chore: update lance dependency to v7.0.0-beta.1

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.30.0-beta.1"
+current_version = "0.28.0-beta.11"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

.github/dependabot.yml

@@ -11,11 +11,6 @@ updates:
     schedule:
       interval: weekly
     open-pull-requests-limit: 10
-    # Only update Cargo.lock, never widen/raise the version requirements in
-    # Cargo.toml. The goal is keeping the lockfile (and the binaries we ship)
-    # current on security fixes, not forcing our library's consumers onto
-    # newer minimum versions.
-    versioning-strategy: lockfile-only
     groups:
       rust-minor-patch:
         update-types:

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

@@ -45,9 +45,7 @@ jobs:
       - name: Set up Node.js
         uses: actions/setup-node@v4
         with:
-          # pnpm 11 (used by the nodejs install step below) requires
-          # Node >= 22.13; use 24 since 22 hits EOL in October.
-          node-version: 24
+          node-version: 20
 
       - name: Install Codex CLI
         run: npm install -g @openai/codex
@@ -81,14 +79,10 @@ 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
-          pnpm install --frozen-lockfile
+          npm ci
 
       - name: Configure git user
         run: |
@@ -143,7 +137,7 @@ 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 && pnpm build && pnpm test -- --testNamePattern='<test_name>'"
+               - For TypeScript test failures: Run "cd nodejs && npm run build && npm 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/java-publish.yml

@@ -43,7 +43,7 @@ jobs:
           server-username: SONATYPE_USER
           server-password: SONATYPE_TOKEN
           gpg-private-key: ${{ secrets.GPG_PRIVATE_KEY }}
-          gpg-passphrase: MAVEN_GPG_PASSPHRASE
+          gpg-passphrase: ${{ secrets.GPG_PASSPHRASE }}
       - name: Set git config
         run: |
           git config --global user.email "dev+gha@lancedb.com"
@@ -58,11 +58,10 @@ jobs:
           echo "use-agent" >> ~/.gnupg/gpg.conf
           echo "pinentry-mode loopback" >> ~/.gnupg/gpg.conf
           export GPG_TTY=$(tty)
-          ./mvnw --batch-mode -DskipTests -DpushChanges=false deploy -pl lancedb-core -am -P deploy-to-ossrh
+          ./mvnw --batch-mode -DskipTests -DpushChanges=false -Dgpg.passphrase=${{ secrets.GPG_PASSPHRASE }} 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/nodejs.yml

@@ -42,17 +42,11 @@ jobs:
       with:
         fetch-depth: 0
         lfs: true
-    - uses: pnpm/action-setup@v4
-      with:
-        version: 11.1.1
     - uses: actions/setup-node@v4
       with:
-        # 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
+        node-version: 20
+        cache: 'npm'
+        cache-dependency-path: nodejs/package-lock.json
     - uses: actions-rust-lang/setup-rust-toolchain@v1
       with:
         components: rustfmt, clippy
@@ -67,13 +61,11 @@ jobs:
       run: cargo clippy --profile ci --all --all-features -- -D warnings
     - name: Lint Typescript
       run: |
-        pnpm install --frozen-lockfile
-        pnpm lint-ci
+        npm ci
+        npm run lint-ci
     - name: Lint examples
       working-directory: nodejs/examples
-      # 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
+      run: npm ci && npm run lint-ci
   linux:
     name: Linux (NodeJS ${{ matrix.node-version }})
     timeout-minutes: 30
@@ -90,18 +82,14 @@ jobs:
       with:
         fetch-depth: 0
         lfs: true
-    - uses: pnpm/action-setup@v4
-      with:
-        version: 11.1.1
     - uses: actions/setup-node@v4
-      name: Setup Node.js 24 for build
+      name: Setup Node.js 20 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
+        # @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
     - uses: Swatinem/rust-cache@v2
     - name: Install dependencies
       run: |
@@ -109,58 +97,48 @@ jobs:
         sudo apt install -y protobuf-compiler libssl-dev
     - name: Build
       run: |
-        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
+        npm ci --include=optional
+        npm run build:debug -- --profile ci
     - 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"
-        # 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
+      run: npm run test
+    - name: Setup examples
+      working-directory: nodejs/examples
+      run: npm ci
     - 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
-        npx jest --testEnvironment jest-environment-node-single-context --verbose
+        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
   macos:
     timeout-minutes: 30
-    # macos-15 ships a newer linker; the older macos-14 linker fails to insert
-    # branch islands when the debug cdylib's __text section exceeds the 128 MB
-    # AArch64 B/BL branch range.
-    runs-on: "macos-15"
+    runs-on: "macos-14"
     defaults:
       run:
         shell: bash
@@ -170,28 +148,20 @@ jobs:
       with:
         fetch-depth: 0
         lfs: true
-    - uses: pnpm/action-setup@v4
-      with:
-        version: 11.1.1
     - 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
-    - uses: dtolnay/rust-toolchain@stable
+        node-version: 20
+        cache: 'npm'
+        cache-dependency-path: nodejs/package-lock.json
     - uses: Swatinem/rust-cache@v2
     - name: Install dependencies
       run: |
         brew install protobuf
     - name: Build
       run: |
-        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
+        npm ci --include=optional
+        npm run build:debug -- --profile ci
+        npm run tsc
     - name: Test
       run: |
-        pnpm test
+        npm run test

.github/workflows/npm-publish.yml

@@ -171,18 +171,13 @@ 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:
-          # 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
+          node-version: 20
+          cache: npm
+          cache-dependency-path: nodejs/package-lock.json
       - name: Install
         uses: dtolnay/rust-toolchain@stable
         if: ${{ !matrix.settings.docker }}
@@ -200,7 +195,7 @@ jobs:
             target/
           key: nodejs-${{ matrix.settings.target }}-cargo-${{ matrix.settings.host }}
       - name: Install dependencies
-        run: pnpm install --frozen-lockfile
+        run: npm ci
       - name: Install Zig
         uses: mlugg/setup-zig@v2
         if: ${{ contains(matrix.settings.target, 'musl') }}
@@ -253,7 +248,7 @@ jobs:
       # one to do the upload.
       - name: Make generic artifacts
         if: ${{ matrix.settings.target == 'aarch64-apple-darwin' }}
-        run: pnpm tsc
+        run: npm run tsc
       - name: Upload Generic Artifacts
         if: ${{ matrix.settings.target == 'aarch64-apple-darwin' }}
         uses: actions/upload-artifact@v4
@@ -288,24 +283,14 @@ 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.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
+      - name: Setup node
         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:
@@ -326,9 +311,7 @@ jobs:
       - name: Move built files
         run: cp dist/native.d.ts dist/native.js dist/*.node lancedb/
       - name: Test bindings
-        # Invoke jest directly because pnpm 11 itself requires Node 22+
-        # while the matrix tests on older Node versions.
-        run: npx jest --verbose
+        run: npm test
   publish:
     name: Publish
     runs-on: ubuntu-latest
@@ -340,19 +323,15 @@ 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: pnpm
-          cache-dependency-path: nodejs/pnpm-lock.yaml
+          cache: npm
+          cache-dependency-path: nodejs/package-lock.json
           registry-url: "https://registry.npmjs.org"
       - name: Install dependencies
-        run: pnpm install --frozen-lockfile
+        run: npm ci
       - uses: actions/download-artifact@v4
         with:
           name: nodejs-dist
@@ -372,7 +351,7 @@ jobs:
       - name: Display structure of downloaded files
         run: find dist && find nodejs-artifacts
       - name: Move artifacts
-        run: pnpm exec napi artifacts -d nodejs-artifacts
+        run: npx napi artifacts -d nodejs-artifacts
       - name: List packages
         run: find npm
       - name: Publish

.github/workflows/python.yml

@@ -205,7 +205,7 @@ jobs:
       - name: Delete wheels
         run: rm -rf target/wheels
   pydantic1x:
-    timeout-minutes: 60
+    timeout-minutes: 30
     runs-on: "ubuntu-24.04"
     defaults:
       run:

.github/workflows/rust.yml

@@ -233,26 +233,6 @@ jobs:
           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
-          # aws-runtime/sigv4/credential-types/types and the aws-smithy-*
-          # crates bumped their MSRV to 1.91.1 in late 2026; pin to the last
-          # 1.91.0-compatible versions. The order matters — each downgrade
-          # only succeeds once everything that still pins it at a higher
-          # version has itself been downgraded.
-          cargo update -p aws-runtime --precise 1.5.12
-          cargo update -p aws-types --precise 1.3.9
-          cargo update -p aws-sigv4 --precise 1.3.5
-          cargo update -p aws-credential-types --precise 1.2.8
-          cargo update -p aws-smithy-checksums --precise 0.63.9
-          cargo update -p aws-smithy-runtime --precise 1.9.3
-          cargo update -p aws-smithy-http --precise 0.62.4
-          cargo update -p aws-smithy-eventstream --precise 0.60.12
-          cargo update -p aws-smithy-http-client --precise 1.1.3
-          cargo update -p aws-smithy-observability --precise 0.1.4
-          cargo update -p aws-smithy-query --precise 0.60.8
-          cargo update -p aws-smithy-runtime-api --precise 1.9.1
-          cargo update -p aws-smithy-async --precise 1.2.6
-          cargo update -p aws-smithy-types --precise 1.3.5
-          cargo update -p aws-smithy-xml --precise 0.60.11
           cargo update -p home --precise 0.5.9
       - name: cargo +${{ matrix.msrv }} check
         env:

AGENTS.md

@@ -17,30 +17,9 @@ Common commands:
 * Run tests: `cargo test --quiet --features remote --tests`
 * Run specific test: `cargo test --quiet --features remote -p <package_name> --test <test_name>`
 * Lint: `cargo clippy --quiet --features remote --tests --examples`
-* Format Rust: `cargo fmt --all`
-* Format Python: `ruff format .`
-* Lint Python: `ruff check .`
-* Bootstrap Python dev env: `cd python && uv run --extra tests --extra dev maturin develop --extras tests,dev`
-* Run Python tests: `cd python && uv run --extra tests pytest python/tests -vv --durations=10 -m "not slow and not s3_test"`
-* Run specific Python test: `cd python && uv run --extra tests pytest python/tests/<test_file>.py::<test_name> -q`
+* Format: `cargo fmt --all`
 
-For Python validation, prefer the uv-managed environment declared by `python/uv.lock`.
-Do not treat system `python`, global `pytest`, or missing editable-install errors as
-final blockers; bootstrap or enter the uv environment instead. If `lancedb._lancedb`
-is missing or stale, or if Rust/PyO3 binding code changed, rebuild the Python
-extension with the bootstrap command above before running tests.
-
-Before committing changes, run formatting for every language you touched. At minimum:
-
-* Rust changes: run `cargo fmt --all`.
-* Python changes: run `ruff format .` and `ruff check .` from the repository root,
-  and run targeted tests through `cd python && uv run ...`.
-* TypeScript changes: run the relevant `npm`/`pnpm` lint, format, build, and docs commands in `nodejs`.
-
-Before creating a PR, make sure the PR title follows Conventional Commits, such as
-`fix: support nested field paths in native index creation` or
-`feat(python): add dataset multiprocessing support`. The semantic-release check uses the
-PR title and body as the merge commit message, so a non-conventional PR title will fail CI.
+Before committing changes, run formatting.
 
 ## Coding tips
 

Cargo.toml

@@ -13,20 +13,20 @@ categories = ["database-implementations"]
 rust-version = "1.91.0"
 
 [workspace.dependencies]
-lance = { "version" = "=7.0.0-beta.13", default-features = false, "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-core = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-datagen = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-file = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-io = { "version" = "=7.0.0-beta.13", default-features = false, "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-index = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-linalg = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-namespace = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-namespace-impls = { "version" = "=7.0.0-beta.13", default-features = false, "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-table = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-testing = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-datafusion = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-encoding = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
-lance-arrow = { "version" = "=7.0.0-beta.13", "tag" = "v7.0.0-beta.13", "git" = "https://github.com/lance-format/lance.git" }
+lance = { "version" = "=7.0.0-beta.1", default-features = false, "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-core = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-datagen = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-file = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-io = { "version" = "=7.0.0-beta.1", default-features = false, "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-index = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-linalg = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-namespace = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-namespace-impls = { "version" = "=7.0.0-beta.1", default-features = false, "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-table = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-testing = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-datafusion = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-encoding = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
+lance-arrow = { "version" = "=7.0.0-beta.1", "tag" = "v7.0.0-beta.1", "git" = "https://github.com/lance-format/lance.git" }
 ahash = "0.8"
 # Note that this one does not include pyarrow
 arrow = { version = "58.0.0", optional = false }
@@ -54,7 +54,7 @@ half = { "version" = "2.7.1", default-features = false, features = [
 futures = "0"
 log = "0.4"
 moka = { version = "0.12", features = ["future"] }
-object_store = "0.13.2"
+object_store = "0.12.0"
 pin-project = "1.0.7"
 rand = "0.9"
 snafu = "0.8"

deny.toml

@@ -51,18 +51,6 @@ ignore = [
     # 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.
@@ -80,17 +68,11 @@ ignore = [
     # 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-pemfile: unmaintained. Reached from two separate chains:
+    # rustls-native-certs 0.6 (via hyper-rustls 0.24) and object_store 0.12.
+    # Both upstream dependencies need to move before we can drop it.
+    # https://rustsec.org/advisories/RUSTSEC-2025-0134
+    { id = "RUSTSEC-2025-0134", reason = "transitive via rustls-native-certs/object_store; waiting on upstream migration" },
 
     # 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
@@ -107,12 +89,6 @@ ignore = [
     # 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" },
 ]
 
 # ---------------------------------------------------------------------------

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>0.30.0-beta.1</version>
+    <version>0.28.0-beta.11</version>
 </dependency>
 ```
 

docs/src/js/_media/CONTRIBUTING.md

@@ -12,22 +12,20 @@ Typescript.
 * `src/`: Rust bindings source code
 * `lancedb/`: Typescript package source code
 * `__test__/`: Unit tests
-* `examples/`: A pnpm package with the examples shown in the documentation
+* `examples/`: An npm 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 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)
+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)
 
 Initial setup:
 
 ```shell
-pnpm install
+npm install
 ```
 
 ### Commit Hooks
@@ -41,38 +39,38 @@ pre-commit install
 
 ## Development
 
-Most common development commands can be run using the pnpm scripts.
+Most common development commands can be run using the npm scripts.
 
 Build the package
 
 ```shell
-pnpm install
-pnpm build
+npm install
+npm run build
 ```
 
 Lint:
 
 ```shell
-pnpm lint
+npm run lint
 ```
 
 Format and fix lints:
 
 ```shell
-pnpm lint-fix
+npm run lint-fix
 ```
 
 Run tests:
 
 ```shell
-pnpm test
+npm test
 ```
 
 To run a single test:
 
 ```shell
 # Single file: table.test.ts
-pnpm test -- table.test.ts
+npm test -- table.test.ts
 # Single test: 'merge insert' in table.test.ts
-pnpm test -- table.test.ts --testNamePattern=merge\ insert
+npm test -- table.test.ts --testNamePattern=merge\ insert
 ```

docs/src/js/classes/Connection.md

@@ -148,33 +148,6 @@ Creates a new empty Table
 
 ***
 
-### createNamespace()
-
-```ts
-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)
@@ -257,29 +230,6 @@ 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
@@ -313,36 +263,6 @@ 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
@@ -379,36 +299,6 @@ 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
@@ -437,39 +327,6 @@ Open a table in the database.
 
 ***
 
-### renameTable()
-
-```ts
-abstract renameTable(
-   currentName,
-   newName,
-   options?): Promise<void>
-```
-
-Rename a table.
-
-Currently only supported by LanceDB Cloud. Local OSS connections and
-namespace-backed connections (via [connectNamespace](../functions/connectNamespace.md)) reject with
-a "not supported" error.
-
-#### Parameters
-
-* **currentName**: `string`
-    The current name of the table.
-
-* **newName**: `string`
-    The new name for the table.
-
-* **options?**: [`RenameTableOptions`](../interfaces/RenameTableOptions.md)
-    Optional namespace paths. When
-    `newNamespacePath` is omitted the table stays in `namespacePath`.
-
-#### Returns
-
-`Promise`&lt;`void`&gt;
-
-***
-
 ### tableNames()
 
 #### tableNames(options)

docs/src/js/classes/Query.md

@@ -343,30 +343,6 @@ This is useful for pagination.
 
 ***
 
-### orderBy()
-
-```ts
-orderBy(ordering): this
-```
-
-Sort the results by the specified column(s).
-
-#### Parameters
-
-* **ordering**: [`ColumnOrdering`](../interfaces/ColumnOrdering.md) \| [`ColumnOrdering`](../interfaces/ColumnOrdering.md)[]
-
-#### Returns
-
-`this`
-
-This query builder.
-
-#### Inherited from
-
-`StandardQueryBase.orderBy`
-
-***
-
 ### outputSchema()
 
 ```ts

docs/src/js/classes/Scannable.md

@@ -1,173 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / Scannable
-
-# Class: Scannable
-
-A data source that can be scanned as a stream of Arrow `RecordBatch`es.
-
-`Scannable` wraps the schema + optional row count + rescannable flag and
-a callback that yields batches one at a time. It is passed to consumers
-(e.g. `Table.add`, `createTable`, `mergeInsert` — follow-up work) that
-need to pull data without materializing the full dataset in JS memory.
-
-Batches cross the JS↔Rust boundary as Arrow IPC Stream messages; a fresh
-writer serializes each batch, and the Rust side decodes it with
-`arrow_ipc::reader::StreamReader`. One batch is in flight at a time.
-
-## Properties
-
-### numRows
-
-```ts
-readonly numRows: null | number;
-```
-
-***
-
-### rescannable
-
-```ts
-readonly rescannable: boolean;
-```
-
-***
-
-### schema
-
-```ts
-readonly schema: Schema<any>;
-```
-
-## Methods
-
-### fromFactory()
-
-```ts
-static fromFactory(
-   schema,
-   factory,
-   opts): Promise<Scannable>
-```
-
-Build a Scannable from an explicit schema and a factory that returns a
-fresh batch iterator on each call.
-
-The factory is invoked once per scan. Each iterator yields
-`RecordBatch`es matching the declared schema. Use this when you need
-direct control over the pull loop — for example, to wrap a streaming
-source whose batches are produced lazily.
-
-#### Parameters
-
-* **schema**: `Schema`&lt;`any`&gt;
-    The Arrow schema of the produced batches.
-
-* **factory**
-    Called at the start of each scan to produce a batch
-    iterator. Must be idempotent when `rescannable` is true.
-
-* **opts**: [`ScannableOptions`](../interfaces/ScannableOptions.md) = `{}`
-    Optional hints. `rescannable` defaults to `true`; set to
-    `false` if calling `factory()` twice would not reproduce the same data.
-
-#### Returns
-
-`Promise`&lt;[`Scannable`](Scannable.md)&gt;
-
-***
-
-### fromIterable()
-
-```ts
-static fromIterable(
-   schema,
-   iter,
-   opts): Promise<Scannable>
-```
-
-Build a Scannable from an iterable of `RecordBatch`es. `rescannable`
-defaults to `false`. Pass an explicit schema so the consumer can
-validate before any batch is pulled.
-
-`opts.rescannable: true` is honest for replayable iterables (Arrays,
-Sets, or custom iterables whose `[Symbol.iterator]()` returns a fresh
-iterator each call). It is rejected for one-shot iterables (generators,
-async generators, or already-an-iterator inputs) because their
-`[Symbol.iterator]()` returns the same exhausted object on the second
-scan. For replayable sources outside this shape, use
-`fromFactory(schema, () => createIter(), { rescannable: true })`.
-
-Note: when `opts.rescannable` is `true`, the constructor calls
-`[Symbol.iterator]()` once on the input to perform the structural check.
-
-#### Parameters
-
-* **schema**: `Schema`&lt;`any`&gt;
-
-* **iter**: `Iterable`&lt;`RecordBatch`&lt;`any`&gt;&gt; \| `AsyncIterable`&lt;`RecordBatch`&lt;`any`&gt;&gt;
-
-* **opts**: [`ScannableOptions`](../interfaces/ScannableOptions.md) = `{}`
-
-#### Returns
-
-`Promise`&lt;[`Scannable`](Scannable.md)&gt;
-
-***
-
-### fromRecordBatchReader()
-
-```ts
-static fromRecordBatchReader(reader, opts): Promise<Scannable>
-```
-
-Build a Scannable from an Arrow `RecordBatchReader`. A reader can only
-be consumed once; `rescannable` defaults to `false`.
-
-The reader must already be opened (via `.open()`) so its `.schema` is
-populated. `RecordBatchReader.from(...)` returns an unopened reader.
-
-`opts.rescannable: true` is rejected because `RecordBatchReader` is a
-self-iterator (its `[Symbol.iterator]()` returns itself), and this
-constructor does not call `reader.reset()` between scans, so a second
-scan would always see an exhausted reader. For genuinely replayable
-sources, use
-`fromFactory(schema, () => openReader(), { rescannable: true })`,
-which mints a fresh reader on each scan.
-
-#### Parameters
-
-* **reader**: `RecordBatchReader`&lt;`any`&gt;
-
-* **opts**: [`ScannableOptions`](../interfaces/ScannableOptions.md) = `{}`
-
-#### Returns
-
-`Promise`&lt;[`Scannable`](Scannable.md)&gt;
-
-***
-
-### fromTable()
-
-```ts
-static fromTable(table, opts): Promise<Scannable>
-```
-
-Build a Scannable from an in-memory Arrow `Table`. Always rescannable;
-the table's batches are replayed on each scan.
-
-The table's row count is authoritative: `opts.numRows` must either be
-omitted or equal to `table.numRows`. `opts.rescannable` of `false` is
-rejected because in-memory Tables are always rescannable.
-
-#### Parameters
-
-* **table**: `Table`&lt;`any`&gt;
-
-* **opts**: [`ScannableOptions`](../interfaces/ScannableOptions.md) = `{}`
-
-#### Returns
-
-`Promise`&lt;[`Scannable`](Scannable.md)&gt;

docs/src/js/classes/Table.md

@@ -501,34 +501,6 @@ 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
@@ -690,74 +662,6 @@ of the given query
 
 ***
 
-### setLsmWriteSpec()
-
-```ts
-abstract setLsmWriteSpec(spec): Promise<void>
-```
-
-Install an [LsmWriteSpec](../interfaces/LsmWriteSpec.md) on this table, selecting Lance's MemWAL
-LSM-style write path for future `mergeInsert` calls.
-
-`LsmWriteSpec` chooses one of three sharding strategies via `specType`:
-
-- `"bucket"` — hash-bucket writes by the single-column unenforced primary
-  key (`column` and `numBuckets` required).
-- `"identity"` — shard by the raw value of a scalar `column`.
-- `"unsharded"` — route every write to a single shard.
-
-All variants require the table to have an unenforced primary key
-([Table#setUnenforcedPrimaryKey](Table.md#setunenforcedprimarykey)); bucket sharding additionally
-requires it to be the single column being bucketed.
-
-#### Parameters
-
-* **spec**: [`LsmWriteSpec`](../interfaces/LsmWriteSpec.md)
-    The sharding spec to install.
-
-#### Returns
-
-`Promise`&lt;`void`&gt;
-
-#### Example
-
-```ts
-await table.setUnenforcedPrimaryKey("id");
-await table.setLsmWriteSpec({
-  specType: "bucket",
-  column: "id",
-  numBuckets: 16,
-  maintainedIndexes: ["id_idx"],
-});
-```
-
-***
-
-### setUnenforcedPrimaryKey()
-
-```ts
-abstract setUnenforcedPrimaryKey(columns): Promise<void>
-```
-
-Set the unenforced primary key for this table to a single column.
-
-"Unenforced" means LanceDB does not check uniqueness on writes; the
-column is recorded in the schema as the primary key for use by features
-such as `merge_insert`. Only single-column primary keys are supported,
-and the key cannot be changed once set.
-
-#### Parameters
-
-* **columns**: `string` \| `string`[]
-    The primary key column. A one-element
-    array is also accepted; passing more than one column is rejected.
-
-#### Returns
-
-`Promise`&lt;`void`&gt;
-
-***
-
 ### stats()
 
 ```ts
@@ -861,23 +765,6 @@ Return the table as an arrow table
 
 ***
 
-### unsetLsmWriteSpec()
-
-```ts
-abstract unsetLsmWriteSpec(): Promise<void>
-```
-
-Remove the [LsmWriteSpec](../interfaces/LsmWriteSpec.md) from this table, reverting to the standard
-`mergeInsert` write path.
-
-Errors if no spec is currently set.
-
-#### Returns
-
-`Promise`&lt;`void`&gt;
-
-***
-
 ### update()
 
 #### update(opts)

docs/src/js/classes/VectorQuery.md

@@ -498,30 +498,6 @@ This is useful for pagination.
 
 ***
 
-### orderBy()
-
-```ts
-orderBy(ordering): this
-```
-
-Sort the results by the specified column(s).
-
-#### Parameters
-
-* **ordering**: [`ColumnOrdering`](../interfaces/ColumnOrdering.md) \| [`ColumnOrdering`](../interfaces/ColumnOrdering.md)[]
-
-#### Returns
-
-`this`
-
-This query builder.
-
-#### Inherited from
-
-`StandardQueryBase.orderBy`
-
-***
-
 ### outputSchema()
 
 ```ts

docs/src/js/functions/connectNamespace.md

@@ -1,131 +0,0 @@
-[**@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

@@ -32,7 +32,6 @@
 - [PhraseQuery](classes/PhraseQuery.md)
 - [Query](classes/Query.md)
 - [QueryBase](classes/QueryBase.md)
-- [Scannable](classes/Scannable.md)
 - [Session](classes/Session.md)
 - [StaticHeaderProvider](classes/StaticHeaderProvider.md)
 - [Table](classes/Table.md)
@@ -51,19 +50,11 @@
 - [AlterColumnsResult](interfaces/AlterColumnsResult.md)
 - [ClientConfig](interfaces/ClientConfig.md)
 - [ColumnAlteration](interfaces/ColumnAlteration.md)
-- [ColumnOrdering](interfaces/ColumnOrdering.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)
@@ -78,19 +69,13 @@
 - [IvfFlatOptions](interfaces/IvfFlatOptions.md)
 - [IvfPqOptions](interfaces/IvfPqOptions.md)
 - [IvfRqOptions](interfaces/IvfRqOptions.md)
-- [ListNamespacesOptions](interfaces/ListNamespacesOptions.md)
-- [ListNamespacesResponse](interfaces/ListNamespacesResponse.md)
-- [LsmWriteSpec](interfaces/LsmWriteSpec.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)
-- [RenameTableOptions](interfaces/RenameTableOptions.md)
-- [RestNamespaceConfig](interfaces/RestNamespaceConfig.md)
 - [RetryConfig](interfaces/RetryConfig.md)
-- [ScannableOptions](interfaces/ScannableOptions.md)
 - [ShuffleOptions](interfaces/ShuffleOptions.md)
 - [SplitCalculatedOptions](interfaces/SplitCalculatedOptions.md)
 - [SplitHashOptions](interfaces/SplitHashOptions.md)
@@ -105,7 +90,6 @@
 - [UpdateResult](interfaces/UpdateResult.md)
 - [Version](interfaces/Version.md)
 - [WriteExecutionOptions](interfaces/WriteExecutionOptions.md)
-- [WriteProgress](interfaces/WriteProgress.md)
 
