC-2025-0141", reason = "transitive via lindera/lindera-dictionary for native Lindera tokenizer" },
+
+    # lru: soundness issue in IterMut. Reached only through aws-sdk-s3 in
+    # LanceDB's dev-dependency graph; LanceDB does not use that iterator
+    # directly. Clearing this requires the AWS SDK chain to update lru.
+    # https://rustsec.org/advisories/RUSTSEC-2026-0002
+    { id = "RUSTSEC-2026-0002", reason = "transitive via aws-sdk-s3 dev-dependency; waiting on AWS SDK lru upgrade" },
+
+    # rustls-webpki 0.101.7 (old major line): name-constraint checks for
+    # URI / wildcard names. Pulled in only via the legacy rustls 0.21 chain
+    # from aws-smithy-http-client. The 0.103 line we actively use is patched.
+    # Clearing the 0.101 copy requires the aws-sdk chain to migrate off
+    # rustls 0.21.
+    # https://rustsec.org/advisories/RUSTSEC-2026-0098
+    # https://rustsec.org/advisories/RUSTSEC-2026-0099
+    { id = "RUSTSEC-2026-0098", reason = "only affects rustls-webpki 0.101 from legacy aws-smithy/rustls 0.21 chain" },
+    { id = "RUSTSEC-2026-0099", reason = "only affects rustls-webpki 0.101 from legacy aws-smithy/rustls 0.21 chain" },
+
+    # rustls-webpki 0.101.7: reachable panic in CRL parsing. Same legacy
+    # rustls 0.21 chain from aws-smithy-http-client as above. The 0.103 line
+    # we actively use is upgraded to 0.103.13 which contains the fix.
+    # https://rustsec.org/advisories/RUSTSEC-2026-0104
+    { id = "RUSTSEC-2026-0104", reason = "only affects rustls-webpki 0.101 from legacy aws-smithy/rustls 0.21 chain" },
+
+    # rand 0.8.5: soundness issue only when ThreadRng reseeds inside a custom
+    # logger. Reached through several transitive chains. LanceDB does not use
+    # rand from a custom logger; upgrade once all pinned chains accept 0.8.6+.
+    # https://rustsec.org/advisories/RUSTSEC-2026-0097
+    { id = "RUSTSEC-2026-0097", reason = "transitive rand 0.8.5; LanceDB does not call ThreadRng from custom logging" },
+]
+
+# ---------------------------------------------------------------------------
+# Licenses: only allow licenses we've reviewed as compatible with Apache-2.0.
+# ---------------------------------------------------------------------------
+[licenses]
+version = 2
+# SPDX identifiers for licenses that are compatible with our Apache-2.0
+# distribution. Additions require legal review.
+allow = [
+    "Apache-2.0",
+    "Apache-2.0 WITH LLVM-exception",
+    "MIT",
+    "BSD-2-Clause",
+    "BSD-3-Clause",
+    "ISC",
+    "Unicode-3.0",
+    "Unicode-DFS-2016",
+    "Zlib",
+    "CC0-1.0",
+    "MPL-2.0",
+    "BSL-1.0",
+    "OpenSSL",
+    # 0BSD ("BSD Zero Clause") is effectively public domain — no attribution
+    # required. Pulled in by `mock_instant`.
+    "0BSD",
+    # bzip2-1.0.6 is the permissive upstream bzip2 license (BSD-like). Pulled
+    # in by `libbz2-rs-sys`, the pure-Rust bzip2 implementation.
+    "bzip2-1.0.6",
+    # CDLA-Permissive-2.0 is a permissive data license used by `webpki-roots`
+    # for the Mozilla CA root bundle. Data-only, distribution-compatible.
+    "CDLA-Permissive-2.0",
+]
+confidence-threshold = 0.8
+# Crates whose license cannot be determined from Cargo metadata but whose
+# license we've manually confirmed from upstream. Keep this list minimal.
+[[licenses.clarify]]
+# polars-arrow-format omits the `license` field in its Cargo.toml, but the
+# upstream repo (pola-rs/polars-arrow-format) is dual-licensed Apache-2.0 OR
+# MIT. See https://github.com/pola-rs/polars-arrow-format/blob/main/LICENSE
+crate = "polars-arrow-format"
+expression = "Apache-2.0 OR MIT"
+license-files = []
+
+# ---------------------------------------------------------------------------
+# Bans: disallow specific crates and flag dependency hygiene issues.
+# ---------------------------------------------------------------------------
+[bans]
+# Warn (not deny) on duplicate versions of the same crate. In a large
+# workspace like this one, duplicates are common and often unavoidable
+# transitively. We surface them to discourage growth, but don't fail CI.
+multiple-versions = "warn"
+# Wildcard version requirements (`foo = "*"`) are a footgun — they let any
+# future release in without review. Ban them outright.
+wildcards = "deny"
+# Internal workspace crates reference each other via `path = "..."`, which
+# cargo-deny sees as a wildcard version. That's fine for private workspace
+# members (not published to crates.io), so allow it specifically for paths.
+allow-wildcard-paths = true
+# Features that, if enabled, should cause the check to fail.
+deny = []
+# Crates to skip when checking for duplicate versions.
+skip = []
+# Similar to `skip`, but also skips the entire transitive subtree.
+skip-tree = []
+
+# ---------------------------------------------------------------------------
+# Sources: restrict where crates can come from.
+# ---------------------------------------------------------------------------
+[sources]
+# Deny any registry other than the ones explicitly listed below.
+unknown-registry = "deny"
+# Deny any git dependency whose host isn't in the allow-list below. This
+# prevents accidental pulls from arbitrary forks.
+unknown-git = "deny"
+allow-registry = ["https://github.com/rust-lang/crates.io-index"]
+# Lance is developed in a sibling repo and pulled as a git dependency until
+# releases are cut to crates.io. Allow that specific host.
+allow-git = [
+    "https://github.com/lance-format/lance",
+]

docker-compose.yml

@@ -1,7 +1,7 @@
 version: "3.9"
 services:
   localstack:
-    image: localstack/localstack:3.3
+    image: localstack/localstack:4.0
     ports:
       - 4566:4566
     environment:

dockerfiles/Dockerfile

@@ -24,4 +24,4 @@ RUN python --version && \
   rustc --version && \
   protoc --version
 
-RUN pip install --no-cache-dir tantivy lancedb
+RUN pip install --no-cache-dir lancedb

docs/README.md

@@ -1,6 +1,6 @@
 # LanceDB Documentation
 
-LanceDB docs are available at [lancedb.com/docs](https://lancedb.com/docs).
+LanceDB docs are available at [docs.lancedb.com](https://docs.lancedb.com).
 
 The SDK docs are 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

docs/requirements.txt

@@ -1,8 +1,8 @@
-mkdocs==1.5.3
+mkdocs==1.6.1
 mkdocs-jupyter==0.24.1
-mkdocs-material==9.5.3
+mkdocs-material==9.6.23
 mkdocs-autorefs>=0.5,<=1.0
-mkdocstrings[python]==0.25.2
+mkdocstrings[python]>=0.24,<1.0
 griffe>=0.40,<1.0
 mkdocs-render-swagger-plugin>=0.1.0
 pydantic>=2.0,<3.0

docs/src/java/java.md

@@ -14,7 +14,7 @@ Add the following dependency to your `pom.xml`:
 <dependency>
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-core</artifactId>
-    <version>0.27.0-beta.5</version>
+    <version>0.28.0-beta.11</version>
 </dependency>
 ```
 
@@ -57,32 +57,32 @@ LanceNamespace namespaceClient = LanceDbNamespaceClientBuilder.newBuilder()
 
 ## Metadata Operations
 
-### Creating a Namespace
+### Creating a Namespace Path
 
-Namespaces organize tables hierarchically. Create a namespace before creating tables within it:
+Namespace paths organize tables hierarchically. Create the desired namespace path before creating tables within it:
 
 ```java
 import org.lance.namespace.model.CreateNamespaceRequest;
 import org.lance.namespace.model.CreateNamespaceResponse;
 
-// Create a child namespace
+// Create a child namespace path
 CreateNamespaceRequest request = new CreateNamespaceRequest();
 request.setId(Arrays.asList("my_namespace"));
 
 CreateNamespaceResponse response = namespaceClient.createNamespace(request);
 ```
 
-You can also create nested namespaces:
+You can also create nested namespace paths:
 
 ```java
-// Create a nested namespace: parent/child
+// Create a nested namespace path: parent/child
 CreateNamespaceRequest request = new CreateNamespaceRequest();
 request.setId(Arrays.asList("parent_namespace", "child_namespace"));
 
 CreateNamespaceResponse response = namespaceClient.createNamespace(request);
 ```
 
-### Describing a Namespace
+### Describing a Namespace Path
 
 ```java
 import org.lance.namespace.model.DescribeNamespaceRequest;
@@ -95,22 +95,22 @@ DescribeNamespaceResponse response = namespaceClient.describeNamespace(request);
 System.out.println("Namespace properties: " + response.getProperties());
 ```
 
-### Listing Namespaces
+### Listing Namespace Paths
 
 ```java
 import org.lance.namespace.model.ListNamespacesRequest;
 import org.lance.namespace.model.ListNamespacesResponse;
 
-// List all namespaces at root level
+// List all namespace paths at the root level
 ListNamespacesRequest request = new ListNamespacesRequest();
 request.setId(Arrays.asList());  // Empty for root
 
 ListNamespacesResponse response = namespaceClient.listNamespaces(request);
 for (String ns : response.getNamespaces()) {
-    System.out.println("Namespace: " + ns);
+    System.out.println("Namespace path: " + ns);
 }
 
-// List child namespaces under a parent
+// List child namespace paths under a parent path
 ListNamespacesRequest childRequest = new ListNamespacesRequest();
 childRequest.setId(Arrays.asList("parent_namespace"));
 
@@ -123,7 +123,7 @@ ListNamespacesResponse childResponse = namespaceClient.listNamespaces(childReque
 import org.lance.namespace.model.ListTablesRequest;
 import org.lance.namespace.model.ListTablesResponse;
 
-// List tables in a namespace
+// List tables in a namespace path
 ListTablesRequest request = new ListTablesRequest();
 request.setId(Arrays.asList("my_namespace"));
 
@@ -133,7 +133,7 @@ for (String table : response.getTables()) {
 }
 ```
 
-### Dropping a Namespace
+### Dropping a Namespace Path
 
 ```java
 import org.lance.namespace.model.DropNamespaceRequest;
@@ -175,7 +175,7 @@ DropTableResponse response = namespaceClient.dropTable(request);
 
 ### Creating a Table
 
-Tables are created within a namespace by providing data in Apache Arrow IPC format:
+Tables are created within a namespace path by providing data in Apache Arrow IPC format:
 
 ```java
 import org.lance.namespace.LanceNamespace;
@@ -242,7 +242,7 @@ try (BufferAllocator allocator = new RootAllocator();
     }
     byte[] tableData = out.toByteArray();
 
-    // Create table in a namespace
+    // Create a table in a namespace path
     CreateTableRequest request = new CreateTableRequest();
     request.setId(Arrays.asList("my_namespace", "my_table"));
     CreateTableResponse response = namespaceClient.createTable(request, tableData);

docs/src/js/README.md

@@ -34,7 +34,7 @@ const results = await table.vectorSearch([0.1, 0.3]).limit(20).toArray();
 console.log(results);
 ```
 
-The [quickstart](https://lancedb.com/docs/quickstart/basic-usage/) contains more complete examples.
+The [quickstart](https://docs.lancedb.com/quickstart/) contains more complete examples.
 
 ## Development
 

docs/src/js/_media/CONTRIBUTING.md

@@ -12,20 +12,22 @@ Typescript.
 * `src/`: Rust bindings source code
 * `lancedb/`: Typescript package source code
 * `__test__/`: Unit tests
-* `examples/`: An npm package with the examples shown in the documentation
+* `examples/`: A pnpm package with the examples shown in the documentation
 
 ## Development environment
 
 To set up your development environment, you will need to install the following:
 
-1. Node.js 14 or later
-2. Rust's package manager, Cargo. Use [rustup](https://rustup.rs/) to install.
-3. [protoc](https://grpc.io/docs/protoc-installation/) (Protocol Buffers compiler)
+1. Node.js 22 or later (required by pnpm 11)
+2. [pnpm](https://pnpm.io/installation) 11 or later (or run via `corepack enable`,
+   which uses the `packageManager` field in `package.json`)
+3. Rust's package manager, Cargo. Use [rustup](https://rustup.rs/) to install.
+4. [protoc](https://grpc.io/docs/protoc-installation/) (Protocol Buffers compiler)
 
 Initial setup:
 
 ```shell
-npm install
+pnpm install
 ```
 
 ### Commit Hooks
@@ -39,38 +41,38 @@ pre-commit install
 
 ## Development
 
-Most common development commands can be run using the npm scripts.
+Most common development commands can be run using the pnpm scripts.
 
 Build the package
 
 ```shell
-npm install
-npm run build
+pnpm install
+pnpm build
 ```
 
 Lint:
 
 ```shell
-npm run lint
+pnpm lint
 ```
 
 Format and fix lints:
 
 ```shell
-npm run lint-fix
+pnpm lint-fix
 ```
 
 Run tests:
 
 ```shell
-npm test
+pnpm test
 ```
 
 To run a single test:
 
 ```shell
 # Single file: table.test.ts
-npm test -- table.test.ts
+pnpm test -- table.test.ts
 # Single test: 'merge insert' in table.test.ts
-npm test -- table.test.ts --testNamePattern=merge\ insert
+pnpm test -- table.test.ts --testNamePattern=merge\ insert
 ```

docs/src/js/classes/Connection.md

@@ -61,8 +61,8 @@ sharing the same data, deletion, and index files.
 * **options.sourceVersion?**: `number`
     The version of the source table to clone.
 
-* **options.targetNamespace?**: `string`[]
-    The namespace for the target table (defaults to root namespace).
+* **options.targetNamespacePath?**: `string`[]
+    The namespace path for the target table (defaults to root namespace).
 
 #### Returns
 
@@ -116,13 +116,13 @@ Creates a new empty Table
 
 `Promise`<[`Table`](Table.md)>
 
-#### createEmptyTable(name, schema, namespace, options)
+#### createEmptyTable(name, schema, namespacePath, options)
 
 ```ts
 abstract createEmptyTable(
    name,
    schema,
-   namespace?,
+   namespacePath?,
    options?): Promise<Table>
 ```
 
@@ -136,8 +136,8 @@ Creates a new empty Table
 * **schema**: [`SchemaLike`](../type-aliases/SchemaLike.md)
     The schema of the table
 
-* **namespace?**: `string`[]
-    The namespace to create the table in (defaults to root namespace)
+* **namespacePath?**: `string`[]
+    The namespace path to create the table in (defaults to root namespace)
 
 * **options?**: `Partial`&lt;[`CreateTableOptions`](../interfaces/CreateTableOptions.md)&gt;
     Additional options
@@ -148,12 +148,39 @@ Creates a new empty Table
 
 ***
 
-### createTable()
-
-#### createTable(options, namespace)
+### createNamespace()
 
 ```ts
-abstract createTable(options, namespace?): Promise<Table>
+abstract createNamespace(namespacePath, options?): Promise<CreateNamespaceResponse>
+```
+
+Create a new namespace at the given path.
+
+#### Parameters
+
+* **namespacePath**: `string`[]
+    The namespace path to create.
+
+* **options?**: `Partial`&lt;[`CreateNamespaceOptions`](../interfaces/CreateNamespaceOptions.md)&gt;
+    Creation `mode`
+    ("create" | "exist_ok" | "overwrite") and optional `properties`
+    to attach to the namespace.
+
+#### Returns
+
+`Promise`&lt;[`CreateNamespaceResponse`](../interfaces/CreateNamespaceResponse.md)&gt;
+
+The properties of the
+  created namespace and an optional transaction id.
+
+***
+
+### createTable()
+
+#### createTable(options, namespacePath)
+
+```ts
+abstract createTable(options, namespacePath?): Promise<Table>
 ```
 
 Creates a new Table and initialize it with new data.
@@ -163,8 +190,8 @@ Creates a new Table and initialize it with new data.
 * **options**: `object` & `Partial`&lt;[`CreateTableOptions`](../interfaces/CreateTableOptions.md)&gt;
     The options object.
 
-* **namespace?**: `string`[]
-    The namespace to create the table in (defaults to root namespace)
+* **namespacePath?**: `string`[]
+    The namespace path to create the table in (defaults to root namespace)
 
 ##### Returns
 
@@ -197,13 +224,13 @@ Creates a new Table and initialize it with new data.
 
 `Promise`&lt;[`Table`](Table.md)&gt;
 
-#### createTable(name, data, namespace, options)
+#### createTable(name, data, namespacePath, options)
 
 ```ts
 abstract createTable(
    name,
    data,
-   namespace?,
+   namespacePath?,
    options?): Promise<Table>
 ```
 
@@ -218,8 +245,8 @@ Creates a new Table and initialize it with new data.
     Non-empty Array of Records
     to be inserted into the table
 
-* **namespace?**: `string`[]
-    The namespace to create the table in (defaults to root namespace)
+* **namespacePath?**: `string`[]
+    The namespace path to create the table in (defaults to root namespace)
 
 * **options?**: `Partial`&lt;[`CreateTableOptions`](../interfaces/CreateTableOptions.md)&gt;
     Additional options
@@ -230,6 +257,29 @@ Creates a new Table and initialize it with new data.
 
