Bump version: 0.25.2-beta.3 → 0.25.2

Expand the custom redirection links list to include storage page

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.22.1"
+current_version = "0.22.2-beta.2"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

.github/actions/create-failure-issue/action.yml

@@ -0,0 +1,45 @@
+name: Create Failure Issue
+description: Creates a GitHub issue if any jobs in the workflow failed
+
+inputs:
+  job-results:
+    description: 'JSON string of job results from needs context'
+    required: true
+  workflow-name:
+    description: 'Name of the workflow'
+    required: true
+
+runs:
+  using: composite
+  steps:
+    - name: Check for failures and create issue
+      shell: bash
+      env:
+        JOB_RESULTS: ${{ inputs.job-results }}
+        WORKFLOW_NAME: ${{ inputs.workflow-name }}
+        RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
+        GH_TOKEN: ${{ github.token }}
+      run: |
+        # Check if any job failed
+        if echo "$JOB_RESULTS" | jq -e 'to_entries | any(.value.result == "failure")' > /dev/null; then
+          echo "Detected job failures, creating issue..."
+
+          # Extract failed job names
+          FAILED_JOBS=$(echo "$JOB_RESULTS" | jq -r 'to_entries | map(select(.value.result == "failure")) | map(.key) | join(", ")')
+
+          # Create issue with workflow name, failed jobs, and run URL
+          gh issue create \
+            --title "$WORKFLOW_NAME Failed ($FAILED_JOBS)" \
+            --body "The workflow **$WORKFLOW_NAME** failed during execution.
+
+        **Failed jobs:** $FAILED_JOBS
+
+        **Run URL:** $RUN_URL
+
+        Please investigate the failed jobs and address any issues." \
+            --label "ci"
+
+          echo "Issue created successfully"
+        else
+          echo "No job failures detected, skipping issue creation"
+        fi

.github/workflows/cargo-publish.yml

@@ -38,3 +38,17 @@ jobs:
       - name: Publish the package
         run: |
           cargo publish -p lancedb --all-features --token ${{ steps.auth.outputs.token }}
+  report-failure:
+    name: Report Workflow Failure
+    runs-on: ubuntu-latest
+    needs: [build]
+    if: always() && (github.event_name == 'release' || github.event_name == 'workflow_dispatch')
+    permissions:
+      contents: read
+      issues: write
+    steps:
+      - uses: actions/checkout@v4
+      - uses: ./.github/actions/create-failure-issue
+        with:
+          job-results: ${{ toJSON(needs) }}
+          workflow-name: ${{ github.workflow }}

.github/workflows/docs.yml

@@ -56,8 +56,9 @@ jobs:
         with:
           node-version: 20
           cache: 'npm'
+          cache-dependency-path: docs/package-lock.json
       - name: Install node dependencies
-        working-directory: node
+        working-directory: nodejs
         run: |
           sudo apt update
           sudo apt install -y protobuf-compiler libssl-dev

.github/workflows/java-publish.yml

@@ -43,7 +43,6 @@ jobs:
       - uses: Swatinem/rust-cache@v2
       - uses: actions-rust-lang/setup-rust-toolchain@v1
         with:
-          toolchain: "1.81.0"
           cache-workspaces: "./java/core/lancedb-jni"
           # Disable full debug symbol generation to speed up CI build and keep memory down
           # "1" means line tables only, which is useful for panic tracebacks.
@@ -112,3 +111,17 @@ jobs:
         env:
           SONATYPE_USER: ${{ secrets.SONATYPE_USER }}
           SONATYPE_TOKEN: ${{ secrets.SONATYPE_TOKEN }}
+  report-failure:
+    name: Report Workflow Failure
+    runs-on: ubuntu-latest
+    needs: [linux-arm64, linux-x86, macos-arm64]
+    if: always() && (github.event_name == 'release' || github.event_name == 'workflow_dispatch')
+    permissions:
+      contents: read
+      issues: write
+    steps:
+      - uses: actions/checkout@v4
+      - uses: ./.github/actions/create-failure-issue
+        with:
+          job-results: ${{ toJSON(needs) }}
+          workflow-name: ${{ github.workflow }}

.github/workflows/nodejs.yml

@@ -6,6 +6,7 @@ on:
       - main
   pull_request:
     paths:
+      - Cargo.toml
       - nodejs/**
       - .github/workflows/nodejs.yml
       - docker-compose.yml

.github/workflows/npm-publish.yml

@@ -365,3 +365,17 @@ jobs:
             ARGS="$ARGS --tag preview"
           fi
           npm publish $ARGS
+  report-failure:
+    name: Report Workflow Failure
+    runs-on: ubuntu-latest
+    needs: [build-lancedb, test-lancedb, publish]
+    if: always() && (github.event_name == 'release' || github.event_name == 'workflow_dispatch')
+    permissions:
+      contents: read
+      issues: write
+    steps:
+      - uses: actions/checkout@v4
+      - uses: ./.github/actions/create-failure-issue
+        with:
+          job-results: ${{ toJSON(needs) }}
+          workflow-name: ${{ github.workflow }}