 ## Type Aliases
 
@@ -123,7 +107,6 @@
 
 - [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/AddDataOptions.md

@@ -19,39 +19,3 @@ mode: "append" | "overwrite";
 If "append" (the default) then the new data will be added to the table
 
 If "overwrite" then the new data will replace the existing data in the table.
-
-***
-
-### progress()
-
-```ts
-progress: (progress) => void;
-```
-
-Optional callback invoked periodically with write progress.
-
-The callback is fired once per batch written and once more with
-`done: true` when the write completes. Calls are dispatched
-asynchronously to the JS event loop and never block the write — a slow
-callback will queue events rather than back-pressure the writer.
-
-Errors thrown from the callback are logged with `console.warn` and
-swallowed — they do not abort the write.
-
-#### Parameters
-
-* **progress**: [`WriteProgress`](WriteProgress.md)
-
-#### Returns
-
-`void`
-
-#### Example
-
-```ts
-await table.add(data, {
-  progress: (p) => {
-    console.log(`${p.outputRows}/${p.totalRows ?? "?"} rows`);
-  },
-});
-```

docs/src/js/interfaces/ColumnOrdering.md

@@ -1,31 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / ColumnOrdering
-
-# Interface: ColumnOrdering
-
-## Properties
-
-### ascending?
-
-```ts
-optional ascending: boolean;
-```
-
-***
-
-### columnName
-
-```ts
-columnName: string;
-```
-
-***
-
-### nullsFirst?
-
-```ts
-optional nullsFirst: boolean;
-```

docs/src/js/interfaces/ConnectNamespaceOptions.md

@@ -1,54 +0,0 @@
-[**@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/CreateNamespaceOptions.md

@@ -1,27 +0,0 @@
-[**@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

@@ -1,23 +0,0 @@
-[**@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/DescribeNamespaceResponse.md

@@ -1,15 +0,0 @@
-[**@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

@@ -1,47 +0,0 @@
-[**@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

@@ -1,27 +0,0 @@
-[**@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

@@ -1,23 +0,0 @@
-[**@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

@@ -1,27 +0,0 @@
-[**@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

@@ -1,23 +0,0 @@
-[**@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/LsmWriteSpec.md

@@ -1,64 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / LsmWriteSpec
-
-# Interface: LsmWriteSpec
-
-Specification selecting Lance's MemWAL LSM-style write path for
-`mergeInsert`.
-
-`specType` is `"bucket"`, `"identity"`, or `"unsharded"`. For `"bucket"`,
-`column` and `numBuckets` are required; for `"identity"`, `column` is
-required.
-
-## Properties
-
-### column?
-
-```ts
-optional column: string;
-```
-
-Bucket and identity variants: the sharding column.
-
-***
-
-### maintainedIndexes?
-
-```ts
-optional maintainedIndexes: string[];
-```
-
-Names of indexes the MemWAL should keep up to date during writes.
-
-***
-
-### numBuckets?
-
-```ts
-optional numBuckets: number;
-```
-
-Bucket variant: the number of buckets, in `[1, 1024]`.
-
-***
-
-### specType
-
-```ts
-specType: "bucket" | "identity" | "unsharded";
-```
-
-One of `"bucket"`, `"identity"`, or `"unsharded"`.
-
-***
-
-### writerConfigDefaults?
-
-```ts
-optional writerConfigDefaults: Record<string, string>;
-```
-
-Default `ShardWriter` configuration recorded in the MemWAL index.

docs/src/js/interfaces/RenameTableOptions.md

@@ -1,29 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / RenameTableOptions
-
-# Interface: RenameTableOptions
-
-## Properties
-
-### namespacePath?
-
-```ts
-optional namespacePath: string[];
-```
-
-The namespace path of the table being renamed. Defaults to the root
-namespace (`[]`) when omitted.
-
-***
-
-### newNamespacePath?
-
-```ts
-optional newNamespacePath: string[];
-```
-
-The namespace path to move the table to as part of the rename. When
-omitted the table stays in `namespacePath`.

docs/src/js/interfaces/RestNamespaceConfig.md

@@ -1,47 +0,0 @@
-[**@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/interfaces/ScannableOptions.md

@@ -1,29 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / ScannableOptions
-
-# Interface: ScannableOptions
-
-## Properties
-
-### numRows?
-
-```ts
-optional numRows: number;
-```
-
-Hint about the number of rows. Not validated against the stream.
-
-***
-
-### rescannable?
-
-```ts
-optional rescannable: boolean;
-```
-
-Whether the source can be scanned more than once. Defaults to `true` for
-`fromTable` / `fromFactory` and `false` for `fromIterable` /
-`fromRecordBatchReader`.

docs/src/js/interfaces/WriteProgress.md

@@ -1,84 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / WriteProgress
-
-# Interface: WriteProgress
-
-Progress snapshot for a write operation, delivered to the `progress`
-callback passed to [Table.add](../classes/Table.md#add).
-
-## Properties
-
-### activeTasks
-
-```ts
-activeTasks: number;
-```
-
-Number of parallel write tasks currently in flight.
-
-***
-
-### done
-
-```ts
-done: boolean;
-```
-
-`true` for the final callback; `false` otherwise.
-
-***
-
-### elapsedSeconds
-
-```ts
-elapsedSeconds: number;
-```
-
-Wall-clock seconds since the write started.
-
-***
-
-### outputBytes
-
-```ts
-outputBytes: number;
-```
-
-Number of bytes written so far.
-
-***
-
-### outputRows
-
-```ts
-outputRows: number;
-```
-
-Number of rows written so far.
-
-***
-
-### totalRows?
-
-```ts
-optional totalRows: number;
-```
-
-Total rows expected, when the input source reports it.
-
-Always set on the final callback (the one with `done: true`), falling
-back to the actual number of rows written when the source could not
-report a row count up front.
-
-***
-
-### totalTasks
-
-```ts
-totalTasks: number;
-```
-
-Total number of parallel write tasks (the write parallelism).

docs/src/python/python.md

@@ -94,11 +94,11 @@ of raw SQL strings with [where][lancedb.query.LanceQueryBuilder.where] and
 
 ## Full text search
 
-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.create_index
 
-::: lancedb.index.FTS
+::: lancedb.fts.populate_index
+
+::: lancedb.fts.search_index
 
 ## Utilities
 
@@ -166,12 +166,6 @@ lists the indices that LanceDb supports.
 
 ::: lancedb.index.IvfFlat
 
-::: lancedb.index.IvfSq
-
-::: lancedb.index.IvfRq
-
-::: lancedb.index.HnswFlat
-
 ::: lancedb.table.IndexStatistics
 
 ## Querying (Asynchronous)

java/lancedb-core/pom.xml

@@ -8,7 +8,7 @@
     <parent>
       <groupId>com.lancedb</groupId>
       <artifactId>lancedb-parent</artifactId>
-      <version>0.30.0-beta.1</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>0.30.0-beta.1</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>7.0.0-beta.13</lance-core.version>
+        <lance-core.version>7.0.0-beta.1</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: `pnpm build`.
+Whenever you change the Rust code, you will need to recompile: `npm run build`.
 
 Common commands:
-* Build: `pnpm build`
-* Lint: `pnpm lint`
-* Fix lints: `pnpm lint-fix`
-* Test: `pnpm test`
-* Run single test file: `pnpm test __test__/arrow.test.ts`
+* 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`

nodejs/CONTRIBUTING.md

@@ -12,22 +12,20 @@ Typescript.
 * `src/`: Rust bindings source code
 * `lancedb/`: Typescript package source code
 * `__test__/`: Unit tests
-* `examples/`: A pnpm package with the examples shown in the documentation
+* `examples/`: An npm 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 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)
+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)
 
 Initial setup:
 
 ```shell
-pnpm install
+npm install
 ```
 
 ### Commit Hooks
@@ -41,38 +39,38 @@ pre-commit install
 
 ## Development
 
-Most common development commands can be run using the pnpm scripts.
+Most common development commands can be run using the npm scripts.
 
 Build the package
 
 ```shell
-pnpm install
-pnpm build
+npm install
+npm run build
 ```
 
 Lint:
 
 ```shell
-pnpm lint
+npm run lint
 ```
 
 Format and fix lints:
 
 ```shell
-pnpm lint-fix
+npm run lint-fix
 ```
 
 Run tests:
 
 ```shell
-pnpm test
+npm test
 ```
 
 To run a single test:
 
 ```shell
 # Single file: table.test.ts
-pnpm test -- table.test.ts
+npm test -- table.test.ts
 # Single test: 'merge insert' in table.test.ts
-pnpm test -- table.test.ts --testNamePattern=merge\ insert
+npm test -- table.test.ts --testNamePattern=merge\ insert
 ```

nodejs/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.30.0-beta.1"
+version = "0.28.0-beta.11"
 publish = false
 license.workspace = true
 description.workspace = true
@@ -22,7 +22,6 @@ 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"

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, connectNamespace } from "../lancedb";
+import { Connection, Table, connect } from "../lancedb";
 import { LocalTable } from "../lancedb/table";
 
 describe("when connecting", () => {
@@ -47,14 +47,6 @@ describe("given a connection", () => {
     await db.close();
     expect(db.isOpen()).toBe(false);
     await expect(db.tableNames()).rejects.toThrow("Connection is closed");
-    await expect(db.renameTable("a", "b")).rejects.toThrow(
-      "Connection is closed",
-    );
-  });
-
-  it("should report renameTable as unsupported on an OSS connection", async () => {
-    await db.createTable("a", [{ id: 1 }]);
-    await expect(db.renameTable("a", "b")).rejects.toThrow(/not supported/);
   });
   it("should be able to create a table from an object arg `createTable(options)`, or args `createTable(name, data, options)`", async () => {
     let tbl = await db.createTable("test", [{ id: 1 }, { id: 2 }]);
@@ -314,186 +306,3 @@ 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__/query.test.ts

@@ -109,209 +109,3 @@ describe("Query outputSchema", () => {
     expect(schema.fields.length).toBe(3);
   });
 });
-
-describe("Query orderBy", () => {
-  let tmpDir: tmp.DirResult;
-  let table: Table;
-
-  beforeEach(async () => {
-    tmpDir = tmp.dirSync({ unsafeCleanup: true });
-    const db = await connect(tmpDir.name);
-
-    // Create table with numeric data for sorting
-    const schema = new Schema([
-      new Field("id", new Int64(), true),
-      new Field("score", new Float32(), true),
-      new Field("name", new Utf8(), true),
-    ]);
-
-    const data = makeArrowTable(
-      [
-        { id: 1n, score: 3.5, name: "charlie" },
-        { id: 2n, score: 1.2, name: "alice" },
-        { id: 3n, score: 2.8, name: "bob" },
-        { id: 4n, score: 0.5, name: "david" },
-        { id: 5n, score: 4.1, name: "eve" },
-      ],
-      { schema },
-    );
-    table = await db.createTable("test", data);
-  });
-
-  afterEach(() => {
-    tmpDir.removeCallback();
-  });
-
-  it("should sort by single column ascending", async () => {
-    const results = await table
-      .query()
-      .orderBy({ columnName: "score", ascending: true, nullsFirst: false })
-      .toArray();
-
-    expect(results.length).toBe(5);
-    // Verify ascending order
-    expect(results[0].score).toBeCloseTo(0.5, 0.001);
-    expect(results[1].score).toBeCloseTo(1.2, 0.001);
-    expect(results[2].score).toBeCloseTo(2.8, 0.001);
-    expect(results[3].score).toBeCloseTo(3.5, 0.001);
-    expect(results[4].score).toBeCloseTo(4.1, 0.001);
-  });
-
-  it("should sort by single column descending", async () => {
-    const results = await table
-      .query()
-      .orderBy({ columnName: "score", ascending: false, nullsFirst: false })
-      .toArray();
-
-    expect(results.length).toBe(5);
-    // Verify descending order
-    expect(results[0].score).toBeCloseTo(4.1, 0.001);
-    expect(results[1].score).toBeCloseTo(3.5, 0.001);
-    expect(results[2].score).toBeCloseTo(2.8, 0.001);
-    expect(results[3].score).toBeCloseTo(1.2, 0.001);
-    expect(results[4].score).toBeCloseTo(0.5, 0.001);
-  });
-
-  it("should use ascending as default direction", async () => {
-    const results = await table
-      .query()
-      .orderBy({ columnName: "score" })
-      .toArray();
-
-    expect(results.length).toBe(5);
-    // Verify ascending order (default)
-    expect(results[0].score).toBeCloseTo(0.5, 0.001);
-    expect(results[1].score).toBeCloseTo(1.2, 0.001);
-    expect(results[2].score).toBeCloseTo(2.8, 0.001);
-    expect(results[3].score).toBeCloseTo(3.5, 0.001);
-    expect(results[4].score).toBeCloseTo(4.1, 0.001);
-  });
-
-  it("should sort by string column", async () => {
-    const results = await table
-      .query()
-      .orderBy({ columnName: "name" })
-      .toArray();
-
-    expect(results.length).toBe(5);
-    // Verify alphabetical order
-    expect(results[0].name).toBe("alice");
-    expect(results[1].name).toBe("bob");
-    expect(results[2].name).toBe("charlie");
-    expect(results[3].name).toBe("david");
-    expect(results[4].name).toBe("eve");
-  });
-
-  it("should support method chaining with where", async () => {
-    const results = await table
-      .query()
-      .where("score > 2.0")
-      .orderBy({ columnName: "score" })
-      .toArray();
-    expect(results.length).toBe(3);
-    // Verify filtered and sorted
-    expect(results[0].score).toBeCloseTo(2.8, 0.001);
-    expect(results[1].score).toBeCloseTo(3.5, 0.001);
-    expect(results[2].score).toBeCloseTo(4.1, 0.001);
-  });
-
-  it("should support method chaining with limit", async () => {
-    const results = await table
-      .query()
-      .orderBy({ columnName: "score", ascending: false })
-      .limit(3)
-      .toArray();
-
-    expect(results.length).toBe(3);
-    // Verify top 3 in descending order
-    expect(results[0].score).toBeCloseTo(4.1, 0.001);
-    expect(results[1].score).toBeCloseTo(3.5, 0.001);
-    expect(results[2].score).toBeCloseTo(2.8, 0.001);
-  });
-
-  it("should support method chaining with offset", async () => {
-    const results = await table
-      .query()
-      .orderBy({ columnName: "score" })
-      .offset(2)
-      .limit(2)
-      .toArray();
-
-    expect(results.length).toBe(2);
-    // Verify results skip first 2 and take next 2
-    expect(results[0].score).toBeCloseTo(2.8, 0.001);
-    expect(results[1].score).toBeCloseTo(3.5, 0.001);
-  });
-
-  it("should support method chaining with select", async () => {
-    const results = await table
-      .query()
-      .orderBy({ columnName: "name" })
-      .select(["name", "score"])
-      .toArray();
-
-    expect(results.length).toBe(5);
-    // Verify only selected columns are present
-    expect(Object.keys(results[0])).toEqual(["name", "score"]);
-    expect(Object.keys(results[4])).toEqual(["name", "score"]);
-    // Verify sorted by name
-    expect(results[0].name).toBe("alice");
-    expect(results[4].name).toBe("eve");
-  });
-
-  it("should support complex method chaining", async () => {
-    const results = await table
-      .query()
-      .where("score > 1.0")
-      .orderBy({ columnName: "score", ascending: false })
-      .limit(3)
-      .select(["id", "score", "name"])
-      .toArray();
-
-    expect(results.length).toBe(3);
-    // Verify filtered, sorted, limited, and projected
-    expect(results[0].score).toBeCloseTo(4.1, 0.001);
-    expect(results[1].score).toBeCloseTo(3.5, 0.001);
-    expect(results[2].score).toBeCloseTo(2.8, 0.001);
-    expect(Object.keys(results[0])).toEqual(["id", "score", "name"]);
-  });
-
-  it("should support multi-column ordering and null placement", async () => {
-    const schema = new Schema([
-      new Field("group", new Int64(), true),
-      new Field("score", new Float32(), true),
-      new Field("name", new Utf8(), true),
-    ]);
-
-    const data = makeArrowTable(
-      [
-        { group: 1n, score: null, name: "z" },
-        { group: 1n, score: 1.0, name: "b" },
-        { group: 1n, score: 1.0, name: "a" },
-        { group: 2n, score: 0.5, name: "c" },
-      ],
-      { schema },
-    );
-    const nullTable = await (await connect(tmpDir.name)).createTable(
-      "test_multi_order",
-      data,
-      { mode: "overwrite" },
-    );
-
-    const results = await nullTable
-      .query()
-      .orderBy([
-        { columnName: "group", ascending: true, nullsFirst: false },
-        { columnName: "score", ascending: true, nullsFirst: true },
-        { columnName: "name", ascending: true, nullsFirst: false },
-      ])
-      .toArray();
-
-    expect(results.map((r) => [r.group, r.score, r.name])).toEqual([
-      [1n, null, "z"],
-      [1n, 1.0, "a"],
-      [1n, 1.0, "b"],
-      [2n, 0.5, "c"],
-    ]);
-  });
-});

nodejs/__test__/remote.test.ts

@@ -617,68 +617,4 @@ describe("remote connection", () => {
       );
     });
   });
-
-  describe("renameTable", () => {
-    async function captureRenameRequest(
-      call: (db: Connection) => Promise<void>,
-    ): Promise<{ url: string; body: Record<string, unknown> }> {
-      let captured: { url: string; body: Record<string, unknown> } | undefined;
-      await withMockDatabase((req, res) => {
-        let raw = "";
-        req.on("data", (chunk) => {
-          raw += chunk;
-        });
-        req.on("end", () => {
-          captured = {
-            url: req.url ?? "",
-            body: raw ? JSON.parse(raw) : {},
-          };
-          res.writeHead(200, { "Content-Type": "application/json" }).end("");
-        });
-      }, call);
-      if (!captured) {
-        throw new Error("mock server never saw a request");
-      }
-      return captured;
-    }
-
-    it("sends rename request for a table in the root namespace", async () => {
-      const { url, body } = await captureRenameRequest(async (db) => {
-        await db.renameTable("table1", "table2");
-      });
-      expect(url).toBe("/v1/table/table1/rename/");
-      // biome-ignore lint/style/useNamingConvention: snake_case mandated by the server wire format
-      expect(body).toEqual({ new_table_name: "table2" });
-    });
-
-    it("omits new_namespace when only the current namespace is supplied", async () => {
-      // Safe-default check: passing namespacePath alone must not send
-      // `new_namespace`, so the server keeps the table in its current
-      // namespace instead of silently moving it to root.
-      const { url, body } = await captureRenameRequest(async (db) => {
-        await db.renameTable("table1", "table2", {
-          namespacePath: ["ns1"],
-        });
-      });
-      expect(url).toBe("/v1/table/ns1$table1/rename/");
-      // biome-ignore lint/style/useNamingConvention: snake_case mandated by the server wire format
-      expect(body).toEqual({ new_table_name: "table2" });
-    });
-
-    it("includes new_namespace in the body for a cross-namespace rename", async () => {
-      const { url, body } = await captureRenameRequest(async (db) => {
-        await db.renameTable("table1", "table2", {
-          namespacePath: ["ns1"],
-          newNamespacePath: ["ns2"],
-        });
-      });
-      expect(url).toBe("/v1/table/ns1$table1/rename/");
-      expect(body).toEqual({
-        // biome-ignore lint/style/useNamingConvention: snake_case mandated by the server wire format
-        new_table_name: "table2",
-        // biome-ignore lint/style/useNamingConvention: snake_case mandated by the server wire format
-        new_namespace: ["ns2"],
-      });
-    });
-  });
 });