 ***
 
+### describeNamespace()
+
+```ts
+abstract describeNamespace(namespacePath): Promise<DescribeNamespaceResponse>
+```
+
+Describe a namespace, returning its properties.
+
+#### Parameters
+
+* **namespacePath**: `string`[]
+    The namespace path to describe, in
+    parent → child order, e.g. `["analytics", "sales"]`.
+
+#### Returns
+
+`Promise`&lt;[`DescribeNamespaceResponse`](../interfaces/DescribeNamespaceResponse.md)&gt;
+
+The namespace's properties
+  (may be undefined if the namespace has none).
+
+***
+
 ### display()
 
 ```ts
@@ -247,15 +297,15 @@ Return a brief description of the connection
 ### dropAllTables()
 
 ```ts
-abstract dropAllTables(namespace?): Promise<void>
+abstract dropAllTables(namespacePath?): Promise<void>
 ```
 
 Drop all tables in the database.
 
 #### Parameters
 
-* **namespace?**: `string`[]
-    The namespace to drop tables from (defaults to root namespace).
+* **namespacePath?**: `string`[]
+    The namespace path to drop tables from (defaults to root namespace).
 
 #### Returns
 
@@ -263,10 +313,40 @@ Drop all tables in the database.
 
 ***
 
+### dropNamespace()
+
+```ts
+abstract dropNamespace(namespacePath, options?): Promise<DropNamespaceResponse>
+```
+
+Drop a namespace.
+
+Use `behavior: "cascade"` to also drop everything contained in the
+namespace (sub-namespaces and tables). The default `"restrict"`
+behavior refuses to drop a non-empty namespace.
+
+#### Parameters
+
+* **namespacePath**: `string`[]
+    The namespace path to drop.
+
+* **options?**: `Partial`&lt;[`DropNamespaceOptions`](../interfaces/DropNamespaceOptions.md)&gt;
+    `mode` ("skip" | "fail"
+    for missing-namespace handling) and `behavior` ("restrict" | "cascade").
+
+#### Returns
+
+`Promise`&lt;[`DropNamespaceResponse`](../interfaces/DropNamespaceResponse.md)&gt;
+
+Any properties returned by
+  the server and an optional transaction id.
+
+***
+
 ### dropTable()
 
 ```ts
-abstract dropTable(name, namespace?): Promise<void>
+abstract dropTable(name, namespacePath?): Promise<void>
 ```
 
 Drop an existing table.
@@ -276,8 +356,8 @@ Drop an existing table.
 * **name**: `string`
     The name of the table to drop.
 
-* **namespace?**: `string`[]
-    The namespace of the table (defaults to root namespace).
+* **namespacePath?**: `string`[]
+    The namespace path of the table (defaults to root namespace).
 
 #### Returns
 
@@ -299,12 +379,42 @@ Return true if the connection has not been closed
 
 ***
 
+### listNamespaces()
+
+```ts
+abstract listNamespaces(namespacePath?, options?): Promise<ListNamespacesResponse>
+```
+
+List the immediate child namespaces under the given parent.
+
+Results may be paginated. To retrieve subsequent pages, pass the
+`pageToken` returned by a previous call.
+
+#### Parameters
+
+* **namespacePath?**: `string`[]
+    The parent namespace path. Defaults
+    to the root namespace if omitted.
+
+* **options?**: `Partial`&lt;[`ListNamespacesOptions`](../interfaces/ListNamespacesOptions.md)&gt;
+    Pagination options
+    (`pageToken`, `limit`).
+
+#### Returns
+
+`Promise`&lt;[`ListNamespacesResponse`](../interfaces/ListNamespacesResponse.md)&gt;
+
+Child namespace names and
+  an optional token for fetching the next page.
+
+***
+
 ### openTable()
 
 ```ts
 abstract openTable(
    name,
-   namespace?,
+   namespacePath?,
    options?): Promise<Table>
 ```
 
@@ -315,8 +425,8 @@ Open a table in the database.
 * **name**: `string`
     The name of the table
 
-* **namespace?**: `string`[]
-    The namespace of the table (defaults to root namespace)
+* **namespacePath?**: `string`[]
+    The namespace path of the table (defaults to root namespace)
 
 * **options?**: `Partial`&lt;[`OpenTableOptions`](../interfaces/OpenTableOptions.md)&gt;
     Additional options
@@ -349,10 +459,10 @@ Tables will be returned in lexicographical order.
 
 `Promise`&lt;`string`[]&gt;
 
-#### tableNames(namespace, options)
+#### tableNames(namespacePath, options)
 
 ```ts
-abstract tableNames(namespace?, options?): Promise<string[]>
+abstract tableNames(namespacePath?, options?): Promise<string[]>
 ```
 
 List all the table names in this database.
@@ -361,8 +471,8 @@ Tables will be returned in lexicographical order.
 
 ##### Parameters
 
-* **namespace?**: `string`[]
-    The namespace to list tables from (defaults to root namespace)
+* **namespacePath?**: `string`[]
+    The namespace path to list tables from (defaults to root namespace)
 
 * **options?**: `Partial`&lt;[`TableNamesOptions`](../interfaces/TableNamesOptions.md)&gt;
     options to control the

docs/src/js/classes/Table.md

@@ -71,11 +71,12 @@ Add new columns with defined values.
 
 #### Parameters
 
-* **newColumnTransforms**: [`AddColumnsSql`](../interfaces/AddColumnsSql.md)[]
-    pairs of column names and
-    the SQL expression to use to calculate the value of the new column. These
-    expressions will be evaluated for each row in the table, and can
-    reference existing columns in the table.
+* **newColumnTransforms**: `Field`<`any`> \| `Field`<`any`>[] \| `Schema`<`any`> \| [`AddColumnsSql`](../interfaces/AddColumnsSql.md)[]
+    Either:
+    - An array of objects with column names and SQL expressions to calculate values
+    - A single Arrow Field defining one column with its data type (column will be initialized with null values)
+    - An array of Arrow Fields defining columns with their data types (columns will be initialized with null values)
+    - An Arrow Schema defining columns with their data types (columns will be initialized with null values)
 
 #### Returns
 
@@ -484,19 +485,7 @@ Modeled after ``VACUUM`` in PostgreSQL.
  - Prune: Removes old versions of the dataset
  - Index: Optimizes the indices, adding new data to existing indices
 
- Experimental API
- ----------------
-
- The optimization process is undergoing active development and may change.
- Our goal with these changes is to improve the performance of optimization and
- reduce the complexity.
-
- That being said, it is essential today to run optimize if you want the best
- performance.  It should be stable and safe to use in production, but it our
- hope that the API may be simplified (or not even need to be called) in the
- future.
-
- The frequency an application shoudl call optimize is based on the frequency of
+ The frequency an application should call optimize is based on the frequency of
  data modifications.  If data is frequently added, deleted, or updated then
  optimize should be run frequently.  A good rule of thumb is to run optimize if
  you have added or modified 100,000 or more records or run more than 20 data
@@ -512,6 +501,34 @@ Modeled after ``VACUUM`` in PostgreSQL.
 
 ***
 
+### prewarmData()
+
+```ts
+abstract prewarmData(columns?): Promise<void>
+```
+
+Prewarm one or more columns of data in the table.
+
+#### Parameters
+
+* **columns?**: `string`[]
+    The columns to prewarm. If undefined, all columns are prewarmed.
+    This will load the column data into the page cache so that future queries that
+    read those columns avoid the initial cold-start latency.  This call initiates
+    prewarming and returns once the request is accepted; the warming itself may
+    continue in the background.  Calling it on already-prewarmed columns is a
+    no-op on the server.
+    Prewarming is generally useful for columns used in filters or projections.
+    Large columns (e.g. high-dimensional vectors or binary data) may not be
+    practical to prewarm.
+    This feature is currently only supported on remote tables.
+
+#### Returns
+
+`Promise`&lt;`void`&gt;
+
+***
+
 ### prewarmIndex()
 
 ```ts

docs/src/js/functions/connectNamespace.md

@@ -0,0 +1,131 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / connectNamespace
+
+# Function: connectNamespace()
+
+## connectNamespace(implName, config, options)
+
+```ts
+function connectNamespace(
+   implName,
+   config,
+   options?): Promise<Connection>
+```
+
+Connect to a LanceDB database through a namespace.
+
+Unlike [connect](connect.md), which routes by URI scheme (local path vs.
+`db://` cloud), `connectNamespace` always returns a namespace-backed
+connection. The `implName` selects the namespace implementation:
+
+- `"dir"` — directory namespace, configured with [DirNamespaceConfig](../interfaces/DirNamespaceConfig.md).
+- `"rest"` — remote REST catalog, configured with [RestNamespaceConfig](../interfaces/RestNamespaceConfig.md).
+- Any other string — full module path for a custom implementation,
+  configured with a free-form string-keyed `properties` map.
+
+### Parameters
+
+* **implName**: `"dir"`
+
+* **config**: [`DirNamespaceConfig`](../interfaces/DirNamespaceConfig.md)
+
+* **options?**: `Partial`&lt;[`ConnectNamespaceOptions`](../interfaces/ConnectNamespaceOptions.md)&gt;
+
+### Returns
+
+`Promise`&lt;[`Connection`](../classes/Connection.md)&gt;
+
+### Examples
+
+```ts
+const db = await connectNamespace("dir", { root: "/path/to/db" });
+await db.createTable("users", [{ id: 1 }]);
+```
+
+```ts
+const db = await connectNamespace("rest", {
+  uri: "https://catalog.example.com",
+  headers: { "x-api-key": process.env.CATALOG_KEY ?? "" },
+});
+```
+
+```ts
+const db = await connectNamespace("my.custom.Namespace", {
+  endpoint: "...",
+});
+```
+
+## connectNamespace(implName, config, options)
+
+```ts
+function connectNamespace(
+   implName,
+   config,
+   options?): Promise<Connection>
+```
+
+Connect through the built-in REST namespace.
+
+Configured with [RestNamespaceConfig](../interfaces/RestNamespaceConfig.md). See the function-level
+documentation above for the full surface, examples, and how this
+relates to [connect](connect.md).
+
+### Parameters
+
+* **implName**: `"rest"`
+
+* **config**: [`RestNamespaceConfig`](../interfaces/RestNamespaceConfig.md)
+
+* **options?**: `Partial`&lt;[`ConnectNamespaceOptions`](../interfaces/ConnectNamespaceOptions.md)&gt;
+
+### Returns
+
+`Promise`&lt;[`Connection`](../classes/Connection.md)&gt;
+
+### Example
+
+```ts
+const db = await connectNamespace("rest", {
+  uri: "https://catalog.example.com",
+  headers: { "x-api-key": process.env.CATALOG_KEY ?? "" },
+});
+```
+
+## connectNamespace(implName, properties, options)
+
+```ts
+function connectNamespace(
+   implName,
+   properties,
+   options?): Promise<Connection>
+```
+
+Connect through a custom namespace implementation by full module path,
+configured with a free-form string-keyed `properties` map. Use the
+typed overloads above for the built-in `"dir"` and `"rest"` impls.
+
+See the function-level documentation above for examples and how this
+relates to [connect](connect.md).
+
+### Parameters
+
+* **implName**: `string`
+
+* **properties**: `Record`&lt;`string`, `string`&gt;
+
+* **options?**: `Partial`&lt;[`ConnectNamespaceOptions`](../interfaces/ConnectNamespaceOptions.md)&gt;
+
+### Returns
+
+`Promise`&lt;[`Connection`](../classes/Connection.md)&gt;
+
+### Example
+
+```ts
+const db = await connectNamespace("my.custom.Namespace", {
+  endpoint: "...",
+});
+```

docs/src/js/globals.md

@@ -51,10 +51,17 @@
 - [ClientConfig](interfaces/ClientConfig.md)
 - [ColumnAlteration](interfaces/ColumnAlteration.md)
 - [CompactionStats](interfaces/CompactionStats.md)
+- [ConnectNamespaceOptions](interfaces/ConnectNamespaceOptions.md)
 - [ConnectionOptions](interfaces/ConnectionOptions.md)
+- [CreateNamespaceOptions](interfaces/CreateNamespaceOptions.md)
+- [CreateNamespaceResponse](interfaces/CreateNamespaceResponse.md)
 - [CreateTableOptions](interfaces/CreateTableOptions.md)
 - [DeleteResult](interfaces/DeleteResult.md)
+- [DescribeNamespaceResponse](interfaces/DescribeNamespaceResponse.md)
+- [DirNamespaceConfig](interfaces/DirNamespaceConfig.md)
 - [DropColumnsResult](interfaces/DropColumnsResult.md)
+- [DropNamespaceOptions](interfaces/DropNamespaceOptions.md)
+- [DropNamespaceResponse](interfaces/DropNamespaceResponse.md)
 - [ExecutableQuery](interfaces/ExecutableQuery.md)
 - [FragmentStatistics](interfaces/FragmentStatistics.md)
 - [FragmentSummaryStats](interfaces/FragmentSummaryStats.md)
@@ -69,12 +76,15 @@
 - [IvfFlatOptions](interfaces/IvfFlatOptions.md)
 - [IvfPqOptions](interfaces/IvfPqOptions.md)
 - [IvfRqOptions](interfaces/IvfRqOptions.md)
+- [ListNamespacesOptions](interfaces/ListNamespacesOptions.md)
+- [ListNamespacesResponse](interfaces/ListNamespacesResponse.md)
 - [MergeResult](interfaces/MergeResult.md)
 - [OpenTableOptions](interfaces/OpenTableOptions.md)
 - [OptimizeOptions](interfaces/OptimizeOptions.md)
 - [OptimizeStats](interfaces/OptimizeStats.md)
 - [QueryExecutionOptions](interfaces/QueryExecutionOptions.md)
 - [RemovalStats](interfaces/RemovalStats.md)
+- [RestNamespaceConfig](interfaces/RestNamespaceConfig.md)
 - [RetryConfig](interfaces/RetryConfig.md)
 - [ShuffleOptions](interfaces/ShuffleOptions.md)
 - [SplitCalculatedOptions](interfaces/SplitCalculatedOptions.md)
@@ -107,6 +117,7 @@
 
 - [RecordBatchIterator](functions/RecordBatchIterator.md)
 - [connect](functions/connect.md)
+- [connectNamespace](functions/connectNamespace.md)
 - [makeArrowTable](functions/makeArrowTable.md)
 - [packBits](functions/packBits.md)
 - [permutationBuilder](functions/permutationBuilder.md)

docs/src/js/interfaces/ClientConfig.md

@@ -53,3 +53,18 @@ optional tlsConfig: TlsConfig;
 ```ts
 optional userAgent: string;
 ```
+
+***
+
+### userId?
+
+```ts
+optional userId: string;
+```
+
+User identifier for tracking purposes.
+
+This is sent as the `x-lancedb-user-id` header in requests to LanceDB Cloud/Enterprise.
+It can be set directly, or via the `LANCEDB_USER_ID` environment variable.
+Alternatively, set `LANCEDB_USER_ID_ENV_KEY` to specify another environment
+variable that contains the user ID value.

docs/src/js/interfaces/ConnectNamespaceOptions.md

@@ -0,0 +1,54 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / ConnectNamespaceOptions
+
+# Interface: ConnectNamespaceOptions
+
+## Properties
+
+### namespaceClientProperties?
+
+```ts
+optional namespaceClientProperties: Record<string, string>;
+```
+
+Extra properties for the backing namespace client.
+
+***
+
+### readConsistencyInterval?
+
+```ts
+optional readConsistencyInterval: number;
+```
+
+The interval, in seconds, at which to check for updates to the table
+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 value for
+eventual consistency.
+
+***
+
+### session?
+
+```ts
+optional session: Session;
+```
+
+The session to use for this connection. Holds shared caches and other
+session-specific state.
+
+***
+
+### storageOptions?
+
+```ts
+optional storageOptions: Record<string, string>;
+```
+
+Configuration for object storage. The available options are described
+at https://docs.lancedb.com/storage/

docs/src/js/interfaces/ConnectionOptions.md

@@ -41,6 +41,29 @@ for testing purposes.
 
 ***
 
+### manifestEnabled?
+
+```ts
+optional manifestEnabled: boolean;
+```
+
+(For LanceDB OSS only): use directory namespace manifests as the source
+of truth for table metadata. Existing directory-listed root tables are
+migrated into the manifest on access.
+
+***
+
+### namespaceClientProperties?
+
+```ts
+optional namespaceClientProperties: Record<string, string>;
+```
+
+(For LanceDB OSS only): extra properties for the backing namespace
+client used by manifest-enabled native connections.
+
+***
+
 ### readConsistencyInterval?
 
 ```ts
@@ -89,4 +112,4 @@ optional storageOptions: Record<string, string>;
 
 (For LanceDB OSS only): configuration for object storage.
 
-The available options are described at https://lancedb.com/docs/storage/
+The available options are described at https://docs.lancedb.com/storage/

docs/src/js/interfaces/CreateNamespaceOptions.md

@@ -0,0 +1,27 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / CreateNamespaceOptions
+
+# Interface: CreateNamespaceOptions
+
+## Properties
+
+### mode?
+
+```ts
+optional mode: "overwrite" | "create" | "exist_ok";
+```
+
+Creation mode.
+
+***
+
+### properties?
+
+```ts
+optional properties: Record<string, string>;
+```
+
+Properties to set on the new namespace.

docs/src/js/interfaces/CreateNamespaceResponse.md

@@ -0,0 +1,23 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / CreateNamespaceResponse
+
+# Interface: CreateNamespaceResponse
+
+## Properties
+
+### properties?
+
+```ts
+optional properties: Record<string, string>;
+```
+
+***
+
+### transactionId?
+
+```ts
+optional transactionId: string;
+```

docs/src/js/interfaces/CreateTableOptions.md

@@ -97,4 +97,4 @@ Configuration for object storage.
 Options already set on the connection will be inherited by the table,
 but can be overridden here.
 
-The available options are described at https://lancedb.com/docs/storage/
+The available options are described at https://docs.lancedb.com/storage/

docs/src/js/interfaces/DescribeNamespaceResponse.md

@@ -0,0 +1,15 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / DescribeNamespaceResponse
+
+# Interface: DescribeNamespaceResponse
+
+## Properties
+
+### properties?
+
+```ts
+optional properties: Record<string, string>;
+```

docs/src/js/interfaces/DirNamespaceConfig.md

@@ -0,0 +1,47 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / DirNamespaceConfig
+
+# Interface: DirNamespaceConfig
+
+Configuration for the built-in directory namespace (`"dir"`).
+
+The directory namespace stores tables under a single root path (local
+filesystem or object storage URI). See
+[https://docs.lancedb.com/namespaces](https://docs.lancedb.com/namespaces) for the documented surface;
+less-common knobs live under [DirNamespaceConfig.extraProperties](DirNamespaceConfig.md#extraproperties).
+
+## Properties
+
+### extraProperties?
+
+```ts
+optional extraProperties: Record<string, string>;
+```
+
+Additional raw properties passed verbatim to the namespace
+implementation (e.g. `storage.*`, `credential_vendor.*`). Typed
+fields above take precedence on key collision.
+
+***
+
+### manifestEnabled?
+
+```ts
+optional manifestEnabled: boolean;
+```
+
+Whether to maintain a namespace manifest at the root. Required for
+child namespaces. Defaults to true on the impl side.
+
+***
+
+### root
+
+```ts
+root: string;
+```
+
+Root path or URI containing the LanceDB tables.

docs/src/js/interfaces/DropNamespaceOptions.md

@@ -0,0 +1,27 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / DropNamespaceOptions
+
+# Interface: DropNamespaceOptions
+
+## Properties
+
+### behavior?
+
+```ts
+optional behavior: "restrict" | "cascade";
+```
+
+Refuse to drop if non-empty (restrict) or drop recursively (cascade).
+
+***
+
+### mode?
+
+```ts
+optional mode: "fail" | "skip";
+```
+
+Whether to skip if the namespace doesn't exist, or fail.

docs/src/js/interfaces/DropNamespaceResponse.md

@@ -0,0 +1,23 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / DropNamespaceResponse
+
+# Interface: DropNamespaceResponse
+
+## Properties
+
+### properties?
+
+```ts
+optional properties: Record<string, string>;
+```
+
+***
+
+### transactionId?
+
+```ts
+optional transactionId: string[];
+```

docs/src/js/interfaces/ListNamespacesOptions.md

@@ -0,0 +1,27 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / ListNamespacesOptions
+
+# Interface: ListNamespacesOptions
+
+## Properties
+
+### limit?
+
+```ts
+optional limit: number;
+```
+
+An optional limit to the number of results to return.
+
+***
+
+### pageToken?
+
+```ts
+optional pageToken: string;
+```
+
+Token from a previous response for pagination.

docs/src/js/interfaces/ListNamespacesResponse.md

@@ -0,0 +1,23 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / ListNamespacesResponse
+
+# Interface: ListNamespacesResponse
+
+## Properties
+
+### namespaces
+
+```ts
+namespaces: string[];
+```
+
+***
+
+### pageToken?
+
+```ts
+optional pageToken: string;
+```

docs/src/js/interfaces/OpenTableOptions.md

@@ -42,4 +42,4 @@ Configuration for object storage.
 Options already set on the connection will be inherited by the table,
 but can be overridden here.
 
-The available options are described at https://lancedb.com/docs/storage/
+The available options are described at https://docs.lancedb.com/storage/

docs/src/js/interfaces/OptimizeOptions.md

@@ -37,3 +37,12 @@ tbl.optimize({cleanupOlderThan: new Date()});
 ```ts
 deleteUnverified: boolean;
 ```
+
+Because they may be part of an in-progress transaction, files newer than
+7 days old are not deleted by default. If you are sure that there are no
+in-progress transactions, then you can set this to true to delete all
+files older than `cleanupOlderThan`.
+
+**WARNING**: This should only be set to true if you can guarantee that
+no other process is currently working on this dataset. Otherwise the
+dataset could be put into a corrupted state.

docs/src/js/interfaces/RestNamespaceConfig.md

@@ -0,0 +1,47 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / RestNamespaceConfig
+
+# Interface: RestNamespaceConfig
+
+Configuration for the built-in REST namespace (`"rest"`).
+
+The REST namespace talks to a remote catalog server over HTTP. See
+[https://docs.lancedb.com/namespaces](https://docs.lancedb.com/namespaces) for the documented surface;
+less-common knobs (TLS, metrics) live under
+[RestNamespaceConfig.extraProperties](RestNamespaceConfig.md#extraproperties).
+
+## Properties
+
+### extraProperties?
+
+```ts
+optional extraProperties: Record<string, string>;
+```
+
+Additional raw properties passed verbatim to the namespace
+implementation (e.g. `tls.*`, `ops_metrics_enabled`, `delimiter`).
+Typed fields above take precedence on key collision.
+
+***
+
+### headers?
+
+```ts
+optional headers: Record<string, string>;
+```
+
+HTTP headers forwarded with each request. Keys are passed through
+as-is (e.g. `"x-api-key"`, `"Authorization"`).
+
+***
+
+### uri
+
+```ts
+uri: string;
+```
+
+Catalog endpoint URL.

docs/src/js/namespaces/embedding/classes/EmbeddingFunction.md

@@ -52,7 +52,7 @@ new EmbeddingFunction<T, M>(): EmbeddingFunction<T, M>
 ### computeQueryEmbeddings()
 