.github/workflows/pypi-publish.yml

@@ -173,3 +173,17 @@ jobs:
           generate_release_notes: false
           name: Python LanceDB v${{ steps.extract_version.outputs.version }}
           body: ${{ steps.python_release_notes.outputs.changelog }}
+  report-failure:
+    name: Report Workflow Failure
+    runs-on: ubuntu-latest
+    needs: [linux, mac, windows]
+    permissions:
+      contents: read
+      issues: write
+    if: always() && (github.event_name == 'release' || github.event_name == 'workflow_dispatch')
+    steps:
+      - uses: actions/checkout@v4
+      - uses: ./.github/actions/create-failure-issue
+        with:
+          job-results: ${{ toJSON(needs) }}
+          workflow-name: ${{ github.workflow }}

.github/workflows/python.yml

@@ -6,6 +6,7 @@ on:
       - main
   pull_request:
     paths:
+      - Cargo.toml
       - python/**
       - .github/workflows/python.yml
 

.github/workflows/rust.yml

@@ -96,6 +96,7 @@ jobs:
       # Need up-to-date compilers for kernels
       CC: clang-18
       CXX: clang++-18
+      GH_TOKEN: ${{ secrets.SOPHON_READ_TOKEN }}
     steps:
       - uses: actions/checkout@v4
         with:
@@ -117,15 +118,17 @@ jobs:
           sudo chmod 600 /swapfile
           sudo mkswap /swapfile
           sudo swapon /swapfile
-      - name: Start S3 integration test environment
-        working-directory: .
-        run: docker compose up --detach --wait
       - name: Build
         run: cargo build --all-features --tests --locked --examples
-      - name: Run tests
-        run: cargo test --all-features --locked
+      - name: Run feature tests
+        run: make -C ./lancedb feature-tests
       - name: Run examples
         run: cargo run --example simple --locked
+      - name: Run remote tests
+        # Running this requires access to secrets, so skip if this is
+        # a PR from a fork.
+        if: github.event_name != 'pull_request' || !github.event.pull_request.head.repo.fork
+        run: make -C ./lancedb remote-tests
 
   macos:
     timeout-minutes: 30

.github/workflows/trigger-vectordb-recipes.yml

@@ -1,26 +0,0 @@
-name: Trigger vectordb-recipers workflow
-on:
-  push:
-    branches: [ main ]
-  pull_request:
-    paths:
-      - .github/workflows/trigger-vectordb-recipes.yml
-  workflow_dispatch:
-
-jobs:
-  build:
-    runs-on: ubuntu-latest
-
-    steps:
-      - name: Trigger vectordb-recipes workflow
-        uses: actions/github-script@v6
-        with:
-          github-token: ${{ secrets.VECTORDB_RECIPES_ACTION_TOKEN }}
-          script: |
-            const result = await github.rest.actions.createWorkflowDispatch({
-                owner: 'lancedb',
-                repo: 'vectordb-recipes',
-                workflow_id: 'examples-test.yml',
-                ref: 'main'
-            });
-            console.log(result);

Cargo.toml

@@ -15,31 +15,30 @@ categories = ["database-implementations"]
 rust-version = "1.78.0"
 