nodejs/__test__/scannable.test.ts

@@ -1,438 +0,0 @@
-// SPDX-License-Identifier: Apache-2.0
-// SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-import {
-  Field,
-  Float16,
-  Int32,
-  type RecordBatch,
-  RecordBatchReader,
-  Schema,
-  tableToIPC,
-} from "apache-arrow";
-import { makeArrowTable, makeEmptyTable } from "../lancedb/arrow";
-import { Scannable } from "../lancedb/scannable";
-
-function makeTable() {
-  return makeArrowTable(
-    [
-      { id: 1, name: "a" },
-      { id: 2, name: "b" },
-      { id: 3, name: "c" },
-    ],
-    { vectorColumns: {} },
-  );
-}
-
-async function makeReader(): Promise<RecordBatchReader> {
-  // `RecordBatchReader.from()` returns an unopened reader; `.schema` is only
-  // populated after `.open()`. Opening sync readers is synchronous.
-  const reader = RecordBatchReader.from(tableToIPC(makeTable()));
-  return reader.open() as RecordBatchReader;
-}
-
-describe("Scannable", () => {
-  describe("fromTable", () => {
-    test("reflects schema, numRows, and defaults rescannable=true", async () => {
-      const table = makeTable();
-      const scannable = await Scannable.fromTable(table);
-
-      expect(scannable.schema).toBe(table.schema);
-      expect(scannable.numRows).toBe(table.numRows);
-      expect(scannable.rescannable).toBe(true);
-    });
-
-    test("throws when opts.numRows does not match table.numRows", async () => {
-      await expect(
-        Scannable.fromTable(makeTable(), { numRows: 42 }),
-      ).rejects.toThrow(/does not match table\.numRows/);
-    });
-
-    test("throws when opts.rescannable is false", async () => {
-      await expect(
-        Scannable.fromTable(makeTable(), { rescannable: false }),
-      ).rejects.toThrow(/always rescannable/);
-    });
-  });
-
-  describe("fromRecordBatchReader", () => {
-    test("reflects schema and defaults numRows=null, rescannable=false", async () => {
-      const reader = await makeReader();
-      const scannable = await Scannable.fromRecordBatchReader(reader);
-
-      expect(scannable.schema).toBe(reader.schema);
-      expect(scannable.numRows).toBeNull();
-      expect(scannable.rescannable).toBe(false);
-    });
-
-    test("honors numRows override", async () => {
-      const scannable = await Scannable.fromRecordBatchReader(
-        await makeReader(),
-        { numRows: 3 },
-      );
-
-      expect(scannable.numRows).toBe(3);
-      expect(scannable.rescannable).toBe(false);
-    });
-
-    test("rescannable: false explicit does not throw", async () => {
-      const reader = await makeReader();
-      const scannable = await Scannable.fromRecordBatchReader(reader, {
-        rescannable: false,
-      });
-      expect(scannable.rescannable).toBe(false);
-    });
-
-    test("throws when opts.rescannable is true", async () => {
-      const reader = await makeReader();
-      await expect(
-        Scannable.fromRecordBatchReader(reader, { rescannable: true }),
-      ).rejects.toThrow(/does not accept rescannable/);
-    });
-
-    test("throws when opts.rescannable is true even alongside numRows", async () => {
-      const reader = await makeReader();
-      await expect(
-        Scannable.fromRecordBatchReader(reader, {
-          numRows: 3,
-          rescannable: true,
-        }),
-      ).rejects.toThrow(/does not accept rescannable/);
-    });
-  });
-
-  describe("fromIterable", () => {
-    test("accepts a sync iterable of batches", async () => {
-      const table = makeTable();
-      const scannable = await Scannable.fromIterable(
-        table.schema,
-        table.batches,
-      );
-
-      expect(scannable.schema).toBe(table.schema);
-      expect(scannable.numRows).toBeNull();
-      expect(scannable.rescannable).toBe(false);
-    });
-
-    test("accepts an async iterable of batches", async () => {
-      const table = makeTable();
-      async function* generator(): AsyncGenerator<RecordBatch> {
-        for (const batch of table.batches) {
-          yield batch;
-        }
-      }
-
-      const scannable = await Scannable.fromIterable(table.schema, generator());
-      expect(scannable.schema).toBe(table.schema);
-      expect(scannable.rescannable).toBe(false);
-    });
-
-    describe("rescannable: true detection", () => {
-      // Replayable inputs: [Symbol.iterator]() / [Symbol.asyncIterator]()
-      // returns a fresh iterator each call. Must NOT throw.
-
-      test("Array passes (fresh ArrayIterator each call)", async () => {
-        const table = makeTable();
-        const scannable = await Scannable.fromIterable(
-          table.schema,
-          table.batches,
-          { rescannable: true },
-        );
-        expect(scannable.rescannable).toBe(true);
-      });
-
-      test("Set passes (fresh SetIterator each call)", async () => {
-        const table = makeTable();
-        const set = new Set<RecordBatch>(table.batches);
-        const scannable = await Scannable.fromIterable(table.schema, set, {
-          rescannable: true,
-        });
-        expect(scannable.rescannable).toBe(true);
-      });
-
-      test("custom Iterable returning a fresh iterator passes", async () => {
-        const table = makeTable();
-        const replayable: Iterable<RecordBatch> = {
-          [Symbol.iterator]() {
-            return table.batches[Symbol.iterator]();
-          },
-        };
-        const scannable = await Scannable.fromIterable(
-          table.schema,
-          replayable,
-          { rescannable: true },
-        );
-        expect(scannable.rescannable).toBe(true);
-      });
-
-      test("object with generator method passes (fresh generator each call)", async () => {
-        const table = makeTable();
-        const replayable: Iterable<RecordBatch> = {
-          *[Symbol.iterator]() {
-            for (const batch of table.batches) yield batch;
-          },
-        };
-        const scannable = await Scannable.fromIterable(
-          table.schema,
-          replayable,
-          { rescannable: true },
-        );
-        expect(scannable.rescannable).toBe(true);
-      });
-
-      test("empty Array passes (replayable degenerate case)", async () => {
-        const schema = makeTable().schema;
-        const scannable = await Scannable.fromIterable(
-          schema,
-          [] as RecordBatch[],
-          { rescannable: true },
-        );
-        expect(scannable.rescannable).toBe(true);
-      });
-
-      // One-shot inputs: [Symbol.iterator]() / [Symbol.asyncIterator]()
-      // returns the same object, or the input is already-an-iterator.
-      // Must throw with a /one-shot/ message.
-
-      test("sync generator throws", async () => {
-        const table = makeTable();
-        function* generator(): Generator<RecordBatch> {
-          for (const batch of table.batches) yield batch;
-        }
-        await expect(
-          Scannable.fromIterable(table.schema, generator(), {
-            rescannable: true,
-          }),
-        ).rejects.toThrow(/one-shot/);
-      });
-
-      test("async generator throws", async () => {
-        const table = makeTable();
-        async function* generator(): AsyncGenerator<RecordBatch> {
-          for (const batch of table.batches) yield batch;
-        }
-        await expect(
-          Scannable.fromIterable(table.schema, generator(), {
-            rescannable: true,
-          }),
-        ).rejects.toThrow(/one-shot/);
-      });
-
-      test("empty generator throws (one-shot degenerate case)", async () => {
-        const schema = makeTable().schema;
-        function* generator(): Generator<RecordBatch> {
-          // intentionally empty; yields nothing.
-        }
-        await expect(
-          Scannable.fromIterable(schema, generator(), { rescannable: true }),
-        ).rejects.toThrow(/one-shot/);
-      });
-
-      test("custom self-iterator throws", async () => {
-        const table = makeTable();
-        const batches = table.batches;
-        let i = 0;
-        const oneShot: Iterable<RecordBatch> & Iterator<RecordBatch> = {
-          [Symbol.iterator]() {
-            return this;
-          },
-          next() {
-            if (i >= batches.length) {
-              return { done: true, value: undefined };
-            }
-            return { done: false, value: batches[i++] };
-          },
-        };
-        await expect(
-          Scannable.fromIterable(table.schema, oneShot, { rescannable: true }),
-        ).rejects.toThrow(/one-shot/);
-      });
-
-      test("Array.values() (IterableIterator) throws", async () => {
-        const table = makeTable();
-        const iter = table.batches.values();
-        await expect(
-          Scannable.fromIterable(table.schema, iter, { rescannable: true }),
-        ).rejects.toThrow(/one-shot/);
-      });
-
-      test("raw iterator (only `.next`) throws", async () => {
-        const table = makeTable();
-        const batches = table.batches;
-        let i = 0;
-        const rawIter = {
-          next(): IteratorResult<RecordBatch> {
-            if (i >= batches.length) {
-              return { done: true, value: undefined };
-            }
-            return { done: false, value: batches[i++] };
-          },
-        };
-        await expect(
-          Scannable.fromIterable(
-            table.schema,
-            rawIter as unknown as Iterable<RecordBatch>,
-            { rescannable: true },
-          ),
-        ).rejects.toThrow(/one-shot/);
-      });
-
-      // Edge: null/undefined must not crash the detection helper. The
-      // null check belongs to `normalizeIterator` and only fires when a
-      // scan starts.
-
-      test("null input does not crash detection at construction", async () => {
-        const schema = makeTable().schema;
-        await expect(
-          Scannable.fromIterable(
-            schema,
-            null as unknown as Iterable<RecordBatch>,
-            {
-              rescannable: true,
-            },
-          ),
-        ).resolves.toBeDefined();
-      });
-
-      test("undefined input does not crash detection at construction", async () => {
-        const schema = makeTable().schema;
-        await expect(
-          Scannable.fromIterable(
-            schema,
-            undefined as unknown as Iterable<RecordBatch>,
-            { rescannable: true },
-          ),
-        ).resolves.toBeDefined();
-      });
-
-      // Default (rescannable omitted) skips the check entirely, so even
-      // pathological inputs construct without throwing here.
-
-      test("rescannable omitted skips detection entirely (generator passes)", async () => {
-        const table = makeTable();
-        function* generator(): Generator<RecordBatch> {
-          for (const batch of table.batches) yield batch;
-        }
-        const scannable = await Scannable.fromIterable(
-          table.schema,
-          generator(),
-        );
-        expect(scannable.rescannable).toBe(false);
-      });
-
-      test("rescannable: false explicit skips detection entirely (generator passes)", async () => {
-        const table = makeTable();
-        function* generator(): Generator<RecordBatch> {
-          for (const batch of table.batches) yield batch;
-        }
-        const scannable = await Scannable.fromIterable(
-          table.schema,
-          generator(),
-          { rescannable: false },
-        );
-        expect(scannable.rescannable).toBe(false);
-      });
-    });
-  });
-
-  describe("fromFactory", () => {
-    test("defaults rescannable=true and does not invoke the factory eagerly", async () => {
-      const table = makeTable();
-      const factory = jest.fn(() => table.batches);
-
-      const scannable = await Scannable.fromFactory(table.schema, factory);
-
-      expect(scannable.schema).toBe(table.schema);
-      expect(scannable.rescannable).toBe(true);
-      expect(factory).not.toHaveBeenCalled();
-    });
-
-    test("honors rescannable and numRows overrides", async () => {
-      const table = makeTable();
-      const scannable = await Scannable.fromFactory(
-        table.schema,
-        () => table.batches,
-        { numRows: 7, rescannable: false },
-      );
-
-      expect(scannable.numRows).toBe(7);
-      expect(scannable.rescannable).toBe(false);
-    });
-  });
-
-  describe("validation", () => {
-    test("throws when numRows is negative", async () => {
-      await expect(
-        Scannable.fromFactory(makeTable().schema, () => [], { numRows: -1 }),
-      ).rejects.toThrow(/non-negative/);
-    });
-
-    test("throws when numRows is not an integer", async () => {
-      await expect(
-        Scannable.fromFactory(makeTable().schema, () => [], { numRows: 3.5 }),
-      ).rejects.toThrow(/integer/);
-    });
-  });
-
-  describe("native handle", () => {
-    test("exposes a native handle via inner", async () => {
-      const scannable = await Scannable.fromTable(makeTable());
-      expect(scannable.inner).toBeDefined();
-      expect(typeof scannable.inner).toBe("object");
-      expect(scannable.inner).not.toBeNull();
-    });
-  });
-
-  // Schema-variety construction tests. Each asserts that construction
-  // succeeds against a richer Arrow schema, which transitively exercises
-  // schema serialization and the Rust-side `ipc_file_to_schema` for types
-  // beyond flat primitives.
-  describe("schema variety", () => {
-    test("accepts an empty table", async () => {
-      const schema = new Schema([new Field("id", new Int32(), true)]);
-      const table = makeEmptyTable(schema);
-      const scannable = await Scannable.fromTable(table);
-
-      expect(scannable.numRows).toBe(0);
-      expect(scannable.schema).toBe(table.schema);
-    });
-
-    test("accepts nested struct and list columns", async () => {
-      const table = makeArrowTable(
-        [
-          { id: 1, point: { x: 0, y: 0 }, tags: ["a", "b"] },
-          { id: 2, point: { x: 1, y: 2 }, tags: ["c"] },
-        ],
-        { vectorColumns: {} },
-      );
-      const scannable = await Scannable.fromTable(table);
-
-      expect(scannable.schema).toBe(table.schema);
-      expect(scannable.numRows).toBe(2);
-    });
-
-    test("accepts a FixedSizeList (vector) column", async () => {
-      const table = makeArrowTable(
-        [
-          { id: 1, vec: [1, 2, 3] },
-          { id: 2, vec: [4, 5, 6] },
-        ],
-        { vectorColumns: { vec: { type: new Float16() } } },
-      );
-      const scannable = await Scannable.fromTable(table);
-
-      expect(scannable.schema).toBe(table.schema);
-      expect(scannable.numRows).toBe(2);
-    });
-
-    test("accepts a table with many columns", async () => {
-      const row: Record<string, number> = {};
-      for (let i = 0; i < 50; i++) row[`c${i}`] = i;
-      const table = makeArrowTable([row, row], { vectorColumns: {} });
-      const scannable = await Scannable.fromTable(table);
-
-      expect(scannable.schema.fields.length).toBe(50);
-      expect(scannable.numRows).toBe(2);
-    });
-  });
-});

nodejs/__test__/table.test.ts

@@ -28,7 +28,6 @@ import {
   List,
   Schema,
   SchemaLike,
-  Struct,
   Type,
   Uint8,
   Utf8,
@@ -116,48 +115,6 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       await expect(table.countRows()).resolves.toBe(1);
     });
 
-    it("should invoke the progress callback", async () => {
-      const events: import("../lancedb").WriteProgress[] = [];
-      await table.add([{ id: 1 }, { id: 2 }, { id: 3 }], {
-        progress: (p) => events.push(p),
-      });
-
-      expect(events.length).toBeGreaterThan(0);
-      const last = events[events.length - 1];
-      expect(last.done).toBe(true);
-      // Earlier callbacks must have done=false.
-      for (const ev of events.slice(0, -1)) {
-        expect(ev.done).toBe(false);
-      }
-      // outputRows reflects the rows added in this call, not table size.
-      expect(last.outputRows).toBe(3);
-      // The input source (an array) reports a row count, so totalRows is set.
-      expect(last.totalRows).toBe(3);
-      // outputRows is monotonic.
-      for (let i = 1; i < events.length; i++) {
-        expect(events[i].outputRows).toBeGreaterThanOrEqual(
-          events[i - 1].outputRows,
-        );
-      }
-    });
-
-    it("should swallow errors thrown from the progress callback", async () => {
-      const warn = jest
-        .spyOn(console, "warn")
-        .mockImplementation(() => undefined);
-      try {
-        const res = await table.add([{ id: 1 }, { id: 2 }], {
-          progress: () => {
-            throw new Error("callback bomb");
-          },
-        });
-        expect(res.version).toBeGreaterThan(0);
-        expect(warn).toHaveBeenCalled();
-      } finally {
-        warn.mockRestore();
-      }
-    });
-
     it("should let me close the table", async () => {
       expect(table.isOpen()).toBe(true);
       table.close();
@@ -781,113 +738,6 @@ describe("When creating an index", () => {
     expect(indices2.length).toBe(0);
   });
 
-  it("should create and search a nested vector index", async () => {
-    const db = await connect(tmpDir.name);
-    const nestedSchema = new Schema([
-      new Field("id", new Int32(), true),
-      new Field(
-        "image",
-        new Struct([
-          new Field(
-            "embedding",
-            new FixedSizeList(2, new Field("item", new Float32(), true)),
-            true,
-          ),
-        ]),
-        true,
-      ),
-    ]);
-    const nestedTable = await db.createTable(
-      "nested_vector",
-      makeArrowTable(
-        Array.from({ length: 300 }, (_, id) => ({
-          id,
-          image: { embedding: [id, id + 1] },
-        })),
-        { schema: nestedSchema },
-      ),
-    );
-
-    await nestedTable.createIndex("image.embedding", {
-      name: "image_embedding_idx",
-    });
-    const indices = await nestedTable.listIndices();
-    expect(indices).toContainEqual({
-      name: "image_embedding_idx",
-      indexType: "IvfPq",
-      columns: ["image.embedding"],
-    });
-
-    const explicit = await nestedTable
-      .query()
-      .nearestTo([0.0, 1.0])
-      .column("image.embedding")
-      .limit(1)
-      .toArray();
-    const inferred = await nestedTable
-      .query()
-      .nearestTo([0.0, 1.0])
-      .limit(1)
-      .toArray();
-    expect(inferred[0].id).toEqual(explicit[0].id);
-  });
-
-  it("should report multiple nested vector candidates", async () => {
-    const db = await connect(tmpDir.name);
-    const nestedSchema = new Schema([
-      new Field(
-        "image",
-        new Struct([
-          new Field(
-            "embedding",
-            new FixedSizeList(2, new Field("item", new Float32(), true)),
-            true,
-          ),
-        ]),
-        true,
-      ),
-      new Field(
-        "text",
-        new Struct([
-          new Field(
-            "embedding",
-            new FixedSizeList(2, new Field("item", new Float32(), true)),
-            true,
-          ),
-        ]),
-        true,
-      ),
-    ]);
-    const nestedTable = await db.createTable(
-      "multiple_nested_vectors",
-      makeArrowTable(
-        [
-          {
-            image: { embedding: [0.0, 1.0] },
-            text: { embedding: [2.0, 3.0] },
-          },
-        ],
-        { schema: nestedSchema },
-      ),
-    );
-
-    await expect(
-      nestedTable.query().nearestTo([0.0, 1.0]).limit(1).toArray(),
-    ).rejects.toThrow(/image\.embedding.*text\.embedding/);
-  });
-
-  it("should report when no default vector column exists", async () => {
-    const db = await connect(tmpDir.name);
-    const noVectorTable = await db.createTable(
-      "no_vector",
-      makeArrowTable([{ id: 0, label: "cat" }]),
-    );
-
-    await expect(
-      noVectorTable.query().nearestTo([0.0, 1.0]).limit(1).toArray(),
-    ).rejects.toThrow(/No vector column/);
-  });
-
   it("should wait for index readiness", async () => {
     // Create an index and then wait for it to be ready
     await tbl.createIndex("vec");
@@ -2020,25 +1870,6 @@ 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 = [
@@ -2498,130 +2329,3 @@ describe("when creating a table with Float32Array vectors", () => {
     expect((fsl.children[0].type as Float32).precision).toBe(1);
   });
 });
-
-describe("setUnenforcedPrimaryKey", () => {
-  let tmpDir: tmp.DirResult;
-
-  beforeEach(() => {
-    tmpDir = tmp.dirSync({ unsafeCleanup: true });
-  });
-  afterEach(() => tmpDir.removeCallback());
-
-  it("sets a single-column primary key (string or one-element array)", async () => {
-    const conn = await connect(tmpDir.name);
-    const schema = new arrow.Schema([
-      new arrow.Field("id", new arrow.Int64(), false),
-    ]);
-    const t1 = await conn.createEmptyTable("t1", schema);
-    await t1.setUnenforcedPrimaryKey("id");
-
-    const t2 = await conn.createEmptyTable("t2", schema);
-    await t2.setUnenforcedPrimaryKey(["id"]);
-  });
-
-  it("rejects a compound primary key", async () => {
-    const conn = await connect(tmpDir.name);
-    const table = await conn.createEmptyTable(
-      "t",
-      new arrow.Schema([
-        new arrow.Field("id", new arrow.Int64(), false),
-        new arrow.Field("name", new arrow.Utf8(), false),
-      ]),
-    );
-    await expect(
-      table.setUnenforcedPrimaryKey(["id", "name"]),
-    ).rejects.toThrow();
-  });
-
-  it("rejects changing the primary key once set", async () => {
-    const conn = await connect(tmpDir.name);
-    const table = await conn.createEmptyTable(
-      "t",
-      new arrow.Schema([
-        new arrow.Field("id", new arrow.Int64(), false),
-        new arrow.Field("name", new arrow.Utf8(), false),
-      ]),
-    );
-    await table.setUnenforcedPrimaryKey("id");
-    await expect(table.setUnenforcedPrimaryKey("name")).rejects.toThrow();
-    await expect(table.setUnenforcedPrimaryKey("id")).rejects.toThrow();
-  });
-});
-
-describe("setLsmWriteSpec / unsetLsmWriteSpec", () => {
-  let tmpDir: tmp.DirResult;
-
-  beforeEach(() => {
-    tmpDir = tmp.dirSync({ unsafeCleanup: true });
-  });
-  afterEach(() => tmpDir.removeCallback());
-
-  async function makeTable(conn: Connection): Promise<Table> {
-    return await conn.createEmptyTable(
-      "t",
-      new arrow.Schema([new arrow.Field("id", new arrow.Int64(), false)]),
-    );
-  }
-
-  it("installs and removes a bucket spec", async () => {
-    const conn = await connect(tmpDir.name);
-    const table = await makeTable(conn);
-
-    await table.setUnenforcedPrimaryKey("id");
-    await table.setLsmWriteSpec({
-      specType: "bucket",
-      column: "id",
-      numBuckets: 4,
-    });
-    await table.unsetLsmWriteSpec();
-    // A second unset errors — there is no spec left to remove.
-    await expect(table.unsetLsmWriteSpec()).rejects.toThrow();
-    // A fresh spec can be installed after unset.
-    await table.setLsmWriteSpec({
-      specType: "bucket",
-      column: "id",
-      numBuckets: 8,
-    });
-  });
-
-  it("installs an unsharded spec", async () => {
-    const conn = await connect(tmpDir.name);
-    const table = await makeTable(conn);
-
-    await table.setUnenforcedPrimaryKey("id");
-    await table.setLsmWriteSpec({ specType: "unsharded" });
-    await table.unsetLsmWriteSpec();
-  });
-
-  it("installs an identity spec", async () => {
-    const conn = await connect(tmpDir.name);
-    const table = await makeTable(conn);
-
-    await table.setUnenforcedPrimaryKey("id");
-    await table.setLsmWriteSpec({ specType: "identity", column: "id" });
-    await table.unsetLsmWriteSpec();
-  });
-
-  it("rejects an invalid spec", async () => {
-    const conn = await connect(tmpDir.name);
-    const table = await makeTable(conn);
-
-    await table.setUnenforcedPrimaryKey("id");
-    // num_buckets out of range.
-    await expect(
-      table.setLsmWriteSpec({
-        specType: "bucket",
-        column: "id",
-        numBuckets: 0,
-      }),
-    ).rejects.toThrow();
-    // Column mismatch.
-    await expect(
-      table.setLsmWriteSpec({
-        specType: "bucket",
-        column: "missing",
-        numBuckets: 4,
-      }),
-    ).rejects.toThrow();
-  });
-});

nodejs/examples/filtering.test.ts

@@ -38,14 +38,5 @@ test("filtering examples", async () => {
     // --8<-- [start:sql_search]
     await tbl.query().where("id = 10").limit(10).toArray();
     // --8<-- [end:sql_search]
-
-    // --8<-- [start:orderby_search]
-    await tbl
-      .query()
-      .where("id > 10")
-      .orderBy({ columnName: "id", ascending: false })
-      .limit(5)
-      .toArray();
-    // --8<-- [end:orderby_search]
   });
 });

nodejs/examples/package.json

@@ -11,17 +11,16 @@
     "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 && pnpm format",
+    "lint-fix": "biome check --write *.ts && npm run 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

@@ -1,13 +0,0 @@
-# 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

@@ -1291,18 +1291,6 @@ export async function fromRecordBatchToBuffer(
   return Buffer.from(await writer.toUint8Array());
 }
 
-/**
- * Create a buffer containing a single record batch using the Arrow IPC Stream
- * serialization. Each call produces a self-contained Stream message (schema +
- * batch + EOS) suitable for incremental decode by `arrow_ipc::reader::StreamReader`.
- */
-export async function fromRecordBatchToStreamBuffer(
-  batch: RecordBatch,
-): Promise<Buffer> {
-  const writer = RecordBatchStreamWriter.writeAll([batch]);
-  return Buffer.from(await writer.toUint8Array());
-}
-
 /**
  * Serialize an Arrow Table into a buffer using the Arrow IPC Stream serialization
  *

nodejs/lancedb/connection.ts

@@ -16,18 +16,6 @@ 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";
 
@@ -122,41 +110,6 @@ 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";
-}
-
-export interface RenameTableOptions {
-  /**
-   * The namespace path of the table being renamed. Defaults to the root
-   * namespace (`[]`) when omitted.
-   */
-  namespacePath?: string[];
-  /**
-   * The namespace path to move the table to as part of the rename. When
-   * omitted the table stays in `namespacePath`.
-   */
-  newNamespacePath?: string[];
-}
-
 /**
  * A LanceDB Connection that allows you to open tables and create new ones.
  *
@@ -315,69 +268,6 @@ export abstract class Connection {
    */
   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.
    *
@@ -404,24 +294,6 @@ export abstract class Connection {
       isShallow?: boolean;
     },
   ): Promise<Table>;
-
-  /**
-   * Rename a table.
-   *
-   * Currently only supported by LanceDB Cloud. Local OSS connections and
-   * namespace-backed connections (via {@link connectNamespace}) reject with
-   * a "not supported" error.
-   *
-   * @param {string} currentName - The current name of the table.
-   * @param {string} newName - The new name for the table.
-   * @param {RenameTableOptions} options - Optional namespace paths. When
-   *   `newNamespacePath` is omitted the table stays in `namespacePath`.
-   */
-  abstract renameTable(
-    currentName: string,
-    newName: string,
-    options?: RenameTableOptions,
-  ): Promise<void>;
 }
 
 /** @hideconstructor */
@@ -643,58 +515,6 @@ export class LocalConnection extends Connection {
   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,
-    );
-  }
-
-  async renameTable(
-    currentName: string,
-    newName: string,
-    options?: RenameTableOptions,
-  ): Promise<void> {
-    return this.inner.renameTable(
-      currentName,
-      newName,
-      options?.namespacePath ?? [],
-      options?.newNamespacePath,
-    );
-  }
 }
 
 /**

nodejs/lancedb/index.ts

@@ -8,7 +8,6 @@ import {
 } from "./connection";
 
 import {
-  ConnectNamespaceOptions,
   ConnectionOptions,
   Connection as LanceDbConnection,
   JsHeaderProvider as NativeJsHeaderProvider,
@@ -23,7 +22,6 @@ export { JsHeaderProvider as NativeJsHeaderProvider } from "./native.js";
 export {
   AddColumnsSql,
   ConnectionOptions,
-  ConnectNamespaceOptions,
   IndexStatistics,
   IndexConfig,
   ClientConfig,
@@ -64,14 +62,6 @@ export {
   CreateTableOptions,
   TableNamesOptions,
   OpenTableOptions,
-  ListNamespacesOptions,
-  CreateNamespaceOptions,
-  DropNamespaceOptions,
-  ListNamespacesResponse,
-  CreateNamespaceResponse,
-  DropNamespaceResponse,
-  DescribeNamespaceResponse,
-  RenameTableOptions,
 } from "./connection";
 
 export { Session } from "./native.js";
@@ -83,7 +73,6 @@ export {
   VectorQuery,
   TakeQuery,
   QueryExecutionOptions,
-  ColumnOrdering,
   FullTextSearchOptions,
   RecordBatchIterator,
   FullTextQuery,
@@ -114,8 +103,6 @@ export {
   UpdateOptions,
   OptimizeOptions,
   Version,
-  WriteProgress,
-  LsmWriteSpec,
   ColumnAlteration,
 } from "./table";
 
@@ -130,7 +117,6 @@ export { MergeInsertBuilder, WriteExecutionOptions } from "./merge";
 
 export * as embedding from "./embedding";
 export { permutationBuilder, PermutationBuilder } from "./permutation";
-export { Scannable, ScannableOptions } from "./scannable";
 export * as rerankers from "./rerankers";
 export {
   SchemaLike,
@@ -307,197 +293,3 @@ 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/query.ts

@@ -79,12 +79,6 @@ export interface QueryExecutionOptions {
   timeoutMs?: number;
 }
 
-export interface ColumnOrdering {
-  columnName: string;
-  ascending?: boolean;
-  nullsFirst?: boolean;
-}
-
 /**
  * Options that control the behavior of a full text search
  */
@@ -423,21 +417,6 @@ export class StandardQueryBase<
     return this;
   }
 
-  /**
-   * Sort the results by the specified column(s).
-   * @returns This query builder.
-   */
-  orderBy(ordering: ColumnOrdering | ColumnOrdering[]): this {
-    const orderings = Array.isArray(ordering) ? ordering : [ordering];
-    const normalized = orderings.map((o) => ({
-      columnName: o.columnName,
-      ascending: o.ascending ?? true,
-      nullsFirst: o.nullsFirst ?? false,
-    }));
-    this.doCall((inner) => inner.orderBy(normalized));
-    return this;
-  }
-
   /**
    * Skip searching un-indexed data. This can make search faster, but will miss
    * any data that is not yet indexed.