 ```ts
-computeQueryEmbeddings(data): Promise<number[] | Float32Array | Float64Array>
+computeQueryEmbeddings(data): Promise<number[] | Uint8Array | Float32Array | Float64Array>
 ```
 
 Compute the embeddings for a single query
@@ -63,7 +63,7 @@ Compute the embeddings for a single query
 
 #### Returns
 
-`Promise`&lt;`number`[] \| `Float32Array` \| `Float64Array`&gt;
+`Promise`&lt;`number`[] \| `Uint8Array` \| `Float32Array` \| `Float64Array`&gt;
 
 ***
 

docs/src/js/namespaces/embedding/classes/TextEmbeddingFunction.md

@@ -37,7 +37,7 @@ new TextEmbeddingFunction<M>(): TextEmbeddingFunction<M>
 ### computeQueryEmbeddings()
 
 ```ts
-computeQueryEmbeddings(data): Promise<number[] | Float32Array | Float64Array>
+computeQueryEmbeddings(data): Promise<number[] | Uint8Array | Float32Array | Float64Array>
 ```
 
 Compute the embeddings for a single query
@@ -48,7 +48,7 @@ Compute the embeddings for a single query
 
 #### Returns
 
-`Promise`&lt;`number`[] \| `Float32Array` \| `Float64Array`&gt;
+`Promise`&lt;`number`[] \| `Uint8Array` \| `Float32Array` \| `Float64Array`&gt;
 
 #### Overrides
 

docs/src/js/type-aliases/IntoVector.md

@@ -7,5 +7,10 @@
 # Type Alias: IntoVector
 
 ```ts
-type IntoVector: Float32Array | Float64Array | number[] | Promise<Float32Array | Float64Array | number[]>;
+type IntoVector:
+  | Float32Array
+  | Float64Array
+  | Uint8Array
+  | number[]
+  | Promise<Float32Array | Float64Array | Uint8Array | number[]>;
 ```

docs/src/python/python.md

@@ -36,6 +36,20 @@ is also an [asynchronous API client](#connections-asynchronous).
 
 ::: lancedb.table.Tags
 
+## Expressions
+
+Type-safe expression builder for filters and projections. Use these instead
+of raw SQL strings with [where][lancedb.query.LanceQueryBuilder.where] and
+[select][lancedb.query.LanceQueryBuilder.select].
+
+::: lancedb.expr.Expr
+
+::: lancedb.expr.col
+
+::: lancedb.expr.lit
+
+::: lancedb.expr.func
+
 ## Querying (Synchronous)
 
 ::: lancedb.query.Query
@@ -80,11 +94,11 @@ is also an [asynchronous API client](#connections-asynchronous).
 
 ## Full text search
 
-::: lancedb.fts.create_index
+Use [lancedb.table.Table.create_fts_index][] for the synchronous API or
+[lancedb.table.AsyncTable.create_index][] with [lancedb.index.FTS][] for the
+asynchronous API.
 
-::: lancedb.fts.populate_index
-
-::: lancedb.fts.search_index
+::: lancedb.index.FTS
 
 ## Utilities
 

java/lancedb-core/pom.xml

@@ -8,7 +8,7 @@
     <parent>
       <groupId>com.lancedb</groupId>
       <artifactId>lancedb-parent</artifactId>
-      <version>0.27.0-beta.5</version>
+      <version>0.28.0-beta.11</version>
       <relativePath>../pom.xml</relativePath>
     </parent>
 

java/pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-parent</artifactId>
-    <version>0.27.0-beta.5</version>
+    <version>0.28.0-beta.11</version>
     <packaging>pom</packaging>
     <name>${project.artifactId}</name>
     <description>LanceDB Java SDK Parent POM</description>
@@ -28,7 +28,7 @@
     <properties>
         <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
         <arrow.version>15.0.0</arrow.version>
-        <lance-core.version>4.0.0-beta.8</lance-core.version>
+        <lance-core.version>7.0.0-beta.7</lance-core.version>
         <spotless.skip>false</spotless.skip>
         <spotless.version>2.30.0</spotless.version>
         <spotless.java.googlejavaformat.version>1.7</spotless.java.googlejavaformat.version>

nodejs/AGENTS.md

@@ -3,11 +3,11 @@ The core Rust library is in the `../rust/lancedb` directory, the rust binding
 code is in the `src/` directory and the typescript bindings are in
 the `lancedb/` directory.
 
-Whenever you change the Rust code, you will need to recompile: `npm run build`.
+Whenever you change the Rust code, you will need to recompile: `pnpm build`.
 
 Common commands:
-* Build: `npm run build`
-* Lint: `npm run lint`
-* Fix lints: `npm run lint-fix`
-* Test: `npm test`
-* Run single test file: `npm test __test__/arrow.test.ts`
+* Build: `pnpm build`
+* Lint: `pnpm lint`
+* Fix lints: `pnpm lint-fix`
+* Test: `pnpm test`
+* Run single test file: `pnpm test __test__/arrow.test.ts`

nodejs/CONTRIBUTING.md

@@ -12,20 +12,22 @@ Typescript.
 * `src/`: Rust bindings source code
 * `lancedb/`: Typescript package source code
 * `__test__/`: Unit tests
-* `examples/`: An npm package with the examples shown in the documentation
+* `examples/`: A pnpm package with the examples shown in the documentation
 
 ## Development environment
 
 To set up your development environment, you will need to install the following:
 
-1. Node.js 14 or later
-2. Rust's package manager, Cargo. Use [rustup](https://rustup.rs/) to install.
-3. [protoc](https://grpc.io/docs/protoc-installation/) (Protocol Buffers compiler)
+1. Node.js 22 or later (required by pnpm 11)
+2. [pnpm](https://pnpm.io/installation) 11 or later (or run via `corepack enable`,
+   which uses the `packageManager` field in `package.json`)
+3. Rust's package manager, Cargo. Use [rustup](https://rustup.rs/) to install.
+4. [protoc](https://grpc.io/docs/protoc-installation/) (Protocol Buffers compiler)
 
 Initial setup:
 
 ```shell
-npm install
+pnpm install
 ```
 
 ### Commit Hooks
@@ -39,38 +41,38 @@ pre-commit install
 
 ## Development
 
-Most common development commands can be run using the npm scripts.
+Most common development commands can be run using the pnpm scripts.
 
 Build the package
 
 ```shell
-npm install
-npm run build
+pnpm install
+pnpm build
 ```
 
 Lint:
 
 ```shell
-npm run lint
+pnpm lint
 ```
 
 Format and fix lints:
 
 ```shell
-npm run lint-fix
+pnpm lint-fix
 ```
 
 Run tests:
 
 ```shell
-npm test
+pnpm test
 ```
 
 To run a single test:
 
 ```shell
 # Single file: table.test.ts
-npm test -- table.test.ts
+pnpm test -- table.test.ts
 # Single test: 'merge insert' in table.test.ts
-npm test -- table.test.ts --testNamePattern=merge\ insert
+pnpm test -- table.test.ts --testNamePattern=merge\ insert
 ```

nodejs/Cargo.toml

@@ -1,7 +1,8 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.27.0-beta.5"
+version = "0.28.0-beta.11"
+publish = false
 license.workspace = true
 description.workspace = true
 repository.workspace = true
@@ -15,10 +16,13 @@ crate-type = ["cdylib"]
 async-trait.workspace = true
 arrow-ipc.workspace = true
 arrow-array.workspace = true
+arrow-buffer = "58.0.0"
+half.workspace = true
 arrow-schema.workspace = true
 env_logger.workspace = true
 futures.workspace = true
 lancedb = { path = "../rust/lancedb", default-features = false }
+lance-namespace.workspace = true
 napi = { version = "3.8.3", default-features = false, features = [
     "napi9",
     "async"
@@ -29,8 +33,8 @@ lzma-sys = { version = "0.1", features = ["static"] }
 log.workspace = true
 
 # Pin to resolve build failures; update periodically for security patches.
-aws-lc-sys = "=0.38.0"
-aws-lc-rs = "=1.16.1"
+aws-lc-sys = "=0.40.0"
+aws-lc-rs = "=1.16.3"
 
 [build-dependencies]
 napi-build = "2.3.1"

nodejs/README.md

@@ -30,7 +30,7 @@ const results = await table.vectorSearch([0.1, 0.3]).limit(20).toArray();
 console.log(results);
 ```
 
-The [quickstart](https://lancedb.com/docs/quickstart/basic-usage/) contains more complete examples.
+The [quickstart](https://docs.lancedb.com/quickstart/) contains more complete examples.
 
 ## Development
 

nodejs/__test__/connection.test.ts

@@ -4,7 +4,7 @@
 import { readdirSync } from "fs";
 import { Field, Float64, Schema } from "apache-arrow";
 import * as tmp from "tmp";
-import { Connection, Table, connect } from "../lancedb";
+import { Connection, Table, connect, connectNamespace } from "../lancedb";
 import { LocalTable } from "../lancedb/table";
 
 describe("when connecting", () => {
@@ -306,3 +306,186 @@ describe("clone table functionality", () => {
     ).rejects.toThrow("Deep clone is not yet implemented");
   });
 });
+
+describe("namespaces", () => {
+  let tmpDir: tmp.DirResult;
+  let db: Connection;
+
+  beforeEach(async () => {
+    tmpDir = tmp.dirSync({ unsafeCleanup: true });
+    // The local DirectoryNamespace backend only supports child namespaces
+    // when manifest mode is enabled (see lance-namespace-impls/src/dir.rs).
+    db = await connect(tmpDir.name, {
+      // biome-ignore lint/style/useNamingConvention: opaque backend property key, must match Rust
+      namespaceClientProperties: { manifest_enabled: "true" },
+    });
+  });
+  afterEach(() => tmpDir.removeCallback());
+
+  it("should create and describe a namespace", async () => {
+    await db.createNamespace(["myns"]);
+    const desc = await db.describeNamespace(["myns"]);
+    expect(desc).toBeDefined();
+  });
+
+  it("should list namespaces created at the root", async () => {
+    await db.createNamespace(["alpha"]);
+    await db.createNamespace(["beta"]);
+    const list = await db.listNamespaces();
+    expect(list.namespaces).toEqual(expect.arrayContaining(["alpha", "beta"]));
+  });
+
+  it("should list child namespaces under a parent", async () => {
+    await db.createNamespace(["parent"]);
+    await db.createNamespace(["parent", "child"]);
+    const list = await db.listNamespaces(["parent"]);
+    expect(list.namespaces).toContain("child");
+  });
+
+  it("should drop a namespace", async () => {
+    await db.createNamespace(["ephemeral"]);
+    await db.dropNamespace(["ephemeral"]);
+    const list = await db.listNamespaces();
+    expect(list.namespaces).not.toContain("ephemeral");
+  });
+
+  it("should raise an error on any namespace op after close", async () => {
+    await db.close();
+    await expect(db.describeNamespace(["foo"])).rejects.toThrow(
+      "Connection is closed",
+    );
+    await expect(db.listNamespaces()).rejects.toThrow("Connection is closed");
+    await expect(db.createNamespace(["foo"])).rejects.toThrow(
+      "Connection is closed",
+    );
+    await expect(db.dropNamespace(["foo"])).rejects.toThrow(
+      "Connection is closed",
+    );
+  });
+
+  it("should raise an understandable error when describing a non-existent namespace", async () => {
+    await expect(db.describeNamespace(["does-not-exist"])).rejects.toThrow(
+      /not found/i,
+    );
+  });
+
+  it("should raise an error when creating a namespace that already exists", async () => {
+    await db.createNamespace(["dup"]);
+    await expect(db.createNamespace(["dup"])).rejects.toThrow();
+  });
+
+  it("should reject an unrecognized createNamespace mode with a clear error", async () => {
+    await expect(
+      // biome-ignore lint/suspicious/noExplicitAny: deliberately bypass TS to test runtime validation
+      db.createNamespace(["x"], { mode: "frobnicate" as any }),
+    ).rejects.toThrow(/Invalid mode 'frobnicate'/);
+  });
+
+  it("should reject an unrecognized dropNamespace mode with a clear error", async () => {
+    await db.createNamespace(["x"]);
+    await expect(
+      // biome-ignore lint/suspicious/noExplicitAny: deliberately bypass TS to test runtime validation
+      db.dropNamespace(["x"], { mode: "frobnicate" as any }),
+    ).rejects.toThrow(/Invalid mode 'frobnicate'/);
+  });
+
+  it("should reject an unrecognized dropNamespace behavior with a clear error", async () => {
+    await db.createNamespace(["x"]);
+    await expect(
+      // biome-ignore lint/suspicious/noExplicitAny: deliberately bypass TS to test runtime validation
+      db.dropNamespace(["x"], { behavior: "frobnicate" as any }),
+    ).rejects.toThrow(/Invalid behavior 'frobnicate'/);
+  });
+});
+
+describe("connectNamespace", () => {
+  let tmpDir: tmp.DirResult;
+  beforeEach(() => {
+    tmpDir = tmp.dirSync({ unsafeCleanup: true });
+  });
+  afterEach(() => tmpDir.removeCallback());
+
+  it("connects via the dir implementation and supports table ops", async () => {
+    const db = await connectNamespace("dir", { root: tmpDir.name });
+    await db.createTable("users", [{ id: 1 }, { id: 2 }]);
+    await expect(db.tableNames()).resolves.toContain("users");
+  });
+
+  it("throws a clear error when implName is empty", async () => {
+    await expect(connectNamespace("", {})).rejects.toThrow(
+      "implName must be a non-empty string",
+    );
+  });
+
+  it("throws when the namespace implementation is unknown", async () => {
+    await expect(connectNamespace("not-a-real-impl", {})).rejects.toThrow();
+  });
+
+  it("passes storage options through to the namespace", async () => {
+    const db = await connectNamespace(
+      "dir",
+      { root: tmpDir.name },
+      { storageOptions: { newTableDataStorageVersion: "stable" } },
+    );
+    await db.createTable("plumbing", [{ id: 1 }]);
+    await expect(db.tableNames()).resolves.toContain("plumbing");
+  });
+
+  it("supports child namespaces when manifestEnabled is true on the dir config", async () => {
+    const writer = await connectNamespace("dir", {
+      root: tmpDir.name,
+      manifestEnabled: true,
+    });
+    await writer.createNamespace(["analytics"]);
+    await writer.createTable("orders", [{ id: 1 }, { id: 2 }], ["analytics"]);
+    await writer.close();
+
+    const reader = await connectNamespace("dir", {
+      root: tmpDir.name,
+      manifestEnabled: true,
+    });
+    await expect(reader.tableNames(["analytics"])).resolves.toContain("orders");
+    const orders = await reader.openTable("orders", ["analytics"]);
+    await expect(orders.countRows()).resolves.toBe(2);
+  });
+
+  it("merges extraProperties into the dir config and is overridden by typed fields", async () => {
+    // Two observable assertions:
+    // - Typed `root` overrides extraProperties.root: createTable would fail
+    //   under the bogus path if the override didn't happen.
+    // - extraProperties.manifest_enabled="false" is honored end-to-end. Child
+    //   namespaces require manifest mode (default true), so explicitly
+    //   disabling it via extraProperties must make createNamespace reject. If
+    //   extraProperties pass-through were silently broken, the default would
+    //   let createNamespace succeed.
+    const db = await connectNamespace("dir", {
+      root: tmpDir.name,
+      extraProperties: {
+        root: "/should/be/overridden",
+        // biome-ignore lint/style/useNamingConvention: backend property key
+        manifest_enabled: "false",
+      },
+    });
+    await db.createTable("base", [{ id: 1 }]);
+    await expect(db.tableNames()).resolves.toContain("base");
+    await expect(db.createNamespace(["analytics"])).rejects.toThrow();
+  });
+
+  it("flows unknown top-level keys through when implName is dynamic (no silent drop)", async () => {
+    // Routes via the third overload because `impl` is `string`, not the
+    // literal `"dir"`. The dispatcher still notices the runtime value is
+    // "dir", but unknown keys like `manifest_enabled` must not be silently
+    // dropped during the conversion.
+    //
+    // Asserting a *negative* outcome (manifest disabled -> createNamespace
+    // rejects) is required for observability, since the backend default for
+    // `manifest_enabled` is true.
+    const impl: string = "dir";
+    const db = await connectNamespace(impl, {
+      root: tmpDir.name,
+      // biome-ignore lint/style/useNamingConvention: backend property key
+      manifest_enabled: "false",
+    });
+    await expect(db.createNamespace(["mixed"])).rejects.toThrow();
+  });
+});

nodejs/__test__/rerankers.test.ts

@@ -1,6 +1,8 @@
 // SPDX-License-Identifier: Apache-2.0
 // SPDX-FileCopyrightText: Copyright The LanceDB Authors
 
+import { spawn } from "node:child_process";
+import * as path from "node:path";
 import { RecordBatch } from "apache-arrow";
 import * as tmp from "tmp";
 import { Connection, Index, Table, connect, makeArrowTable } from "../lancedb";
@@ -76,4 +78,91 @@ describe("rerankers", function () {
 
     expect(result).toHaveLength(2);
   });
+
+  it("does not keep process alive after rerank query", async function () {
+    const script = `
+import * as lancedb from "./dist/index.js";
+import * as os from "node:os";
+import * as path from "node:path";
+import * as fs from "node:fs/promises";
+
+const dir = await fs.mkdtemp(path.join(os.tmpdir(), "lancedb-rerank-exit-"));
+const db = await lancedb.connect(dir);
+const table = await db.createTable("test", [{ text: "hello", vector: [1, 2, 3] }], {
+  mode: "overwrite",
+});
+await table.createIndex("text", { config: lancedb.Index.fts() });
+await table.waitForIndex(["text_idx"], 30);
+
+const reranker = await lancedb.rerankers.RRFReranker.create();
+await table
+  .query()
+  .nearestTo([1, 2, 3])
+  .fullTextSearch("hello")
+  .rerank(reranker)
+  .toArray();
+
+table.close();
+db.close();
+`;
+
+    await new Promise<void>((resolve, reject) => {
+      const child = spawn(
+        process.execPath,
+        ["--input-type=module", "-e", script],
+        {
+          cwd: path.resolve(__dirname, ".."),
+          stdio: ["ignore", "pipe", "pipe"],
+        },
+      );
+
+      let stdout = "";
+      let stderr = "";
+
+      child.stdout.on("data", (chunk) => {
+        stdout += chunk.toString();
+      });
+
+      child.stderr.on("data", (chunk) => {
+        stderr += chunk.toString();
+      });
+
+      const timeout = setTimeout(() => {
+        child.kill();
+        reject(
+          new Error(
+            `child process did not exit in time\nstdout:\n${stdout}\nstderr:\n${stderr}`,
+          ),
+        );
+      }, 20_000);
+
+      child.on("error", (err) => {
+        clearTimeout(timeout);
+        reject(err);
+      });
+
+      child.on("exit", (code, signal) => {
+        clearTimeout(timeout);
+        if (signal !== null) {
+          reject(
+            new Error(
+              `child process exited with signal ${signal}\nstdout:\n${stdout}\nstderr:\n${stderr}`,
+            ),
+          );
+          return;
+        }
+
+        if (code !== 0) {
+          reject(
+            new Error(
+              `child process exited with code ${code}\nstdout:\n${stdout}\nstderr:\n${stderr}`,
+            ),
+          );
+          return;
+        }
+
+        resolve();
+      });
+    });
+  });
 });

nodejs/__test__/table.test.ts

@@ -103,7 +103,7 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
         },
         numIndices: 0,
         numRows: 3,
-        totalBytes: 24,
+        totalBytes: 44,
       });
     });
 
@@ -115,6 +115,12 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       await expect(table.countRows()).resolves.toBe(1);
     });
 
+    it("should accept skipAutoCleanup on add()", async () => {
+      await table.add([{ id: 1 }], { skipAutoCleanup: true });
+      await table.add([{ id: 2 }], { skipAutoCleanup: true });
+      await expect(table.countRows()).resolves.toBe(2);
+    });
+
     it("should let me close the table", async () => {
       expect(table.isOpen()).toBe(true);
       table.close();
@@ -1259,6 +1265,98 @@ describe("schema evolution", function () {
     expect(await table.schema()).toEqual(expectedSchema);
   });
 