 [workspace.dependencies]
-lance = { "version" = "=0.37.0", default-features = false, "features" = ["dynamodb"] }
-lance-io = { "version" = "=0.37.0", default-features = false }
-lance-index = "=0.37.0"
-lance-linalg = "=0.37.0"
-lance-table = "=0.37.0"
-lance-testing = "=0.37.0"
-lance-datafusion = "=0.37.0"
-lance-encoding = "=0.37.0"
-lance-namespace = "0.0.15"
+lance = { "version" = "=0.38.2", default-features = false, "features" = ["dynamodb"] }
+lance-io = { "version" = "=0.38.2", default-features = false }
+lance-index = "=0.38.2"
+lance-linalg = "=0.38.2"
+lance-table = "=0.38.2"
+lance-testing = "=0.38.2"
+lance-datafusion = "=0.38.2"
+lance-encoding = "=0.38.2"
+lance-namespace = "0.0.18"
 # Note that this one does not include pyarrow
-arrow = { version = "55.1", optional = false }
-arrow-array = "55.1"
-arrow-data = "55.1"
-arrow-ipc = "55.1"
-arrow-ord = "55.1"
-arrow-schema = "55.1"
-arrow-arith = "55.1"
-arrow-cast = "55.1"
+arrow = { version = "56.2", optional = false }
+arrow-array = "56.2"
+arrow-data = "56.2"
+arrow-ipc = "56.2"
+arrow-ord = "56.2"
+arrow-schema = "56.2"
+arrow-cast = "56.2"
 async-trait = "0"
-datafusion = { version = "49.0", default-features = false }
-datafusion-catalog = "49.0"
-datafusion-common = { version = "49.0", default-features = false }
-datafusion-execution = "49.0"
-datafusion-expr = "49.0"
-datafusion-physical-plan = "49.0"
+datafusion = { version = "50.1", default-features = false }
+datafusion-catalog = "50.1"
+datafusion-common = { version = "50.1", default-features = false }
+datafusion-execution = "50.1"
+datafusion-expr = "50.1"
+datafusion-physical-plan = "50.1"
 env_logger = "0.11"
 half = { "version" = "2.6.0", default-features = false, features = [
     "num-traits",
@@ -52,7 +51,6 @@ pin-project = "1.0.7"
 snafu = "0.8"
 url = "2"
 num-traits = "0.2"
-rand = "0.9"
 regex = "1.10"
 lazy_static = "1"
 semver = "1.0.25"
@@ -60,7 +58,17 @@ crunchy = "0.2.4"
 # Temporary pins to work around downstream issues
 # https://github.com/apache/arrow-rs/commit/2fddf85afcd20110ce783ed5b4cdeb82293da30b
 chrono = "=0.4.41"
-# https://github.com/RustCrypto/formats/issues/1684
-base64ct = "=1.6.0"
 # Workaround for: https://github.com/Lokathor/bytemuck/issues/306
 bytemuck_derive = ">=1.8.1, <1.9.0"
+
+# This is only needed when we reference preview releases of lance
+# [patch.crates-io]
+# # Force to use the same lance version as the rest of the project to avoid duplicate dependencies
+# lance = { "version" = "=0.38.0", "tag" = "v0.38.0", "git" = "https://github.com/lancedb/lance.git" }
+# lance-io = { "version" = "=0.38.0", "tag" = "v0.38.0", "git" = "https://github.com/lancedb/lance.git" }
+# lance-index = { "version" = "=0.38.0", "tag" = "v0.38.0", "git" = "https://github.com/lancedb/lance.git" }
+# lance-linalg = { "version" = "=0.38.0", "tag" = "v0.38.0", "git" = "https://github.com/lancedb/lance.git" }
+# lance-table = { "version" = "=0.38.0", "tag" = "v0.38.0", "git" = "https://github.com/lancedb/lance.git" }
+# lance-testing = { "version" = "=0.38.0", "tag" = "v0.38.0", "git" = "https://github.com/lancedb/lance.git" }
+# lance-datafusion = { "version" = "=0.38.0", "tag" = "v0.38.0", "git" = "https://github.com/lancedb/lance.git" }
+# lance-encoding = { "version" = "=0.38.0", "tag" = "v0.38.0", "git" = "https://github.com/lancedb/lance.git" }

ci/create_lancedb_test_connection.sh

@@ -0,0 +1,4 @@
+#!/usr/bin/env bash
+
+export RUST_LOG=info
+exec ./lancedb server --port 0 --sql-port 0  --data-dir "${1}"

ci/run_with_docker_compose.sh

@@ -0,0 +1,18 @@
+#!/usr/bin/env bash
+
+#
+# A script for running the given command together with a docker compose environment.
+#
+
+# Bring down the docker setup once the command is done running.
+tear_down() {
+    docker compose -p fixture down
+}
+trap tear_down EXIT
+
+set +xe
+
+# Clean up any existing docker setup and bring up a new one.
+docker compose -p fixture up --detach --wait || exit 1
+
+"${@}"