nodejs/lancedb/scannable.ts

@@ -1,274 +0,0 @@
-// SPDX-License-Identifier: Apache-2.0
-// SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-import {
-  Table as ArrowTable,
-  RecordBatch,
-  RecordBatchReader,
-  Schema,
-} from "apache-arrow";
-import {
-  fromRecordBatchToStreamBuffer,
-  fromTableToBuffer,
-  makeEmptyTable,
-} from "./arrow";
-import { NapiScannable } from "./native.js";
-
-export interface ScannableOptions {
-  /** Hint about the number of rows. Not validated against the stream. */
-  numRows?: number;
-  /**
-   * Whether the source can be scanned more than once. Defaults to `true` for
-   * `fromTable` / `fromFactory` and `false` for `fromIterable` /
-   * `fromRecordBatchReader`.
-   */
-  rescannable?: boolean;
-}
-
-/**
- * A data source that can be scanned as a stream of Arrow `RecordBatch`es.
- *
- * `Scannable` wraps the schema + optional row count + rescannable flag and
- * a callback that yields batches one at a time. It is passed to consumers
- * (e.g. `Table.add`, `createTable`, `mergeInsert` — follow-up work) that
- * need to pull data without materializing the full dataset in JS memory.
- *
- * Batches cross the JS↔Rust boundary as Arrow IPC Stream messages; a fresh
- * writer serializes each batch, and the Rust side decodes it with
- * `arrow_ipc::reader::StreamReader`. One batch is in flight at a time.
- */
-export class Scannable {
-  readonly schema: Schema;
-  readonly numRows: number | null;
-  readonly rescannable: boolean;
-
-  /** @hidden */
-  private readonly native: NapiScannable;
-
-  private constructor(
-    native: NapiScannable,
-    schema: Schema,
-    numRows: number | null,
-    rescannable: boolean,
-  ) {
-    this.native = native;
-    this.schema = schema;
-    this.numRows = numRows;
-    this.rescannable = rescannable;
-  }
-
-  /** @hidden Access the native handle for passing through to Rust consumers. */
-  get inner(): NapiScannable {
-    return this.native;
-  }
-
-  /**
-   * Build a Scannable from an explicit schema and a factory that returns a
-   * fresh batch iterator on each call.
-   *
-   * The factory is invoked once per scan. Each iterator yields
-   * `RecordBatch`es matching the declared schema. Use this when you need
-   * direct control over the pull loop — for example, to wrap a streaming
-   * source whose batches are produced lazily.
-   *
-   * @param schema - The Arrow schema of the produced batches.
-   * @param factory - Called at the start of each scan to produce a batch
-   *   iterator. Must be idempotent when `rescannable` is true.
-   * @param opts - Optional hints. `rescannable` defaults to `true`; set to
-   *   `false` if calling `factory()` twice would not reproduce the same data.
-   */
-  static async fromFactory(
-    schema: Schema,
-    factory: () =>
-      | AsyncIterable<RecordBatch>
-      | Iterable<RecordBatch>
-      | AsyncIterator<RecordBatch>
-      | Iterator<RecordBatch>,
-    opts: ScannableOptions = {},
-  ): Promise<Scannable> {
-    const numRows = opts.numRows ?? null;
-    if (numRows != null && !Number.isInteger(numRows)) {
-      throw new TypeError("numRows must be an integer");
-    }
-    const rescannable = opts.rescannable ?? true;
-
-    let iter: AsyncIterator<RecordBatch> | Iterator<RecordBatch> | null = null;
-    const getNextBatch = async (isStart: boolean): Promise<Buffer | null> => {
-      // `isStart` is true on the first pull of every new scan_as_stream.
-      // Drop any cached iterator so factory() is re-invoked for the next scan
-      if (isStart) {
-        iter = null;
-      }
-      if (iter === null) {
-        iter = normalizeIterator(factory());
-      }
-      const result = await iter.next();
-      if (result.done) {
-        iter = null;
-        return null;
-      }
-      return fromRecordBatchToStreamBuffer(result.value);
-    };
-
-    const schemaBuf = await fromTableToBuffer(makeEmptyTable(schema));
-    const native = new NapiScannable(
-      schemaBuf,
-      numRows,
-      rescannable,
-      getNextBatch,
-    );
-    return new Scannable(native, schema, numRows, rescannable);
-  }
-
-  /**
-   * Build a Scannable from an in-memory Arrow `Table`. Always rescannable;
-   * the table's batches are replayed on each scan.
-   *
-   * The table's row count is authoritative: `opts.numRows` must either be
-   * omitted or equal to `table.numRows`. `opts.rescannable` of `false` is
-   * rejected because in-memory Tables are always rescannable.
-   */
-  static async fromTable(
-    table: ArrowTable,
-    opts: ScannableOptions = {},
-  ): Promise<Scannable> {
-    if (opts.numRows != null && opts.numRows !== table.numRows) {
-      throw new TypeError(
-        `opts.numRows (${opts.numRows}) does not match table.numRows (${table.numRows}). ` +
-          `The table's row count is authoritative; omit numRows or pass the matching value.`,
-      );
-    }
-    if (opts.rescannable === false) {
-      throw new TypeError(
-        `fromTable does not accept rescannable: false. ` +
-          `In-memory Arrow Tables are always rescannable; omit the option or pass true.`,
-      );
-    }
-    return Scannable.fromFactory(table.schema, () => table.batches, {
-      numRows: table.numRows,
-      rescannable: true,
-    });
-  }
-
-  /**
-   * Build a Scannable from an iterable of `RecordBatch`es. `rescannable`
-   * defaults to `false`. Pass an explicit schema so the consumer can
-   * validate before any batch is pulled.
-   *
-   * `opts.rescannable: true` is honest for replayable iterables (Arrays,
-   * Sets, or custom iterables whose `[Symbol.iterator]()` returns a fresh
-   * iterator each call). It is rejected for one-shot iterables (generators,
-   * async generators, or already-an-iterator inputs) because their
-   * `[Symbol.iterator]()` returns the same exhausted object on the second
-   * scan. For replayable sources outside this shape, use
-   * `fromFactory(schema, () => createIter(), { rescannable: true })`.
-   *
-   * Note: when `opts.rescannable` is `true`, the constructor calls
-   * `[Symbol.iterator]()` once on the input to perform the structural check.
-   */
-  static async fromIterable(
-    schema: Schema,
-    iter: AsyncIterable<RecordBatch> | Iterable<RecordBatch>,
-    opts: ScannableOptions = {},
-  ): Promise<Scannable> {
-    if (opts.rescannable === true && isOneShotIterable(iter)) {
-      throw new TypeError(
-        `fromIterable: rescannable: true is not honest for one-shot iterables ` +
-          `(generators, async generators, or iterators where [Symbol.iterator]() ` +
-          `returns the same object). The source would be exhausted after the first scan. ` +
-          `Use fromFactory(schema, () => createIter(), { rescannable: true }) for sources ` +
-          `where each call mints a fresh iterator.`,
-      );
-    }
-    return Scannable.fromFactory(schema, () => iter, {
-      numRows: opts.numRows,
-      rescannable: opts.rescannable ?? false,
-    });
-  }
-
-  /**
-   * Build a Scannable from an Arrow `RecordBatchReader`. A reader can only
-   * be consumed once; `rescannable` defaults to `false`.
-   *
-   * The reader must already be opened (via `.open()`) so its `.schema` is
-   * populated. `RecordBatchReader.from(...)` returns an unopened reader.
-   *
-   * `opts.rescannable: true` is rejected because `RecordBatchReader` is a
-   * self-iterator (its `[Symbol.iterator]()` returns itself), and this
-   * constructor does not call `reader.reset()` between scans, so a second
-   * scan would always see an exhausted reader. For genuinely replayable
-   * sources, use
-   * `fromFactory(schema, () => openReader(), { rescannable: true })`,
-   * which mints a fresh reader on each scan.
-   */
-  static async fromRecordBatchReader(
-    reader: RecordBatchReader,
-    opts: ScannableOptions = {},
-  ): Promise<Scannable> {
-    if (opts.rescannable === true) {
-      throw new TypeError(
-        `fromRecordBatchReader does not accept rescannable: true. ` +
-          `RecordBatchReader is a self-iterator (its [Symbol.iterator]() ` +
-          `returns itself) and would be exhausted after the first scan. ` +
-          `Use fromFactory(schema, () => openReader(), { rescannable: true }) ` +
-          `for sources where each call mints a fresh reader.`,
-      );
-    }
-    return Scannable.fromFactory(reader.schema, () => reader, {
-      numRows: opts.numRows,
-      rescannable: false,
-    });
-  }
-}
-
-function normalizeIterator<T>(
-  source: AsyncIterable<T> | Iterable<T> | AsyncIterator<T> | Iterator<T>,
-): AsyncIterator<T> | Iterator<T> {
-  if (source == null) {
-    throw new TypeError("Scannable factory returned null/undefined");
-  }
-  if (
-    typeof (source as AsyncIterable<T>)[Symbol.asyncIterator] === "function"
-  ) {
-    return (source as AsyncIterable<T>)[Symbol.asyncIterator]();
-  }
-  if (typeof (source as Iterable<T>)[Symbol.iterator] === "function") {
-    return (source as Iterable<T>)[Symbol.iterator]();
-  }
-  // Already an iterator (has `.next`).
-  if (typeof (source as Iterator<T>).next === "function") {
-    return source as Iterator<T>;
-  }
-  throw new TypeError("Scannable factory returned a non-iterable value");
-}
-
-// A "self-iterator" returns the same object from `[Symbol.iterator]()` /
-// `[Symbol.asyncIterator]()`. Generators behave this way, so they exhaust
-// after one pass. Replayable iterables (Array, Set, custom) return a fresh
-// iterator each call. Detection mirrors `normalizeIterator`'s ordering so
-// classification matches scan-time behavior.
-function isOneShotIterable(
-  source: AsyncIterable<unknown> | Iterable<unknown>,
-): boolean {
-  // null/undefined are not one-shot in any meaningful sense; let
-  // `normalizeIterator` raise the actual error at scan time.
-  if (source == null) return false;
-  const ref = source as unknown;
-  if (
-    typeof (source as AsyncIterable<unknown>)[Symbol.asyncIterator] ===
-    "function"
-  ) {
-    const it = (source as AsyncIterable<unknown>)[
-      Symbol.asyncIterator
-    ]() as unknown;
-    return it === ref;
-  }
-  if (typeof (source as Iterable<unknown>)[Symbol.iterator] === "function") {
-    const it = (source as Iterable<unknown>)[Symbol.iterator]() as unknown;
-    return it === ref;
-  }
-  // Already-an-iterator (has `.next` but no `Symbol.iterator`) is by
-  // definition one-shot.
-  if (typeof (source as { next?: unknown }).next === "function") return true;
-  return false;
-}

nodejs/lancedb/table.ts

@@ -46,33 +46,6 @@ import { sanitizeType } from "./sanitize";
 import { IntoSql, toSQL } from "./util";
 export { IndexConfig } from "./native";
 
-/**
- * Progress snapshot for a write operation, delivered to the `progress`
- * callback passed to {@link Table.add}.
- */
-export interface WriteProgress {
-  /** Number of rows written so far. */
-  outputRows: number;
-  /** Number of bytes written so far. */
-  outputBytes: number;
-  /**
-   * Total rows expected, when the input source reports it.
-   *
-   * Always set on the final callback (the one with `done: true`), falling
-   * back to the actual number of rows written when the source could not
-   * report a row count up front.
-   */
-  totalRows?: number;
-  /** Wall-clock seconds since the write started. */
-  elapsedSeconds: number;
-  /** Number of parallel write tasks currently in flight. */
-  activeTasks: number;
-  /** Total number of parallel write tasks (the write parallelism). */
-  totalTasks: number;
-  /** `true` for the final callback; `false` otherwise. */
-  done: boolean;
-}
-
 /**
  * Options for adding data to a table.
  */
@@ -83,28 +56,6 @@ export interface AddDataOptions {
    * If "overwrite" then the new data will replace the existing data in the table.
    */
   mode: "append" | "overwrite";
-
-  /**
-   * Optional callback invoked periodically with write progress.
-   *
-   * The callback is fired once per batch written and once more with
-   * `done: true` when the write completes. Calls are dispatched
-   * asynchronously to the JS event loop and never block the write — a slow
-   * callback will queue events rather than back-pressure the writer.
-   *
-   * Errors thrown from the callback are logged with `console.warn` and
-   * swallowed — they do not abort the write.
-   *
-   * @example
-   * ```ts
-   * await table.add(data, {
-   *   progress: (p) => {
-   *     console.log(`${p.outputRows}/${p.totalRows ?? "?"} rows`);
-   *   },
-   * });
-   * ```
-   */
-  progress: (progress: WriteProgress) => void;
 }
 
 export interface UpdateOptions {
@@ -155,27 +106,6 @@ export interface Version {
   metadata: Record<string, string>;
 }
 
-/**
- * Specification selecting Lance's MemWAL LSM-style write path for
- * `mergeInsert`.
- *
- * `specType` is `"bucket"`, `"identity"`, or `"unsharded"`. For `"bucket"`,
- * `column` and `numBuckets` are required; for `"identity"`, `column` is
- * required.
- */
-export interface LsmWriteSpec {
-  /** One of `"bucket"`, `"identity"`, or `"unsharded"`. */
-  specType: "bucket" | "identity" | "unsharded";
-  /** Bucket and identity variants: the sharding column. */
-  column?: string;
-  /** Bucket variant: the number of buckets, in `[1, 1024]`. */
-  numBuckets?: number;
-  /** Names of indexes the MemWAL should keep up to date during writes. */
-  maintainedIndexes?: string[];
-  /** Default `ShardWriter` configuration recorded in the MemWAL index. */
-  writerConfigDefaults?: Record<string, string>;
-}
-
 /**
  * A Table is a collection of Records in a LanceDB Database.
  *
@@ -355,25 +285,6 @@ 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.
    *
@@ -519,54 +430,6 @@ export abstract class Table {
    * containing the new version number of the table after dropping the columns.
    */
   abstract dropColumns(columnNames: string[]): Promise<DropColumnsResult>;
-  /**
-   * Set the unenforced primary key for this table to a single column.
-   *
-   * "Unenforced" means LanceDB does not check uniqueness on writes; the
-   * column is recorded in the schema as the primary key for use by features
-   * such as `merge_insert`. Only single-column primary keys are supported,
-   * and the key cannot be changed once set.
-   * @param {string | string[]} columns The primary key column. A one-element
-   * array is also accepted; passing more than one column is rejected.
-   * @returns {Promise<void>}
-   */
-  abstract setUnenforcedPrimaryKey(columns: string | string[]): Promise<void>;
-  /**
-   * Install an {@link LsmWriteSpec} on this table, selecting Lance's MemWAL
-   * LSM-style write path for future `mergeInsert` calls.
-   *
-   * `LsmWriteSpec` chooses one of three sharding strategies via `specType`:
-   *
-   * - `"bucket"` — hash-bucket writes by the single-column unenforced primary
-   *   key (`column` and `numBuckets` required).
-   * - `"identity"` — shard by the raw value of a scalar `column`.
-   * - `"unsharded"` — route every write to a single shard.
-   *
-   * All variants require the table to have an unenforced primary key
-   * ({@link Table#setUnenforcedPrimaryKey}); bucket sharding additionally
-   * requires it to be the single column being bucketed.
-   * @param {LsmWriteSpec} spec The sharding spec to install.
-   * @returns {Promise<void>}
-   * @example
-   * ```ts
-   * await table.setUnenforcedPrimaryKey("id");
-   * await table.setLsmWriteSpec({
-   *   specType: "bucket",
-   *   column: "id",
-   *   numBuckets: 16,
-   *   maintainedIndexes: ["id_idx"],
-   * });
-   * ```
-   */
-  abstract setLsmWriteSpec(spec: LsmWriteSpec): Promise<void>;
-  /**
-   * Remove the {@link LsmWriteSpec} from this table, reverting to the standard
-   * `mergeInsert` write path.
-   *
-   * Errors if no spec is currently set.
-   * @returns {Promise<void>}
-   */
-  abstract unsetLsmWriteSpec(): Promise<void>;
   /** Retrieve the version of the table */
 
   abstract version(): Promise<number>;
@@ -754,20 +617,7 @@ export class LocalTable extends Table {
     const schema = await this.schema();
 
     const buffer = await fromDataToBuffer(data, undefined, schema);
-    // Wrap the user callback so a thrown error doesn't surface as an
-    // unhandled exception (the callback fires from a napi threadsafe
-    // function — exceptions there crash the process).
-    const userProgress = options?.progress;
-    const progress = userProgress
-      ? (p: WriteProgress) => {
-          try {
-            userProgress(p);
-          } catch (e) {
-            console.warn("Table.add progress callback threw:", e);
-          }
-        }
-      : undefined;
-    return await this.inner.add(buffer, mode, progress);
+    return await this.inner.add(buffer, mode);
   }
 
   async update(
@@ -860,10 +710,6 @@ 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,
@@ -1028,19 +874,6 @@ export class LocalTable extends Table {
     return await this.inner.dropColumns(columnNames);
   }
 
-  async setUnenforcedPrimaryKey(columns: string | string[]): Promise<void> {
-    const cols = typeof columns === "string" ? [columns] : columns;
-    return await this.inner.setUnenforcedPrimaryKey(cols);
-  }
-
-  async setLsmWriteSpec(spec: LsmWriteSpec): Promise<void> {
-    return await this.inner.setLsmWriteSpec(spec);
-  }
-
-  async unsetLsmWriteSpec(): Promise<void> {
-    return await this.inner.unsetLsmWriteSpec();
-  }
-
   async version(): Promise<number> {
     return await this.inner.version();
   }

nodejs/npm/darwin-arm64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-arm64",
-	"version": "0.30.0-beta.1",
+	"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": "0.30.0-beta.1",
+	"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": "0.30.0-beta.1",
+	"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": "0.30.0-beta.1",
+	"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": "0.30.0-beta.1",
+	"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": "0.30.0-beta.1",
+  "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": "0.30.0-beta.1",
+	"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": "0.30.0-beta.1",
+  "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": "3.1003.0",
-    "@aws-sdk/client-kms": "3.1003.0",
-    "@aws-sdk/client-s3": "3.1003.0",
+    "@aws-sdk/client-dynamodb": "^3.33.0",
+    "@aws-sdk/client-kms": "^3.33.0",
+    "@aws-sdk/client-s3": "^3.33.0",
     "@biomejs/biome": "^1.7.3",
     "@jest/globals": "^29.7.0",
-    "@napi-rs/cli": "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/ && node -e \"require('fs').writeFileSync('dist/package.json', JSON.stringify({name:'@lancedb/lancedb',type:'commonjs'}))\"",
+    "postbuild:debug": "shx mkdir -p dist && shx cp lancedb/*.node dist/",
     "build:release": "napi build --platform --release --dts ../lancedb/native.d.ts --js ../lancedb/native.js --output-dir dist",
-    "build": "pnpm build:debug && pnpm tsc",
-    "build-release": "pnpm build:release && pnpm tsc",
+    "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",
     "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 pnpm test",
+    "integration": "S3_TEST=1 npm run 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

@@ -1,18 +0,0 @@
-# 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

@@ -8,16 +8,12 @@ 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, connect_namespace};
+use lancedb::connection::{ConnectBuilder, Connection as LanceDBConnection};
 
-use lance_namespace::models::{
-    CreateNamespaceRequest, DescribeNamespaceRequest, DropNamespaceRequest, ListNamespacesRequest,
-};
 use lancedb::ipc::{ipc_file_to_batches, ipc_file_to_schema};
 
 #[napi]
@@ -25,29 +21,6 @@ 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) }
@@ -133,39 +106,6 @@ 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())
@@ -333,149 +273,4 @@ impl Connection {
         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()?;
-        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,
-        })
-    }
-
-    /// Rename a table. `current_namespace_path` and `new_namespace_path` default to
-    /// the root namespace when omitted; the caller is expected to either pass both
-    /// or pass neither.
-    #[napi(catch_unwind)]
-    pub async fn rename_table(
-        &self,
-        current_name: String,
-        new_name: String,
-        current_namespace_path: Option<Vec<String>>,
-        new_namespace_path: Option<Vec<String>>,
-    ) -> napi::Result<()> {
-        let cur_ns = current_namespace_path.unwrap_or_default();
-        let new_ns = new_namespace_path.unwrap_or_default();
-        self.get_inner()?
-            .rename_table(&current_name, &new_name, &cur_ns, &new_ns)
-            .await
-            .default_error()
-    }
 }

nodejs/src/lib.rs

@@ -16,7 +16,6 @@ pub mod permutation;
 mod query;
 pub mod remote;
 mod rerankers;
-mod scannable;
 mod session;
 mod table;
 mod util;
@@ -68,26 +67,6 @@ 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/query.rs

@@ -3,12 +3,6 @@
 
 use std::sync::Arc;
 
-use crate::error::NapiErrorExt;
-use crate::error::convert_error;
-use crate::iterator::RecordBatchIterator;
-use crate::rerankers::RerankHybridCallbackArgs;
-use crate::rerankers::Reranker;
-use crate::util::{parse_distance_type, schema_to_buffer};
 use arrow_array::{
     Array, Float16Array as ArrowFloat16Array, Float32Array as ArrowFloat32Array,
     Float64Array as ArrowFloat64Array, UInt8Array as ArrowUInt8Array,
@@ -25,27 +19,16 @@ use lancedb::query::QueryBase;
 use lancedb::query::QueryExecutionOptions;
 use lancedb::query::Select;
 use lancedb::query::TakeQuery as LanceDbTakeQuery;
-use lancedb::query::{ColumnOrdering as LanceDbColumnOrdering, VectorQuery as LanceDbVectorQuery};
+use lancedb::query::VectorQuery as LanceDbVectorQuery;
 use napi::bindgen_prelude::*;
 use napi_derive::napi;
 
-#[napi(object)]
-pub struct ColumnOrdering {
-    pub ascending: bool,
-    pub nulls_first: bool,
-    pub column_name: String,
-}
-
-impl From<ColumnOrdering> for LanceDbColumnOrdering {
-    fn from(value: ColumnOrdering) -> Self {
-        match (value.ascending, value.nulls_first) {
-            (true, true) => Self::asc_nulls_first(value.column_name),
-            (true, false) => Self::asc_nulls_last(value.column_name),
-            (false, true) => Self::desc_nulls_first(value.column_name),
-            (false, false) => Self::desc_nulls_last(value.column_name),
-        }
-    }
-}
+use crate::error::NapiErrorExt;
+use crate::error::convert_error;
+use crate::iterator::RecordBatchIterator;
+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());
@@ -145,18 +128,6 @@ impl Query {
         self.inner = self.inner.clone().with_row_id();
     }
 
-    #[napi]
-    pub fn order_by(&mut self, ordering: Option<Vec<ColumnOrdering>>) -> napi::Result<()> {
-        let ordering = ordering.map(|ordering| {
-            ordering
-                .into_iter()
-                .map(LanceDbColumnOrdering::from)
-                .collect()
-        });
-        self.inner = self.inner.clone().order_by(ordering);
-        Ok(())
-    }
-
     #[napi(catch_unwind)]
     pub async fn output_schema(&self) -> napi::Result<Buffer> {
         let schema = self.inner.output_schema().await.default_error()?;
@@ -357,18 +328,6 @@ impl VectorQuery {
         Ok(())
     }
 
-    #[napi]
-    pub fn order_by(&mut self, ordering: Option<Vec<ColumnOrdering>>) -> napi::Result<()> {
-        let ordering = ordering.map(|ordering| {
-            ordering
-                .into_iter()
-                .map(LanceDbColumnOrdering::from)
-                .collect()
-        });
-        self.inner = self.inner.clone().order_by(ordering);
-        Ok(())
-    }
-
     #[napi(catch_unwind)]
     pub async fn output_schema(&self) -> napi::Result<Buffer> {
         let schema = self.inner.output_schema().await.default_error()?;

nodejs/src/scannable.rs

@@ -1,253 +0,0 @@
-// SPDX-License-Identifier: Apache-2.0
-// SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-//! NodeJS binding for the [`lancedb::data::scannable::Scannable`] trait.
-//!
-//! The JS side supplies a `getNextBatch(isStart)` callback that returns the
-//! next Arrow `RecordBatch` encoded as a self-contained Arrow IPC Stream
-//! message (schema message + record batch message + EOS marker) wrapped in a
-//! `Buffer`, or `null` when the stream is exhausted. The Rust side parses
-//! each buffer with `arrow_ipc::reader::StreamReader`, validates every
-//! standalone batch stream against the declared schema, and yields decoded
-//! `RecordBatch`es as a [`SendableRecordBatchStream`].
-//!
-//! `isStart` is `true` on the first `getNextBatch` call of each new
-//! `scan_as_stream` and `false` thereafter. JS uses it to drop any cached
-//! iterator and re-invoke its factory at scan boundaries, so retries
-//! triggered by mid-stream failures restart at batch 0.
-
-use std::io::Cursor;
-use std::sync::Arc;
-
-use arrow_array::RecordBatch;
-use arrow_ipc::reader::StreamReader;
-use arrow_schema::SchemaRef;
-use futures::stream::once;
-use lancedb::arrow::{SendableRecordBatchStream, SimpleRecordBatchStream};
-use lancedb::data::scannable::Scannable as LanceScannable;
-use lancedb::ipc::ipc_file_to_schema;
-use lancedb::{Error, Result as LanceResult};
-use napi::bindgen_prelude::*;
-use napi::threadsafe_function::ThreadsafeFunction;
-use napi_derive::napi;
-
-/// Threadsafe handle to the JS `getNextBatch` callback. The callback takes a
-/// single boolean `isStart` (`true` on the first call of each new scan) and
-/// returns a Promise that resolves to a `Buffer` containing one IPC Stream
-/// message, or `null` at end-of-stream.
-type GetNextBatchFn = ThreadsafeFunction<bool, Promise<Option<Buffer>>, bool, Status, false>;
-
-/// A Rust-side view of a JS-constructed `Scannable`.
-///
-/// Held in JS as the return value of the `Scannable` class constructor. When
-/// passed to a consumer that accepts `impl lancedb::data::scannable::Scannable`,
-/// the consumer invokes `scan_as_stream()` to pull batches through the JS
-/// callback.
-#[napi]
-pub struct NapiScannable {
-    schema: SchemaRef,
-    num_rows: Option<usize>,
-    rescannable: bool,
-    // `ThreadsafeFunction` is not `Clone`; wrap in `Arc` so the stream
-    // returned by `scan_as_stream` can own a handle independent of `self`.
-    get_next_batch: Arc<GetNextBatchFn>,
-    // Tracks whether a scan has already started; used to enforce one-shot
-    // semantics on non-rescannable sources.
-    scanned: bool,
-}
-
-#[napi]
-impl NapiScannable {
-    /// Construct a new `NapiScannable`.
-    ///
-    /// - `schema_buf` — Arrow IPC File buffer carrying only the schema (no batches).
-    /// - `num_rows` — optional row count hint; not validated against the stream.
-    /// - `rescannable` — whether `get_next_batch` may be re-driven after the
-    ///   scan completes.
-    /// - `get_next_batch` -- JS callback that yields the next batch as an Arrow
-    ///   IPC Stream message wrapped in a `Buffer`, or `null` at EOF. The
-    ///   `isStart` argument is `true` on the first call of each new scan;
-    ///   JS uses it to discard any cached iterator before pulling.
-    #[napi(constructor)]
-    pub fn new(
-        schema_buf: Buffer,
-        num_rows: Option<i64>,
-        rescannable: bool,
-        get_next_batch: Function<bool, Promise<Option<Buffer>>>,
-    ) -> napi::Result<Self> {
-        let schema = ipc_file_to_schema(schema_buf.to_vec())
-            .map_err(|e| napi::Error::from_reason(format!("Invalid schema buffer: {}", e)))?;
-        let num_rows = num_rows
-            .map(|n| {
-                usize::try_from(n)
-                    .map_err(|_| napi::Error::from_reason("num_rows must be non-negative"))
-            })
-            .transpose()?;
-        let get_next_batch = Arc::new(get_next_batch.build_threadsafe_function().build()?);
-        Ok(Self {
-            schema,
-            num_rows,
-            rescannable,
-            get_next_batch,
-            scanned: false,
-        })
-    }
-}
-
-impl std::fmt::Debug for NapiScannable {
-    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
-        f.debug_struct("NapiScannable")
-            .field("schema", &self.schema)
-            .field("num_rows", &self.num_rows)
-            .field("rescannable", &self.rescannable)
-            .finish()
-    }
-}
-
-impl LanceScannable for NapiScannable {
-    fn schema(&self) -> SchemaRef {
-        self.schema.clone()
-    }
-
-    fn scan_as_stream(&mut self) -> SendableRecordBatchStream {
-        let schema = self.schema.clone();
-
-        // One-shot enforcement for non-rescannable sources: return a stream
-        // whose first item is an error.
-        if self.scanned && !self.rescannable {
-            let err_stream = once(async {
-                Err(Error::InvalidInput {
-                    message: "Scannable has already been consumed (non-rescannable source)"
-                        .to_string(),
-                })
-            });
-            return Box::pin(SimpleRecordBatchStream::new(err_stream, schema));
-        }
-        self.scanned = true;
-
-        let tsfn = Arc::clone(&self.get_next_batch);
-        let declared_schema = schema.clone();
-
-        // State threaded through the unfold. `is_first_pull` starts true so
-        // the first call into JS signals a new-scan boundary; JS uses it to
-        // reset any cached iterator before factory()-ing a fresh one.
-        let initial = State {
-            tsfn,
-            batch_index: 0,
-            declared_schema,
-            errored: false,
-            is_first_pull: true,
-        };
-
-        let stream = futures::stream::unfold(initial, |mut state| async move {
-            if state.errored {
-                return None;
-            }
-
-            // Pull the next IPC Stream buffer from JS. `is_first_pull` is
-            // consumed here and cleared so subsequent pulls continue the
-            // same scan rather than restarting it.
-            let is_start = state.is_first_pull;
-            state.is_first_pull = false;
-            let buf = match pull_next(&state.tsfn, is_start).await {
-                Ok(Some(buf)) => buf,
-                Ok(None) => return None,
-                Err(e) => {
-                    state.errored = true;
-                    return Some((Err(e), state));
-                }
-            };
-
-            match decode_one_batch(buf.as_ref(), &state.declared_schema) {
-                Ok(batch) => {
-                    state.batch_index += 1;
-                    Some((Ok(batch), state))
-                }
-                Err(e) => {
-                    let tagged = Error::Runtime {
-                        message: format!(
-                            "[scannable/rust-bridge] failure at batch index {}: {}",
-                            state.batch_index, e
-                        ),
-                    };
-                    state.errored = true;
-                    Some((Err(tagged), state))
-                }
-            }
-        });
-
-        Box::pin(SimpleRecordBatchStream::new(stream, schema))
-    }
-
-    fn num_rows(&self) -> Option<usize> {
-        self.num_rows
-    }
-
-    fn rescannable(&self) -> bool {
-        self.rescannable
-    }
-}
-
-struct State {
-    tsfn: Arc<GetNextBatchFn>,
-    batch_index: usize,
-    declared_schema: SchemaRef,
-    errored: bool,
-    /// True for the very first pull of a new scan. Forwarded to JS so the
-    /// callback can drop any cached iterator and call its factory fresh,
-    /// which makes rescannable sources restart at batch 0 even when the
-    /// previous scan ended mid-stream.
-    is_first_pull: bool,
-}
-
-/// Invoke the JS callback and await its Promise. `is_start` is forwarded to
-/// the JS side as the `isStart` argument so it can reset its iterator at the
-/// scan boundary. Errors on the JS side surface here as rejected promises
-/// and are tunneled back as `lancedb::Error::Runtime`.
-async fn pull_next(tsfn: &GetNextBatchFn, is_start: bool) -> LanceResult<Option<Buffer>> {
-    let promise = tsfn
-        .call_async(is_start)
-        .await
-        .map_err(|e| Error::Runtime {
-            message: format!(
-                "[scannable/js-factory] napi error status={}, reason={}",
-                e.status, e.reason
-            ),
-        })?;
-    promise.await.map_err(|e| Error::Runtime {
-        message: format!(
-            "[scannable/js-iterator] napi error status={}, reason={}",
-            e.status, e.reason
-        ),
-    })
-}
-
-/// Decode one IPC Stream buffer (schema + batch + EOS) into a `RecordBatch`.
-/// Each buffer is a standalone IPC stream, so every decoded stream schema must
-/// match the one declared at construction.
-fn decode_one_batch(buf: &[u8], declared: &SchemaRef) -> LanceResult<RecordBatch> {
-    let reader = StreamReader::try_new(Cursor::new(buf), None).map_err(|e| Error::Runtime {
-        message: format!("failed to open IPC stream reader: {}", e),
-    })?;
-
-    let actual = reader.schema();
-    if actual.as_ref() != declared.as_ref() {
-        return Err(Error::InvalidInput {
-            message: format!(
-                "declared schema does not match stream schema: declared={:?} actual={:?}",
-                declared, actual
-            ),
-        });
-    }
-
-    let mut iter = reader;
-    let batch = iter
-        .next()
-        .ok_or_else(|| Error::Runtime {
-            message: "IPC stream contained schema but no record batch".to_string(),
-        })?
-        .map_err(|e| Error::Runtime {
-            message: format!("failed to decode record batch: {}", e),
-        })?;
-    Ok(batch)
-}

nodejs/src/table.rs

@@ -9,7 +9,6 @@ use lancedb::table::{
     OptimizeAction, OptimizeOptions, Table as LanceDbTable,
 };
 use napi::bindgen_prelude::*;
-use napi::threadsafe_function::{ThreadsafeFunction, ThreadsafeFunctionCallMode};
 use napi_derive::napi;
 
 use crate::error::NapiErrorExt;
@@ -68,16 +67,8 @@ impl Table {
         schema_to_buffer(&schema)
     }
 