+  it("can add columns with schema for explicit data types", async function () {
+    const con = await connect(tmpDir.name);
+    const table = await con.createTable("vectors", [
+      { id: 1n, vector: [0.1, 0.2] },
+    ]);
+
+    // Define schema for new columns with explicit data types
+    // Note: All columns must be nullable when using addColumns with Schema
+    // because they are initially populated with null values
+    const newColumnsSchema = new Schema([
+      new Field("price", new Float64(), true),
+      new Field("category", new Utf8(), true),
+      new Field("rating", new Int32(), true),
+    ]);
+
+    const result = await table.addColumns(newColumnsSchema);
+    expect(result).toHaveProperty("version");
+    expect(result.version).toBe(2);
+
+    const expectedSchema = new Schema([
+      new Field("id", new Int64(), true),
+      new Field(
+        "vector",
+        new FixedSizeList(2, new Field("item", new Float32(), true)),
+        true,
+      ),
+      new Field("price", new Float64(), true),
+      new Field("category", new Utf8(), true),
+      new Field("rating", new Int32(), true),
+    ]);
+    expect(await table.schema()).toEqual(expectedSchema);
+
+    // Verify that new columns are populated with null values
+    const results = await table.query().toArray();
+    expect(results).toHaveLength(1);
+    expect(results[0].price).toBeNull();
+    expect(results[0].category).toBeNull();
+    expect(results[0].rating).toBeNull();
+  });
+
+  it("can add a single column using Field", async function () {
+    const con = await connect(tmpDir.name);
+    const table = await con.createTable("vectors", [
+      { id: 1n, vector: [0.1, 0.2] },
+    ]);
+
+    // Add a single field
+    const priceField = new Field("price", new Float64(), true);
+    const result = await table.addColumns(priceField);
+    expect(result).toHaveProperty("version");
+    expect(result.version).toBe(2);
+
+    const expectedSchema = new Schema([
+      new Field("id", new Int64(), true),
+      new Field(
+        "vector",
+        new FixedSizeList(2, new Field("item", new Float32(), true)),
+        true,
+      ),
+      new Field("price", new Float64(), true),
+    ]);
+    expect(await table.schema()).toEqual(expectedSchema);
+  });
+
+  it("can add multiple columns using array of Fields", async function () {
+    const con = await connect(tmpDir.name);
+    const table = await con.createTable("vectors", [
+      { id: 1n, vector: [0.1, 0.2] },
+    ]);
+
+    // Add multiple fields as array
+    const fields = [
+      new Field("price", new Float64(), true),
+      new Field("category", new Utf8(), true),
+    ];
+    const result = await table.addColumns(fields);
+    expect(result).toHaveProperty("version");
+    expect(result.version).toBe(2);
+
+    const expectedSchema = new Schema([
+      new Field("id", new Int64(), true),
+      new Field(
+        "vector",
+        new FixedSizeList(2, new Field("item", new Float32(), true)),
+        true,
+      ),
+      new Field("price", new Float64(), true),
+      new Field("category", new Utf8(), true),
+    ]);
+    expect(await table.schema()).toEqual(expectedSchema);
+  });
+
   it("can alter the columns in the schema", async function () {
     const con = await connect(tmpDir.name);
     const schema = new Schema([
@@ -1778,6 +1876,25 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       expect(results.length).toBe(3);
     });
 
+    test("prewarmData errors on local tables", async () => {
+      const db = await connect(tmpDir.name);
+      const data = [
+        { text: "alpha", vector: [0.1, 0.2, 0.3] },
+        { text: "beta", vector: [0.4, 0.5, 0.6] },
+      ];
+      const table = await db.createTable("prewarm_data_test", data);
+
+      // prewarmData is only supported on remote tables. We verify the call
+      // is wired through napi and surfaces the expected error for both
+      // arg shapes (undefined and string[]).
+      await expect(table.prewarmData()).rejects.toThrow(
+        "prewarm_data is currently only supported on remote tables",
+      );
+      await expect(table.prewarmData(["text"])).rejects.toThrow(
+        "prewarm_data is currently only supported on remote tables",
+      );
+    });
+
     test("full text index on list", async () => {
       const db = await connect(tmpDir.name);
       const data = [

nodejs/__test__/vector_types.test.ts

@@ -0,0 +1,110 @@
+// SPDX-License-Identifier: Apache-2.0
+// SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+import * as tmp from "tmp";
+
+import { type Table, connect } from "../lancedb";
+import {
+  Field,
+  FixedSizeList,
+  Float32,
+  Int64,
+  Schema,
+  makeArrowTable,
+} from "../lancedb/arrow";
+
+describe("Vector query with different typed arrays", () => {
+  let tmpDir: tmp.DirResult;
+
+  afterEach(() => {
+    tmpDir?.removeCallback();
+  });
+
+  async function createFloat32Table(): Promise<Table> {
+    tmpDir = tmp.dirSync({ unsafeCleanup: true });
+    const db = await connect(tmpDir.name);
+    const schema = new Schema([
+      new Field("id", new Int64(), true),
+      new Field(
+        "vec",
+        new FixedSizeList(2, new Field("item", new Float32())),
+        true,
+      ),
+    ]);
+    const data = makeArrowTable(
+      [
+        { id: 1n, vec: [1.0, 0.0] },
+        { id: 2n, vec: [0.0, 1.0] },
+        { id: 3n, vec: [1.0, 1.0] },
+      ],
+      { schema },
+    );
+    return db.createTable("test_f32", data);
+  }
+
+  it("should search with Float32Array (baseline)", async () => {
+    const table = await createFloat32Table();
+    const results = await table
+      .query()
+      .nearestTo(new Float32Array([1.0, 0.0]))
+      .limit(1)
+      .toArray();
+
+    expect(results.length).toBe(1);
+    expect(Number(results[0].id)).toBe(1);
+  });
+
+  it("should search with number[] (backward compat)", async () => {
+    const table = await createFloat32Table();
+    const results = await table
+      .query()
+      .nearestTo([1.0, 0.0])
+      .limit(1)
+      .toArray();
+
+    expect(results.length).toBe(1);
+    expect(Number(results[0].id)).toBe(1);
+  });
+
+  it("should search with Float64Array via raw path", async () => {
+    const table = await createFloat32Table();
+    const results = await table
+      .query()
+      .nearestTo(new Float64Array([1.0, 0.0]))
+      .limit(1)
+      .toArray();
+
+    expect(results.length).toBe(1);
+    expect(Number(results[0].id)).toBe(1);
+  });
+
+  it("should add multiple query vectors with Float64Array", async () => {
+    const table = await createFloat32Table();
+    const results = await table
+      .query()
+      .nearestTo(new Float64Array([1.0, 0.0]))
+      .addQueryVector(new Float64Array([0.0, 1.0]))
+      .limit(2)
+      .toArray();
+
+    expect(results.length).toBeGreaterThanOrEqual(2);
+  });
+
+  // Float16Array is only available in Node 22+; not in TypeScript's standard lib yet
+  const float16ArrayCtor = (globalThis as unknown as Record<string, unknown>)
+    .Float16Array as (new (values: number[]) => unknown) | undefined;
+  const hasFloat16 = float16ArrayCtor !== undefined;
+  const f16it = hasFloat16 ? it : it.skip;
+
+  f16it("should search with Float16Array via raw path", async () => {
+    const table = await createFloat32Table();
+    const results = await table
+      .query()
+      .nearestTo(new float16ArrayCtor!([1.0, 0.0]) as Float32Array)
+      .limit(1)
+      .toArray();
+
+    expect(results.length).toBe(1);
+    expect(Number(results[0].id)).toBe(1);
+  });
+});

nodejs/examples/package.json

@@ -11,16 +11,17 @@
     "test": "node --experimental-vm-modules node_modules/.bin/jest --testEnvironment jest-environment-node-single-context --verbose",
     "lint": "biome check *.ts && biome format *.ts",
     "lint-ci": "biome ci .",
-    "lint-fix": "biome check --write *.ts && npm run format",
+    "lint-fix": "biome check --write *.ts && pnpm format",
     "format": "biome format --write *.ts"
   },
   "author": "Lance Devs",
   "license": "Apache-2.0",
+  "packageManager": "pnpm@11.1.1",
   "dependencies": {
-    "@huggingface/transformers": "^3.0.2",
+    "@huggingface/transformers": "3.0.2",
     "@lancedb/lancedb": "file:../dist",
-    "openai": "^4.29.2",
-    "sharp": "^0.33.5"
+    "openai": "4.29.2",
+    "sharp": "0.33.5"
   },
   "devDependencies": {
     "@biomejs/biome": "^1.7.3",

nodejs/examples/pnpm-workspace.yaml

@@ -0,0 +1,13 @@
+# Block resolution of versions less than 24h old (Shai-Hulud window).
+# This is the pnpm 11 default but pinned here so it's visible to
+# reviewers and survives a future pnpm major flipping the default.
+minimumReleaseAge: 1440
+
+# Fail install if a transitive dep tries to run an unapproved script.
+strictDepBuilds: true
+
+allowBuilds:
+  '@biomejs/biome': true
+  onnxruntime-node: true
+  protobufjs: true
+  sharp: true

nodejs/lancedb/arrow.ts

@@ -117,8 +117,9 @@ export type TableLike =
 export type IntoVector =
   | Float32Array
   | Float64Array
+  | Uint8Array
   | number[]
-  | Promise<Float32Array | Float64Array | number[]>;
+  | Promise<Float32Array | Float64Array | Uint8Array | number[]>;
 
 export type MultiVector = IntoVector[];
 
@@ -126,14 +127,48 @@ export function isMultiVector(value: unknown): value is MultiVector {
   return Array.isArray(value) && isIntoVector(value[0]);
 }
 
+// Float16Array is not in TypeScript's standard lib yet; access dynamically
+type Float16ArrayCtor = new (
+  ...args: unknown[]
+) => { buffer: ArrayBuffer; byteOffset: number; byteLength: number };
+const float16ArrayCtor = (globalThis as unknown as Record<string, unknown>)
+  .Float16Array as Float16ArrayCtor | undefined;
+
 export function isIntoVector(value: unknown): value is IntoVector {
   return (
     value instanceof Float32Array ||
     value instanceof Float64Array ||
+    value instanceof Uint8Array ||
+    (float16ArrayCtor !== undefined && value instanceof float16ArrayCtor) ||
     (Array.isArray(value) && !Array.isArray(value[0]))
   );
 }
 
+/**
+ * Extract the underlying byte buffer and data type from a typed array
+ * for passing to the Rust NAPI layer without precision loss.
+ */
+export function extractVectorBuffer(
+  vector: Float32Array | Float64Array | Uint8Array,
+): { data: Uint8Array; dtype: string } | null {
+  if (float16ArrayCtor !== undefined && vector instanceof float16ArrayCtor) {
+    return {
+      data: new Uint8Array(vector.buffer, vector.byteOffset, vector.byteLength),
+      dtype: "float16",
+    };
+  }
+  if (vector instanceof Float64Array) {
+    return {
+      data: new Uint8Array(vector.buffer, vector.byteOffset, vector.byteLength),
+      dtype: "float64",
+    };
+  }
+  if (vector instanceof Uint8Array && !(vector instanceof Float32Array)) {
+    return { data: vector, dtype: "uint8" };
+  }
+  return null;
+}
+
 export function isArrowTable(value: object): value is TableLike {
   if (value instanceof ArrowTable) return true;
   return "schema" in value && "batches" in value;

nodejs/lancedb/connection.ts

@@ -16,6 +16,18 @@ import {
 } from "./arrow";
 import { EmbeddingFunctionConfig, getRegistry } from "./embedding/registry";
 import { Connection as LanceDbConnection } from "./native";
+import type {
+  CreateNamespaceResponse,
+  DescribeNamespaceResponse,
+  DropNamespaceResponse,
+  ListNamespacesResponse,
+} from "./native";
+export type {
+  CreateNamespaceResponse,
+  DescribeNamespaceResponse,
+  DropNamespaceResponse,
+  ListNamespacesResponse,
+};
 import { sanitizeTable } from "./sanitize";
 import { LocalTable, Table } from "./table";
 
@@ -42,7 +54,7 @@ export interface CreateTableOptions {
    * Options already set on the connection will be inherited by the table,
    * but can be overridden here.
    *
-   * The available options are described at https://lancedb.com/docs/storage/
+   * The available options are described at https://docs.lancedb.com/storage/
    */
   storageOptions?: Record<string, string>;
 
@@ -78,7 +90,7 @@ export interface OpenTableOptions {
    * Options already set on the connection will be inherited by the table,
    * but can be overridden here.
    *
-   * The available options are described at https://lancedb.com/docs/storage/
+   * The available options are described at https://docs.lancedb.com/storage/
    */
   storageOptions?: Record<string, string>;
   /**
@@ -110,6 +122,28 @@ export interface TableNamesOptions {
   /** An optional limit to the number of results to return. */
   limit?: number;
 }
+
+export interface ListNamespacesOptions {
+  /** Token from a previous response for pagination. */
+  pageToken?: string;
+  /** An optional limit to the number of results to return. */
+  limit?: number;
+}
+
+export interface CreateNamespaceOptions {
+  /** Creation mode. */
+  mode?: "create" | "exist_ok" | "overwrite";
+  /** Properties to set on the new namespace. */
+  properties?: Record<string, string>;
+}
+
+export interface DropNamespaceOptions {
+  /** Whether to skip if the namespace doesn't exist, or fail. */
+  mode?: "skip" | "fail";
+  /** Refuse to drop if non-empty (restrict) or drop recursively (cascade). */
+  behavior?: "restrict" | "cascade";
+}
+
 /**
  * A LanceDB Connection that allows you to open tables and create new ones.
  *
@@ -166,25 +200,25 @@ export abstract class Connection {
    * List all the table names in this database.
    *
    * Tables will be returned in lexicographical order.
-   * @param {string[]} namespace - The namespace to list tables from (defaults to root namespace)
+   * @param {string[]} namespacePath - The namespace path to list tables from (defaults to root namespace)
    * @param {Partial<TableNamesOptions>} options - options to control the
    * paging / start point
    *
    */
   abstract tableNames(
-    namespace?: string[],
+    namespacePath?: string[],
     options?: Partial<TableNamesOptions>,
   ): Promise<string[]>;
 
   /**
    * Open a table in the database.
    * @param {string} name - The name of the table
-   * @param {string[]} namespace - The namespace of the table (defaults to root namespace)
+   * @param {string[]} namespacePath - The namespace path of the table (defaults to root namespace)
    * @param {Partial<OpenTableOptions>} options - Additional options
    */
   abstract openTable(
     name: string,
-    namespace?: string[],
+    namespacePath?: string[],
     options?: Partial<OpenTableOptions>,
   ): Promise<Table>;
 
@@ -193,7 +227,7 @@ export abstract class Connection {
    * @param {object} options - The options object.
    * @param {string} options.name - The name of the table.
    * @param {Data} options.data - Non-empty Array of Records to be inserted into the table
-   * @param {string[]} namespace - The namespace to create the table in (defaults to root namespace)
+   * @param {string[]} namespacePath - The namespace path to create the table in (defaults to root namespace)
    *
    */
   abstract createTable(
@@ -201,7 +235,7 @@ export abstract class Connection {
       name: string;
       data: Data;
     } & Partial<CreateTableOptions>,
-    namespace?: string[],
+    namespacePath?: string[],
   ): Promise<Table>;
   /**
    * Creates a new Table and initialize it with new data.
@@ -220,13 +254,13 @@ export abstract class Connection {
    * @param {string} name - The name of the table.
    * @param {Record<string, unknown>[] | TableLike} data - Non-empty Array of Records
    * to be inserted into the table
-   * @param {string[]} namespace - The namespace to create the table in (defaults to root namespace)
+   * @param {string[]} namespacePath - The namespace path to create the table in (defaults to root namespace)
    * @param {Partial<CreateTableOptions>} options - Additional options
    */
   abstract createTable(
     name: string,
     data: Record<string, unknown>[] | TableLike,
-    namespace?: string[],
+    namespacePath?: string[],
     options?: Partial<CreateTableOptions>,
   ): Promise<Table>;
 
@@ -245,28 +279,91 @@ export abstract class Connection {
    * Creates a new empty Table
    * @param {string} name - The name of the table.
    * @param {Schema} schema - The schema of the table
-   * @param {string[]} namespace - The namespace to create the table in (defaults to root namespace)
+   * @param {string[]} namespacePath - The namespace path to create the table in (defaults to root namespace)
    * @param {Partial<CreateTableOptions>} options - Additional options
    */
   abstract createEmptyTable(
     name: string,
     schema: import("./arrow").SchemaLike,
-    namespace?: string[],
+    namespacePath?: string[],
     options?: Partial<CreateTableOptions>,
   ): Promise<Table>;
 
   /**
    * Drop an existing table.
    * @param {string} name The name of the table to drop.
-   * @param {string[]} namespace The namespace of the table (defaults to root namespace).
+   * @param {string[]} namespacePath The namespace path of the table (defaults to root namespace).
    */
-  abstract dropTable(name: string, namespace?: string[]): Promise<void>;
+  abstract dropTable(name: string, namespacePath?: string[]): Promise<void>;
 
   /**
    * Drop all tables in the database.
-   * @param {string[]} namespace The namespace to drop tables from (defaults to root namespace).
+   * @param {string[]} namespacePath The namespace path to drop tables from (defaults to root namespace).
    */
-  abstract dropAllTables(namespace?: string[]): Promise<void>;
+  abstract dropAllTables(namespacePath?: string[]): Promise<void>;
+
+  /**
+   * Describe a namespace, returning its properties.
+   *
+   * @param {string[]} namespacePath - The namespace path to describe, in
+   *   parent → child order, e.g. `["analytics", "sales"]`.
+   * @returns {Promise<DescribeNamespaceResponse>} The namespace's properties
+   *   (may be undefined if the namespace has none).
+   */
+  abstract describeNamespace(
+    namespacePath: string[],
+  ): Promise<DescribeNamespaceResponse>;
+
+  /**
+   * List the immediate child namespaces under the given parent.
+   *
+   * Results may be paginated. To retrieve subsequent pages, pass the
+   * `pageToken` returned by a previous call.
+   *
+   * @param {string[]} namespacePath - The parent namespace path. Defaults
+   *   to the root namespace if omitted.
+   * @param {Partial<ListNamespacesOptions>} options - Pagination options
+   *   (`pageToken`, `limit`).
+   * @returns {Promise<ListNamespacesResponse>} Child namespace names and
+   *   an optional token for fetching the next page.
+   */
+  abstract listNamespaces(
+    namespacePath?: string[],
+    options?: Partial<ListNamespacesOptions>,
+  ): Promise<ListNamespacesResponse>;
+
+  /**
+   * Create a new namespace at the given path.
+   *
+   * @param {string[]} namespacePath - The namespace path to create.
+   * @param {Partial<CreateNamespaceOptions>} options - Creation `mode`
+   *   ("create" | "exist_ok" | "overwrite") and optional `properties`
+   *   to attach to the namespace.
+   * @returns {Promise<CreateNamespaceResponse>} The properties of the
+   *   created namespace and an optional transaction id.
+   */
+  abstract createNamespace(
+    namespacePath: string[],
+    options?: Partial<CreateNamespaceOptions>,
+  ): Promise<CreateNamespaceResponse>;
+
+  /**
+   * Drop a namespace.
+   *
+   * Use `behavior: "cascade"` to also drop everything contained in the
+   * namespace (sub-namespaces and tables). The default `"restrict"`
+   * behavior refuses to drop a non-empty namespace.
+   *
+   * @param {string[]} namespacePath - The namespace path to drop.
+   * @param {Partial<DropNamespaceOptions>} options - `mode` ("skip" | "fail"
+   *   for missing-namespace handling) and `behavior` ("restrict" | "cascade").
+   * @returns {Promise<DropNamespaceResponse>} Any properties returned by
+   *   the server and an optional transaction id.
+   */
+  abstract dropNamespace(
+    namespacePath: string[],
+    options?: Partial<DropNamespaceOptions>,
+  ): Promise<DropNamespaceResponse>;
 