ci/run_with_test_connection.sh

@@ -0,0 +1,68 @@
+#!/usr/bin/env bash
+
+#
+# A script for running the given command together with the lancedb cli.
+#
+
+die() {
+    echo $?
+    exit 1
+}
+
+check_command_exists() {
+    command="${1}"
+    which ${command} &> /dev/null || \
+        die "Unable to locate command: ${command}. Did you install it?"
+}
+
+if [[ ! -e ./lancedb ]]; then
+    if [[ -v SOPHON_READ_TOKEN ]]; then
+        INPUT="lancedb-linux-x64"
+        gh release \
+            --repo lancedb/lancedb \
+            download ci-support-binaries \
+            --pattern "${INPUT}" \
+            || die "failed to fetch cli."
+        check_command_exists openssl
+        openssl enc -aes-256-cbc \
+            -d -pbkdf2 \
+            -pass "env:SOPHON_READ_TOKEN" \
+            -in "${INPUT}" \
+            -out ./lancedb-linux-x64.tar.gz \
+            || die "openssl failed"
+        TARGET="${INPUT}.tar.gz"
+    else
+        ARCH="x64"
+        if [[ $OSTYPE == 'darwin'* ]]; then
+            UNAME=$(uname -m)
+            if [[ $UNAME == 'arm64' ]]; then
+                ARCH='arm64'
+            fi
+            OSTYPE="macos"
+        elif [[ $OSTYPE == 'linux'* ]]; then
+            if [[ $UNAME == 'aarch64' ]]; then
+                ARCH='arm64'
+            fi
+            OSTYPE="linux"
+        else
+            die "unknown OSTYPE: $OSTYPE"
+        fi
+
+        check_command_exists gh
+        TARGET="lancedb-${OSTYPE}-${ARCH}.tar.gz"
+        gh release \
+            --repo lancedb/sophon \
+            download lancedb-cli-v0.0.3 \
+            --pattern "${TARGET}" \
+            || die "failed to fetch cli."
+    fi
+
+    check_command_exists tar
+    tar xvf "${TARGET}" || die "tar failed."
+    [[ -e ./lancedb ]] || die "failed to extract lancedb."
+fi
+
+SCRIPT_DIR=$(dirname "$(readlink -f "$0")")
+export CREATE_LANCEDB_TEST_CONNECTION_SCRIPT="${SCRIPT_DIR}/create_lancedb_test_connection.sh"
+
+"${@}"

ci/set_lance_version.py

@@ -117,7 +117,7 @@ def update_cargo_toml(line_updater):
     lance_line = ""
     is_parsing_lance_line = False
     for line in lines:
-        if line.startswith("lance"):
+        if line.startswith("lance") and not line.startswith("lance-namespace"):
             # Check if this is a single-line or multi-line entry
             # Single-line entries either:
             # 1. End with } (complete inline table)

docs/mkdocs.yml

@@ -70,6 +70,23 @@ plugins:
   - mkdocs-jupyter
   - render_swagger:
       allow_arbitrary_locations: true
+  - redirects:
+      redirect_maps:
+        # Redirect the home page and other top-level markdown files. This enables maximum SEO benefit
+        # other sub-pages are handled by the ingected js in overrides/partials/header.html
+        'index.md': 'https://lancedb.com/docs/'
+        'guides/tables.md': 'https://lancedb.com/docs/tables/'
+        'ann_indexes.md': 'https://lancedb.com/docs/indexing/'
+        'basic.md': 'https://lancedb.com/docs/quickstart/'
+        'faq.md': 'https://lancedb.com/docs/faq/'
+        'embeddings/understanding_embeddings.md': 'https://lancedb.com/docs/embedding/'
+        'integrations.md': 'https://lancedb.com/docs/integrations/'
+        'examples.md': 'https://lancedb.com/docs/tutorials/'
+        'concepts/vector_search.md': 'https://lancedb.com/docs/search/vector-search/'
+        'troubleshooting.md': 'https://lancedb.com/docs/troubleshooting/'
+        'guides/storage.md': 'https://lancedb.com/docs/storage/integrations'
+
+
 
 markdown_extensions:
   - admonition

docs/overrides/partials/header.html

@@ -19,7 +19,13 @@
   FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
   IN THE SOFTWARE.
 -->
-
+<div id="deprecation-banner" style="background-color: #f8d7da; color: #721c24; padding: 1em; text-align: center;">                                                           
+    <p style="margin: 0; font-size: 1.1em;">                                                                                                           
+        <strong>This documentation site is deprecated.</strong>                                                                           
+        Please visit our new documentation site at <a href="https://lancedb.com/docs" style="color: #721c24; text-decoration: underline;">
+          lancedb.com/docs</a> for the latest information.                                                                                        
+    </p>                                                                                                                                               
+  </div>  
 {% set class = "md-header" %}
 {% if "navigation.tabs.sticky" in features %}
   {% set class = class ~ " md-header--shadow md-header--lifted" %}
@@ -150,9 +156,9 @@
 