-    #[napi(
-        catch_unwind,
-        ts_args_type = "buf: Buffer, mode: string, progressCallback?: (progress: WriteProgressInfo) => void"
-    )]
-    pub async fn add(
-        &self,
-        buf: Buffer,
-        mode: String,
-        progress_callback: Option<ProgressFn>,
-    ) -> napi::Result<AddResult> {
+    #[napi(catch_unwind)]
+    pub async fn add(&self, buf: Buffer, mode: String) -> 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
@@ -101,19 +92,6 @@ impl Table {
             return Err(napi::Error::from_reason(format!("Invalid mode: {}", mode)));
         };
 
-        if let Some(tsfn) = progress_callback {
-            op = op.progress(move |p| {
-                // NonBlocking: dispatch onto the JS event loop without
-                // blocking the writer thread.  With napi-rs's default
-                // unbounded queue, events are not dropped — a slow JS
-                // callback will just queue them.
-                tsfn.call(
-                    WriteProgressInfo::from(p),
-                    ThreadsafeFunctionCallMode::NonBlocking,
-                );
-            });
-        }
-
         let res = op.execute().await.default_error()?;
         Ok(res.into())
     }
@@ -181,14 +159,6 @@ 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());
@@ -366,31 +336,6 @@ impl Table {
         Ok(res.into())
     }
 
-    #[napi(catch_unwind)]
-    pub async fn set_unenforced_primary_key(&self, columns: Vec<String>) -> napi::Result<()> {
-        self.inner_ref()?
-            .set_unenforced_primary_key(columns)
-            .await
-            .default_error()
-    }
-
-    #[napi(catch_unwind)]
-    pub async fn set_lsm_write_spec(&self, spec: LsmWriteSpec) -> napi::Result<()> {
-        let native_spec = lancedb::table::LsmWriteSpec::try_from(spec)?;
-        self.inner_ref()?
-            .set_lsm_write_spec(native_spec)
-            .await
-            .default_error()
-    }
-
-    #[napi(catch_unwind)]
-    pub async fn unset_lsm_write_spec(&self) -> napi::Result<()> {
-        self.inner_ref()?
-            .unset_lsm_write_spec()
-            .await
-            .default_error()
-    }
-
     #[napi(catch_unwind)]
     pub async fn version(&self) -> napi::Result<i64> {
         self.inner_ref()?
@@ -585,63 +530,6 @@ impl From<lancedb::index::IndexConfig> for IndexConfig {
     }
 }
 
-/// Specification selecting Lance's MemWAL LSM-style write path for
-/// `mergeInsert`.
-///
-/// `specType` must be `"bucket"`, `"identity"`, or `"unsharded"`. For
-/// `"bucket"`, `column` and `numBuckets` are required; for `"identity"`,
-/// `column` is required.
-#[napi(object)]
-#[derive(Clone, Debug)]
-pub struct LsmWriteSpec {
-    /// One of `"bucket"`, `"identity"`, or `"unsharded"`.
-    pub spec_type: String,
-    /// Bucket and identity variants: the sharding column.
-    pub column: Option<String>,
-    /// Bucket variant: the number of buckets, in `[1, 1024]`.
-    pub num_buckets: Option<u32>,
-    /// Names of indexes the MemWAL should keep up to date during writes.
-    pub maintained_indexes: Option<Vec<String>>,
-    /// Default `ShardWriter` configuration recorded in the MemWAL index.
-    pub writer_config_defaults: Option<HashMap<String, String>>,
-}
-
-impl TryFrom<LsmWriteSpec> for lancedb::table::LsmWriteSpec {
-    type Error = napi::Error;
-
-    fn try_from(value: LsmWriteSpec) -> napi::Result<Self> {
-        let maintained = value.maintained_indexes.unwrap_or_default();
-        let writer_config_defaults = value.writer_config_defaults.unwrap_or_default();
-        let spec = match value.spec_type.as_str() {
-            "bucket" => {
-                let column = value.column.ok_or_else(|| {
-                    napi::Error::from_reason("LsmWriteSpec bucket requires `column`")
-                })?;
-                let num_buckets = value.num_buckets.ok_or_else(|| {
-                    napi::Error::from_reason("LsmWriteSpec bucket requires `numBuckets`")
-                })?;
-                Self::bucket(column, num_buckets)
-            }
-            "identity" => {
-                let column = value.column.ok_or_else(|| {
-                    napi::Error::from_reason("LsmWriteSpec identity requires `column`")
-                })?;
-                Self::identity(column)
-            }
-            "unsharded" => Self::unsharded(),
-            other => {
-                return Err(napi::Error::from_reason(format!(
-                    "LsmWriteSpec `specType` must be 'bucket', 'identity', or 'unsharded', got '{}'",
-                    other
-                )));
-            }
-        };
-        Ok(spec
-            .with_maintained_indexes(maintained)
-            .with_writer_config_defaults(writer_config_defaults))
-    }
-}
-
 /// Statistics about a compaction operation.
 #[napi(object)]
 #[derive(Clone, Debug)]
@@ -676,44 +564,6 @@ pub struct OptimizeStats {
     pub prune: RemovalStats,
 }
 
-/// Progress snapshot for a write operation, delivered to the JS callback
-/// passed to `Table.add`.
-#[napi(object)]
-#[derive(Clone, Debug)]
-pub struct WriteProgressInfo {
-    /// Number of rows written so far.
-    pub output_rows: i64,
-    /// Number of bytes written so far.
-    pub output_bytes: i64,
-    /// Total rows expected, if the input source reports it.
-    /// Always set on the final callback (where `done` is `true`).
-    pub total_rows: Option<i64>,
-    /// Wall-clock seconds since monitoring started.
-    pub elapsed_seconds: f64,
-    /// Number of parallel write tasks currently in flight.
-    pub active_tasks: i64,
-    /// Total number of parallel write tasks (the write parallelism).
-    pub total_tasks: i64,
-    /// `true` for the final callback; `false` otherwise.
-    pub done: bool,
-}
-
-impl From<&lancedb::table::write_progress::WriteProgress> for WriteProgressInfo {
-    fn from(p: &lancedb::table::write_progress::WriteProgress) -> Self {
-        Self {
-            output_rows: p.output_rows() as i64,
-            output_bytes: p.output_bytes() as i64,
-            total_rows: p.total_rows().map(|n| n as i64),
-            elapsed_seconds: p.elapsed().as_secs_f64(),
-            active_tasks: p.active_tasks() as i64,
-            total_tasks: p.total_tasks() as i64,
-            done: p.done(),
-        }
-    }
-}
-
-type ProgressFn = ThreadsafeFunction<WriteProgressInfo, (), WriteProgressInfo, Status, false>;
-
 ///  A definition of a column alteration. The alteration changes the column at
 /// `path` to have the new name `name`, to be nullable if `nullable` is true,
 /// and to have the data type `data_type`. At least one of `rename` or `nullable`

python/.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.33.0-beta.1"
+current_version = "0.31.0-beta.11"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

python/AGENTS.md

@@ -4,26 +4,16 @@ code is in the `src/` directory and the Python bindings are in the `lancedb/` di
 
 Common commands:
 
-* Bootstrap dev env: `uv run --extra tests --extra dev maturin develop --extras tests,dev`
 * Build: `make develop`
 * Format: `make format`
 * Lint: `make check`
 * Fix lints: `make fix`
-* Test: `uv run --extra tests pytest python/tests -vv --durations=10 -m "not slow and not s3_test"`
-* Run specific test: `uv run --extra tests pytest python/tests/<test_file>.py::<test_name> -q`
-* Doc test: `uv run --extra tests pytest --doctest-modules python/lancedb`
-
-Use the uv-managed environment declared by `uv.lock` for Python validation. Do
-not treat system `python`, global `pytest`, or missing editable-install errors
-as final blockers; bootstrap or enter the uv environment instead. `make test`
-and `make doctest` assume the development environment is already prepared.
+* Test: `make test`
+* Doc test: `make doctest`
 
 Before committing changes, run lints and then formatting.
 
-When you change the Rust code, PyO3 binding code, or see a missing/stale
-`lancedb._lancedb`, recompile the Python bindings with
-`uv run --extra tests --extra dev maturin develop --extras tests,dev` before
-running tests.
+When you change the Rust code, you will need to recompile the Python bindings: `make develop`.
 
 When you export new types from Rust to Python, you must manually update `python/lancedb/_lancedb.pyi`
 with the corresponding type hints. You can run `pyright` to check for type errors in the Python code.

python/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-python"
-version = "0.33.0-beta.1"
+version = "0.31.0-beta.11"
 publish = false
 edition.workspace = true
 description = "Python bindings for LanceDB"
@@ -19,7 +19,6 @@ arrow = { version = "58.0.0", features = ["pyarrow"] }
 async-trait = "0.1"
 bytes = "1"
 lancedb = { path = "../rust/lancedb", default-features = false }
-datafusion-common.workspace = true
 lance-core.workspace = true
 lance-namespace.workspace = true
 lance-namespace-impls.workspace = true
@@ -36,8 +35,7 @@ futures.workspace = true
 serde = "1"
 serde_json = "1"
 snafu.workspace = true
-tokio = { version = "1.40", features = ["sync", "rt-multi-thread"] }
-libc = "0.2"
+tokio = { version = "1.40", features = ["sync"] }
 
 [build-dependencies]
 pyo3-build-config = { version = "0.28", features = [

python/python/lancedb/__init__.py

@@ -7,6 +7,7 @@ import os
 from concurrent.futures import ThreadPoolExecutor
 from datetime import timedelta
 from typing import Dict, Optional, Union, Any, List
+import warnings
 
 __version__ = importlib.metadata.version("lancedb")
 
@@ -437,3 +438,13 @@ __all__ = [
     "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

@@ -12,7 +12,6 @@ from .index import (
     LabelList,
     HnswPq,
     HnswSq,
-    HnswFlat,
     FTS,
 )
 from lance_namespace import (
@@ -26,7 +25,6 @@ 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)."""
@@ -51,7 +49,7 @@ class PyExpr:
     def to_sql(self) -> str: ...
 
 def expr_col(name: str) -> PyExpr: ...
-def expr_lit(value: Union[bool, int, float, str, bytes]) -> PyExpr: ...
+def expr_lit(value: Union[bool, int, float, str]) -> PyExpr: ...
 def expr_func(name: str, args: List[PyExpr]) -> PyExpr: ...
 
 class Session:
@@ -182,7 +180,6 @@ class Table:
             IvfPq,
             HnswPq,
             HnswSq,
-            HnswFlat,
             BTree,
             Bitmap,
             LabelList,
@@ -217,9 +214,6 @@ class Table:
     async def uri(self) -> str: ...
     async def initial_storage_options(self) -> Optional[Dict[str, str]]: ...
     async def latest_storage_options(self) -> Optional[Dict[str, str]]: ...
-    async def set_unenforced_primary_key(self, columns: List[str]) -> None: ...
-    async def set_lsm_write_spec(self, spec: LsmWriteSpec) -> None: ...
-    async def unset_lsm_write_spec(self) -> None: ...
     @property
     def tags(self) -> Tags: ...
     def query(self) -> Query: ...
@@ -258,11 +252,6 @@ class RecordBatchStream:
     def __aiter__(self) -> "RecordBatchStream": ...
     async def __anext__(self) -> pa.RecordBatch: ...
 
-class ColumnOrdering(TypedDict):
-    column_name: str
-    ascending: bool
-    nulls_first: bool
-
 class Query:
     def where(self, filter: str): ...
     def where_expr(self, expr: PyExpr): ...
@@ -276,7 +265,6 @@ class Query:
     def postfilter(self): ...
     def nearest_to(self, query_vec: pa.Array) -> VectorQuery: ...
     def nearest_to_text(self, query: dict) -> FTSQuery: ...
-    def order_by(self, ordering: Optional[List[ColumnOrdering]]): ...
     async def output_schema(self) -> pa.Schema: ...
     async def execute(
         self, max_batch_length: Optional[int], timeout: Optional[timedelta]
@@ -305,7 +293,6 @@ class FTSQuery:
     def get_query(self) -> str: ...
     def add_query_vector(self, query_vec: pa.Array) -> None: ...
     def nearest_to(self, query_vec: pa.Array) -> HybridQuery: ...
-    def order_by(self, ordering: Optional[List[ColumnOrdering]]): ...
     async def output_schema(self) -> pa.Schema: ...
     async def execute(
         self, max_batch_length: Optional[int], timeout: Optional[timedelta]
@@ -331,7 +318,6 @@ class VectorQuery:
     def maximum_nprobes(self, maximum_nprobes: int): ...
     def bypass_vector_index(self): ...
     def nearest_to_text(self, query: dict) -> HybridQuery: ...
-    def order_by(self, ordering: Optional[List[ColumnOrdering]]): ...
     def to_query_request(self) -> PyQueryRequest: ...
 
 class HybridQuery:
@@ -350,7 +336,6 @@ class HybridQuery:
     def minimum_nprobes(self, minimum_nprobes: int): ...
     def maximum_nprobes(self, maximum_nprobes: int): ...
     def bypass_vector_index(self): ...
-    def order_by(self, ordering: Optional[List[ColumnOrdering]]): ...
     def to_vector_query(self) -> VectorQuery: ...
     def to_fts_query(self) -> FTSQuery: ...
     def get_limit(self) -> int: ...
@@ -380,7 +365,6 @@ class PyQueryRequest:
     bypass_vector_index: Optional[bool]
     postfilter: Optional[bool]
     norm: Optional[str]
-    order_by: Optional[List[ColumnOrdering]]
 
 class CompactionStats:
     fragments_removed: int
@@ -421,37 +405,6 @@ class MergeResult:
     num_deleted_rows: int
     num_attempts: int
 
-class LsmWriteSpec:
-    """Specification selecting Lance's MemWAL LSM-style write path for
-    `merge_insert`."""
-
-    @staticmethod
-    def bucket(column: str, num_buckets: int) -> "LsmWriteSpec": ...
-    @staticmethod
-    def identity(column: str) -> "LsmWriteSpec": ...
-    @staticmethod
-    def unsharded() -> "LsmWriteSpec": ...
-    def with_maintained_indexes(self, indexes: List[str]) -> "LsmWriteSpec":
-        """Return a copy of this spec asking the MemWAL to keep the named
-        indexes up to date as rows are appended."""
-        ...
-    def with_writer_config_defaults(self, defaults: Dict[str, str]) -> "LsmWriteSpec":
-        """Return a copy of this spec recording the given default
-        `ShardWriter` configuration in the MemWAL index."""
-        ...
-    @property
-    def spec_type(self) -> str:
-        """One of 'bucket', 'identity', or 'unsharded'."""
-        ...
-    @property
-    def column(self) -> Optional[str]: ...
-    @property
-    def num_buckets(self) -> Optional[int]: ...
-    @property
-    def maintained_indexes(self) -> List[str]: ...
-    @property
-    def writer_config_defaults(self) -> Dict[str, str]: ...
-
 class AddColumnsResult:
     version: int
 
@@ -489,7 +442,7 @@ class AsyncPermutationBuilder:
     async def execute(self) -> Table: ...
 
 def async_permutation_builder(
-    table: Table,
+    table: Table, dest_table_name: str
 ) -> AsyncPermutationBuilder: ...
 def fts_query_to_json(query: Any) -> str: ...
 

python/python/lancedb/background_loop.py

@@ -2,9 +2,7 @@
 # SPDX-FileCopyrightText: Copyright The LanceDB Authors
 
 import asyncio
-import os
 import threading
-import warnings
 
 
 class BackgroundEventLoop:
@@ -15,9 +13,6 @@ 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,
@@ -36,30 +31,3 @@ 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/db.py

@@ -8,17 +8,7 @@ from abc import abstractmethod
 from datetime import timedelta
 from pathlib import Path
 import sys
-from typing import (
-    TYPE_CHECKING,
-    Any,
-    Dict,
-    Generator,
-    Iterable,
-    List,
-    Literal,
-    Optional,
-    Union,
-)
+from typing import TYPE_CHECKING, Any, Dict, Iterable, List, Literal, Optional, Union
 
 if sys.version_info >= (3, 12):
     from typing import override
@@ -323,7 +313,7 @@ class DBConnection(EnforceOverrides):
         >>> data = [{"vector": [1.1, 1.2], "lat": 45.5, "long": -122.7},
         ...         {"vector": [0.2, 1.8], "lat": 40.1, "long":  -74.1}]
         >>> db.create_table("my_table", data)
-        LanceTable(name='my_table', ...)
+        LanceTable(name='my_table', version=1, ...)
         >>> db["my_table"].head()
         pyarrow.Table
         vector: fixed_size_list<item: float>[2]
@@ -344,7 +334,7 @@ class DBConnection(EnforceOverrides):
         ...    "long": [-122.7, -74.1]
         ... })
         >>> db.create_table("table2", data)
-        LanceTable(name='table2', ...)
+        LanceTable(name='table2', version=1, ...)
         >>> db["table2"].head()
         pyarrow.Table
         vector: fixed_size_list<item: float>[2]
@@ -367,7 +357,7 @@ class DBConnection(EnforceOverrides):
         ...   pa.field("long", pa.float32())
         ... ])
         >>> db.create_table("table3", data, schema = custom_schema)
-        LanceTable(name='table3', ...)
+        LanceTable(name='table3', version=1, ...)
         >>> db["table3"].head()
         pyarrow.Table
         vector: fixed_size_list<item: float>[2]
@@ -401,7 +391,7 @@ class DBConnection(EnforceOverrides):
         ...     pa.field("price", pa.float32()),
         ... ])
         >>> db.create_table("table4", make_batches(), schema=schema)
-        LanceTable(name='table4', ...)
+        LanceTable(name='table4', version=1, ...)
 
         """
         raise NotImplementedError
@@ -578,15 +568,15 @@ class LanceDBConnection(DBConnection):
     >>> db = lancedb.connect("./.lancedb")
     >>> db.create_table("my_table", data=[{"vector": [1.1, 1.2], "b": 2},
     ...                                   {"vector": [0.5, 1.3], "b": 4}])
-    LanceTable(name='my_table', ...)
+    LanceTable(name='my_table', version=1, ...)
     >>> db.create_table("another_table", data=[{"vector": [0.4, 0.4], "b": 6}])
-    LanceTable(name='another_table', ...)
+    LanceTable(name='another_table', version=1, ...)
     >>> sorted(db.table_names())
     ['another_table', 'my_table']
     >>> len(db)
     2
     >>> db["my_table"]
-    LanceTable(name='my_table', ...)
+    LanceTable(name='my_table', version=1, ...)
     >>> "my_table" in db
     True
     >>> db.drop_table("my_table")
@@ -857,20 +847,11 @@ class LanceDBConnection(DBConnection):
             )
         )
 
-    def _all_table_names(self) -> Generator[str, None, None]:
-        page_token = None
-        while True:
-            response = self.list_tables(page_token=page_token)
-            yield from response.tables
-            page_token = response.page_token
-            if not page_token:
-                return
-
     def __len__(self) -> int:
-        return sum(1 for _ in self._all_table_names())
+        return len(self.table_names())
 
     def __contains__(self, name: str) -> bool:
-        return name in self._all_table_names()
+        return name in self.table_names()
 
     @override
     def create_table(

python/python/lancedb/expr.py

@@ -63,7 +63,7 @@ def _coerce(value: "ExprLike") -> "Expr":
 
 
 # Type alias used in annotations.
-ExprLike = Union["Expr", bool, int, float, str, bytes]
+ExprLike = Union["Expr", bool, int, float, str]
 
 
 class Expr:
@@ -261,13 +261,13 @@ def col(name: str) -> Expr:
     return Expr(expr_col(name))
 
 
-def lit(value: Union[bool, int, float, str, bytes]) -> Expr:
+def lit(value: Union[bool, int, float, str]) -> Expr:
     """Create a literal (constant) value expression.
 
     Parameters
     ----------
     value:
-        A Python ``bool``, ``int``, ``float``, ``str``, or ``bytes``.
+        A Python ``bool``, ``int``, ``float``, or ``str``.
 