   /**
    * Clone a table from a source table.
@@ -279,7 +376,7 @@ export abstract class Connection {
    * @param {string} targetTableName - The name of the target table to create.
    * @param {string} sourceUri - The URI of the source table to clone from.
    * @param {object} options - Clone options.
-   * @param {string[]} options.targetNamespace - The namespace for the target table (defaults to root namespace).
+   * @param {string[]} options.targetNamespacePath - The namespace path for the target table (defaults to root namespace).
    * @param {number} options.sourceVersion - The version of the source table to clone.
    * @param {string} options.sourceTag - The tag of the source table to clone.
    * @param {boolean} options.isShallow - Whether to perform a shallow clone (defaults to true).
@@ -288,7 +385,7 @@ export abstract class Connection {
     targetTableName: string,
     sourceUri: string,
     options?: {
-      targetNamespace?: string[];
+      targetNamespacePath?: string[];
       sourceVersion?: number;
       sourceTag?: string;
       isShallow?: boolean;
@@ -319,25 +416,25 @@ export class LocalConnection extends Connection {
   }
 
   async tableNames(
-    namespaceOrOptions?: string[] | Partial<TableNamesOptions>,
+    namespacePathOrOptions?: string[] | Partial<TableNamesOptions>,
     options?: Partial<TableNamesOptions>,
   ): Promise<string[]> {
-    // Detect if first argument is namespace array or options object
-    let namespace: string[] | undefined;
+    // Detect if first argument is namespacePath array or options object
+    let namespacePath: string[] | undefined;
     let tableNamesOptions: Partial<TableNamesOptions> | undefined;
 
-    if (Array.isArray(namespaceOrOptions)) {
-      // First argument is namespace array
-      namespace = namespaceOrOptions;
+    if (Array.isArray(namespacePathOrOptions)) {
+      // First argument is namespacePath array
+      namespacePath = namespacePathOrOptions;
       tableNamesOptions = options;
     } else {
       // First argument is options object (backwards compatibility)
-      namespace = undefined;
-      tableNamesOptions = namespaceOrOptions;
+      namespacePath = undefined;
+      tableNamesOptions = namespacePathOrOptions;
     }
 
     return this.inner.tableNames(
-      namespace ?? [],
+      namespacePath ?? [],
       tableNamesOptions?.startAfter,
       tableNamesOptions?.limit,
     );
@@ -345,12 +442,12 @@ export class LocalConnection extends Connection {
 
   async openTable(
     name: string,
-    namespace?: string[],
+    namespacePath?: string[],
     options?: Partial<OpenTableOptions>,
   ): Promise<Table> {
     const innerTable = await this.inner.openTable(
       name,
-      namespace ?? [],
+      namespacePath ?? [],
       cleanseStorageOptions(options?.storageOptions),
       options?.indexCacheSize,
     );
@@ -362,7 +459,7 @@ export class LocalConnection extends Connection {
     targetTableName: string,
     sourceUri: string,
     options?: {
-      targetNamespace?: string[];
+      targetNamespacePath?: string[];
       sourceVersion?: number;
       sourceTag?: string;
       isShallow?: boolean;
@@ -371,7 +468,7 @@ export class LocalConnection extends Connection {
     const innerTable = await this.inner.cloneTable(
       targetTableName,
       sourceUri,
-      options?.targetNamespace ?? [],
+      options?.targetNamespacePath ?? [],
       options?.sourceVersion ?? null,
       options?.sourceTag ?? null,
       options?.isShallow ?? true,
@@ -406,42 +503,42 @@ export class LocalConnection extends Connection {
     nameOrOptions:
       | string
       | ({ name: string; data: Data } & Partial<CreateTableOptions>),
-    dataOrNamespace?: Record<string, unknown>[] | TableLike | string[],
-    namespaceOrOptions?: string[] | Partial<CreateTableOptions>,
+    dataOrNamespacePath?: Record<string, unknown>[] | TableLike | string[],
+    namespacePathOrOptions?: string[] | Partial<CreateTableOptions>,
     options?: Partial<CreateTableOptions>,
   ): Promise<Table> {
     if (typeof nameOrOptions !== "string" && "name" in nameOrOptions) {
-      // First overload: createTable(options, namespace?)
+      // First overload: createTable(options, namespacePath?)
       const { name, data, ...createOptions } = nameOrOptions;
-      const namespace = dataOrNamespace as string[] | undefined;
-      return this._createTableImpl(name, data, namespace, createOptions);
+      const namespacePath = dataOrNamespacePath as string[] | undefined;
+      return this._createTableImpl(name, data, namespacePath, createOptions);
     }
 
-    // Second overload: createTable(name, data, namespace?, options?)
+    // Second overload: createTable(name, data, namespacePath?, options?)
     const name = nameOrOptions;
-    const data = dataOrNamespace as Record<string, unknown>[] | TableLike;
+    const data = dataOrNamespacePath as Record<string, unknown>[] | TableLike;
 
-    // Detect if third argument is namespace array or options object
-    let namespace: string[] | undefined;
+    // Detect if third argument is namespacePath array or options object
+    let namespacePath: string[] | undefined;
     let createOptions: Partial<CreateTableOptions> | undefined;
 
-    if (Array.isArray(namespaceOrOptions)) {
-      // Third argument is namespace array
-      namespace = namespaceOrOptions;
+    if (Array.isArray(namespacePathOrOptions)) {
+      // Third argument is namespacePath array
+      namespacePath = namespacePathOrOptions;
       createOptions = options;
     } else {
       // Third argument is options object (backwards compatibility)
-      namespace = undefined;
-      createOptions = namespaceOrOptions;
+      namespacePath = undefined;
+      createOptions = namespacePathOrOptions;
     }
 
-    return this._createTableImpl(name, data, namespace, createOptions);
+    return this._createTableImpl(name, data, namespacePath, createOptions);
   }
 
   private async _createTableImpl(
     name: string,
     data: Data,
-    namespace?: string[],
+    namespacePath?: string[],
     options?: Partial<CreateTableOptions>,
   ): Promise<Table> {
     if (data === undefined) {
@@ -455,7 +552,7 @@ export class LocalConnection extends Connection {
       name,
       buf,
       mode,
-      namespace ?? [],
+      namespacePath ?? [],
       storageOptions,
     );
 
@@ -465,21 +562,21 @@ export class LocalConnection extends Connection {
   async createEmptyTable(
     name: string,
     schema: import("./arrow").SchemaLike,
-    namespaceOrOptions?: string[] | Partial<CreateTableOptions>,
+    namespacePathOrOptions?: string[] | Partial<CreateTableOptions>,
     options?: Partial<CreateTableOptions>,
   ): Promise<Table> {
-    // Detect if third argument is namespace array or options object
-    let namespace: string[] | undefined;
+    // Detect if third argument is namespacePath array or options object
+    let namespacePath: string[] | undefined;
     let createOptions: Partial<CreateTableOptions> | undefined;
 
-    if (Array.isArray(namespaceOrOptions)) {
-      // Third argument is namespace array
-      namespace = namespaceOrOptions;
+    if (Array.isArray(namespacePathOrOptions)) {
+      // Third argument is namespacePath array
+      namespacePath = namespacePathOrOptions;
       createOptions = options;
     } else {
       // Third argument is options object (backwards compatibility)
-      namespace = undefined;
-      createOptions = namespaceOrOptions;
+      namespacePath = undefined;
+      createOptions = namespacePathOrOptions;
     }
 
     let mode: string = createOptions?.mode ?? "create";
@@ -502,18 +599,57 @@ export class LocalConnection extends Connection {
       name,
       buf,
       mode,
-      namespace ?? [],
+      namespacePath ?? [],
       storageOptions,
     );
     return new LocalTable(innerTable);
   }
 
-  async dropTable(name: string, namespace?: string[]): Promise<void> {
-    return this.inner.dropTable(name, namespace ?? []);
+  async dropTable(name: string, namespacePath?: string[]): Promise<void> {
+    return this.inner.dropTable(name, namespacePath ?? []);
   }
 
-  async dropAllTables(namespace?: string[]): Promise<void> {
-    return this.inner.dropAllTables(namespace ?? []);
+  async dropAllTables(namespacePath?: string[]): Promise<void> {
+    return this.inner.dropAllTables(namespacePath ?? []);
+  }
+
+  describeNamespace(
+    namespacePath: string[],
+  ): Promise<DescribeNamespaceResponse> {
+    return this.inner.describeNamespace(namespacePath);
+  }
+
+  listNamespaces(
+    namespacePath?: string[],
+    options?: Partial<ListNamespacesOptions>,
+  ): Promise<ListNamespacesResponse> {
+    return this.inner.listNamespaces(
+      namespacePath ?? [],
+      options?.pageToken,
+      options?.limit,
+    );
+  }
+
+  createNamespace(
+    namespacePath: string[],
+    options?: Partial<CreateNamespaceOptions>,
+  ): Promise<CreateNamespaceResponse> {
+    return this.inner.createNamespace(
+      namespacePath,
+      options?.mode,
+      options?.properties,
+    );
+  }
+
+  dropNamespace(
+    namespacePath: string[],
+    options?: Partial<DropNamespaceOptions>,
+  ): Promise<DropNamespaceResponse> {
+    return this.inner.dropNamespace(
+      namespacePath,
+      options?.mode,
+      options?.behavior,
+    );
   }
 }
 

nodejs/lancedb/index.ts

@@ -8,6 +8,7 @@ import {
 } from "./connection";
 
 import {
+  ConnectNamespaceOptions,
   ConnectionOptions,
   Connection as LanceDbConnection,
   JsHeaderProvider as NativeJsHeaderProvider,
@@ -22,6 +23,7 @@ export { JsHeaderProvider as NativeJsHeaderProvider } from "./native.js";
 export {
   AddColumnsSql,
   ConnectionOptions,
+  ConnectNamespaceOptions,
   IndexStatistics,
   IndexConfig,
   ClientConfig,
@@ -62,6 +64,13 @@ export {
   CreateTableOptions,
   TableNamesOptions,
   OpenTableOptions,
+  ListNamespacesOptions,
+  CreateNamespaceOptions,
+  DropNamespaceOptions,
+  ListNamespacesResponse,
+  CreateNamespaceResponse,
+  DropNamespaceResponse,
+  DescribeNamespaceResponse,
 } from "./connection";
 
 export { Session } from "./native.js";
@@ -293,3 +302,197 @@ export async function connect(
   );
   return new LocalConnection(nativeConn);
 }
+
+/**
+ * Configuration for the built-in directory namespace (`"dir"`).
+ *
+ * The directory namespace stores tables under a single root path (local
+ * filesystem or object storage URI). See
+ * {@link https://docs.lancedb.com/namespaces} for the documented surface;
+ * less-common knobs live under {@link DirNamespaceConfig.extraProperties}.
+ */
+export interface DirNamespaceConfig {
+  /** Root path or URI containing the LanceDB tables. */
+  root: string;
+  /**
+   * Whether to maintain a namespace manifest at the root. Required for
+   * child namespaces. Defaults to true on the impl side.
+   */
+  manifestEnabled?: boolean;
+  /**
+   * Additional raw properties passed verbatim to the namespace
+   * implementation (e.g. `storage.*`, `credential_vendor.*`). Typed
+   * fields above take precedence on key collision.
+   */
+  extraProperties?: Record<string, string>;
+}
+
+/**
+ * Configuration for the built-in REST namespace (`"rest"`).
+ *
+ * The REST namespace talks to a remote catalog server over HTTP. See
+ * {@link https://docs.lancedb.com/namespaces} for the documented surface;
+ * less-common knobs (TLS, metrics) live under
+ * {@link RestNamespaceConfig.extraProperties}.
+ */
+export interface RestNamespaceConfig {
+  /** Catalog endpoint URL. */
+  uri: string;
+  /**
+   * HTTP headers forwarded with each request. Keys are passed through
+   * as-is (e.g. `"x-api-key"`, `"Authorization"`).
+   */
+  headers?: Record<string, string>;
+  /**
+   * Additional raw properties passed verbatim to the namespace
+   * implementation (e.g. `tls.*`, `ops_metrics_enabled`, `delimiter`).
+   * Typed fields above take precedence on key collision.
+   */
+  extraProperties?: Record<string, string>;
+}
+
+function dirConfigToProperties(
+  config: DirNamespaceConfig,
+): Record<string, string> {
+  // Spread the whole input so that unknown keys (e.g. a raw `manifest_enabled`
+  // passed via the dynamic-impl path) flow through instead of being dropped.
+  // Typed transformations layer on top.
+  const { manifestEnabled, extraProperties, ...rest } = config;
+  const properties: Record<string, string> = {
+    ...(extraProperties ?? {}),
+    ...(rest as Record<string, string>),
+  };
+  if (manifestEnabled !== undefined) {
+    properties.manifest_enabled = String(manifestEnabled);
+  }
+  return properties;
+}
+
+function restConfigToProperties(
+  config: RestNamespaceConfig,
+): Record<string, string> {
+  const { headers, extraProperties, ...rest } = config;
+  const properties: Record<string, string> = {
+    ...(extraProperties ?? {}),
+    ...(rest as Record<string, string>),
+  };
+  if (headers) {
+    for (const [name, value] of Object.entries(headers)) {
+      properties[`headers.${name}`] = value;
+    }
+  }
+  return properties;
+}
+
+/**
+ * Connect to a LanceDB database through a namespace.
+ *
+ * Unlike {@link connect}, which routes by URI scheme (local path vs.
+ * `db://` cloud), `connectNamespace` always returns a namespace-backed
+ * connection. The `implName` selects the namespace implementation:
+ *
+ * - `"dir"` — directory namespace, configured with {@link DirNamespaceConfig}.
+ * - `"rest"` — remote REST catalog, configured with {@link RestNamespaceConfig}.
+ * - Any other string — full module path for a custom implementation,
+ *   configured with a free-form string-keyed `properties` map.
+ *
+ * @example Typed dir namespace
+ * ```ts
+ * const db = await connectNamespace("dir", { root: "/path/to/db" });
+ * await db.createTable("users", [{ id: 1 }]);
+ * ```
+ *
+ * @example Typed REST namespace with auth headers
+ * ```ts
+ * const db = await connectNamespace("rest", {
+ *   uri: "https://catalog.example.com",
+ *   headers: { "x-api-key": process.env.CATALOG_KEY ?? "" },
+ * });
+ * ```
+ *
+ * @example Custom implementation with raw properties
+ * ```ts
+ * const db = await connectNamespace("my.custom.Namespace", {
+ *   endpoint: "...",
+ * });
+ * ```
+ */
+export function connectNamespace(
+  implName: "dir",
+  config: DirNamespaceConfig,
+  options?: Partial<ConnectNamespaceOptions>,
+): Promise<Connection>;
+/**
+ * Connect through the built-in REST namespace.
+ *
+ * Configured with {@link RestNamespaceConfig}. See the function-level
+ * documentation above for the full surface, examples, and how this
+ * relates to {@link connect}.
+ *
+ * @example
+ * ```ts
+ * const db = await connectNamespace("rest", {
+ *   uri: "https://catalog.example.com",
+ *   headers: { "x-api-key": process.env.CATALOG_KEY ?? "" },
+ * });
+ * ```
+ */
+export function connectNamespace(
+  implName: "rest",
+  config: RestNamespaceConfig,
+  options?: Partial<ConnectNamespaceOptions>,
+): Promise<Connection>;
+/**
+ * Connect through a custom namespace implementation by full module path,
+ * configured with a free-form string-keyed `properties` map. Use the
+ * typed overloads above for the built-in `"dir"` and `"rest"` impls.
+ *
+ * See the function-level documentation above for examples and how this
+ * relates to {@link connect}.
+ *
+ * @example
+ * ```ts
+ * const db = await connectNamespace("my.custom.Namespace", {
+ *   endpoint: "...",
+ * });
+ * ```
+ */
+export function connectNamespace(
+  implName: string,
+  properties: Record<string, string>,
+  options?: Partial<ConnectNamespaceOptions>,
+): Promise<Connection>;
+export async function connectNamespace(
+  implName: string,
+  configOrProperties:
+    | DirNamespaceConfig
+    | RestNamespaceConfig
+    | Record<string, string>,
+  options?: Partial<ConnectNamespaceOptions>,
+): Promise<Connection> {
+  let properties: Record<string, string>;
+  if (implName === "dir") {
+    properties = dirConfigToProperties(
+      configOrProperties as DirNamespaceConfig,
+    );
+  } else if (implName === "rest") {
+    properties = restConfigToProperties(
+      configOrProperties as RestNamespaceConfig,
+    );
+  } else {
+    properties = configOrProperties as Record<string, string>;
+  }
+
+  const finalOptions: ConnectNamespaceOptions = (options ??
+    {}) as ConnectNamespaceOptions;
+  finalOptions.storageOptions = cleanseStorageOptions(
+    finalOptions.storageOptions,
+  );
+
+  const nativeConn = await LanceDbConnection.newWithNamespace(
+    implName,
+    properties,
+    finalOptions,
+  );
+  return new LocalConnection(nativeConn);
+}

nodejs/lancedb/merge.ts

@@ -87,6 +87,23 @@ export class MergeInsertBuilder {
       this.#schema,
     );
   }
+
+  /**
+   * Skip the automatic cleanup of old dataset versions that would otherwise
+   * run as part of this merge insert's commit. Forwards to
+   * `MergeInsertBuilder::skip_auto_cleanup` in lance-core.
+   *
+   * Useful for high-frequency writers that prefer to manage version cleanup
+   * themselves, or writers without delete permissions on the underlying storage.
+   *
+   * @param skip - If true, the auto-cleanup step is skipped at commit time.
+   */
+  skipAutoCleanup(skip: boolean): MergeInsertBuilder {
+    return new MergeInsertBuilder(
+      this.#native.skipAutoCleanup(skip),
+      this.#schema,
+    );
+  }
   /**
    * Executes the merge insert operation
    *

nodejs/lancedb/query.ts

@@ -5,6 +5,7 @@ import {
   Table as ArrowTable,
   type IntoVector,
   RecordBatch,
+  extractVectorBuffer,
   fromBufferToRecordBatch,
   fromRecordBatchToBuffer,
   tableFromIPC,
@@ -661,10 +662,8 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
       const res = (async () => {
         try {
           const v = await vector;
-          const arr = Float32Array.from(v);
-          //
           // biome-ignore lint/suspicious/noExplicitAny: we need to get the `inner`, but js has no package scoping
-          const value: any = this.addQueryVector(arr);
+          const value: any = this.addQueryVector(v);
           const inner = value.inner as
             | NativeVectorQuery
             | Promise<NativeVectorQuery>;
@@ -676,7 +675,12 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
       return new VectorQuery(res);
     } else {
       super.doCall((inner) => {
-        inner.addQueryVector(Float32Array.from(vector));
+        const raw = Array.isArray(vector) ? null : extractVectorBuffer(vector);
+        if (raw) {
+          inner.addQueryVectorRaw(raw.data, raw.dtype);
+        } else {
+          inner.addQueryVector(Float32Array.from(vector as number[]));
+        }
       });
       return this;
     }
@@ -765,14 +769,23 @@ export class Query extends StandardQueryBase<NativeQuery> {
    * a default `limit` of 10 will be used.  @see {@link Query#limit}
    */
   nearestTo(vector: IntoVector): VectorQuery {
+    const callNearestTo = (
+      inner: NativeQuery,
+      resolved: Float32Array | Float64Array | Uint8Array | number[],
+    ): NativeVectorQuery => {
+      const raw = Array.isArray(resolved)
+        ? null
+        : extractVectorBuffer(resolved);
+      if (raw) {
+        return inner.nearestToRaw(raw.data, raw.dtype);
+      }
+      return inner.nearestTo(Float32Array.from(resolved as number[]));
+    };
+
     if (this.inner instanceof Promise) {
       const nativeQuery = this.inner.then(async (inner) => {
-        if (vector instanceof Promise) {
-          const arr = await vector.then((v) => Float32Array.from(v));
-          return inner.nearestTo(arr);
-        } else {
-          return inner.nearestTo(Float32Array.from(vector));
-        }
+        const resolved = vector instanceof Promise ? await vector : vector;
+        return callNearestTo(inner, resolved);
       });
       return new VectorQuery(nativeQuery);
     }
@@ -780,10 +793,8 @@ export class Query extends StandardQueryBase<NativeQuery> {
       const res = (async () => {
         try {
           const v = await vector;
-          const arr = Float32Array.from(v);
-          //
           // biome-ignore lint/suspicious/noExplicitAny: we need to get the `inner`, but js has no package scoping
-          const value: any = this.nearestTo(arr);
+          const value: any = this.nearestTo(v);
           const inner = value.inner as
             | NativeVectorQuery
             | Promise<NativeVectorQuery>;
@@ -794,7 +805,7 @@ export class Query extends StandardQueryBase<NativeQuery> {
       })();
       return new VectorQuery(res);
     } else {
-      const vectorQuery = this.inner.nearestTo(Float32Array.from(vector));
+      const vectorQuery = callNearestTo(this.inner, vector);
       return new VectorQuery(vectorQuery);
     }
   }

nodejs/lancedb/table.ts

@@ -5,12 +5,15 @@ import {
   Table as ArrowTable,
   Data,
   DataType,
+  Field,
   IntoVector,
   MultiVector,
   Schema,
   dataTypeToJson,
   fromDataToBuffer,
+  fromTableToBuffer,
   isMultiVector,
+  makeEmptyTable,
   tableFromIPC,
 } from "./arrow";
 