     <div style="margin-left: 10px; margin-right: 5px;">
         <a href="https://discord.com/invite/zMM32dvNtd" target="_blank" rel="noopener noreferrer">
-            <svg fill="#FFFFFF" xmlns="http://www.w3.org/2000/svg"  viewBox="0 0 50 50" width="25px" height="25px"><path d="M 41.625 10.769531 C 37.644531 7.566406 31.347656 7.023438 31.078125 7.003906 C 30.660156 6.96875 30.261719 7.203125 30.089844 7.589844 C 30.074219 7.613281 29.9375 7.929688 29.785156 8.421875 C 32.417969 8.867188 35.652344 9.761719 38.578125 11.578125 C 39.046875 11.867188 39.191406 12.484375 38.902344 12.953125 C 38.710938 13.261719 38.386719 13.429688 38.050781 13.429688 C 37.871094 13.429688 37.6875 13.378906 37.523438 13.277344 C 32.492188 10.15625 26.210938 10 25 10 C 23.789063 10 17.503906 10.15625 12.476563 13.277344 C 12.007813 13.570313 11.390625 13.425781 11.101563 12.957031 C 10.808594 12.484375 10.953125 11.871094 11.421875 11.578125 C 14.347656 9.765625 17.582031 8.867188 20.214844 8.425781 C 20.0625 7.929688 19.925781 7.617188 19.914063 7.589844 C 19.738281 7.203125 19.34375 6.960938 18.921875 7.003906 C 18.652344 7.023438 12.355469 7.566406 8.320313 10.8125 C 6.214844 12.761719 2 24.152344 2 34 C 2 34.175781 2.046875 34.34375 2.132813 34.496094 C 5.039063 39.605469 12.972656 40.941406 14.78125 41 C 14.789063 41 14.800781 41 14.8125 41 C 15.132813 41 15.433594 40.847656 15.621094 40.589844 L 17.449219 38.074219 C 12.515625 36.800781 9.996094 34.636719 9.851563 34.507813 C 9.4375 34.144531 9.398438 33.511719 9.765625 33.097656 C 10.128906 32.683594 10.761719 32.644531 11.175781 33.007813 C 11.234375 33.0625 15.875 37 25 37 C 34.140625 37 38.78125 33.046875 38.828125 33.007813 C 39.242188 32.648438 39.871094 32.683594 40.238281 33.101563 C 40.601563 33.515625 40.5625 34.144531 40.148438 34.507813 C 40.003906 34.636719 37.484375 36.800781 32.550781 38.074219 L 34.378906 40.589844 C 34.566406 40.847656 34.867188 41 35.1875 41 C 35.199219 41 35.210938 41 35.21875 41 C 37.027344 40.941406 44.960938 39.605469 47.867188 34.496094 C 47.953125 34.34375 48 34.175781 48 34 C 48 24.152344 43.785156 12.761719 41.625 10.769531 Z M 18.5 30 C 16.566406 30 15 28.210938 15 26 C 15 23.789063 16.566406 22 18.5 22 C 20.433594 22 22 23.789063 22 26 C 22 28.210938 20.433594 30 18.5 30 Z M 31.5 30 C 29.566406 30 28 28.210938 28 26 C 28 23.789063 29.566406 22 31.5 22 C 33.433594 22 35 23.789063 35 26 C 35 28.210938 33.433594 30 31.5 30 Z"/></svg>
-        </a>
-    </div>
+          <svg fill="#FFFFFF" xmlns="http://www.w3.org/2000/svg"  viewBox="0 0 50 50" width="25px" height="25px"><path d="M 41.625 10.769531 C 37.644531 7.566406 31.347656 7.023438 31.078125 7.003906 C 30.660156 6.96875 30.261719 7.203125 30.089844 7.589844 C 30.074219 7.613281 29.9375 7.929688 29.785156 8.421875 C 32.417969 8.867188 35.652344 9.761719 38.578125 11.578125 C 39.046875 11.867188 39.191406 12.484375 38.902344 12.953125 C 38.710938 13.261719 38.386719 13.429688 38.050781 13.429688 C 37.871094 13.429688 37.6875 13.378906 37.523438 13.277344 C 32.492188 10.15625 26.210938 10 25 10 C 23.789063 10 17.503906 10.15625 12.476563 13.277344 C 12.007813 13.570313 11.390625 13.425781 11.101563 12.957031 C 10.808594 12.484375 10.953125 11.871094 11.421875 11.578125 C 14.347656 9.765625 17.582031 8.867188 20.214844 8.425781 C 20.0625 7.929688 19.925781 7.617188 19.914063 7.589844 C 19.738281 7.203125 19.34375 6.960938 18.921875 7.003906 C 18.652344 7.023438 12.355469 7.566406 8.320313 10.8125 C 6.214844 12.761719 2 24.152344 2 34 C 2 34.175781 2.046875 34.34375 2.132813 34.496094 C 5.039063 39.605469 12.972656 40.941406 14.78125 41 C 14.789063 41 14.800781 41 14.8125 41 C 15.132813 41 15.433594 40.847656 15.621094 40.589844 L 17.449219 38.074219 C 12.515625 36.800781 9.996094 34.636719 9.851563 34.507813 C 9.4375 34.144531 9.398438 33.511719 9.765625 33.097656 C 10.128906 32.683594 10.761719 32.644531 11.175781 33.007813 C 11.234375 33.0625 15.875 37 25 37 C 34.140625 37 38.78125 33.046875 38.828125 33.007813 C 39.242188 32.648438 39.871094 32.683594 40.238281 33.101563 C 40.601563 33.515625 40.5625 34.144531 40.148438 34.507813 C 40.003906 34.636719 37.484375 36.800781 32.550781 38.074219 L 34.378906 40.589844 C 34.566406 40.847656 34.867188 41 35.1875 41 C 35.199219 41 35.210938 41 35.21875 41 C 37.027344 40.941406 44.960938 39.605469 47.867188 34.496094 C 47.953125 34.34375 48 34.175781 48 34 C 48 24.152344 43.785156 12.761719 41.625 10.769531 Z M 18.5 30 C 16.566406 30 15 28.210938 15 26 C 15 23.789063 16.566406 22 18.5 22 C 20.433594 22 22 23.789063 22 26 C 22 28.210938 20.433594 30 18.5 30 Z M 31.5 30 C 29.566406 30 28 28.210938 28 26 C 28 23.789063 29.566406 22 31.5 22 C 33.433594 22 35 23.789063 35 26 C 35 28.210938 33.433594 30 31.5 30 Z"/></svg>