     Examples
     --------

python/python/lancedb/index.py

@@ -7,7 +7,6 @@ from typing import Literal, Optional
 from ._lancedb import (
     IndexConfig,
 )
-from .types import BaseTokenizerType
 
 lang_mapping = {
     "ar": "Arabic",
@@ -112,12 +111,8 @@ class FTS:
         - "simple": Splits text by whitespace and punctuation.
         - "whitespace": Split text by whitespace, but not punctuation.
         - "raw": No tokenization. The entire text is treated as a single token.
-        - "ngram": N-gram tokenizer for substring-style matching.
-        - "jieba/*": Jieba tokenizer loaded from Lance's language model home.
-        - "lindera/*": Lindera tokenizer loaded from Lance's language model home.
     language : str, default "English"
-        The language to use for stemming and stop-word removal. This is not the
-        primary way to enable CJK tokenization.
+        The language to use for tokenization.
     max_token_length : int, default 40
         The maximum token length to index. Tokens longer than this length will be
         ignored.
@@ -132,17 +127,10 @@ class FTS:
     ascii_folding : bool, default True
         Whether to fold ASCII characters. This converts accented characters to
         their ASCII equivalent. For example, "café" would be converted to "cafe".
-
-    Notes
-    -----
-    Model-backed tokenizers such as ``jieba/default`` and ``lindera/ipadic``
-    require tokenizer models in Lance's language model home. Set
-    ``LANCE_LANGUAGE_MODEL_HOME`` to override the default platform data
-    directory under ``lance/language_models``.
     """
 
     with_position: bool = False
-    base_tokenizer: BaseTokenizerType = "simple"
+    base_tokenizer: Literal["simple", "raw", "whitespace"] = "simple"
     language: str = "English"
     max_token_length: Optional[int] = 40
     lower_case: bool = True
@@ -388,98 +376,9 @@ class HnswSq:
     target_partition_size: Optional[int] = None
 
 
-@dataclass
-class HnswFlat:
-    """Describe a HNSW-FLAT index configuration.
-
-    HNSW-FLAT stands for Hierarchical Navigable Small World without quantization.
-    It stores raw vectors in the HNSW graph, providing the highest recall among
-    the IVF_HNSW family at the cost of more memory and disk space compared to
-    :class:`HnswSq` or :class:`HnswPq`.
-
-    Parameters
-    ----------
-
-    distance_type: str, default "l2"
-
-        The distance metric used to train the index.
-
-        The following distance types are available:
-
-        "l2" - Euclidean distance. This is a very common distance metric that
-        accounts for both magnitude and direction when determining the distance
-        between vectors. l2 distance has a range of [0, ∞).
-
-        "cosine" - Cosine distance.  Cosine distance is a distance metric
-        calculated from the cosine similarity between two vectors. Cosine
-        similarity is a measure of similarity between two non-zero vectors of an
-        inner product space. It is defined to equal the cosine of the angle
-        between them.  Unlike l2, the cosine distance is not affected by the
-        magnitude of the vectors.  Cosine distance has a range of [0, 2].
-
-        "dot" - Dot product. Dot distance is the dot product of two vectors. Dot
-        distance has a range of (-∞, ∞). If the vectors are normalized (i.e. their
-        l2 norm is 1), then dot distance is equivalent to the cosine distance.
-
-    num_partitions, default sqrt(num_rows)
-
-        The number of IVF partitions to create.
-
-        For HNSW, we recommend a small number of partitions. Setting this to 1
-        works well for most tables. For very large tables, training just one HNSW
-        graph will require too much memory. Each partition becomes its own HNSW
-        graph, so setting this value higher reduces the peak memory use of
-        training.
-
-    max_iterations, default 50
-
-        Max iterations to train kmeans.
-
-        When training an IVF index we use kmeans to calculate the partitions.
-        This parameter controls how many iterations of kmeans to run.
-
-    sample_rate, default 256
-
-        The rate used to calculate the number of training vectors for kmeans.
-
-    m, default 20
-
-        The number of neighbors to select for each vector in the HNSW graph.
-
-        This value controls the tradeoff between search speed and accuracy.
-        The higher the value the more accurate the search but the slower it
-        will be.
-
-    ef_construction, default 300
-
-        The number of candidates to evaluate during the construction of the HNSW
-        graph.
-
-        This value controls the tradeoff between build speed and accuracy.
-        The higher the value the more accurate the build but the slower it will
-        be.  150 to 300 is the typical range. 100 is a minimum for good quality
-        search results. In most cases, there is no benefit to setting this higher
-        than 500.  This value should be set to a value that is not less than `ef`
-        in the search phase.
-
-    target_partition_size, default is 1,048,576
-
-        The target size of each partition.
-    """
-
-    distance_type: Literal["l2", "cosine", "dot"] = "l2"
-    num_partitions: Optional[int] = None
-    max_iterations: int = 50
-    sample_rate: int = 256
-    m: int = 20
-    ef_construction: int = 300
-    target_partition_size: Optional[int] = None
-
-
 # Backwards-compatible aliases
 IvfHnswPq = HnswPq
 IvfHnswSq = HnswSq
-IvfHnswFlat = HnswFlat
 
 
 @dataclass
@@ -799,13 +698,11 @@ __all__ = [
     "IvfPq",
     "IvfHnswPq",
     "IvfHnswSq",
-    "IvfHnswFlat",
     "IvfSq",
     "IvfRq",
     "IvfFlat",
     "HnswPq",
     "HnswSq",
-    "HnswFlat",
     "IndexConfig",
     "FTS",
     "Bitmap",

python/python/lancedb/namespace_utils.py

@@ -6,44 +6,22 @@
 from typing import Optional
 
 
-_CREATE_NAMESPACE_MODES = frozenset({"create", "exist_ok", "overwrite"})
-_DROP_NAMESPACE_MODES = frozenset({"SKIP", "FAIL"})
-_DROP_NAMESPACE_BEHAVIORS = frozenset({"RESTRICT", "CASCADE"})
-
-
 def _normalize_create_namespace_mode(mode: Optional[str]) -> Optional[str]:
     """Normalize create namespace mode to lowercase (API expects lowercase)."""
     if mode is None:
         return None
-    normalized = mode.lower()
-    if normalized not in _CREATE_NAMESPACE_MODES:
-        raise ValueError(
-            f"Invalid create namespace mode {mode!r}: "
-            f"expected one of 'create', 'exist_ok', 'overwrite'"
-        )
-    return normalized
+    return mode.lower()
 
 
 def _normalize_drop_namespace_mode(mode: Optional[str]) -> Optional[str]:
     """Normalize drop namespace mode to uppercase (API expects uppercase)."""
     if mode is None:
         return None
-    normalized = mode.upper()
-    if normalized not in _DROP_NAMESPACE_MODES:
-        raise ValueError(
-            f"Invalid drop namespace mode {mode!r}: expected one of 'skip', 'fail'"
-        )
-    return normalized
+    return mode.upper()
 
 
 def _normalize_drop_namespace_behavior(behavior: Optional[str]) -> Optional[str]:
     """Normalize drop namespace behavior to uppercase (API expects uppercase)."""
     if behavior is None:
         return None
-    normalized = behavior.upper()
-    if normalized not in _DROP_NAMESPACE_BEHAVIORS:
-        raise ValueError(
-            f"Invalid drop namespace behavior {behavior!r}: "
-            f"expected one of 'restrict', 'cascade'"
-        )
-    return normalized
+    return behavior.upper()

python/python/lancedb/permutation.py

@@ -1,11 +1,10 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright The LanceDB Authors
 
-import copy
-import json
-
 from deprecation import deprecated
+from lancedb import AsyncConnection, DBConnection
 import pyarrow as pa
+import json
 
 from ._lancedb import async_permutation_builder, PermutationReader
 from .table import LanceTable
@@ -37,7 +36,10 @@ class PermutationBuilder:
     be referenced by name in the future.  If names are not provided then they can only
     be referenced by their ordinal index.  There is no requirement to name every split.
 
-    The permutation is stored in memory and will be lost when the program exits.
+    By default, the permutation will be stored in memory and will be lost when the
+    program exits.  To persist the permutation (for very large datasets or to share
+    the permutation across multiple workers) use the [persist](#persist) method to
+    create a permanent table.
     """
 
     def __init__(self, table: LanceTable):
@@ -49,6 +51,15 @@ class PermutationBuilder:
         """
         self._async = async_permutation_builder(table)
 
+    def persist(
+        self, database: Union[DBConnection, AsyncConnection], table_name: str
+    ) -> "PermutationBuilder":
+        """
+        Persist the permutation to the given database.
+        """
+        self._async.persist(database, table_name)
+        return self
+
     def split_random(
         self,
         *,
@@ -369,44 +380,20 @@ class Permutation:
 
     def __init__(
         self,
-        base_table: LanceTable,
-        permutation_table: Optional[LanceTable],
-        split: int,
+        reader: PermutationReader,
         selection: dict[str, str],
         batch_size: int,
         transform_fn: Callable[pa.RecordBatch, Any],
-        offset: Optional[int] = None,
-        limit: Optional[int] = None,
-        connection_factory: Optional[Callable[[str], LanceTable]] = None,
-        _reader: Optional[PermutationReader] = None,
     ):
         """
         Internal constructor.  Use [from_tables](#from_tables) instead.
         """
-        assert base_table is not None, "base_table is required"
+        assert reader is not None, "reader is required"
         assert selection is not None, "selection is required"
-        self.base_table = base_table
-        self.permutation_table = permutation_table
-        self.split = split
+        self.reader = reader
         self.selection = selection
         self.transform_fn = transform_fn
         self.batch_size = batch_size
-        self.offset = offset
-        self.limit = limit
-        self.connection_factory = connection_factory
-        if _reader is None:
-            _reader = LOOP.run(self._build_reader())
-        self.reader: PermutationReader = _reader
-
-    async def _build_reader(self) -> PermutationReader:
-        reader = await PermutationReader.from_tables(
-            self.base_table, self.permutation_table, self.split
-        )
-        if self.offset is not None:
-            reader = await reader.with_offset(self.offset)
-        if self.limit is not None:
-            reader = await reader.with_limit(self.limit)
-        return reader
 
     def _with_selection(self, selection: dict[str, str]) -> "Permutation":
         """
@@ -415,97 +402,21 @@ class Permutation:
         Does not validation of the selection and it replaces it entirely.  This is not
         intended for public use.
         """
-        new = copy.copy(self)
-        new.selection = selection
-        return new
+        return Permutation(self.reader, selection, self.batch_size, self.transform_fn)
+
+    def _with_reader(self, reader: PermutationReader) -> "Permutation":
+        """
+        Creates a new permutation with the given reader
+
+        This is an internal method and should not be used directly.
+        """
+        return Permutation(reader, self.selection, self.batch_size, self.transform_fn)
 
     def with_batch_size(self, batch_size: int) -> "Permutation":
         """
         Creates a new permutation with the given batch size
         """
-        new = copy.copy(self)
-        new.batch_size = batch_size
-        return new
-
-    def with_connection_factory(
-        self, connection_factory: Callable[[str], LanceTable]
-    ) -> "Permutation":
-        """
-        Creates a new permutation that will use ``connection_factory`` to reopen
-        the base table when this permutation is unpickled in a worker process.
-
-        The factory is a callable that takes a single argument — the base table
-        name — and returns a [LanceTable]. It must be picklable; the worker
-        will pickle it via standard ``pickle`` and call it to recover the base
-        table. Picklable callables in practice means top-level (module-level)
-        functions, ``functools.partial`` of such functions, or instances of
-        picklable classes implementing ``__call__``. Lambdas and closures over
-        local variables don't pickle with the default protocol.
-
-        Setting a factory is necessary when the URI alone is not enough to
-        re-open the connection — most importantly for LanceDB Cloud (``db://``)
-        connections, where ``api_key`` and ``region`` aren't recoverable from
-        the connection object after construction.
-
-        For local file or cloud-storage paths the factory is optional: if not
-        set, ``__getstate__`` falls back to capturing
-        ``(uri, storage_options, namespace_path)`` and re-opening via
-        ``lancedb.connect(uri, storage_options=...)``.
-
-        Examples
-        --------
-        Basic native (file-system path), parameterized via ``functools.partial``::
-
-            import functools, lancedb
-            from lancedb.permutation import Permutation
-
-            def open_native_table(uri: str, table_name: str):
-                return lancedb.connect(uri).open_table(table_name)
-
-            factory = functools.partial(open_native_table, "/data/lance_db")
-            permutation = Permutation.identity(
-                factory("training")
-            ).with_connection_factory(factory)
-
-        Native via :func:`lancedb.connect_namespace` (e.g. a directory- or
-        REST-backed namespace client). The factory takes the
-        implementation name and properties dict as partial-bound args so
-        the worker can rebuild the same namespace connection::
-
-            def open_via_namespace(
-                impl: str, properties: dict[str, str], table_name: str,
-            ):
-                return lancedb.connect_namespace(impl, properties).open_table(
-                    table_name,
-                )
-
-            factory = functools.partial(
-                open_via_namespace,
-                "dir",
-                {"root": "/data/lance_db"},
-            )
-
-        LanceDB Cloud, reading credentials from env vars at worker startup
-        so secrets aren't pickled into the dataset::
-
-            import os, lancedb
-
-            def open_remote_table(table_name: str):
-                db = lancedb.connect(
-                    "db://my-database",
-                    api_key=os.environ["LANCEDB_API_KEY"],
-                    region=os.environ.get("LANCEDB_REGION", "us-east-1"),
-                )
-                return db.open_table(table_name)
-
-            permutation = Permutation.identity(
-                open_remote_table("training")
-            ).with_connection_factory(open_remote_table)
-        """
-        assert connection_factory is not None, "connection_factory is required"
-        new = copy.copy(self)
-        new.connection_factory = connection_factory
-        return new
+        return Permutation(self.reader, self.selection, batch_size, self.transform_fn)
 
     @classmethod
     def identity(cls, table: LanceTable) -> "Permutation":
@@ -578,126 +489,11 @@ class Permutation:
             schema = await reader.output_schema(None)
             initial_selection = {name: name for name in schema.names}
             return cls(
-                base_table,
-                permutation_table,
-                split,
-                initial_selection,
-                DEFAULT_BATCH_SIZE,
-                Transforms.arrow2python,
-                _reader=reader,
+                reader, initial_selection, DEFAULT_BATCH_SIZE, Transforms.arrow2python
             )
 
         return LOOP.run(do_from_tables())
 
-    def __getstate__(self) -> dict[str, Any]:
-        """Build a picklable state dict for this permutation.
-
-        The base table is captured either via a user-supplied
-        ``connection_factory`` (see [with_connection_factory]) or, as a
-        fallback, by introspecting ``(uri, storage_options, namespace_path)``
-        on the connection. The permutation table — always an in-memory
-        LanceDB table — is captured as a pyarrow Table (which pickles via
-        Arrow IPC natively). The reader is dropped from the wire format;
-        ``__setstate__`` rebuilds it from the restored tables.
-        """
-        permutation_data: Optional[pa.Table] = None
-        if self.permutation_table is not None:
-            permutation_data = self.permutation_table.to_arrow()
-
-        common = {
-            "base_table_name": self.base_table.name,
-            "permutation_data": permutation_data,
-            "split": self.split,
-            "selection": self.selection,
-            "batch_size": self.batch_size,
-            "transform_fn": self.transform_fn,
-            "offset": self.offset,
-            "limit": self.limit,
-            "connection_factory": self.connection_factory,
-        }
-
-        if self.connection_factory is not None:
-            # The factory carries enough state to recover the base table on
-            # its own; we don't need to capture the URI / storage options /
-            # namespace from the existing connection.
-            return common
-
-        # URI-introspection fallback: only viable for native (OSS) connections
-        # where (uri, storage_options) is enough to reopen. Remote / cloud
-        # connections don't expose recoverable api_key / region — those users
-        # must call with_connection_factory().
-        try:
-            base_uri = self.base_table._conn.uri
-            storage_options = self.base_table._conn.storage_options
-        except AttributeError as e:
-            raise ValueError(
-                "Cannot pickle this Permutation: the base table's connection "
-                "does not expose a uri/storage_options, which usually means it "
-                "is a remote (LanceDB Cloud) connection. Call "
-                "Permutation.with_connection_factory(...) first to provide a "
-                "picklable callable that re-opens the base table from a worker "
-                "process."
-            ) from e
-
-        if base_uri.startswith("memory://"):
-            # In-memory base tables don't exist in any worker process by
-            # default, so dump the entire base table into the pickle. This
-            # can be expensive for large datasets — users with large
-            # in-memory base tables should either persist them or set a
-            # connection_factory.
-            return {
-                **common,
-                "base_table_data": self.base_table.to_arrow(),
-            }
-
-        return {
-            **common,
-            "base_table_uri": base_uri,
-            "base_table_namespace": self.base_table._namespace_path,
-            "base_table_storage_options": storage_options,
-        }
-
-    def __setstate__(self, state: dict[str, Any]) -> None:
-        from . import connect
-
-        connection_factory = state["connection_factory"]
-        if connection_factory is not None:
-            base_table = connection_factory(state["base_table_name"])
-        elif "base_table_data" in state:
-            # In-memory base table inlined into the pickle; rebuild the same
-            # way we rebuild the in-memory permutation table.
-            mem_db = connect("memory://")
-            base_table = mem_db.create_table(
-                state["base_table_name"], state["base_table_data"]
-            )
-        else:
-            base_db = connect(
-                state["base_table_uri"],
-                storage_options=state["base_table_storage_options"],
-            )
-            base_table = base_db.open_table(
-                state["base_table_name"],
-                namespace_path=state["base_table_namespace"] or None,
-            )
-
-        permutation_table: Optional[LanceTable] = None
-        if state["permutation_data"] is not None:
-            mem_db = connect("memory://")
-            permutation_table = mem_db.create_table(
-                "permutation", state["permutation_data"]
-            )
-
-        self.base_table = base_table
-        self.permutation_table = permutation_table
-        self.split = state["split"]
-        self.selection = state["selection"]
-        self.batch_size = state["batch_size"]
-        self.transform_fn = state["transform_fn"]
-        self.offset = state["offset"]
-        self.limit = state["limit"]
-        self.connection_factory = connection_factory
-        self.reader = LOOP.run(self._build_reader())
-
     @property
     def schema(self) -> pa.Schema:
         async def do_output_schema():
@@ -964,36 +760,24 @@ class Permutation:
         for expensive operations such as image decoding.
         """
         assert transform is not None, "transform is required"
-        new = copy.copy(self)
-        new.transform_fn = transform
-        return new
-
-    def take_offsets(self, offsets: list[int]) -> Any:
-        """
-        Take rows from the permutation by offset
-
-        The returned value is passed through the permutation's current transform,
-        so `with_format` and `with_transform` affect this method in the same way
-        they affect iteration.
-        """
-
-        async def do_take_offsets():
-            return await self.reader.take_offsets(offsets, selection=self.selection)
-
-        batch = LOOP.run(do_take_offsets())
-        return self.transform_fn(batch)
+        return Permutation(self.reader, self.selection, self.batch_size, transform)
 
     def __getitem__(self, index: int) -> Any:
         """
         Returns a single row from the permutation by offset
         """
-        return self.take_offsets([index])
+        return self.__getitems__([index])
 
     def __getitems__(self, indices: list[int]) -> Any:
         """
         Returns rows from the permutation by offset
         """
-        return self.take_offsets(indices)
+
+        async def do_getitems():
+            return await self.reader.take_offsets(indices, selection=self.selection)
+
+        batch = LOOP.run(do_getitems())
+        return self.transform_fn(batch)
 
     @deprecated(details="Use with_skip instead")
     def skip(self, skip: int) -> "Permutation":
@@ -1011,10 +795,12 @@ class Permutation:
         """
         Skip the first `skip` rows of the permutation
         """
-        new = copy.copy(self)
-        new.offset = skip
-        new.reader = LOOP.run(new._build_reader())
-        return new
+
+        async def do_with_skip():
+            reader = await self.reader.with_offset(skip)
+            return self._with_reader(reader)
+
+        return LOOP.run(do_with_skip())
 
     @deprecated(details="Use with_take instead")
     def take(self, limit: int) -> "Permutation":
@@ -1032,10 +818,12 @@ class Permutation:
         """
         Limit the permutation to `limit` rows (following any `skip`)
         """
-        new = copy.copy(self)
-        new.limit = limit
-        new.reader = LOOP.run(new._build_reader())
-        return new
+
+        async def do_with_take():
+            reader = await self.reader.with_limit(limit)
+            return self._with_reader(reader)
+
+        return LOOP.run(do_with_take())
 
     @deprecated(details="Use with_repeat instead")
     def repeat(self, times: int) -> "Permutation":

python/python/lancedb/query.py

@@ -3,14 +3,12 @@
 
 from __future__ import annotations
 
-import asyncio
 from abc import ABC, abstractmethod
 from concurrent.futures import ThreadPoolExecutor
-from datetime import timedelta
 from enum import Enum
+from datetime import timedelta
 from typing import (
     TYPE_CHECKING,
-    Any,
     Dict,
     List,
     Literal,
@@ -19,40 +17,41 @@ from typing import (
     Type,
     TypeVar,
     Union,
+    Any,
 )
 
+import asyncio
 import deprecation
 import numpy as np
 import pyarrow as pa
 import pyarrow.compute as pc
 import pydantic
-from typing_extensions import Annotated
 
-from lancedb._lancedb import fts_query_to_json
-from lancedb.background_loop import LOOP
 from lancedb.pydantic import PYDANTIC_VERSION
+from lancedb.background_loop import LOOP
 
 from . import __version__
 from .arrow import AsyncRecordBatchReader
 from .dependencies import pandas as pd
-from .expr import Expr
 from .rerankers.base import Reranker
 from .rerankers.rrf import RRFReranker
 from .rerankers.util import check_reranker_result
 from .util import flatten_columns
+from .expr import Expr
+from lancedb._lancedb import fts_query_to_json
+from typing_extensions import Annotated
 
 if TYPE_CHECKING:
     import sys
-
     import PIL
     import polars as pl
 
+    from ._lancedb import Query as LanceQuery
     from ._lancedb import FTSQuery as LanceFTSQuery
     from ._lancedb import HybridQuery as LanceHybridQuery
-    from ._lancedb import PyQueryRequest
-    from ._lancedb import Query as LanceQuery
-    from ._lancedb import TakeQuery as LanceTakeQuery
     from ._lancedb import VectorQuery as LanceVectorQuery
+    from ._lancedb import TakeQuery as LanceTakeQuery
+    from ._lancedb import PyQueryRequest
     from .common import VEC
     from .pydantic import LanceModel
     from .table import Table
@@ -93,12 +92,6 @@ def ensure_vector_query(
         return val
 
 
-class ColumnOrdering(pydantic.BaseModel):
-    column_name: str
-    ascending: bool = True
-    nulls_first: bool = False
-
-
 class FullTextQueryType(str, Enum):
     MATCH = "match"
     MATCH_PHRASE = "match_phrase"
@@ -511,8 +504,6 @@ class Query(pydantic.BaseModel):
     # Bypass the vector index and use a brute force search
     bypass_vector_index: Optional[bool] = None
 
-    order_by: Optional[List[ColumnOrdering]] = None
-
     @classmethod
     def from_inner(cls, req: PyQueryRequest) -> Self:
         query = cls()
@@ -533,8 +524,6 @@ class Query(pydantic.BaseModel):
         query.refine_factor = req.refine_factor
         query.bypass_vector_index = req.bypass_vector_index
         query.postfilter = req.postfilter
-        if req.order_by is not None:
-            query.order_by = [ColumnOrdering(**o) for o in req.order_by]
         if req.full_text_search is not None:
             query.full_text_query = FullTextSearchQuery(
                 columns=None,
@@ -583,22 +572,9 @@ class LanceQueryBuilder(ABC):
             If "auto", the query type is inferred based on the query.
         vector_column_name: str
             The name of the vector column to use for vector search.
-        ordering_field_name: Optional[str]
-            .. deprecated:: 0.27.0
-                Use ``order_by()`` method instead.
-        fts_columns: Optional[Union[str, List[str]]]
-            The columns to search in for full text search.
         fast_search: bool
             Skip flat search of unindexed data.
         """
-        if ordering_field_name is not None:
-            import warnings
-
-            warnings.warn(
-                "ordering_field_name is deprecated, use .order_by() method instead.",
-                DeprecationWarning,
-                stacklevel=2,
-            )
         # Check hybrid search first as it supports empty query pattern
         if query_type == "hybrid":
             # hybrid fts and vector query
@@ -695,7 +671,6 @@ class LanceQueryBuilder(ABC):
         self._text = None
         self._ef = None
         self._bypass_vector_index = None
-        self._order_by = None
 
     @deprecation.deprecated(
         deprecated_in="0.3.1",
@@ -719,7 +694,6 @@ class LanceQueryBuilder(ABC):
         flatten: Optional[Union[int, bool]] = None,
         *,
         timeout: Optional[timedelta] = None,
-        **kwargs,
     ) -> "pd.DataFrame":
         """
         Execute the query and return the results as a pandas DataFrame.
@@ -737,12 +711,9 @@ class LanceQueryBuilder(ABC):
         timeout: Optional[timedelta]
             The maximum time to wait for the query to complete.
             If None, wait indefinitely.
-        **kwargs
-            Forwarded to pyarrow.Table.to_pandas after query execution and
-            optional flattening.
         """
         tbl = flatten_columns(self.to_arrow(timeout=timeout), flatten)
-        return tbl.to_pandas(**kwargs)
+        return tbl.to_pandas()
 
     @abstractmethod
     def to_arrow(self, *, timeout: Optional[timedelta] = None) -> pa.Table:
@@ -976,24 +947,6 @@ class LanceQueryBuilder(ABC):
         """  # noqa: E501
         return self._table._explain_plan(self.to_query_object(), verbose=verbose)
 
-    def order_by(self, ordering: Optional[List[ColumnOrdering]]) -> Self:
-        """
-        Set the ordering for the results.
-
-        Parameters
-        ----------
-        ordering: Optional[List[ColumnOrdering]]
-            The ordering to use for the results.  If None, then the default ordering
-            will be used.
-
-        Returns
-        -------
-        LanceQueryBuilder
-            The LanceQueryBuilder object.
-        """
-        self._order_by = ordering
-        return self
-
     def analyze_plan(self) -> str:
         """
         Run the query and return its execution plan with runtime metrics.
@@ -1361,7 +1314,6 @@ class LanceVectorQueryBuilder(LanceQueryBuilder):
             fast_search=self._fast_search,
             ef=self._ef,
             bypass_vector_index=self._bypass_vector_index,
-            order_by=self._order_by,
         )
 
     def to_batches(
@@ -1513,9 +1465,7 @@ class LanceFtsQueryBuilder(LanceQueryBuilder):
         super().__init__(table)
         self._query = query
         self._phrase_query = False
-        # Deprecated compatibility parameter. Native FTS ordering is now
-        # configured through order_by(); LanceQueryBuilder.create emits the warning.
-        _ = ordering_field_name
+        self.ordering_field_name = ordering_field_name
         self._reranker = None
         self._fast_search = fast_search
         if isinstance(fts_columns, str):
@@ -1564,7 +1514,6 @@ class LanceFtsQueryBuilder(LanceQueryBuilder):
             ),
             offset=self._offset,
             fast_search=self._fast_search,
-            order_by=self._order_by,
         )
 
     def output_schema(self) -> pa.Schema:
@@ -1630,7 +1579,6 @@ class LanceEmptyQueryBuilder(LanceQueryBuilder):
             limit=self._limit,
             with_row_id=self._with_row_id,
             offset=self._offset,
-            order_by=self._order_by,
         )
 
     def output_schema(self) -> pa.Schema:
@@ -2357,7 +2305,6 @@ class AsyncQueryBase(object):
         self,
         flatten: Optional[Union[int, bool]] = None,
         timeout: Optional[timedelta] = None,
-        **kwargs,
     ) -> "pd.DataFrame":
         """
         Execute the query and collect the results into a pandas DataFrame.
@@ -2390,13 +2337,10 @@ class AsyncQueryBase(object):
             The maximum time to wait for the query to complete.
             If not specified, no timeout is applied. If the query does not
             complete within the specified time, an error will be raised.
-        **kwargs
-            Forwarded to pyarrow.Table.to_pandas after query execution and
-            optional flattening.
         """
         return (
             flatten_columns(await self.to_arrow(timeout=timeout), flatten)
-        ).to_pandas(**kwargs)
+        ).to_pandas()
 
     async def to_polars(
         self,
@@ -2558,27 +2502,6 @@ class AsyncStandardQuery(AsyncQueryBase):
         self._inner.offset(offset)
         return self
 
-    def order_by(self, ordering: Optional[List[ColumnOrdering]]) -> Self:
-        """
-        Set the ordering for the results.
-
-        Parameters
-        ----------
-        ordering: Optional[List[ColumnOrdering]]
-            The ordering to use for the results.  If None, then the default ordering
-            will be used.
-        """
-        if ordering is None:
-            self._inner.order_by(None)
-        else:
-            self._inner.order_by(
-                [
-                    o.model_dump() if hasattr(o, "model_dump") else o.dict()
-                    for o in ordering
-                ]
-            )
-        return self
-
     def fast_search(self) -> Self:
         """
         Skip searching un-indexed data.
@@ -3349,18 +3272,16 @@ class BaseQueryBuilder(object):
             If not specified, no timeout is applied. If the query does not
             complete within the specified time, an error will be raised.
         """
-        async_reader = LOOP.run(
-            self._inner.to_batches(max_batch_length=max_batch_length, timeout=timeout)
-        )
+        async_iter = LOOP.run(self._inner.execute(max_batch_length, timeout))
 
         def iter_sync():
             try:
                 while True:
-                    yield LOOP.run(async_reader.__anext__())
+                    yield LOOP.run(async_iter.__anext__())
             except StopAsyncIteration:
                 return
 
-        return pa.RecordBatchReader.from_batches(async_reader.schema, iter_sync())
+        return pa.RecordBatchReader.from_batches(async_iter.schema, iter_sync())
 
     def to_arrow(self, timeout: Optional[timedelta] = None) -> pa.Table:
         """
@@ -3400,7 +3321,6 @@ class BaseQueryBuilder(object):
         self,
         flatten: Optional[Union[int, bool]] = None,
         timeout: Optional[timedelta] = None,
-        **kwargs,
     ) -> "pd.DataFrame":
         """
         Execute the query and collect the results into a pandas DataFrame.
@@ -3433,11 +3353,8 @@ class BaseQueryBuilder(object):
             The maximum time to wait for the query to complete.
             If not specified, no timeout is applied. If the query does not
             complete within the specified time, an error will be raised.
-        **kwargs
-            Forwarded to pyarrow.Table.to_pandas after query execution and
-            optional flattening.
         """
-        return LOOP.run(self._inner.to_pandas(flatten, timeout, **kwargs))
+        return LOOP.run(self._inner.to_pandas(flatten, timeout))
 
     def to_polars(
         self,

python/python/lancedb/remote/table.py

@@ -14,7 +14,6 @@ from lancedb._lancedb import (
     DeleteResult,
     DropColumnsResult,
     IndexConfig,
-    LsmWriteSpec,
     MergeResult,
     UpdateResult,
 )
@@ -23,7 +22,6 @@ from lancedb.index import (
     FTS,
     BTree,
     Bitmap,
-    HnswFlat,
     HnswSq,
     IvfFlat,
     IvfPq,
@@ -40,8 +38,7 @@ from lancedb.embeddings import EmbeddingFunctionRegistry
 from lancedb.table import _normalize_progress
 
 from ..query import LanceVectorQueryBuilder, LanceQueryBuilder, LanceTakeQueryBuilder
-from ..table import AsyncTable, BlobMode, IndexStatistics, Query, Table, Tags
-from ..types import BaseTokenizerType
+from ..table import AsyncTable, IndexStatistics, Query, Table, Tags
 
 
 class RemoteTable(Table):
@@ -101,7 +98,7 @@ class RemoteTable(Table):
         """to_arrow() is not yet supported on LanceDB cloud."""
         raise NotImplementedError("to_arrow() is not yet supported on LanceDB cloud.")
 