@@ -53,6 +56,18 @@ export interface AddDataOptions {
    * If "overwrite" then the new data will replace the existing data in the table.
    */
   mode: "append" | "overwrite";
+  /**
+   * If true, skip the automatic cleanup of old dataset versions that would
+   * otherwise run as part of this write's commit. Forwards to
+   * `WriteParams.skip_auto_cleanup` in lance-core.
+   *
+   * Useful for high-frequency writers that prefer to manage version cleanup
+   * themselves (for example, via a separate periodic optimize job), or for
+   * writers that don't have delete permissions on the underlying storage.
+   *
+   * Defaults to false.
+   */
+  skipAutoCleanup?: boolean;
 }
 
 export interface UpdateOptions {
@@ -84,6 +99,16 @@ export interface OptimizeOptions {
    * tbl.optimize({cleanupOlderThan: new Date()});
    */
   cleanupOlderThan: Date;
+  /**
+   * Because they may be part of an in-progress transaction, files newer than
+   * 7 days old are not deleted by default. If you are sure that there are no
+   * in-progress transactions, then you can set this to true to delete all
+   * files older than `cleanupOlderThan`.
+   *
+   * **WARNING**: This should only be set to true if you can guarantee that
+   * no other process is currently working on this dataset. Otherwise the
+   * dataset could be put into a corrupted state.
+   */
   deleteUnverified: boolean;
 }
 
@@ -272,6 +297,25 @@ export abstract class Table {
    */
   abstract prewarmIndex(name: string): Promise<void>;
 
+  /**
+   * Prewarm one or more columns of data in the table.
+   *
+   * @param columns The columns to prewarm. If undefined, all columns are prewarmed.
+   *
+   * This will load the column data into the page cache so that future queries that
+   * read those columns avoid the initial cold-start latency.  This call initiates
+   * prewarming and returns once the request is accepted; the warming itself may
+   * continue in the background.  Calling it on already-prewarmed columns is a
+   * no-op on the server.
+   *
+   * Prewarming is generally useful for columns used in filters or projections.
+   * Large columns (e.g. high-dimensional vectors or binary data) may not be
+   * practical to prewarm.
+   *
+   * This feature is currently only supported on remote tables.
+   */
+  abstract prewarmData(columns?: string[]): Promise<void>;
+
   /**
    * Waits for asynchronous indexing to complete on the table.
    *
@@ -381,15 +425,16 @@ export abstract class Table {
   abstract vectorSearch(vector: IntoVector | MultiVector): VectorQuery;
   /**
    * Add new columns with defined values.
-   * @param {AddColumnsSql[]} newColumnTransforms pairs of column names and
-   * the SQL expression to use to calculate the value of the new column. These
-   * expressions will be evaluated for each row in the table, and can
-   * reference existing columns in the table.
+   * @param {AddColumnsSql[] | Field | Field[] | Schema} newColumnTransforms Either:
+   *   - An array of objects with column names and SQL expressions to calculate values
+   *   - A single Arrow Field defining one column with its data type (column will be initialized with null values)
+   *   - An array of Arrow Fields defining columns with their data types (columns will be initialized with null values)
+   *   - An Arrow Schema defining columns with their data types (columns will be initialized with null values)
    * @returns {Promise<AddColumnsResult>} A promise that resolves to an object
    * containing the new version number of the table after adding the columns.
    */
   abstract addColumns(
-    newColumnTransforms: AddColumnsSql[],
+    newColumnTransforms: AddColumnsSql[] | Field | Field[] | Schema,
   ): Promise<AddColumnsResult>;
 
   /**
@@ -501,19 +546,7 @@ export abstract class Table {
    *  - Index: Optimizes the indices, adding new data to existing indices
    *
    *
-   *  Experimental API
-   *  ----------------
-   *
-   *  The optimization process is undergoing active development and may change.
-   *  Our goal with these changes is to improve the performance of optimization and
-   *  reduce the complexity.
-   *
-   *  That being said, it is essential today to run optimize if you want the best
-   *  performance.  It should be stable and safe to use in production, but it our
-   *  hope that the API may be simplified (or not even need to be called) in the
-   *  future.
-   *
-   *  The frequency an application shoudl call optimize is based on the frequency of
+   *  The frequency an application should call optimize is based on the frequency of
    *  data modifications.  If data is frequently added, deleted, or updated then
    *  optimize should be run frequently.  A good rule of thumb is to run optimize if
    *  you have added or modified 100,000 or more records or run more than 20 data
@@ -615,7 +648,7 @@ export class LocalTable extends Table {
     const schema = await this.schema();
 
     const buffer = await fromDataToBuffer(data, undefined, schema);
-    return await this.inner.add(buffer, mode);
+    return await this.inner.add(buffer, mode, options?.skipAutoCleanup);
   }
 
   async update(
@@ -708,6 +741,10 @@ export class LocalTable extends Table {
     await this.inner.prewarmIndex(name);
   }
 
+  async prewarmData(columns?: string[]): Promise<void> {
+    await this.inner.prewarmData(columns);
+  }
+
   async waitForIndex(
     indexNames: string[],
     timeoutSeconds: number,
@@ -806,9 +843,40 @@ export class LocalTable extends Table {
   // TODO: Support BatchUDF
 
   async addColumns(
-    newColumnTransforms: AddColumnsSql[],
+    newColumnTransforms: AddColumnsSql[] | Field | Field[] | Schema,
   ): Promise<AddColumnsResult> {
-    return await this.inner.addColumns(newColumnTransforms);
+    // Handle single Field -> convert to array of Fields
+    if (newColumnTransforms instanceof Field) {
+      newColumnTransforms = [newColumnTransforms];
+    }
+
+    // Handle array of Fields -> convert to Schema
+    if (
+      Array.isArray(newColumnTransforms) &&
+      newColumnTransforms.length > 0 &&
+      newColumnTransforms[0] instanceof Field
+    ) {
+      const fields = newColumnTransforms as Field[];
+      newColumnTransforms = new Schema(fields);
+    }
+
+    // Handle Schema -> use schema-based approach
+    if (newColumnTransforms instanceof Schema) {
+      const schema = newColumnTransforms;
+      // Convert schema to buffer using Arrow IPC format
+      const emptyTable = makeEmptyTable(schema);
+      const schemaBuf = await fromTableToBuffer(emptyTable);
+      return await this.inner.addColumnsWithSchema(schemaBuf);
+    }
+
+    // Handle SQL expressions (existing functionality)
+    if (Array.isArray(newColumnTransforms)) {
+      return await this.inner.addColumns(
+        newColumnTransforms as AddColumnsSql[],
+      );
+    }
+
+    throw new Error("Invalid input type for addColumns");
   }
 
   async alterColumns(

nodejs/npm/darwin-arm64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-arm64",
-	"version": "0.27.0-beta.5",
+	"version": "0.28.0-beta.11",
 	"os": ["darwin"],
 	"cpu": ["arm64"],
 	"main": "lancedb.darwin-arm64.node",

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-gnu",
-	"version": "0.27.0-beta.5",
+	"version": "0.28.0-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": "0.27.0-beta.5",
+	"version": "0.28.0-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": "0.27.0-beta.5",
+	"version": "0.28.0-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": "0.27.0-beta.5",
+	"version": "0.28.0-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": "0.27.0-beta.5",
+  "version": "0.28.0-beta.11",
   "os": [
     "win32"
   ],

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-win32-x64-msvc",
-	"version": "0.27.0-beta.5",
+	"version": "0.28.0-beta.11",
 	"os": ["win32"],
 	"cpu": ["x64"],
 	"main": "lancedb.win32-x64-msvc.node",

nodejs/package.json

@@ -11,7 +11,7 @@
     "ann"
   ],
   "private": false,
-  "version": "0.27.0-beta.5",
+  "version": "0.28.0-beta.11",
   "main": "dist/index.js",
   "exports": {
     ".": "./dist/index.js",
@@ -38,15 +38,15 @@
     "url": "https://github.com/lancedb/lancedb"
   },
   "devDependencies": {
-    "@aws-sdk/client-dynamodb": "^3.33.0",
-    "@aws-sdk/client-kms": "^3.33.0",
-    "@aws-sdk/client-s3": "^3.33.0",
+    "@aws-sdk/client-dynamodb": "3.1003.0",
+    "@aws-sdk/client-kms": "3.1003.0",
+    "@aws-sdk/client-s3": "3.1003.0",
     "@biomejs/biome": "^1.7.3",
     "@jest/globals": "^29.7.0",
-    "@napi-rs/cli": "^3.5.1",
+    "@napi-rs/cli": "3.5.1",
     "@types/axios": "^0.14.0",
     "@types/jest": "^29.1.2",
-    "@types/node": "^22.7.4",
+    "@types/node": "22.7.4",
     "@types/tmp": "^0.2.6",
     "apache-arrow-15": "npm:apache-arrow@15.0.0",
     "apache-arrow-16": "npm:apache-arrow@16.0.0",
@@ -57,9 +57,9 @@
     "shx": "^0.3.4",
     "tmp": "^0.2.3",
     "ts-jest": "^29.1.2",
-    "typedoc": "^0.26.4",
-    "typedoc-plugin-markdown": "^4.2.1",
-    "typescript": "^5.5.4",
+    "typedoc": "0.26.4",
+    "typedoc-plugin-markdown": "4.2.1",
+    "typescript": "5.5.4",
     "typescript-eslint": "^7.1.0"
   },
   "ava": {
@@ -68,16 +68,16 @@
   "engines": {
     "node": ">= 18"
   },
+  "packageManager": "pnpm@11.1.1",
   "cpu": ["x64", "arm64"],
   "os": ["darwin", "linux", "win32"],
   "scripts": {
     "artifacts": "napi artifacts",
     "build:debug": "napi build --platform --dts ../lancedb/native.d.ts --js ../lancedb/native.js --output-dir lancedb",
-    "postbuild:debug": "shx mkdir -p dist && shx cp lancedb/*.node dist/",
+    "postbuild:debug": "shx mkdir -p dist && shx cp lancedb/*.node dist/ && node -e \"require('fs').writeFileSync('dist/package.json', JSON.stringify({name:'@lancedb/lancedb',type:'commonjs'}))\"",
     "build:release": "napi build --platform --release --dts ../lancedb/native.d.ts --js ../lancedb/native.js --output-dir dist",
-    "postbuild:release": "shx mkdir -p dist && shx cp lancedb/*.node dist/",
-    "build": "npm run build:debug && npm run tsc",
-    "build-release": "npm run build:release && npm run tsc",
+    "build": "pnpm build:debug && pnpm tsc",
+    "build-release": "pnpm build:release && pnpm tsc",
     "tsc": "tsc -b",
     "posttsc": "shx cp lancedb/native.d.ts dist/native.d.ts",
     "lint-ci": "biome ci .",
@@ -87,7 +87,7 @@
     "lint-fix": "biome check --write . && biome format --write .",
     "prepublishOnly": "napi prepublish -t npm",
     "test": "jest --verbose",
-    "integration": "S3_TEST=1 npm run test",
+    "integration": "S3_TEST=1 pnpm test",
     "universal": "napi universalize",
     "version": "napi version"
   },
@@ -95,8 +95,8 @@
     "reflect-metadata": "^0.2.2"
   },
   "optionalDependencies": {
-    "@huggingface/transformers": "^3.0.2",
-    "openai": "^4.29.2"
+    "@huggingface/transformers": "3.0.2",
+    "openai": "4.29.2"
   },
   "peerDependencies": {
     "apache-arrow": ">=15.0.0 <=18.1.0"

nodejs/pnpm-workspace.yaml

@@ -0,0 +1,18 @@
+# Flat node_modules layout. The @napi-rs/cli build step fails to locate
+# the cdylib artifact under pnpm's isolated layout; the hoisted linker
+# mirrors npm's structure and unblocks the native build.
+nodeLinker: hoisted
+
+# Block resolution of versions less than 24h old (Shai-Hulud window).
+# This is the pnpm 11 default but pinned here so it's visible to
+# reviewers and survives a future pnpm major flipping the default.
+minimumReleaseAge: 1440
+
+# Fail install if a transitive dep tries to run an unapproved script.
+strictDepBuilds: true
+
+allowBuilds:
+  '@biomejs/biome': true
+  onnxruntime-node: true
+  protobufjs: true
+  sharp: true

nodejs/src/connection.rs

@@ -8,12 +8,16 @@ use lancedb::database::{CreateTableMode, Database};
 use napi::bindgen_prelude::*;
 use napi_derive::*;
 
+use crate::ConnectNamespaceOptions;
 use crate::ConnectionOptions;
 use crate::error::NapiErrorExt;
 use crate::header::JsHeaderProvider;
 use crate::table::Table;
-use lancedb::connection::{ConnectBuilder, Connection as LanceDBConnection};
+use lancedb::connection::{ConnectBuilder, Connection as LanceDBConnection, connect_namespace};
 
+use lance_namespace::models::{
+    CreateNamespaceRequest, DescribeNamespaceRequest, DropNamespaceRequest, ListNamespacesRequest,
+};
 use lancedb::ipc::{ipc_file_to_batches, ipc_file_to_schema};
 
 #[napi]
@@ -21,6 +25,29 @@ pub struct Connection {
     inner: Option<LanceDBConnection>,
 }
 
+#[napi(object)]
+pub struct DescribeNamespaceResponse {
+    pub properties: Option<HashMap<String, String>>,
+}
+
+#[napi(object)]
+pub struct ListNamespacesResponse {
+    pub namespaces: Vec<String>,
+    pub page_token: Option<String>,
+}
+
+#[napi(object)]
+pub struct CreateNamespaceResponse {
+    pub properties: Option<HashMap<String, String>>,
+    pub transaction_id: Option<String>,
+}
+
+#[napi(object)]
+pub struct DropNamespaceResponse {
+    pub properties: Option<HashMap<String, String>>,
+    pub transaction_id: Option<Vec<String>>,
+}
+
 impl Connection {
     pub(crate) fn inner_new(inner: LanceDBConnection) -> Self {
         Self { inner: Some(inner) }
@@ -67,6 +94,12 @@ impl Connection {
                 builder = builder.storage_option(key, value);
             }
         }
+        if let Some(manifest_enabled) = options.manifest_enabled {
+            builder = builder.manifest_enabled(manifest_enabled);
+        }
+        if let Some(namespace_client_properties) = options.namespace_client_properties {
+            builder = builder.namespace_client_properties(namespace_client_properties);
+        }
 
         // Create client config, optionally with header provider
         let client_config = options.client_config.unwrap_or_default();
@@ -100,6 +133,39 @@ impl Connection {
         Ok(Self::inner_new(builder.execute().await.default_error()?))
     }
 
+    /// Create a new Connection instance backed by a namespace implementation.
+    #[napi(factory)]
+    pub async fn new_with_namespace(
+        impl_name: String,
+        properties: HashMap<String, String>,
+        options: ConnectNamespaceOptions,
+    ) -> napi::Result<Self> {
+        if impl_name.is_empty() {
+            return Err(napi::Error::from_reason(
+                "implName must be a non-empty string",
+            ));
+        }
+
+        let mut builder = connect_namespace(&impl_name, properties);
+        if let Some(interval) = options.read_consistency_interval {
+            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 {
+                builder = builder.storage_option(key, value);
+            }
+        }
+        if let Some(namespace_client_properties) = options.namespace_client_properties {
+            builder = builder.namespace_client_properties(namespace_client_properties);
+        }
+        if let Some(session) = options.session {
+            builder = builder.session(session.inner.clone());
+        }
+
+        Ok(Self::inner_new(builder.execute().await.default_error()?))
+    }
+
     #[napi]
     pub fn display(&self) -> napi::Result<String> {
         Ok(self.get_inner()?.to_string())
@@ -119,12 +185,12 @@ impl Connection {
     #[napi(catch_unwind)]
     pub async fn table_names(
         &self,
-        namespace: Vec<String>,
+        namespace_path: Option<Vec<String>>,
         start_after: Option<String>,
         limit: Option<u32>,
     ) -> napi::Result<Vec<String>> {
         let mut op = self.get_inner()?.table_names();
-        op = op.namespace(namespace);
+        op = op.namespace(namespace_path.unwrap_or_default());
         if let Some(start_after) = start_after {
             op = op.start_after(start_after);
         }
@@ -146,7 +212,7 @@ impl Connection {
         name: String,
         buf: Buffer,
         mode: String,
-        namespace: Vec<String>,
+        namespace_path: Option<Vec<String>>,
         storage_options: Option<HashMap<String, String>>,
     ) -> napi::Result<Table> {
         let batches = ipc_file_to_batches(buf.to_vec())
@@ -154,7 +220,7 @@ impl Connection {
         let mode = Self::parse_create_mode_str(&mode)?;
         let mut builder = self.get_inner()?.create_table(&name, batches).mode(mode);
 
-        builder = builder.namespace(namespace);
+        builder = builder.namespace(namespace_path.unwrap_or_default());
 
         if let Some(storage_options) = storage_options {
             for (key, value) in storage_options {
@@ -171,7 +237,7 @@ impl Connection {
         name: String,
         schema_buf: Buffer,
         mode: String,
-        namespace: Vec<String>,
+        namespace_path: Option<Vec<String>>,
         storage_options: Option<HashMap<String, String>>,
     ) -> napi::Result<Table> {
         let schema = ipc_file_to_schema(schema_buf.to_vec()).map_err(|e| {
@@ -183,7 +249,7 @@ impl Connection {
             .create_empty_table(&name, schema)
             .mode(mode);
 
-        builder = builder.namespace(namespace);
+        builder = builder.namespace(namespace_path.unwrap_or_default());
 
         if let Some(storage_options) = storage_options {
             for (key, value) in storage_options {
@@ -198,13 +264,13 @@ impl Connection {
     pub async fn open_table(
         &self,
         name: String,
-        namespace: Vec<String>,
+        namespace_path: Option<Vec<String>>,
         storage_options: Option<HashMap<String, String>>,
         index_cache_size: Option<u32>,
     ) -> napi::Result<Table> {
         let mut builder = self.get_inner()?.open_table(&name);
 
-        builder = builder.namespace(namespace);
+        builder = builder.namespace(namespace_path.unwrap_or_default());
 
         if let Some(storage_options) = storage_options {
             for (key, value) in storage_options {
@@ -223,7 +289,7 @@ impl Connection {
         &self,
         target_table_name: String,
         source_uri: String,
-        target_namespace: Vec<String>,
+        target_namespace_path: Option<Vec<String>>,
         source_version: Option<i64>,
         source_tag: Option<String>,
         is_shallow: bool,
@@ -232,7 +298,7 @@ impl Connection {
             .get_inner()?
             .clone_table(&target_table_name, &source_uri);
 
-        builder = builder.target_namespace(target_namespace);
+        builder = builder.target_namespace(target_namespace_path.unwrap_or_default());
 
         if let Some(version) = source_version {
             builder = builder.source_version(version as u64);
@@ -250,18 +316,147 @@ impl Connection {
 
     /// Drop table with the name. Or raise an error if the table does not exist.
     #[napi(catch_unwind)]
-    pub async fn drop_table(&self, name: String, namespace: Vec<String>) -> napi::Result<()> {
+    pub async fn drop_table(
+        &self,
+        name: String,
+        namespace_path: Option<Vec<String>>,
+    ) -> napi::Result<()> {
+        let ns = namespace_path.unwrap_or_default();
         self.get_inner()?
-            .drop_table(&name, &namespace)
+            .drop_table(&name, &ns)
             .await
             .default_error()
     }
 