+      </a>
+  </div>
     <div style="margin-left: 5px; margin-right: 5px;">
         <a href="https://twitter.com/lancedb" target="_blank" rel="noopener noreferrer">
             <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0,0,256,256" width="25px" height="25px" fill-rule="nonzero"><g fill-opacity="0" fill="#ffffff" fill-rule="nonzero" stroke="none" stroke-width="1" stroke-linecap="butt" stroke-linejoin="miter" stroke-miterlimit="10" stroke-dasharray="" stroke-dashoffset="0" font-family="none" font-weight="none" font-size="none" text-anchor="none" style="mix-blend-mode: normal"><path d="M0,256v-256h256v256z" id="bgRectangle"></path></g><g fill="#ffffff" fill-rule="nonzero" stroke="none" stroke-width="1" stroke-linecap="butt" stroke-linejoin="miter" stroke-miterlimit="10" stroke-dasharray="" stroke-dashoffset="0" font-family="none" font-weight="none" font-size="none" text-anchor="none" style="mix-blend-mode: normal"><g transform="scale(4,4)"><path d="M57,17.114c-1.32,1.973 -2.991,3.707 -4.916,5.097c0.018,0.423 0.028,0.847 0.028,1.274c0,13.013 -9.902,28.018 -28.016,28.018c-5.562,0 -12.81,-1.948 -15.095,-4.423c0.772,0.092 1.556,0.138 2.35,0.138c4.615,0 8.861,-1.575 12.23,-4.216c-4.309,-0.079 -7.946,-2.928 -9.199,-6.84c1.96,0.308 4.447,-0.17 4.447,-0.17c0,0 -7.7,-1.322 -7.899,-9.779c2.226,1.291 4.46,1.231 4.46,1.231c0,0 -4.441,-2.734 -4.379,-8.195c0.037,-3.221 1.331,-4.953 1.331,-4.953c8.414,10.361 20.298,10.29 20.298,10.29c0,0 -0.255,-1.471 -0.255,-2.243c0,-5.437 4.408,-9.847 9.847,-9.847c2.832,0 5.391,1.196 7.187,3.111c2.245,-0.443 4.353,-1.263 6.255,-2.391c-0.859,3.44 -4.329,5.448 -4.329,5.448c0,0 2.969,-0.329 5.655,-1.55z"></path></g></g></svg>
@@ -173,4 +179,77 @@
       {% include "partials/tabs.html" %}
     {% endif %}
   {% endif %}
-</header>
+</header>
+
+<script>
+  (function() {
+    function checkPathAndRedirect() {
+      var banner = document.getElementById('deprecation-banner');
+
+      if (document.querySelector('meta[http-equiv="refresh"]')) {
+        return; // The redirects plugin is already handling this page.
+      }
+
+      var currentPath = window.location.pathname;
+
+      var cleanPath = currentPath.endsWith('/') && currentPath.length > 1
+        ? currentPath.slice(0, -1)
+        : currentPath;
+
+      // These are the ONLY paths that should remain on the old site
+      var apiPaths = [
+        '/lancedb/python', 
+        '/lancedb/javascript', 
+        '/lancedb/js',
+        '/lancedb/api_reference'
+      ];
+      
+      var isApiPage = apiPaths.some(function(apiPath) {
+        return cleanPath.startsWith(apiPath);
+      });
+
+      if (isApiPage) {
+        if (banner) {
+          banner.style.display = 'none';
+        }
+      } else {
+        if (banner) {
+          banner.style.display = 'block';
+        }
+        
+        // Add noindex meta tag to prevent indexing of old docs for seo
+        var noindexMeta = document.createElement('meta');
+        noindexMeta.setAttribute('name', 'robots');
+        noindexMeta.setAttribute('content', 'noindex, follow');
+        document.head.appendChild(noindexMeta);
+
+        // Add canonical link to point to the new docs to reward new site for seo
+        var canonicalLink = document.createElement('link');
+        canonicalLink.setAttribute('rel', 'canonical');
+        canonicalLink.setAttribute('href', 'https://lancedb.com/docs');
+        document.head.appendChild(canonicalLink);
+        
+        window.location.replace('https://lancedb.com/docs');
+      }
+    }
+
+    // Run the check only if doc is ready. This makes sure we catch the initial load
+    // and redirect.
+    if (document.readyState === 'loading') {
+      document.addEventListener('DOMContentLoaded', checkPathAndRedirect);
+    } else {
+      checkPathAndRedirect();
+    }
+
+    // Use an interval to handle subsequent navigation clicks.
+    var lastPath = window.location.pathname;
+    setInterval(function() {
+      if (window.location.pathname !== lastPath) {
+        lastPath = window.location.pathname;
+        checkPathAndRedirect();
+      }
+    }, 2000); // keeping it 2 second to make it easy for user to understand
+              // what's happening
+
+  })();
+</script>