-    def to_pandas(self, blob_mode: BlobMode = "lazy", **kwargs):
+    def to_pandas(self):
         """to_pandas() is not yet supported on LanceDB cloud."""
         raise NotImplementedError("to_pandas() is not yet supported on LanceDB cloud.")
 
@@ -170,7 +167,7 @@ class RemoteTable(Table):
         wait_timeout: Optional[timedelta] = None,
         with_position: bool = False,
         # tokenizer configs:
-        base_tokenizer: BaseTokenizerType = "simple",
+        base_tokenizer: str = "simple",
         language: str = "English",
         max_token_length: Optional[int] = 40,
         lower_case: bool = True,
@@ -287,15 +284,13 @@ class RemoteTable(Table):
             )
         elif index_type == "IVF_HNSW_SQ":
             config = HnswSq(distance_type=metric, num_partitions=num_partitions)
-        elif index_type == "IVF_HNSW_FLAT":
-            config = HnswFlat(distance_type=metric, num_partitions=num_partitions)
         elif index_type == "IVF_FLAT":
             config = IvfFlat(distance_type=metric, num_partitions=num_partitions)
         else:
             raise ValueError(
                 f"Unknown vector index type: {index_type}. Valid options are"
                 " 'IVF_FLAT', 'IVF_PQ', 'IVF_RQ', 'IVF_SQ',"
-                " 'IVF_HNSW_PQ', 'IVF_HNSW_SQ', 'IVF_HNSW_FLAT'"
+                " 'IVF_HNSW_PQ', 'IVF_HNSW_SQ'"
             )
 
         LOOP.run(
@@ -656,18 +651,6 @@ class RemoteTable(Table):
     def drop_columns(self, columns: Iterable[str]) -> DropColumnsResult:
         return LOOP.run(self._table.drop_columns(columns))
 
-    def set_unenforced_primary_key(self, columns: Union[str, Iterable[str]]) -> None:
-        """Not supported on LanceDB Cloud."""
-        return LOOP.run(self._table.set_unenforced_primary_key(columns))
-
-    def set_lsm_write_spec(self, spec: "LsmWriteSpec") -> None:
-        """Not supported on LanceDB Cloud."""
-        return LOOP.run(self._table.set_lsm_write_spec(spec))
-
-    def unset_lsm_write_spec(self) -> None:
-        """Not supported on LanceDB Cloud."""
-        return LOOP.run(self._table.unset_lsm_write_spec())
-
     def drop_index(self, index_name: str):
         return LOOP.run(self._table.drop_index(index_name))
 

python/python/lancedb/table.py

@@ -57,7 +57,6 @@ from .index import (
     LabelList,
     HnswPq,
     HnswSq,
-    HnswFlat,
     FTS,
 )
 from .merge import LanceMergeInsertBuilder
@@ -87,61 +86,6 @@ from .util import (
 )
 from .index import lang_mapping
 
-BlobMode = Literal["lazy", "bytes", "descriptions"]
-
-_MODEL_BACKED_TOKENIZER_PREFIXES = ("jieba", "lindera")
-_MODEL_BACKED_TOKENIZER_ERRORS = (
-    "unknown base tokenizer",
-    "Invalid directory path:",
-    "Failed to load Jieba",
-    "Failed to load tokenizer config",
-    "Failed to initialize default tokenizer",
-)
-
-
-def _add_unique_note(exception: BaseException, note: str) -> None:
-    existing_notes = getattr(exception, "__notes__", ()) or ()
-    message = (
-        exception.args[0]
-        if exception.args and isinstance(exception.args[0], str)
-        else ""
-    )
-    if note not in existing_notes and note not in message:
-        add_note(exception, note)
-
-
-def _is_model_backed_tokenizer(base_tokenizer: str) -> bool:
-    return any(
-        base_tokenizer == prefix or base_tokenizer.startswith(f"{prefix}/")
-        for prefix in _MODEL_BACKED_TOKENIZER_PREFIXES
-    )
-
-
-def _maybe_add_fts_error_note(
-    exception: BaseException, *, base_tokenizer: str, language: Optional[str] = None
-) -> None:
-    message = str(exception)
-    if language is not None and "not support the requested language" in message:
-        supported_langs = ", ".join(lang_mapping.values())
-        _add_unique_note(exception, f"Supported languages: {supported_langs}")
-        return
-
-    if not _is_model_backed_tokenizer(base_tokenizer):
-        return
-
-    if not any(marker in message for marker in _MODEL_BACKED_TOKENIZER_ERRORS):
-        return
-
-    _add_unique_note(
-        exception,
-        "Model-backed tokenizers such as 'jieba/default' and 'lindera/ipadic' "
-        "require tokenizer models in Lance's language model home. Set "
-        "LANCE_LANGUAGE_MODEL_HOME to override the default platform data "
-        "directory under 'lance/language_models'. Expected layouts include "
-        "'<model-home>/jieba/default/...' and "
-        "'<model-home>/lindera/ipadic/...'.",
-    )
-
 
 if TYPE_CHECKING:
     from .db import LanceDBConnection
@@ -156,7 +100,6 @@ if TYPE_CHECKING:
         AlterColumnsResult,
         DeleteResult,
         DropColumnsResult,
-        LsmWriteSpec,
         MergeResult,
         UpdateResult,
     )
@@ -762,22 +705,14 @@ class Table(ABC):
         """
         raise NotImplementedError
 
-    def to_pandas(self, blob_mode: BlobMode = "lazy", **kwargs) -> "pandas.DataFrame":
+    def to_pandas(self) -> "pandas.DataFrame":
         """Return the table as a pandas DataFrame.
 
-        Parameters
-        ----------
-        blob_mode: str, default "lazy"
-            Controls how blob columns are returned for backends that support
-            Lance blob-aware pandas conversion.
-        **kwargs
-            Forwarded to PyArrow / Lance pandas conversion.
-
         Returns
         -------
         pd.DataFrame
         """
-        return self.to_arrow().to_pandas(**kwargs)
+        return self.to_arrow().to_pandas()
 
     @abstractmethod
     def to_arrow(self) -> pa.Table:
@@ -1023,10 +958,7 @@ class Table(ABC):
         tokenizer_name: str, default "default"
             A compatibility alias for native tokenizer configs. Can be "raw",
             "default" or the 2 letter language code followed by "_stem". So
-            for english it would be "en_stem". For new native FTS indexes, use
-            ``base_tokenizer`` directly; ``tokenizer_name`` is a legacy
-            compatibility alias and does not expose model-backed tokenizer names
-            such as ``jieba/default`` or ``lindera/ipadic``.
+            for english it would be "en_stem".
         use_tantivy: bool, default False
             Deprecated legacy Tantivy parameter. Setting this to True raises an
             error.
@@ -1040,11 +972,8 @@ class Table(ABC):
             - "whitespace": Split text by whitespace, but not punctuation.
             - "raw": No tokenization. The entire text is treated as a single token.
             - "ngram": N-Gram tokenizer.
-            - "jieba/*": Jieba tokenizer loaded from Lance's language model home.
-            - "lindera/*": Lindera tokenizer loaded from Lance's language model home.
         language : str, default "English"
-            The language to use for stemming and stop-word removal. This is not
-            the primary way to enable CJK tokenization.
+            The language to use for tokenization.
         max_token_length : int, default 40
             The maximum token length to index. Tokens longer than this length will be
             ignored.
@@ -1070,13 +999,6 @@ class Table(ABC):
             The timeout to wait if indexing is asynchronous.
         name: str, optional
             The name of the index. If not provided, a default name will be generated.
-
-        Notes
-        -----
-        Model-backed tokenizers such as ``jieba/default`` and ``lindera/ipadic``
-        require tokenizer models in Lance's language model home. Set
-        ``LANCE_LANGUAGE_MODEL_HOME`` to override the default platform data
-        directory under ``lance/language_models``.
         """
         raise NotImplementedError
 
@@ -2178,7 +2100,7 @@ class LanceTable(Table):
         return LOOP.run(self._table.count_rows(filter))
 
     def __repr__(self) -> str:
-        val = f"{self.__class__.__name__}(name={self.name!r}"
+        val = f"{self.__class__.__name__}(name={self.name!r}, version={self.version}"
         if self._conn.read_consistency_interval is not None:
             val += ", read_consistency_interval={!r}".format(
                 self._conn.read_consistency_interval
@@ -2193,27 +2115,14 @@ class LanceTable(Table):
         """Return the first n rows of the table."""
         return LOOP.run(self._table.head(n))
 
-    def to_pandas(self, blob_mode: BlobMode = "lazy", **kwargs) -> "pd.DataFrame":
+    def to_pandas(self) -> "pd.DataFrame":
         """Return the table as a pandas DataFrame.
 
-        Parameters
-        ----------
-        blob_mode: str, default "lazy"
-            Controls how Lance blob columns are returned.
-        **kwargs
-            Forwarded to Lance pandas conversion.
-
         Returns
         -------
         pd.DataFrame
         """
-        if blob_mode == "lazy" and (
-            self._namespace_client is not None
-            or get_uri_scheme(self._dataset_path) == "memory"
-        ):
-            return self.to_arrow().to_pandas(**kwargs)
-
-        return self.to_lance().to_pandas(blob_mode=blob_mode, **kwargs)
+        return self.to_arrow().to_pandas()
 
     def to_arrow(self) -> pa.Table:
         """Return the table as a pyarrow Table.
@@ -2261,13 +2170,7 @@ class LanceTable(Table):
         index_cache_size: Optional[int] = None,
         num_bits: int = 8,
         index_type: Literal[
-            "IVF_FLAT",
-            "IVF_SQ",
-            "IVF_PQ",
-            "IVF_RQ",
-            "IVF_HNSW_SQ",
-            "IVF_HNSW_PQ",
-            "IVF_HNSW_FLAT",
+            "IVF_FLAT", "IVF_SQ", "IVF_PQ", "IVF_RQ", "IVF_HNSW_SQ", "IVF_HNSW_PQ"
         ] = "IVF_PQ",
         max_iterations: int = 50,
         sample_rate: int = 256,
@@ -2354,16 +2257,6 @@ class LanceTable(Table):
                 ef_construction=ef_construction,
                 target_partition_size=target_partition_size,
             )
-        elif index_type == "IVF_HNSW_FLAT":
-            config = HnswFlat(
-                distance_type=metric,
-                num_partitions=num_partitions,
-                max_iterations=max_iterations,
-                sample_rate=sample_rate,
-                m=m,
-                ef_construction=ef_construction,
-                target_partition_size=target_partition_size,
-            )
         else:
             raise ValueError(f"Unknown index type {index_type}")
 
@@ -2542,6 +2435,11 @@ class LanceTable(Table):
                 "at a time. To search over multiple text fields, create a "
                 "separate FTS index for each field."
             )
+        if "." in field_names:
+            raise ValueError(
+                "Native FTS indexes can only be created on top-level fields. "
+                f"Received nested field path: {field_names!r}."
+            )
 
         if tokenizer_name is None:
             tokenizer_configs = {
@@ -2564,22 +2462,14 @@ class LanceTable(Table):
             **tokenizer_configs,
         )
 
-        try:
-            LOOP.run(
-                self._table.create_index(
-                    field_names,
-                    replace=replace,
-                    config=config,
-                    name=name,
-                )
+        LOOP.run(
+            self._table.create_index(
+                field_names,
+                replace=replace,
+                config=config,
+                name=name,
             )
-        except (ValueError, RuntimeError) as e:
-            _maybe_add_fts_error_note(
-                e,
-                base_tokenizer=config.base_tokenizer,
-                language=config.language,
-            )
-            raise e
+        )
 
     @staticmethod
     def infer_tokenizer_configs(tokenizer_name: str) -> dict:
@@ -3282,21 +3172,6 @@ class LanceTable(Table):
     def drop_columns(self, columns: Iterable[str]) -> DropColumnsResult:
         return LOOP.run(self._table.drop_columns(columns))
 
-    def set_unenforced_primary_key(self, columns: Union[str, Iterable[str]]) -> None:
-        """Set the unenforced primary key. See
-        [`AsyncTable.set_unenforced_primary_key`][lancedb.AsyncTable.set_unenforced_primary_key]."""
-        return LOOP.run(self._table.set_unenforced_primary_key(columns))
-
-    def set_lsm_write_spec(self, spec: "LsmWriteSpec") -> None:
-        """Install an LsmWriteSpec. See
-        [`AsyncTable.set_lsm_write_spec`][lancedb.AsyncTable.set_lsm_write_spec]."""
-        return LOOP.run(self._table.set_lsm_write_spec(spec))
-
-    def unset_lsm_write_spec(self) -> None:
-        """Remove the LsmWriteSpec. See
-        [`AsyncTable.unset_lsm_write_spec`][lancedb.AsyncTable.unset_lsm_write_spec]."""
-        return LOOP.run(self._table.unset_lsm_write_spec())
-
     def uses_v2_manifest_paths(self) -> bool:
         """
         Check if the table is using the new v2 manifest paths.
@@ -3842,69 +3717,6 @@ class AsyncTable:
         Any attempt to use the table after it has been closed will raise an error."""
         return self._inner.close()
 
-    async def set_unenforced_primary_key(
-        self, columns: Union[str, Iterable[str]]
-    ) -> None:
-        """Set the unenforced primary key for this table to the given
-        ordered list of columns.
-
-        "Unenforced" means LanceDB does not check uniqueness on writes; the
-        columns are recorded in the schema as the primary key so that
-        features such as `merge_insert` can use them. Calling this again
-        replaces any previously-set primary key.
-
-        Parameters
-        ----------
-        columns : str or Iterable[str]
-            Either a single column name (single-column key) or an ordered
-            iterable of column names (composite key). Each column dtype
-            must be one of: int32, int64, utf8, large_utf8, binary,
-            large_binary, fixed_size_binary.
-        """
-        if isinstance(columns, str):
-            columns = [columns]
-        else:
-            columns = list(columns)
-        await self._inner.set_unenforced_primary_key(columns)
-
-    async def set_lsm_write_spec(self, spec: "LsmWriteSpec") -> None:
-        """Install an LsmWriteSpec on this table.
-
-        The spec selects Lance's MemWAL LSM-style write path for future
-        `merge_insert` calls. ``LsmWriteSpec`` chooses one of three sharding
-        strategies:
-
-        - ``LsmWriteSpec.bucket(column, num_buckets)`` — hash-bucket writes by
-          the single-column unenforced primary key.
-        - ``LsmWriteSpec.identity(column)`` — shard by the raw value of a
-          scalar column.
-        - ``LsmWriteSpec.unsharded()`` — route every write to a single shard.
-
-        All variants require the table to have an unenforced primary key set
-        via [`set_unenforced_primary_key`]; bucket sharding additionally
-        requires it to be the single column being bucketed.
-
-        Parameters
-        ----------
-        spec : LsmWriteSpec
-            The sharding spec to install.
-
-        Examples
-        --------
-        >>> from lancedb._lancedb import LsmWriteSpec
-        >>> # table.set_unenforced_primary_key("id")
-        >>> # table.set_lsm_write_spec(LsmWriteSpec.bucket("id", 16))
-        """
-        await self._inner.set_lsm_write_spec(spec)
-
-    async def unset_lsm_write_spec(self) -> None:
-        """Remove the LsmWriteSpec from this table.
-
-        Reverts to the standard `merge_insert` write path. Errors if no spec
-        is currently set.
-        """
-        await self._inner.unset_lsm_write_spec()
-
     @property
     def name(self) -> str:
         """The name of the table."""
@@ -3963,39 +3775,14 @@ class AsyncTable:
         """
         return AsyncQuery(self._inner.query())
 
-    async def _to_lance(self, **kwargs) -> lance.LanceDataset:
-        try:
-            import lance
-        except ImportError:
-            raise ImportError(
-                "The lance library is required to use this function. "
-                "Please install with `pip install pylance`."
-            )
-
-        return lance.dataset(
-            await self.uri(),
-            version=await self.version(),
-            storage_options=await self.latest_storage_options(),
-            **kwargs,
-        )
-
-    async def to_pandas(self, blob_mode: BlobMode = "lazy", **kwargs) -> "pd.DataFrame":
+    async def to_pandas(self) -> "pd.DataFrame":
         """Return the table as a pandas DataFrame.
 
-        Parameters
-        ----------
-        blob_mode: str, default "lazy"
-            Controls how Lance blob columns are returned.
-        **kwargs
-            Forwarded to PyArrow / Lance pandas conversion.
-
         Returns
         -------
         pd.DataFrame
         """
-        if blob_mode == "lazy":
-            return (await self.to_arrow()).to_pandas(**kwargs)
-        return (await self._to_lance()).to_pandas(blob_mode=blob_mode, **kwargs)
+        return (await self.to_arrow()).to_pandas()
 
     async def to_arrow(self) -> pa.Table:
         """Return the table as a pyarrow Table.
@@ -4012,18 +3799,7 @@ class AsyncTable:
         *,
         replace: Optional[bool] = None,
         config: Optional[
-            Union[
-                IvfFlat,
-                IvfPq,
-                IvfRq,
-                HnswPq,
-                HnswSq,
-                HnswFlat,
-                BTree,
-                Bitmap,
-                LabelList,
-                FTS,
-            ]
+            Union[IvfFlat, IvfPq, IvfRq, HnswPq, HnswSq, BTree, Bitmap, LabelList, FTS]
         ] = None,
         wait_timeout: Optional[timedelta] = None,
         name: Optional[str] = None,
@@ -4070,7 +3846,6 @@ class AsyncTable:
                     IvfRq,
                     HnswPq,
                     HnswSq,
-                    HnswFlat,
                     BTree,
                     Bitmap,
                     LabelList,
@@ -4090,13 +3865,11 @@ class AsyncTable:
                 name=name,
                 train=train,
             )
-        except (ValueError, RuntimeError) as e:
-            if isinstance(config, FTS):
-                _maybe_add_fts_error_note(
-                    e,
-                    base_tokenizer=config.base_tokenizer,
-                    language=config.language,
-                )
+        except ValueError as e:
+            if "not support the requested language" in str(e):
+                supported_langs = ", ".join(lang_mapping.values())
+                help_msg = f"Supported languages: {supported_langs}"
+                add_note(e, help_msg)
             raise e
 
     async def drop_index(self, name: str) -> None:
@@ -4634,8 +4407,6 @@ class AsyncTable:
             async_query = async_query.fast_search()
         if query.with_row_id:
             async_query = async_query.with_row_id()
-        if query.order_by:
-            async_query = async_query.order_by(query.order_by)
 
         if query.vector:
             async_query = async_query.nearest_to(query.vector).distance_range(
@@ -5243,7 +5014,6 @@ class IndexStatistics:
         "IVF_RQ",
         "IVF_HNSW_SQ",
         "IVF_HNSW_PQ",
-        "IVF_HNSW_FLAT",
         "FTS",
         "BTREE",
         "BITMAP",

python/python/lancedb/types.py

@@ -24,7 +24,6 @@ VectorIndexType = Literal[
     "IVF_PQ",
     "IVF_HNSW_SQ",
     "IVF_HNSW_PQ",
-    "IVF_HNSW_FLAT",
     "IVF_RQ",
 ]
 ScalarIndexType = Literal["BTREE", "BITMAP", "LABEL_LIST"]
@@ -32,7 +31,6 @@ IndexType = Literal[
     "IVF_PQ",
     "IVF_HNSW_PQ",
     "IVF_HNSW_SQ",
-    "IVF_HNSW_FLAT",
     "IVF_SQ",
     "FTS",
     "BTREE",
@@ -42,5 +40,4 @@ IndexType = Literal[
 ]
 
 # Tokenizer literals
-BuiltinTokenizerType = Literal["simple", "raw", "whitespace", "ngram"]
-BaseTokenizerType = BuiltinTokenizerType | str
+BaseTokenizerType = Literal["simple", "raw", "whitespace", "ngram"]

python/python/lancedb/util.py

@@ -10,7 +10,7 @@ import pathlib
 import warnings
 from datetime import date, datetime
 from functools import singledispatch
-from typing import Tuple, Union, Optional, Any, List
+from typing import Tuple, Union, Optional, Any
 from urllib.parse import urlparse
 
 import numpy as np
@@ -189,33 +189,7 @@ def flatten_columns(tbl: pa.Table, flatten: Optional[Union[int, bool]] = None):
     return tbl
 
 
-def _format_field_path(path: List[str]) -> str:
-    def format_segment(segment: str) -> str:
-        if all(char.isalnum() or char == "_" for char in segment):
-            return segment
-        return f"`{segment.replace('`', '``')}`"
-
-    return ".".join(format_segment(segment) for segment in path)
-
-
-def _iter_vector_columns(
-    field: pa.Field, path: List[str], dim: Optional[int] = None
-) -> List[str]:
-    field_path = [*path, field.name]
-    if is_vector_column(field.type):
-        vector_dim = infer_vector_column_dim(field.type)
-        if dim is None or vector_dim == dim:
-            return [_format_field_path(field_path)]
-        return []
-    if pa.types.is_struct(field.type):
-        columns = []
-        for idx in range(field.type.num_fields):
-            columns.extend(_iter_vector_columns(field.type.field(idx), field_path, dim))
-        return columns
-    return []
-
-
-def inf_vector_column_query(schema: pa.Schema, dim: Optional[int] = None) -> str:
+def inf_vector_column_query(schema: pa.Schema) -> str:
     """
     Get the vector column name
 
@@ -228,21 +202,26 @@ def inf_vector_column_query(schema: pa.Schema, dim: Optional[int] = None) -> str
     -------
     str: the vector column name.
     """
-    vector_col_names = []
-    for field in schema:
-        vector_col_names.extend(_iter_vector_columns(field, [], dim))
-    if len(vector_col_names) > 1:
-        raise ValueError(
-            "Schema has more than one vector column. "
-            "Please specify the vector column name "
-            f"for vector search. Candidates: {vector_col_names}"
-        )
-    if len(vector_col_names) == 0:
+    vector_col_name = ""
+    vector_col_count = 0
+    for field_name in schema.names:
+        field = schema.field(field_name)
+        if is_vector_column(field.type):
+            vector_col_count += 1
+            if vector_col_count > 1:
+                raise ValueError(
+                    "Schema has more than one vector column. "
+                    "Please specify the vector column name "
+                    "for vector search"
+                )
+            elif vector_col_count == 1:
+                vector_col_name = field_name
+    if vector_col_count == 0:
         raise ValueError(
             "There is no vector column in the data. "
             "Please specify the vector column name for vector search"
         )
-    return vector_col_names[0]
+    return vector_col_name
 
 
 def is_vector_column(data_type: pa.DataType) -> bool:
@@ -268,29 +247,6 @@ def is_vector_column(data_type: pa.DataType) -> bool:
     return False
 
 
-def infer_vector_column_dim(data_type: pa.DataType) -> Optional[int]:
-    if pa.types.is_fixed_size_list(data_type):
-        return data_type.list_size
-    if pa.types.is_list(data_type):
-        return infer_vector_column_dim(data_type.value_type)
-    return None
-
-
-def _query_vector_dim(query: Optional[Any]) -> Optional[int]:
-    if query is None:
-        return None
-    if isinstance(query, np.ndarray):
-        if query.ndim == 0:
-            return None
-        return query.shape[-1]
-    if isinstance(query, list) and query:
-        first = query[0]
-        if isinstance(first, (list, tuple, np.ndarray)):
-            return len(first)
-        return len(query)
-    return None
-
-
 def infer_vector_column_name(
     schema: pa.Schema,
     query_type: str,
@@ -306,9 +262,7 @@ def infer_vector_column_name(
 
     if query is not None or query_type == "hybrid":
         try:
-            vector_column_name = inf_vector_column_query(
-                schema, dim=_query_vector_dim(query)
-            )
+            vector_column_name = inf_vector_column_query(schema)
         except Exception as e:
             raise e
 

python/python/tests/models/jieba/default/dict.txt

@@ -1,8 +0,0 @@
-我们 98740 r
-都 202780 d
-有 423765 v
-光明 1219 n
-的 318825 uj
-前途 1263 n
-前 62779 f
-途 857 n

python/python/tests/models/lindera/ipadic/config.yml

@@ -1,3 +0,0 @@
-segmenter:
-  mode: "normal"
-  dictionary: "./python/tests/models/lindera/ipadic/main"

python/python/tests/test_db.py

@@ -6,7 +6,6 @@ import re
 import sys
 from datetime import timedelta
 import os
-from types import SimpleNamespace
 
 import lancedb
 import numpy as np
@@ -189,43 +188,6 @@ def test_table_names(tmp_db: lancedb.DBConnection):
     assert len(result) == 3
 