     #[napi(catch_unwind)]
-    pub async fn drop_all_tables(&self, namespace: Vec<String>) -> napi::Result<()> {
-        self.get_inner()?
-            .drop_all_tables(&namespace)
+    pub async fn drop_all_tables(&self, namespace_path: Option<Vec<String>>) -> napi::Result<()> {
+        let ns = namespace_path.unwrap_or_default();
+        self.get_inner()?.drop_all_tables(&ns).await.default_error()
+    }
+
+    #[napi(catch_unwind)]
+    /// Describe a namespace and return its properties.
+    pub async fn describe_namespace(
+        &self,
+        namespace_path: Vec<String>,
+    ) -> napi::Result<DescribeNamespaceResponse> {
+        let req = DescribeNamespaceRequest {
+            id: Some(namespace_path),
+            ..Default::default()
+        };
+        let resp = self
+            .get_inner()?
+            .describe_namespace(req)
             .await
-            .default_error()
+            .default_error()?;
+        Ok(DescribeNamespaceResponse {
+            properties: resp.properties,
+        })
+    }
+
+    #[napi(catch_unwind)]
+    /// List child namespaces under the given namespace path
+    pub async fn list_namespaces(
+        &self,
+        namespace_path: Option<Vec<String>>,
+        page_token: Option<String>,
+        limit: Option<u32>,
+    ) -> napi::Result<ListNamespacesResponse> {
+        let req = ListNamespacesRequest {
+            id: namespace_path,
+            page_token,
+            limit: limit.map(|l| l as i32),
+            ..Default::default()
+        };
+        let resp = self
+            .get_inner()?
+            .list_namespaces(req)
+            .await
+            .default_error()?;
+        Ok(ListNamespacesResponse {
+            namespaces: resp.namespaces,
+            page_token: resp.page_token,
+        })
+    }
+
+    #[napi(catch_unwind)]
+    /// Create a new namespace with optional properties.
+    pub async fn create_namespace(
+        &self,
+        namespace_path: Vec<String>,
+        mode: Option<String>,
+        properties: Option<HashMap<String, String>>,
+    ) -> napi::Result<CreateNamespaceResponse> {
+        let mode_str = mode
+            .map(|m| match m.to_lowercase().as_str() {
+                "create" => Ok("Create".to_string()),
+                "exist_ok" => Ok("ExistOk".to_string()),
+                "overwrite" => Ok("Overwrite".to_string()),
+                _ => Err(napi::Error::from_reason(format!(
+                    "Invalid mode '{}': expected one of 'create', 'exist_ok', 'overwrite'",
+                    m
+                ))),
+            })
+            .transpose()?;
+        let req = CreateNamespaceRequest {
+            id: Some(namespace_path),
+            mode: mode_str,
+            properties,
+            ..Default::default()
+        };
+        let resp = self
+            .get_inner()?
+            .create_namespace(req)
+            .await
+            .default_error()?;
+        Ok(CreateNamespaceResponse {
+            properties: resp.properties,
+            transaction_id: resp.transaction_id,
+        })
+    }
+
+    #[napi(catch_unwind)]
+    /// Drop a namespace.
+    pub async fn drop_namespace(
+        &self,
+        namespace_path: Vec<String>,
+        mode: Option<String>,
+        behavior: Option<String>,
+    ) -> napi::Result<DropNamespaceResponse> {
+        let mode_str = mode
+            .map(|m| match m.to_lowercase().as_str() {
+                "skip" => Ok("Skip".to_string()),
+                "fail" => Ok("Fail".to_string()),
+                _ => Err(napi::Error::from_reason(format!(
+                    "Invalid mode '{}': expected one of 'skip', 'fail'",
+                    m
+                ))),
+            })
+            .transpose()?;
+        let behavior_str = behavior
+            .map(|b| match b.to_lowercase().as_str() {
+                "restrict" => Ok("Restrict".to_string()),
+                "cascade" => Ok("Cascade".to_string()),
+                _ => Err(napi::Error::from_reason(format!(
+                    "Invalid behavior '{}': expected one of 'restrict', 'cascade'",
+                    b
+                ))),
+            })
+            .transpose()?;
+        let req = DropNamespaceRequest {
+            id: Some(namespace_path),
+            mode: mode_str,
+            behavior: behavior_str,
+            ..Default::default()
+        };
+        let resp = self
+            .get_inner()?
+            .drop_namespace(req)
+            .await
+            .default_error()?;
+        Ok(DropNamespaceResponse {
+            properties: resp.properties,
+            transaction_id: resp.transaction_id,
+        })
     }
 }

nodejs/src/lib.rs

@@ -35,8 +35,15 @@ pub struct ConnectionOptions {
     pub read_consistency_interval: Option<f64>,
     /// (For LanceDB OSS only): configuration for object storage.
     ///
-    /// The available options are described at https://lancedb.com/docs/storage/
+    /// The available options are described at https://docs.lancedb.com/storage/
     pub storage_options: Option<HashMap<String, String>>,
+    /// (For LanceDB OSS only): use directory namespace manifests as the source
+    /// of truth for table metadata. Existing directory-listed root tables are
+    /// migrated into the manifest on access.
+    pub manifest_enabled: Option<bool>,
+    /// (For LanceDB OSS only): extra properties for the backing namespace
+    /// client used by manifest-enabled native connections.
+    pub namespace_client_properties: Option<HashMap<String, String>>,
     /// (For LanceDB OSS only): the session to use for this connection. Holds
     /// shared caches and other session-specific state.
     pub session: Option<session::Session>,
@@ -60,6 +67,26 @@ pub struct OpenTableOptions {
     pub storage_options: Option<HashMap<String, String>>,
 }
 
+#[napi(object)]
+#[derive(Debug)]
+pub struct ConnectNamespaceOptions {
+    /// The interval, in seconds, at which to check for updates to the table
+    /// 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 value for
+    /// eventual consistency.
+    pub read_consistency_interval: Option<f64>,
+    /// Configuration for object storage. The available options are described
+    /// at https://docs.lancedb.com/storage/
+    pub storage_options: Option<HashMap<String, String>>,
+    /// Extra properties for the backing namespace client.
+    pub namespace_client_properties: Option<HashMap<String, String>>,
+    /// The session to use for this connection. Holds shared caches and other
+    /// session-specific state.
+    pub session: Option<session::Session>,
+}
+
 #[napi_derive::module_init]
 fn init() {
     let env = Env::new()

nodejs/src/merge.rs

@@ -50,6 +50,13 @@ impl NativeMergeInsertBuilder {
         this
     }
 
+    #[napi]
+    pub fn skip_auto_cleanup(&self, skip: bool) -> Self {
+        let mut this = self.clone();
+        this.inner.skip_auto_cleanup(skip);
+        this
+    }
+
     #[napi(catch_unwind)]
     pub async fn execute(&self, buf: Buffer) -> napi::Result<MergeResult> {
         let data = ipc_file_to_batches(buf.to_vec())

nodejs/src/query.rs

@@ -3,6 +3,12 @@
 
 use std::sync::Arc;
 
+use arrow_array::{
+    Array, Float16Array as ArrowFloat16Array, Float32Array as ArrowFloat32Array,
+    Float64Array as ArrowFloat64Array, UInt8Array as ArrowUInt8Array,
+};
+use arrow_buffer::ScalarBuffer;
+use half::f16;
 use lancedb::index::scalar::{
     BooleanQuery, BoostQuery, FtsQuery, FullTextSearchQuery, MatchQuery, MultiMatchQuery, Occur,
     Operator, PhraseQuery,
@@ -24,6 +30,33 @@ use crate::rerankers::RerankHybridCallbackArgs;
 use crate::rerankers::Reranker;
 use crate::util::{parse_distance_type, schema_to_buffer};
 
+fn bytes_to_arrow_array(data: Uint8Array, dtype: String) -> napi::Result<Arc<dyn Array>> {
+    let buf = arrow_buffer::Buffer::from(data.to_vec());
+    let num_bytes = buf.len();
+    match dtype.as_str() {
+        "float16" => {
+            let scalar_buf = ScalarBuffer::<f16>::new(buf, 0, num_bytes / 2);
+            Ok(Arc::new(ArrowFloat16Array::new(scalar_buf, None)))
+        }
+        "float32" => {
+            let scalar_buf = ScalarBuffer::<f32>::new(buf, 0, num_bytes / 4);
+            Ok(Arc::new(ArrowFloat32Array::new(scalar_buf, None)))
+        }
+        "float64" => {
+            let scalar_buf = ScalarBuffer::<f64>::new(buf, 0, num_bytes / 8);
+            Ok(Arc::new(ArrowFloat64Array::new(scalar_buf, None)))
+        }
+        "uint8" => {
+            let scalar_buf = ScalarBuffer::<u8>::new(buf, 0, num_bytes);
+            Ok(Arc::new(ArrowUInt8Array::new(scalar_buf, None)))
+        }
+        _ => Err(napi::Error::from_reason(format!(
+            "Unsupported vector dtype: {}. Expected one of: float16, float32, float64, uint8",
+            dtype
+        ))),
+    }
+}
+
 #[napi]
 pub struct Query {
     inner: LanceDbQuery,
@@ -78,6 +111,13 @@ impl Query {
         Ok(VectorQuery { inner })
     }
 
+    #[napi]
+    pub fn nearest_to_raw(&mut self, data: Uint8Array, dtype: String) -> Result<VectorQuery> {
+        let array = bytes_to_arrow_array(data, dtype)?;
+        let inner = self.inner.clone().nearest_to(array).default_error()?;
+        Ok(VectorQuery { inner })
+    }
+
     #[napi]
     pub fn fast_search(&mut self) {
         self.inner = self.inner.clone().fast_search();
@@ -163,6 +203,13 @@ impl VectorQuery {
         Ok(())
     }
 
+    #[napi]
+    pub fn add_query_vector_raw(&mut self, data: Uint8Array, dtype: String) -> Result<()> {
+        let array = bytes_to_arrow_array(data, dtype)?;
+        self.inner = self.inner.clone().add_query_vector(array).default_error()?;
+        Ok(())
+    }
+
     #[napi]
     pub fn distance_type(&mut self, distance_type: String) -> napi::Result<()> {
         let distance_type = parse_distance_type(distance_type)?;

nodejs/src/remote.rs

@@ -92,6 +92,13 @@ pub struct ClientConfig {
     pub extra_headers: Option<HashMap<String, String>>,
     pub id_delimiter: Option<String>,
     pub tls_config: Option<TlsConfig>,
+    /// User identifier for tracking purposes.
+    ///
+    /// This is sent as the `x-lancedb-user-id` header in requests to LanceDB Cloud/Enterprise.
+    /// It can be set directly, or via the `LANCEDB_USER_ID` environment variable.
+    /// Alternatively, set `LANCEDB_USER_ID_ENV_KEY` to specify another environment
+    /// variable that contains the user ID value.
+    pub user_id: Option<String>,
 }
 
 impl From<TimeoutConfig> for lancedb::remote::TimeoutConfig {
@@ -145,6 +152,7 @@ impl From<ClientConfig> for lancedb::remote::ClientConfig {
             id_delimiter: config.id_delimiter,
             tls_config: config.tls_config.map(Into::into),
             header_provider: None, // the header provider is set separately later
+            user_id: config.user_id,
         }
     }
 }

nodejs/src/rerankers.rs

@@ -18,6 +18,7 @@ type RerankHybridFn = ThreadsafeFunction<
     RerankHybridCallbackArgs,
     Status,
     false,
+    true,
 >;
 
 /// Reranker implementation that "wraps" a NodeJS Reranker implementation.
@@ -32,7 +33,10 @@ impl Reranker {
     pub fn new(
         rerank_hybrid: Function<RerankHybridCallbackArgs, Promise<Buffer>>,
     ) -> napi::Result<Self> {
-        let rerank_hybrid = rerank_hybrid.build_threadsafe_function().build()?;
+        let rerank_hybrid = rerank_hybrid
+            .build_threadsafe_function()
+            .weak::<true>()
+            .build()?;
         Ok(Self { rerank_hybrid })
     }
 }

nodejs/src/table.rs

@@ -3,10 +3,10 @@
 
 use std::collections::HashMap;
 
-use lancedb::ipc::ipc_file_to_batches;
+use lancedb::ipc::{ipc_file_to_batches, ipc_file_to_schema};
 use lancedb::table::{
     AddDataMode, ColumnAlteration as LanceColumnAlteration, Duration, NewColumnTransform,
-    OptimizeAction, OptimizeOptions, Table as LanceDbTable,
+    OptimizeAction, OptimizeOptions, Table as LanceDbTable, WriteOptions,
 };
 use napi::bindgen_prelude::*;
 use napi_derive::napi;
@@ -68,7 +68,12 @@ impl Table {
     }
 
     #[napi(catch_unwind)]
-    pub async fn add(&self, buf: Buffer, mode: String) -> napi::Result<AddResult> {
+    pub async fn add(
+        &self,
+        buf: Buffer,
+        mode: String,
+        skip_auto_cleanup: Option<bool>,
+    ) -> napi::Result<AddResult> {
         let batches = ipc_file_to_batches(buf.to_vec())
             .map_err(|e| napi::Error::from_reason(format!("Failed to read IPC file: {}", e)))?;
         let batches = batches
@@ -92,6 +97,13 @@ impl Table {
             return Err(napi::Error::from_reason(format!("Invalid mode: {}", mode)));
         };
 
+        if skip_auto_cleanup.unwrap_or(false) {
+            op = op.write_options(WriteOptions {
+                skip_auto_cleanup: true,
+                ..Default::default()
+            });
+        }
+
         let res = op.execute().await.default_error()?;
         Ok(res.into())
     }
@@ -159,6 +171,14 @@ impl Table {
             .default_error()
     }
 
+    #[napi(catch_unwind)]
+    pub async fn prewarm_data(&self, columns: Option<Vec<String>>) -> napi::Result<()> {
+        self.inner_ref()?
+            .prewarm_data(columns)
+            .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());
@@ -279,6 +299,23 @@ impl Table {
         Ok(res.into())
     }
 
+    #[napi(catch_unwind)]
+    pub async fn add_columns_with_schema(
+        &self,
+        schema_buf: Buffer,
+    ) -> napi::Result<AddColumnsResult> {
+        let schema = ipc_file_to_schema(schema_buf.to_vec())
+            .map_err(|e| napi::Error::from_reason(format!("Failed to read IPC schema: {}", e)))?;
+
+        let transforms = NewColumnTransform::AllNulls(schema);
+        let res = self
+            .inner_ref()?
+            .add_columns(transforms, None)
+            .await
+            .default_error()?;
+        Ok(res.into())
+    }
+
     #[napi(catch_unwind)]
     pub async fn alter_columns(
         &self,

python/.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.30.0-beta.5"
+current_version = "0.31.0-beta.11"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

python/.gitignore

@@ -1,3 +1,5 @@
 # Test data created by some example tests
 data/
 _lancedb.pyd
+# macOS debug symbols bundle generated during build
+*.dSYM/

python/Cargo.toml

@@ -1,6 +1,7 @@
 [package]
 name = "lancedb-python"
-version = "0.30.0-beta.5"
+version = "0.31.0-beta.11"
+publish = false
 edition.workspace = true
 description = "Python bindings for LanceDB"
 license.workspace = true
@@ -14,7 +15,7 @@ name = "_lancedb"
 crate-type = ["cdylib"]
 
 [dependencies]
-arrow = { version = "57.2", features = ["pyarrow"] }
+arrow = { version = "58.0.0", features = ["pyarrow"] }
 async-trait = "0.1"
 bytes = "1"
 lancedb = { path = "../rust/lancedb", default-features = false }
@@ -23,8 +24,9 @@ lance-namespace.workspace = true
 lance-namespace-impls.workspace = true
 lance-io.workspace = true
 env_logger.workspace = true
-pyo3 = { version = "0.26", features = ["extension-module", "abi3-py39"] }
-pyo3-async-runtimes = { version = "0.26", features = [
+log.workspace = true
+pyo3 = { version = "0.28", features = ["extension-module", "abi3-py39"] }
+pyo3-async-runtimes = { version = "0.28", features = [
     "attributes",
     "tokio-runtime",
 ] }
@@ -33,10 +35,11 @@ futures.workspace = true
 serde = "1"
 serde_json = "1"
 snafu.workspace = true
-tokio = { version = "1.40", features = ["sync"] }
+tokio = { version = "1.40", features = ["sync", "rt-multi-thread"] }
+libc = "0.2"
 
 [build-dependencies]
-pyo3-build-config = { version = "0.26", features = [
+pyo3-build-config = { version = "0.28", features = [
     "extension-module",
     "abi3-py39",
 ] }

python/PYTHON_THIRD_PARTY_LICENSES.md

@@ -183,7 +183,6 @@
 | stack-data                     | 0.6.3           | MIT License                                                                                      | http://github.com/alexmojaki/stack_data                                                           |
 | sympy                          | 1.14.0          | BSD License                                                                                      | https://sympy.org                                                                                 |
 | tabulate                       | 0.9.0           | MIT License                                                                                      | https://github.com/astanin/python-tabulate                                                        |
-| tantivy                        | 0.25.1          | UNKNOWN                                                                                          | UNKNOWN                                                                                           |
 | threadpoolctl                  | 3.6.0           | BSD License                                                                                      | https://github.com/joblib/threadpoolctl                                                           |
 | timm                           | 1.0.24          | Apache Software License                                                                          | https://github.com/huggingface/pytorch-image-models                                               |
 | tinycss2                       | 1.4.0           | BSD License                                                                                      | https://www.courtbouillon.org/tinycss2                                                            |

python/pyproject.toml

@@ -45,7 +45,7 @@ repository = "https://github.com/lancedb/lancedb"
 
 [project.optional-dependencies]
 pylance = [
-    "pylance>=4.0.0b7",
+    "pylance>=5.0.0b5",
 ]
 tests = [
     "aiohttp>=3.9.0",
@@ -57,9 +57,8 @@ tests = [
     "duckdb>=0.9.0",
     "pytz>=2023.3",
     "polars>=0.19, <=1.3.0",
-    "tantivy>=0.20.0",
     "pyarrow-stubs>=16.0",
-    "pylance>=4.0.0b7",
+    "pylance>=5.0.0b5",
     "requests>=2.31.0",
     "datafusion>=52,<53",
 ]
@@ -83,7 +82,7 @@ embeddings = [
     "colpali-engine>=0.3.10",
     "huggingface_hub>=0.19.0",
     "InstructorEmbedding>=1.0.1",
-    "google.generativeai>=0.3.0",
+    "google-genai>=1.0.0",
     "boto3>=1.28.57",
     "awscli>=1.44.38",
     "botocore>=1.31.57",

python/python/lancedb/__init__.py

@@ -6,8 +6,7 @@ import importlib.metadata
 import os
 from concurrent.futures import ThreadPoolExecutor
 from datetime import timedelta
-from typing import Dict, Optional, Union, Any
-import warnings
+from typing import Dict, Optional, Union, Any, List
 
 __version__ = importlib.metadata.version("lancedb")
 
@@ -15,9 +14,9 @@ from ._lancedb import connect as lancedb_connect
 from .common import URI, sanitize_uri
 from urllib.parse import urlparse
 from .db import AsyncConnection, DBConnection, LanceDBConnection
-from .io import StorageOptionsProvider
 from .remote import ClientConfig
 from .remote.db import RemoteDBConnection
+from .expr import Expr, col, lit, func
 from .schema import vector
 from .table import AsyncTable, Table
 from ._lancedb import Session
@@ -63,7 +62,7 @@ def _check_s3_bucket_with_dots(
 
 
 def connect(
-    uri: URI,
+    uri: Optional[URI] = None,
     *,
     api_key: Optional[str] = None,
     region: str = "us-east-1",
@@ -73,14 +72,19 @@ def connect(
     client_config: Union[ClientConfig, Dict[str, Any], None] = None,
     storage_options: Optional[Dict[str, str]] = None,
     session: Optional[Session] = None,
+    manifest_enabled: bool = False,
+    namespace_client_impl: Optional[str] = None,
+    namespace_client_properties: Optional[Dict[str, str]] = None,
+    namespace_client_pushdown_operations: Optional[List[str]] = None,
     **kwargs: Any,
 ) -> DBConnection:
     """Connect to a LanceDB database.
 