docs/requirements.txt

@@ -5,3 +5,4 @@ mkdocstrings[python]==0.25.2
 griffe
 mkdocs-render-swagger-plugin
 pydantic
+mkdocs-redirects

docs/src/js/classes/MergeInsertBuilder.md

@@ -52,6 +52,30 @@ the merge result
 
 ***
 
+### useIndex()
+
+```ts
+useIndex(useIndex): MergeInsertBuilder
+```
+
+Controls whether to use indexes for the merge operation.
+
+When set to `true` (the default), the operation will use an index if available
+on the join key for improved performance. When set to `false`, it forces a full
+table scan even if an index exists. This can be useful for benchmarking or when
+the query optimizer chooses a suboptimal path.
+
+#### Parameters
+
+* **useIndex**: `boolean`
+    Whether to use indices for the merge operation. Defaults to `true`.
+
+#### Returns
+
+[`MergeInsertBuilder`](MergeInsertBuilder.md)
+
+***
+
 ### whenMatchedUpdateAll()
 
 ```ts

java/core/pom.xml

@@ -8,7 +8,7 @@
     <parent>
         <groupId>com.lancedb</groupId>
         <artifactId>lancedb-parent</artifactId>
-        <version>0.22.1-final.0</version>
+        <version>0.22.2-beta.2</version>
         <relativePath>../pom.xml</relativePath>
     </parent>
 

java/lance-namespace/pom.xml

@@ -8,7 +8,7 @@
     <parent>
         <groupId>com.lancedb</groupId>
         <artifactId>lancedb-parent</artifactId>
-        <version>0.22.1-final.0</version>
+        <version>0.22.2-beta.2</version>
         <relativePath>../pom.xml</relativePath>
     </parent>
 

java/pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-parent</artifactId>
-    <version>0.22.1-final.0</version>
+    <version>0.22.2-beta.2</version>
     <packaging>pom</packaging>
     <name>${project.artifactId}</name>
     <description>LanceDB Java SDK Parent POM</description>

nodejs/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.22.1"
+version = "0.22.2-beta.2"
 license.workspace = true
 description.workspace = true
 repository.workspace = true

nodejs/__test__/sanitize.test.ts

@@ -0,0 +1,184 @@
+// SPDX-License-Identifier: Apache-2.0
+// SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+import * as arrow from "../lancedb/arrow";
+import { sanitizeField, sanitizeType } from "../lancedb/sanitize";
+
+describe("sanitize", function () {
+  describe("sanitizeType function", function () {
+    it("should handle type objects", function () {
+      const type = new arrow.Int32();
+      const result = sanitizeType(type);
+
+      expect(result.typeId).toBe(arrow.Type.Int);
+      expect((result as arrow.Int).bitWidth).toBe(32);
+      expect((result as arrow.Int).isSigned).toBe(true);
+
+      const floatType = {
+        typeId: 3, // Type.Float = 3
+        precision: 2,
+        toString: () => "Float",
+        isFloat: true,
+        isFixedWidth: true,
+      };
+
+      const floatResult = sanitizeType(floatType);
+      expect(floatResult).toBeInstanceOf(arrow.DataType);
+      expect(floatResult.typeId).toBe(arrow.Type.Float);
+
+      const floatResult2 = sanitizeType({ ...floatType, typeId: () => 3 });
+      expect(floatResult2).toBeInstanceOf(arrow.DataType);
+      expect(floatResult2.typeId).toBe(arrow.Type.Float);
+    });
+
+    const allTypeNameTestCases = [
+      ["null", new arrow.Null()],
+      ["binary", new arrow.Binary()],
+      ["utf8", new arrow.Utf8()],
+      ["bool", new arrow.Bool()],
+      ["int8", new arrow.Int8()],
+      ["int16", new arrow.Int16()],
+      ["int32", new arrow.Int32()],
+      ["int64", new arrow.Int64()],
+      ["uint8", new arrow.Uint8()],
+      ["uint16", new arrow.Uint16()],
+      ["uint32", new arrow.Uint32()],
+      ["uint64", new arrow.Uint64()],
+      ["float16", new arrow.Float16()],
+      ["float32", new arrow.Float32()],
+      ["float64", new arrow.Float64()],
+      ["datemillisecond", new arrow.DateMillisecond()],
+      ["dateday", new arrow.DateDay()],
+      ["timenanosecond", new arrow.TimeNanosecond()],
+      ["timemicrosecond", new arrow.TimeMicrosecond()],
+      ["timemillisecond", new arrow.TimeMillisecond()],
+      ["timesecond", new arrow.TimeSecond()],