 
-def test_db_contains_and_len_include_all_table_name_pages(tmp_db: lancedb.DBConnection):
-    for idx in range(20):
-        tmp_db.create_table(f"table_{idx}", data=[{"id": idx}])
-
-    assert len(tmp_db) == 20
-    for idx in range(20):
-        assert f"table_{idx}" in tmp_db
-    assert "does_not_exist" not in tmp_db
-
-
-def test_db_contains_stops_after_matching_table_page(
-    tmp_db: lancedb.DBConnection, monkeypatch
-):
-    calls = []
-    pages = {
-        None: SimpleNamespace(tables=["table_0", "table_1"], page_token="next"),
-        "next": SimpleNamespace(tables=["table_2"], page_token=None),
-    }
-
-    def list_tables(*, page_token=None, **_kwargs):
-        calls.append(page_token)
-        return pages[page_token]
-
-    monkeypatch.setattr(tmp_db, "list_tables", list_tables)
-
-    assert "table_1" in tmp_db
-    assert calls == [None]
-
-    calls.clear()
-    assert "table_2" in tmp_db
-    assert calls == [None, "next"]
-
-    calls.clear()
-    assert len(tmp_db) == 3
-    assert calls == [None, "next"]
-
-
 @pytest.mark.asyncio
 async def test_table_names_async(tmp_path):
     db = lancedb.connect(tmp_path)
@@ -952,29 +914,6 @@ def test_local_namespace_operations(tmp_path):
     assert db.list_namespaces().namespaces == []
 
 
-def test_create_namespace_invalid_mode_raises(tmp_path):
-    """Unrecognized create namespace modes raise a clear error."""
-    db = lancedb.connect(tmp_path)
-    with pytest.raises(ValueError, match="Invalid create namespace mode"):
-        db.create_namespace(["child"], mode="frobnicate")
-
-
-def test_drop_namespace_invalid_mode_raises(tmp_path):
-    """Unrecognized drop namespace modes raise a clear error."""
-    db = lancedb.connect(tmp_path)
-    db.create_namespace(["child"])
-    with pytest.raises(ValueError, match="Invalid drop namespace mode"):
-        db.drop_namespace(["child"], mode="frobnicate")
-
-
-def test_drop_namespace_invalid_behavior_raises(tmp_path):
-    """Unrecognized drop namespace behaviors raise a clear error."""
-    db = lancedb.connect(tmp_path)
-    db.create_namespace(["child"])
-    with pytest.raises(ValueError, match="Invalid drop namespace behavior"):
-        db.drop_namespace(["child"], behavior="frobnicate")
-
-
 def test_clone_table_latest_version(tmp_path):
     """Test cloning a table with the latest version (default behavior)"""
     import os

python/python/tests/test_fts.py

@@ -15,10 +15,7 @@
 #  limitations under the License.
 import os
 import random
-import shutil
 from unittest import mock
-from pathlib import Path
-import zipfile
 
 import lancedb as ldb
 from lancedb.db import DBConnection
@@ -29,7 +26,6 @@ from lancedb.query import (
     MultiMatchQuery,
     PhraseQuery,
     BooleanQuery,
-    ColumnOrdering,
     Occur,
     LanceFtsQueryBuilder,
 )
@@ -40,8 +36,6 @@ import pytest
 import pytest_asyncio
 from utils import exception_output
 
-TEST_LANGUAGE_MODEL_HOME = Path(__file__).parent / "models"
-
 
 @pytest.fixture
 def table(tmp_path) -> ldb.table.LanceTable:
@@ -95,39 +89,6 @@ def table(tmp_path) -> ldb.table.LanceTable:
     return table
 
 
-@pytest.fixture
-def language_model_home(monkeypatch, tmp_path):
-    model_home = tmp_path / "language-models"
-    shutil.copytree(TEST_LANGUAGE_MODEL_HOME, model_home)
-    monkeypatch.setenv("LANCE_LANGUAGE_MODEL_HOME", str(model_home))
-    return model_home
-
-
-@pytest.fixture
-def lindera_ipadic(language_model_home):
-    model_path = language_model_home / "lindera" / "ipadic"
-    extracted_model = model_path / "main"
-    config_path = model_path / "config.yml"
-
-    if extracted_model.exists():
-        shutil.rmtree(extracted_model)
-
-    with zipfile.ZipFile(model_path / "main.zip", "r") as zip_ref:
-        zip_ref.extractall(model_path)
-    config_path.write_text(
-        "segmenter:\n"
-        '  mode: "normal"\n'
-        f'  dictionary: "{extracted_model.resolve().as_posix()}"\n',
-        encoding="utf-8",
-    )
-
-    try:
-        yield
-    finally:
-        if extracted_model.exists():
-            shutil.rmtree(extracted_model)
-
-
 @pytest_asyncio.fixture
 async def async_table(tmp_path) -> ldb.table.AsyncTable:
     # Use local random state to avoid affecting other tests
@@ -500,36 +461,6 @@ async def test_search_fts_specify_column_async(async_table):
         pass
 
 
-def test_search_order_by_descending(table):
-    table.create_fts_index("text")
-    rows = (
-        table.search("puppy")
-        .order_by([ColumnOrdering(column_name="count", ascending=False)])
-        .limit(20)
-        .select(["text", "count"])
-        .to_list()
-    )
-
-    for r in rows:
-        assert "puppy" in r["text"]
-    assert sorted(rows, key=lambda x: x["count"], reverse=True) == rows
-
-
-def test_search_order_by_ascending(table):
-    table.create_fts_index("text")
-    rows = (
-        table.search("puppy")
-        .order_by([ColumnOrdering(column_name="count", ascending=True)])
-        .limit(20)
-        .select(["text", "count"])
-        .to_list()
-    )
-
-    for r in rows:
-        assert "puppy" in r["text"]
-    assert sorted(rows, key=lambda x: x["count"]) == rows
-
-
 def test_create_index_from_table(tmp_path, table):
     table.create_fts_index("text")
     df = table.search("puppy").limit(5).select(["text"]).to_pandas()
@@ -563,111 +494,8 @@ def test_create_index_multiple_columns(tmp_path, table):
 
 
 def test_nested_schema(tmp_path, table):
-    table.create_fts_index("nested.text", with_position=True)
-    indices = table.list_indices()
-    assert len(indices) == 1
-    assert indices[0].index_type == "FTS"
-    assert indices[0].columns == ["nested.text"]
-
-    results = (
-        table.search("puppy", query_type="fts", fts_columns="nested.text")
-        .limit(5)
-        .to_list()
-    )
-    assert len(results) > 0
-    assert all("puppy" in row["nested"]["text"] for row in results)
-
-    results = table.search(MatchQuery("puppy", "nested.text")).limit(5).to_list()
-    assert len(results) > 0
-    assert all("puppy" in row["nested"]["text"] for row in results)
-
-    phrase_results = (
-        table.search(PhraseQuery("puppy runs", "nested.text")).limit(5).to_list()
-    )
-    assert len(phrase_results) > 0
-    assert all("puppy runs" in row["nested"]["text"] for row in phrase_results)
-
-    hybrid_results = (
-        table.search(query_type="hybrid", fts_columns="nested.text")
-        .vector([0 for _ in range(128)])
-        .text("puppy")
-        .limit(5)
-        .to_list()
-    )
-    assert len(hybrid_results) > 0
-
-
-@pytest.mark.asyncio
-async def test_nested_schema_async(async_table):
-    await async_table.create_index("nested.text", config=FTS(with_position=True))
-    indices = await async_table.list_indices()
-    assert len(indices) == 1
-    assert indices[0].index_type == "FTS"
-    assert indices[0].columns == ["nested.text"]
-
-    results = await (
-        async_table.query()
-        .nearest_to_text("puppy", columns="nested.text")
-        .limit(5)
-        .to_list()
-    )
-    assert len(results) > 0
-    assert all("puppy" in row["nested"]["text"] for row in results)
-
-    results = await (
-        async_table.query()
-        .nearest_to_text(MatchQuery("puppy", "nested.text"))
-        .limit(5)
-        .to_list()
-    )
-    assert len(results) > 0
-    assert all("puppy" in row["nested"]["text"] for row in results)
-
-    phrase_results = await (
-        async_table.query()
-        .nearest_to_text(PhraseQuery("puppy runs", "nested.text"))
-        .limit(5)
-        .to_list()
-    )
-    assert len(phrase_results) > 0
-    assert all("puppy runs" in row["nested"]["text"] for row in phrase_results)
-
-    hybrid_results = await (
-        async_table.query()
-        .nearest_to([0 for _ in range(128)])
-        .nearest_to_text("puppy", columns="nested.text")
-        .limit(5)
-        .to_list()
-    )
-    assert len(hybrid_results) > 0
-
-
-def test_nested_schema_rejects_invalid_fts_fields(tmp_path):
-    db = ldb.connect(tmp_path)
-    data = pa.table(
-        {
-            "payload": pa.array(
-                [
-                    {"text": "puppy runs", "count": 1},
-                    {"text": "car drives", "count": 2},
-                ]
-            ),
-            "vector": pa.array(
-                [[0.1, 0.1], [0.2, 0.2]],
-                type=pa.list_(pa.float32(), list_size=2),
-            ),
-        }
-    )
-    table = db.create_table("test", data=data)
-
-    with pytest.raises(ValueError, match="FTS index cannot be created.*payload"):
-        table.create_fts_index("payload")
-
-    with pytest.raises(ValueError, match="FTS index cannot be created.*count"):
-        table.create_fts_index("payload.count")
-
-    with pytest.raises(ValueError, match="Field path `payload.missing` not found"):
-        table.create_fts_index("payload.missing")
+    with pytest.raises(ValueError, match="top-level fields"):
+        table.create_fts_index("nested.text")
 
 
 def test_search_index_with_filter(table):
@@ -856,90 +684,6 @@ def test_fts_ngram(mem_db: DBConnection):
     assert set(r["text"] for r in results) == {"lance database", "lance is cool"}
 
 
-def test_fts_jieba_tokenizer(mem_db: DBConnection, language_model_home):
-    data = pa.table({"text": ["我们都有光明的前途", "光明的前途"]})
-    table = mem_db.create_table("test_jieba", data=data)
-    table.create_fts_index(
-        "text",
-        base_tokenizer="jieba/default",
-        stem=False,
-        remove_stop_words=False,
-        ascii_folding=False,
-    )
-
-    results = table.search("我们", query_type="fts").limit(10).to_list()
-    assert [row["text"] for row in results] == ["我们都有光明的前途"]
-
-
-def test_fts_jieba_missing_language_model_note(
-    mem_db: DBConnection, monkeypatch, tmp_path
-):
-    missing_root = tmp_path / "missing-language-models"
-    monkeypatch.setenv("LANCE_LANGUAGE_MODEL_HOME", str(missing_root))
-    table = mem_db.create_table(
-        "test_missing_jieba_model",
-        data=pa.table({"text": ["我们都有光明的前途"]}),
-    )
-
-    with pytest.raises((ValueError, RuntimeError)) as e:
-        table.create_fts_index(
-            "text",
-            base_tokenizer="jieba/default",
-            stem=False,
-            remove_stop_words=False,
-            ascii_folding=False,
-        )
-
-    output = exception_output(e)
-    assert "Invalid directory path:" in output
-    assert "LANCE_LANGUAGE_MODEL_HOME" in output
-    assert "jieba/default" in output
-
-
-@pytest.mark.asyncio
-async def test_fts_jieba_missing_language_model_note_async(monkeypatch, tmp_path):
-    missing_root = tmp_path / "missing-language-models"
-    monkeypatch.setenv("LANCE_LANGUAGE_MODEL_HOME", str(missing_root))
-    db = await ldb.connect_async(tmp_path / "async-db")
-    table = await db.create_table(
-        "test_missing_jieba_model_async",
-        data=pa.table({"text": ["我们都有光明的前途"]}),
-    )
-
-    with pytest.raises((ValueError, RuntimeError)) as e:
-        await table.create_index(
-            "text",
-            config=FTS(
-                base_tokenizer="jieba/default",
-                stem=False,
-                remove_stop_words=False,
-                ascii_folding=False,
-            ),
-        )
-
-    output = exception_output(e)
-    assert "Invalid directory path:" in output
-    assert "LANCE_LANGUAGE_MODEL_HOME" in output
-    assert "jieba/default" in output
-
-
-def test_fts_lindera_tokenizer(
-    mem_db: DBConnection, language_model_home, lindera_ipadic
-):
-    data = pa.table({"text": ["成田国際空港", "東京国際空港", "羽田空港"]})
-    table = mem_db.create_table("test_lindera", data=data)
-    table.create_fts_index(
-        "text",
-        base_tokenizer="lindera/ipadic",
-        stem=False,
-        remove_stop_words=False,
-        ascii_folding=False,
-    )
-
-    results = table.search("成田", query_type="fts").limit(10).to_list()
-    assert [row["text"] for row in results] == ["成田国際空港"]
-
-
 def test_fts_query_to_json():
     """Test that FTS query to_json() produces valid JSON strings with exact format."""
 

python/python/tests/test_index.py

@@ -16,13 +16,11 @@ from lancedb.index import (
     IvfSq,
     IvfHnswPq,
     IvfHnswSq,
-    IvfHnswFlat,
     IvfRq,
     Bitmap,
     LabelList,
     HnswPq,
     HnswSq,
-    HnswFlat,
     FTS,
 )
 from lancedb.table import IndexStatistics
@@ -105,46 +103,6 @@ async def test_create_scalar_index(some_table: AsyncTable):
     assert len(indices) == 0
 
 
-@pytest.mark.asyncio
-async def test_create_nested_scalar_index_lists_canonical_paths(db_async):
-    metadata_type = pa.struct(
-        [
-            pa.field("user_id", pa.int32()),
-            pa.field("user.id", pa.int32()),
-        ]
-    )
-    data = pa.Table.from_arrays(
-        [
-            pa.array([1, 2, 3], type=pa.int32()),
-            pa.array(
-                [
-                    {"user_id": 10, "user.id": 100},
-                    {"user_id": 20, "user.id": 200},
-                    {"user_id": 30, "user.id": 300},
-                ],
-                type=metadata_type,
-            ),
-        ],
-        names=["user_id", "metadata"],
-    )
-    table = await db_async.create_table("nested_scalar_index", data)
-
-    await table.create_index("user_id", config=BTree(), name="top_user_id_idx")
-    await table.create_index(
-        "metadata.user_id", config=BTree(), name="nested_user_id_idx"
-    )
-    await table.create_index(
-        "metadata.`user.id`", config=BTree(), name="escaped_user_id_idx"
-    )
-
-    columns_by_name = {
-        index.name: index.columns for index in await table.list_indices()
-    }
-    assert columns_by_name["top_user_id_idx"] == ["user_id"]
-    assert columns_by_name["nested_user_id_idx"] == ["metadata.user_id"]
-    assert columns_by_name["escaped_user_id_idx"] == ["metadata.`user.id`"]
-
-
 @pytest.mark.asyncio
 async def test_create_fixed_size_binary_index(some_table: AsyncTable):
     await some_table.create_index("fsb", config=BTree())
@@ -292,21 +250,6 @@ async def test_create_hnswpq_alias_index(some_table: AsyncTable):
     assert indices[0].index_type in {"HnswPq", "IvfHnswPq"}
 
 
-@pytest.mark.asyncio
-async def test_create_hnswflat_index(some_table: AsyncTable):
-    await some_table.create_index("vector", config=HnswFlat(num_partitions=10))
-    indices = await some_table.list_indices()
-    assert len(indices) == 1
-
-
-@pytest.mark.asyncio
-async def test_create_hnswflat_alias_index(some_table: AsyncTable):
-    await some_table.create_index("vector", config=IvfHnswFlat(num_partitions=5))
-    indices = await some_table.list_indices()
-    assert len(indices) == 1
-    assert indices[0].index_type in {"HnswFlat", "IvfHnswFlat"}
-
-
 @pytest.mark.asyncio
 async def test_create_ivfsq_index(some_table: AsyncTable):
     await some_table.create_index("vector", config=IvfSq(num_partitions=10))
@@ -352,7 +295,6 @@ def test_index_statistics_index_type_lists_all_supported_values():
         "IVF_RQ",
         "IVF_HNSW_SQ",
         "IVF_HNSW_PQ",
-        "IVF_HNSW_FLAT",
         "FTS",
         "BTREE",
         "BITMAP",

python/python/tests/test_lsm_write_spec.py

@@ -1,149 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-"""Tests for installing and clearing an LsmWriteSpec via
-`Table.set_lsm_write_spec` / `Table.unset_lsm_write_spec`.
-"""
-
-from datetime import timedelta
-
-import lancedb
-import pyarrow as pa
-import pytest
-from lancedb._lancedb import LsmWriteSpec
-
-SCHEMA = pa.schema(
-    [
-        pa.field("id", pa.utf8(), nullable=False),
-        pa.field("v", pa.int32(), nullable=False),
-    ]
-)
-
-
-def _batch(ids, vs):
-    return pa.RecordBatch.from_arrays(
-        [pa.array(ids, type=pa.utf8()), pa.array(vs, type=pa.int32())],
-        schema=SCHEMA,
-    )
-
-
-def _reader(ids, vs):
-    return pa.RecordBatchReader.from_batches(SCHEMA, [_batch(ids, vs)])
-
-
-def _make_table(tmp_path):
-    db = lancedb.connect(tmp_path, read_consistency_interval=timedelta(seconds=0))
-    table = db.create_table("t", _reader(["seed"], [0]))
-    return db, table
-
-
-def test_set_lsm_write_spec_validates(tmp_path):
-    _db, table = _make_table(tmp_path)
-
-    # No PK set yet.
-    with pytest.raises(Exception, match="primary key"):
-        table.set_lsm_write_spec(LsmWriteSpec.bucket("id", 4))
-
-    table.set_unenforced_primary_key("id")
-
-    # Column mismatch.
-    with pytest.raises(Exception, match="match"):
-        table.set_lsm_write_spec(LsmWriteSpec.bucket("v", 4))
-
-    # Out-of-range num_buckets.
-    with pytest.raises(Exception, match="num_buckets"):
-        table.set_lsm_write_spec(LsmWriteSpec.bucket("id", 0))
-    with pytest.raises(Exception, match="num_buckets"):
-        table.set_lsm_write_spec(LsmWriteSpec.bucket("id", 1025))
-
-    # Happy path then mutation rejected.
-    table.set_lsm_write_spec(LsmWriteSpec.bucket("id", 4))
-    with pytest.raises(Exception, match="mutation"):
-        table.set_lsm_write_spec(LsmWriteSpec.bucket("id", 8))
-
-
-def test_unset_lsm_write_spec(tmp_path):
-    _db, table = _make_table(tmp_path)
-
-    # unset errors when no spec is set.
-    with pytest.raises(Exception, match="no LSM write spec"):
-        table.unset_lsm_write_spec()
-
-    # Install a spec, then remove it; afterwards a fresh spec can be set.
-    table.set_unenforced_primary_key("id")
-    table.set_lsm_write_spec(LsmWriteSpec.bucket("id", 4))
-    table.unset_lsm_write_spec()
-    # A second unset errors — there is no spec left to remove.
-    with pytest.raises(Exception, match="no LSM write spec"):
-        table.unset_lsm_write_spec()
-    table.set_lsm_write_spec(LsmWriteSpec.bucket("id", 8))
-
-
-def test_set_unsharded_spec(tmp_path):
-    _db, table = _make_table(tmp_path)
-    # Lance MemWAL still requires a primary key on the dataset; Unsharded
-    # just skips per-row hashing.
-    table.set_unenforced_primary_key("id")
-    table.set_lsm_write_spec(LsmWriteSpec.unsharded())
-    table.unset_lsm_write_spec()
-
-
-def test_lsm_write_spec_repr():
-    s = LsmWriteSpec.bucket("id", 4)
-    assert s.spec_type == "bucket"
-    assert s.column == "id"
-    assert s.num_buckets == 4
-    assert s.maintained_indexes == []
-    assert "bucket" in repr(s)
-    assert "id" in repr(s)
-    assert "4" in repr(s)
-
-    u = LsmWriteSpec.unsharded()
-    assert u.spec_type == "unsharded"
-    assert u.column is None
-    assert u.num_buckets is None
-    assert "unsharded" in repr(u)
-
-
-def test_lsm_write_spec_with_maintained_indexes():
-    s = LsmWriteSpec.bucket("id", 4).with_maintained_indexes(["idx_a", "idx_b"])
-    assert s.maintained_indexes == ["idx_a", "idx_b"]
-
-
-@pytest.mark.asyncio
-async def test_async_set_unset_lsm_write_spec(tmp_path):
-    db = await lancedb.connect_async(
-        tmp_path, read_consistency_interval=timedelta(seconds=0)
-    )
-    table = await db.create_table(
-        "t",
-        pa.RecordBatchReader.from_batches(SCHEMA, [_batch(["seed"], [0])]),
-    )
-
-    await table.set_unenforced_primary_key("id")
-    await table.set_lsm_write_spec(LsmWriteSpec.bucket("id", 4))
-    await table.unset_lsm_write_spec()
-    # A second unset errors.
-    with pytest.raises(Exception, match="no LSM write spec"):
-        await table.unset_lsm_write_spec()
-
-
-def test_set_identity_spec(tmp_path):
-    _db, table = _make_table(tmp_path)
-    # Identity sharding still requires an unenforced primary key on the
-    # table; it shards by the raw value of the given column.
-    table.set_unenforced_primary_key("id")
-    table.set_lsm_write_spec(LsmWriteSpec.identity("v"))
-    table.unset_lsm_write_spec()
-
-
-def test_lsm_write_spec_identity_and_writer_config_defaults():
-    s = LsmWriteSpec.identity("v")
-    assert s.spec_type == "identity"
-    assert s.column == "v"
-    assert s.num_buckets is None
-    assert "identity" in repr(s)
-
-    s = s.with_writer_config_defaults({"durable_write": "false"})
-    assert s.writer_config_defaults == {"durable_write": "false"}
-    assert "durable_write" in repr(s)

python/python/tests/test_permutation.py

@@ -9,6 +9,21 @@ from lancedb import DBConnection, Table, connect
 from lancedb.permutation import Permutation, Permutations, permutation_builder
 
 
+def test_permutation_persistence(tmp_path):
+    db = connect(tmp_path)
+    tbl = db.create_table("test_table", pa.table({"x": range(100), "y": range(100)}))
+
+    permutation_tbl = (
+        permutation_builder(tbl).shuffle().persist(db, "test_permutation").execute()
+    )
+    assert permutation_tbl.count_rows() == 100
+
+    re_open = db.open_table("test_permutation")
+    assert re_open.count_rows() == 100
+
+    assert permutation_tbl.to_arrow() == re_open.to_arrow()
+
+
 def test_split_random_ratios(mem_db):
     """Test random splitting with ratios."""
     tbl = mem_db.create_table(
@@ -1080,29 +1095,3 @@ def test_getitems_invalid_offset(some_permutation: Permutation):
     """Test __getitems__ with an out-of-range offset raises an error."""
     with pytest.raises(Exception):
         some_permutation.__getitems__([999999])
-
-
-def test_take_offsets(some_permutation: Permutation):
-    result = some_permutation.take_offsets([0, 1, 2])
-
-    assert isinstance(result, list)
-    assert "id" in result[0]
-    assert "value" in result[0]
-    assert len(result) == 3
-
-
-def test_take_offsets_empty_identity_permutation(mem_db):
-    tbl = mem_db.create_table(
-        "test_table", pa.table({"id": range(10), "value": range(10)})
-    )
-    permutation = Permutation.identity(tbl)
-
-    result = permutation.take_offsets([])
-
-    assert result == []
-
-
-def test_take_offsets_empty_permutation(some_permutation: Permutation):
-    result = some_permutation.take_offsets([])
-
-    assert result == []

python/python/tests/test_primary_key.py

@@ -1,79 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-"""Tests for Table.set_unenforced_primary_key."""
-
-from datetime import timedelta
-
-import lancedb
-import pyarrow as pa
-import pytest
-
-
-def _empty_table(path, schema):
-    db = lancedb.connect(path, read_consistency_interval=timedelta(seconds=0))
-    return db.create_table("t", schema=schema)
-
-
-def test_set_unenforced_primary_key_accepts_string_or_one_element_list(tmp_path):
-    schema = pa.schema([pa.field("id", pa.int64(), nullable=False)])
-
-    # Bare string.
-    table = _empty_table(tmp_path / "s", schema)
-    table.set_unenforced_primary_key("id")
-
-    # One-element list.
-    table = _empty_table(tmp_path / "l", schema)
-    table.set_unenforced_primary_key(["id"])
-
-
-def test_set_unenforced_primary_key_rejects_compound_and_empty(tmp_path):
-    table = _empty_table(
-        tmp_path,
-        pa.schema(
-            [
-                pa.field("a", pa.utf8(), nullable=False),
-                pa.field("b", pa.int64(), nullable=False),
-            ]
-        ),
-    )
-    # Compound keys are not supported.
-    with pytest.raises(Exception, match="compound"):
-        table.set_unenforced_primary_key(["a", "b"])
-    # Empty input.
-    with pytest.raises(Exception, match="required"):
-        table.set_unenforced_primary_key([])
-
-
-def test_set_unenforced_primary_key_is_immutable(tmp_path):
-    table = _empty_table(
-        tmp_path,
-        pa.schema(
-            [
-                pa.field("a", pa.utf8(), nullable=False),
-                pa.field("b", pa.int64(), nullable=False),
-            ]
-        ),
-    )
-    table.set_unenforced_primary_key("a")
-    # The primary key cannot be changed or re-set once installed.
-    with pytest.raises(Exception, match="already set"):
-        table.set_unenforced_primary_key("b")
-    with pytest.raises(Exception, match="already set"):
-        table.set_unenforced_primary_key("a")
-
-
-def test_set_unenforced_primary_key_validates(tmp_path):
-    table = _empty_table(
-        tmp_path / "t", pa.schema([pa.field("id", pa.utf8(), nullable=False)])
-    )
-    # Unknown column.
-    with pytest.raises(Exception, match="not found"):
-        table.set_unenforced_primary_key("nonexistent")
-
-    # Unsupported dtype (Float32 not in the supported set).
-    bad = _empty_table(
-        tmp_path / "bad", pa.schema([pa.field("id", pa.float32(), nullable=False)])
-    )
-    with pytest.raises(Exception, match="not supported"):
-        bad.set_unenforced_primary_key("id")