     Parameters
     ----------
-    uri: str or Path
-        The uri of the database.
+    uri: str or Path, optional
+        The uri of the database. When ``namespace_client_impl`` is provided you may
+        omit ``uri`` and connect through a namespace client instead.
     api_key: str, optional
         If presented, connect to LanceDB cloud.
         Otherwise, connect to a database on file system or cloud storage.
@@ -106,13 +110,29 @@ def connect(
         default configuration is used.
     storage_options: dict, optional
         Additional options for the storage backend. See available options at
-        <https://lancedb.com/docs/storage/>
+        <https://docs.lancedb.com/storage/>
+    manifest_enabled : bool, default False
+        When true for local/native connections, use directory namespace
+        manifests as the source of truth for table metadata. Existing
+        directory-listed root tables are migrated into the manifest on access.
     session: Session, optional
         (For LanceDB OSS only)
         A session to use for this connection. Sessions allow you to configure
         cache sizes for index and metadata caches, which can significantly
         impact memory use and performance. They can also be re-used across
         multiple connections to share the same cache state.
+    namespace_client_impl : str, optional
+        When provided along with ``namespace_client_properties``, ``connect``
+        returns a namespace-backed connection by delegating to
+        :func:`connect_namespace`. The value identifies which namespace
+        implementation to load (e.g., ``"dir"`` or ``"rest"``).
+    namespace_client_properties : dict, optional
+        Configuration to pass to the namespace client implementation. Required
+        when ``namespace_client_impl`` is set.
+    namespace_client_pushdown_operations : list[str], optional
+        Only used when ``namespace_client_properties`` is provided. Forwards to
+        :func:`connect_namespace` to control which operations are executed on the
+        namespace service (e.g., ``["QueryTable", "CreateTable"]``).
 
     Examples
     --------
@@ -132,11 +152,48 @@ def connect(
     >>> db = lancedb.connect("db://my_database", api_key="ldb_...",
     ...                      client_config={"retry_config": {"retries": 5}})
 
+    Connect to a namespace-backed database:
+
+    >>> db = lancedb.connect(namespace_client_impl="dir",
+    ...                      namespace_client_properties={"root": "/tmp/ns"})
+
     Returns
     -------
     conn : DBConnection
         A connection to a LanceDB database.
     """
+    if namespace_client_impl is not None:
+        if namespace_client_properties is None:
+            raise ValueError(
+                "namespace_client_properties must be provided when "
+                "namespace_client_impl is set"
+            )
+        if kwargs:
+            raise ValueError(f"Unknown keyword arguments: {kwargs}")
+        return connect_namespace(
+            namespace_client_impl,
+            namespace_client_properties,
+            read_consistency_interval=read_consistency_interval,
+            storage_options=storage_options,
+            session=session,
+            namespace_client_pushdown_operations=namespace_client_pushdown_operations,
+        )
+
+    if namespace_client_properties is not None and not manifest_enabled:
+        raise ValueError(
+            "namespace_client_impl must be provided when using "
+            "namespace_client_properties unless manifest_enabled=True"
+        )
+
+    if namespace_client_pushdown_operations is not None:
+        raise ValueError(
+            "namespace_client_pushdown_operations is only valid when "
+            "connecting through a namespace"
+        )
+    if uri is None:
+        raise ValueError(
+            "uri is required when not connecting through a namespace client"
+        )
     if isinstance(uri, str) and uri.startswith("db://"):
         if api_key is None:
             api_key = os.environ.get("LANCEDB_API_KEY")
@@ -165,9 +222,92 @@ def connect(
         read_consistency_interval=read_consistency_interval,
         storage_options=storage_options,
         session=session,
+        manifest_enabled=manifest_enabled,
+        namespace_client_properties=namespace_client_properties,
     )
 
 
+WORKER_PROPERTY_PREFIX = "_lancedb_worker_"
+
+
+def _apply_worker_overrides(props: dict[str, str]) -> dict[str, str]:
+    """Apply worker property overrides.
+
+    Any key starting with ``_lancedb_worker_`` is extracted, the prefix
+    is stripped, and the resulting key-value pair is put back into the
+    map (overriding the existing value if present).  The original
+    prefixed key is removed.
+    """
+    worker_keys = [k for k in props if k.startswith(WORKER_PROPERTY_PREFIX)]
+    if not worker_keys:
+        return props
+    result = dict(props)
+    for key in worker_keys:
+        value = result.pop(key)
+        real_key = key[len(WORKER_PROPERTY_PREFIX) :]
+        result[real_key] = value
+    return result
+
+
+def deserialize_conn(
+    data: str,
+    *,
+    for_worker: bool = False,
+) -> DBConnection:
+    """Reconstruct a DBConnection from a serialized string.
+
+    The string must have been produced by
+    :meth:`DBConnection.serialize`.
+
+    Parameters
+    ----------
+    data : str
+        String produced by ``serialize()``.
+    for_worker : bool, default False
+        When ``True``, any namespace client property whose key starts
+        with ``_lancedb_worker_`` has that prefix stripped and the
+        value overrides the corresponding property.  For example,
+        ``_lancedb_worker_uri`` replaces ``uri``.
+
+    Returns
+    -------
+    DBConnection
+        A new connection matching the serialized state.
+    """
+    import json
+
+    parsed = json.loads(data)
+    connection_type = parsed.get("connection_type")
+
+    rci_secs = parsed.get("read_consistency_interval_seconds")
+    rci = timedelta(seconds=rci_secs) if rci_secs is not None else None
+    storage_options = parsed.get("storage_options")
+
+    if connection_type == "namespace":
+        props = dict(parsed.get("namespace_client_properties") or {})
+        if for_worker:
+            props = _apply_worker_overrides(props)
+        return connect_namespace(
+            namespace_client_impl=parsed["namespace_client_impl"],
+            namespace_client_properties=props,
+            read_consistency_interval=rci,
+            storage_options=storage_options,
+            namespace_client_pushdown_operations=parsed.get(
+                "namespace_client_pushdown_operations"
+            ),
+        )
+    elif connection_type == "local":
+        return LanceDBConnection(
+            parsed["uri"],
+            read_consistency_interval=rci,
+            storage_options=storage_options,
+            manifest_enabled=parsed.get("manifest_enabled", False),
+            namespace_client_properties=parsed.get("namespace_client_properties"),
+        )
+    else:
+        raise ValueError(f"Unknown connection_type: {connection_type}")
+
+
 async def connect_async(
     uri: URI,
     *,
@@ -178,6 +318,8 @@ async def connect_async(
     client_config: Optional[Union[ClientConfig, Dict[str, Any]]] = None,
     storage_options: Optional[Dict[str, str]] = None,
     session: Optional[Session] = None,
+    manifest_enabled: bool = False,
+    namespace_client_properties: Optional[Dict[str, str]] = None,
 ) -> AsyncConnection:
     """Connect to a LanceDB database.
 
@@ -210,13 +352,20 @@ async def connect_async(
         default configuration is used.
     storage_options: dict, optional
         Additional options for the storage backend. See available options at
-        <https://lancedb.com/docs/storage/>
+        <https://docs.lancedb.com/storage/>
     session: Session, optional
         (For LanceDB OSS only)
         A session to use for this connection. Sessions allow you to configure
         cache sizes for index and metadata caches, which can significantly
         impact memory use and performance. They can also be re-used across
         multiple connections to share the same cache state.
+    manifest_enabled : bool, default False
+        When true for local/native connections, use directory namespace
+        manifests as the source of truth for table metadata. Existing
+        directory-listed root tables are migrated into the manifest on access.
+    namespace_client_properties : dict, optional
+        Additional directory namespace client properties to use with
+        ``manifest_enabled=True``.
 
     Examples
     --------
@@ -259,6 +408,8 @@ async def connect_async(
             client_config,
             storage_options,
             session,
+            manifest_enabled,
+            namespace_client_properties,
         )
     )
 
@@ -271,6 +422,10 @@ __all__ = [
     "AsyncConnection",
     "AsyncLanceNamespaceDBConnection",
     "AsyncTable",
+    "col",
+    "Expr",
+    "func",
+    "lit",
     "URI",
     "sanitize_uri",
     "vector",
@@ -279,17 +434,6 @@ __all__ = [
     "LanceNamespaceDBConnection",
     "RemoteDBConnection",
     "Session",
-    "StorageOptionsProvider",
     "Table",
     "__version__",
 ]
-
-
-def __warn_on_fork():
-    warnings.warn(
-        "lance is not fork-safe. If you are using multiprocessing, use spawn instead.",
-    )
-
-
-if hasattr(os, "register_at_fork"):
-    os.register_at_fork(before=__warn_on_fork)  # type: ignore[attr-defined]

python/python/lancedb/_lancedb.pyi

@@ -12,9 +12,9 @@ from .index import (
     LabelList,
     HnswPq,
     HnswSq,
+    HnswFlat,
     FTS,
 )
-from .io import StorageOptionsProvider
 from lance_namespace import (
     ListNamespacesResponse,
     CreateNamespaceResponse,
@@ -26,6 +26,33 @@ from .remote import ClientConfig
 
 IvfHnswPq: type[HnswPq] = HnswPq
 IvfHnswSq: type[HnswSq] = HnswSq
+IvfHnswFlat: type[HnswFlat] = HnswFlat
+
+class PyExpr:
+    """A type-safe DataFusion expression node (Rust-side handle)."""
+
+    def eq(self, other: "PyExpr") -> "PyExpr": ...
+    def ne(self, other: "PyExpr") -> "PyExpr": ...
+    def lt(self, other: "PyExpr") -> "PyExpr": ...
+    def lte(self, other: "PyExpr") -> "PyExpr": ...
+    def gt(self, other: "PyExpr") -> "PyExpr": ...
+    def gte(self, other: "PyExpr") -> "PyExpr": ...
+    def and_(self, other: "PyExpr") -> "PyExpr": ...
+    def or_(self, other: "PyExpr") -> "PyExpr": ...
+    def not_(self) -> "PyExpr": ...
+    def add(self, other: "PyExpr") -> "PyExpr": ...
+    def sub(self, other: "PyExpr") -> "PyExpr": ...
+    def mul(self, other: "PyExpr") -> "PyExpr": ...
+    def div(self, other: "PyExpr") -> "PyExpr": ...
+    def lower(self) -> "PyExpr": ...
+    def upper(self) -> "PyExpr": ...
+    def contains(self, substr: "PyExpr") -> "PyExpr": ...
+    def cast(self, data_type: pa.DataType) -> "PyExpr": ...
+    def to_sql(self) -> str: ...
+
+def expr_col(name: str) -> PyExpr: ...
+def expr_lit(value: Union[bool, int, float, str]) -> PyExpr: ...
+def expr_func(name: str, args: List[PyExpr]) -> PyExpr: ...
 
 class Session:
     def __init__(
@@ -46,35 +73,35 @@ class Connection(object):
     async def close(self): ...
     async def list_namespaces(
         self,
-        namespace: Optional[List[str]] = None,
+        namespace_path: Optional[List[str]] = None,
         page_token: Optional[str] = None,
         limit: Optional[int] = None,
     ) -> ListNamespacesResponse: ...
     async def create_namespace(
         self,
-        namespace: List[str],
+        namespace_path: List[str],
         mode: Optional[str] = None,
         properties: Optional[Dict[str, str]] = None,
     ) -> CreateNamespaceResponse: ...
     async def drop_namespace(
         self,
-        namespace: List[str],
+        namespace_path: List[str],
         mode: Optional[str] = None,
         behavior: Optional[str] = None,
     ) -> DropNamespaceResponse: ...
     async def describe_namespace(
         self,
-        namespace: List[str],
+        namespace_path: List[str],
     ) -> DescribeNamespaceResponse: ...
     async def list_tables(
         self,
-        namespace: Optional[List[str]] = None,
+        namespace_path: Optional[List[str]] = None,
         page_token: Optional[str] = None,
         limit: Optional[int] = None,
     ) -> ListTablesResponse: ...
     async def table_names(
         self,
-        namespace: Optional[List[str]],
+        namespace_path: Optional[List[str]],
         start_after: Optional[str],
         limit: Optional[int],
     ) -> list[str]: ...  # Deprecated: Use list_tables instead
@@ -83,9 +110,8 @@ class Connection(object):
         name: str,
         mode: str,
         data: pa.RecordBatchReader,
-        namespace: Optional[List[str]] = None,
+        namespace_path: Optional[List[str]] = None,
         storage_options: Optional[Dict[str, str]] = None,
-        storage_options_provider: Optional[StorageOptionsProvider] = None,
         location: Optional[str] = None,
     ) -> Table: ...
     async def create_empty_table(
@@ -93,17 +119,15 @@ class Connection(object):
         name: str,
         mode: str,
         schema: pa.Schema,
-        namespace: Optional[List[str]] = None,
+        namespace_path: Optional[List[str]] = None,
         storage_options: Optional[Dict[str, str]] = None,
-        storage_options_provider: Optional[StorageOptionsProvider] = None,
         location: Optional[str] = None,
     ) -> Table: ...
     async def open_table(
         self,
         name: str,
-        namespace: Optional[List[str]] = None,
+        namespace_path: Optional[List[str]] = None,
         storage_options: Optional[Dict[str, str]] = None,
-        storage_options_provider: Optional[StorageOptionsProvider] = None,
         index_cache_size: Optional[int] = None,
         location: Optional[str] = None,
     ) -> Table: ...
@@ -111,7 +135,7 @@ class Connection(object):
         self,
         target_table_name: str,
         source_uri: str,
-        target_namespace: Optional[List[str]] = None,
+        target_namespace_path: Optional[List[str]] = None,
         source_version: Optional[int] = None,
         source_tag: Optional[str] = None,
         is_shallow: bool = True,
@@ -120,13 +144,18 @@ class Connection(object):
         self,
         cur_name: str,
         new_name: str,
-        cur_namespace: Optional[List[str]] = None,
-        new_namespace: Optional[List[str]] = None,
+        cur_namespace_path: Optional[List[str]] = None,
+        new_namespace_path: Optional[List[str]] = None,
     ) -> None: ...
     async def drop_table(
-        self, name: str, namespace: Optional[List[str]] = None
+        self, name: str, namespace_path: Optional[List[str]] = None
     ) -> None: ...
-    async def drop_all_tables(self, namespace: Optional[List[str]] = None) -> None: ...
+    async def drop_all_tables(
+        self, namespace_path: Optional[List[str]] = None
+    ) -> None: ...
+    async def namespace_client_config(
+        self,
+    ) -> Dict[str, Any]: ...
 
 class Table:
     def name(self) -> str: ...
@@ -135,7 +164,10 @@ class Table:
     def close(self) -> None: ...
     async def schema(self) -> pa.Schema: ...
     async def add(
-        self, data: pa.RecordBatchReader, mode: Literal["append", "overwrite"]
+        self,
+        data: pa.RecordBatchReader,
+        mode: Literal["append", "overwrite"],
+        progress: Optional[Any] = None,
     ) -> AddResult: ...
     async def update(
         self, updates: Dict[str, str], where: Optional[str]
@@ -150,6 +182,7 @@ class Table:
             IvfPq,
             HnswPq,
             HnswSq,
+            HnswFlat,
             BTree,
             Bitmap,
             LabelList,
@@ -166,6 +199,8 @@ class Table:
     async def checkout(self, version: Union[int, str]): ...
     async def checkout_latest(self): ...
     async def restore(self, version: Optional[Union[int, str]] = None): ...
+    async def prewarm_index(self, index_name: str) -> None: ...
+    async def prewarm_data(self, columns: Optional[List[str]] = None) -> None: ...
     async def list_indices(self) -> list[IndexConfig]: ...
     async def delete(self, filter: str) -> DeleteResult: ...
     async def add_columns(self, columns: list[tuple[str, str]]) -> AddColumnsResult: ...
@@ -210,6 +245,8 @@ async def connect(
     client_config: Optional[Union[ClientConfig, Dict[str, Any]]],
     storage_options: Optional[Dict[str, str]],
     session: Optional[Session],
+    manifest_enabled: bool = False,
+    namespace_client_properties: Optional[Dict[str, str]] = None,
 ) -> Connection: ...
 
 class RecordBatchStream:
@@ -220,7 +257,9 @@ class RecordBatchStream:
 
 class Query:
     def where(self, filter: str): ...
-    def select(self, columns: Tuple[str, str]): ...
+    def where_expr(self, expr: PyExpr): ...
+    def select(self, columns: List[Tuple[str, str]]): ...
+    def select_expr(self, columns: List[Tuple[str, PyExpr]]): ...
     def select_columns(self, columns: List[str]): ...
     def limit(self, limit: int): ...
     def offset(self, offset: int): ...
@@ -246,7 +285,9 @@ class TakeQuery:
 
 class FTSQuery:
     def where(self, filter: str): ...
-    def select(self, columns: List[str]): ...
+    def where_expr(self, expr: PyExpr): ...
+    def select(self, columns: List[Tuple[str, str]]): ...
+    def select_expr(self, columns: List[Tuple[str, PyExpr]]): ...
     def limit(self, limit: int): ...
     def offset(self, offset: int): ...
     def fast_search(self): ...
@@ -265,7 +306,9 @@ class VectorQuery:
     async def output_schema(self) -> pa.Schema: ...
     async def execute(self) -> RecordBatchStream: ...
     def where(self, filter: str): ...
-    def select(self, columns: List[str]): ...
+    def where_expr(self, expr: PyExpr): ...
+    def select(self, columns: List[Tuple[str, str]]): ...
+    def select_expr(self, columns: List[Tuple[str, PyExpr]]): ...
     def select_with_projection(self, columns: Tuple[str, str]): ...
     def limit(self, limit: int): ...
     def offset(self, offset: int): ...
@@ -282,7 +325,9 @@ class VectorQuery:
 
 class HybridQuery:
     def where(self, filter: str): ...
-    def select(self, columns: List[str]): ...
+    def where_expr(self, expr: PyExpr): ...
+    def select(self, columns: List[Tuple[str, str]]): ...
+    def select_expr(self, columns: List[Tuple[str, PyExpr]]): ...
     def limit(self, limit: int): ...
     def offset(self, offset: int): ...
     def fast_search(self): ...
@@ -400,7 +445,7 @@ class AsyncPermutationBuilder:
     async def execute(self) -> Table: ...
 
 def async_permutation_builder(
-    table: Table, dest_table_name: str
+    table: Table,
 ) -> AsyncPermutationBuilder: ...
 def fts_query_to_json(query: Any) -> str: ...
 

python/python/lancedb/background_loop.py

@@ -2,7 +2,9 @@
 # SPDX-FileCopyrightText: Copyright The LanceDB Authors
 
 import asyncio
+import os
 import threading
+import warnings
 
 
 class BackgroundEventLoop:
@@ -13,6 +15,9 @@ class BackgroundEventLoop:
     """
 
     def __init__(self):
+        self._start()
+
+    def _start(self):
         self.loop = asyncio.new_event_loop()
         self.thread = threading.Thread(
             target=self.loop.run_forever,
@@ -31,3 +36,30 @@ class BackgroundEventLoop:
 
 
 LOOP = BackgroundEventLoop()
+
+_FORK_WARNED = False
+
+
+def _reset_after_fork():
+    # Threads do not survive fork(), so the asyncio loop in LOOP.thread is
+    # dead in the child. Re-initialize the singleton in place so existing
+    # `from .background_loop import LOOP` references in other modules see
+    # the new state. The Rust-side tokio runtime is reset analogously by a
+    # pthread_atfork hook installed in the _lancedb extension.
+    LOOP._start()
+    global _FORK_WARNED
+    if not _FORK_WARNED:
+        _FORK_WARNED = True
+        warnings.warn(
+            "lancedb fork support is experimental: the internal async "
+            "runtime has been reset in the forked child, but a small chance "
+            "of deadlock remains if other state was mid-operation at fork "
+            "time. The 'forkserver' or 'spawn' multiprocessing start method "
+            "is likely a safer alternative.",
+            RuntimeWarning,
+            stacklevel=2,
+        )
+
+
+if hasattr(os, "register_at_fork"):
+    os.register_at_fork(after_in_child=_reset_after_fork)

python/python/lancedb/common.py

@@ -96,7 +96,7 @@ def data_to_reader(
             f"Unknown data type {type(data)}. "
             "Supported types: list of dicts, pandas DataFrame, polars DataFrame, "
             "pyarrow Table/RecordBatch, or Pydantic models. "
-            "See https://lancedb.com/docs/tables/ for examples."
+            "See https://docs.lancedb.com/tables/ for examples."
         )