+      ["intervaldaytime", new arrow.IntervalDayTime()],
+      ["intervalyearmonth", new arrow.IntervalYearMonth()],
+      ["durationnanosecond", new arrow.DurationNanosecond()],
+      ["durationmicrosecond", new arrow.DurationMicrosecond()],
+      ["durationmillisecond", new arrow.DurationMillisecond()],
+      ["durationsecond", new arrow.DurationSecond()],
+    ] as const;
+
+    it.each(allTypeNameTestCases)(
+      'should map type name "%s" to %s',
+      function (name, expected) {
+        const result = sanitizeType(name);
+        expect(result).toBeInstanceOf(expected.constructor);
+      },
+    );
+
+    const caseVariationTestCases = [
+      ["NULL", new arrow.Null()],
+      ["Utf8", new arrow.Utf8()],
+      ["FLOAT32", new arrow.Float32()],
+      ["DaTedAy", new arrow.DateDay()],
+    ] as const;
+
+    it.each(caseVariationTestCases)(
+      'should be case insensitive for type name "%s" mapped to %s',
+      function (name, expected) {
+        const result = sanitizeType(name);
+        expect(result).toBeInstanceOf(expected.constructor);
+      },
+    );
+
+    it("should throw error for unrecognized type name", function () {
+      expect(() => sanitizeType("invalid_type")).toThrow(
+        "Unrecognized type name in schema: invalid_type",
+      );
+    });
+  });
+
+  describe("sanitizeField function", function () {
+    it("should handle field with string type name", function () {
+      const field = sanitizeField({
+        name: "string_field",
+        type: "utf8",
+        nullable: true,
+        metadata: new Map([["key", "value"]]),
+      });
+
+      expect(field).toBeInstanceOf(arrow.Field);
+      expect(field.name).toBe("string_field");
+      expect(field.type).toBeInstanceOf(arrow.Utf8);
+      expect(field.nullable).toBe(true);
+      expect(field.metadata?.get("key")).toBe("value");
+    });
+
+    it("should handle field with type object", function () {
+      const floatType = {
+        typeId: 3, // Float
+        precision: 32,
+      };
+
+      const field = sanitizeField({
+        name: "float_field",
+        type: floatType,
+        nullable: false,
+      });
+
+      expect(field).toBeInstanceOf(arrow.Field);
+      expect(field.name).toBe("float_field");
+      expect(field.type).toBeInstanceOf(arrow.DataType);
+      expect(field.type.typeId).toBe(arrow.Type.Float);
+      expect((field.type as arrow.Float64).precision).toBe(32);
+      expect(field.nullable).toBe(false);
+    });
+
+    it("should handle field with direct Type instance", function () {
+      const field = sanitizeField({
+        name: "bool_field",
+        type: new arrow.Bool(),
+        nullable: true,
+      });
+
+      expect(field).toBeInstanceOf(arrow.Field);
+      expect(field.name).toBe("bool_field");
+      expect(field.type).toBeInstanceOf(arrow.Bool);
+      expect(field.nullable).toBe(true);
+    });
+
+    it("should throw error for invalid field object", function () {
+      expect(() =>
+        sanitizeField({
+          type: "int32",
+          nullable: true,
+        }),
+      ).toThrow(
+        "The field passed in is missing a `type`/`name`/`nullable` property",
+      );
+
+      // Invalid type
+      expect(() =>
+        sanitizeField({
+          name: "invalid",
+          type: { invalid: true },
+          nullable: true,
+        }),
+      ).toThrow("Expected a Type to have a typeId property");
+
+      // Invalid nullable
+      expect(() =>
+        sanitizeField({
+          name: "invalid_nullable",
+          type: "int32",
+          nullable: "not a boolean",
+        }),
+      ).toThrow("The field passed in had a non-boolean `nullable` property");
+    });
+
+    it("should report error for invalid type name", function () {
+      expect(() =>
+        sanitizeField({
+          name: "invalid_field",
+          type: "invalid_type",
+          nullable: true,
+        }),
+      ).toThrow(
+        "Unable to sanitize type for field: invalid_field due to error: Error: Unrecognized type name in schema: invalid_type",
+      );
+    });
+  });
+});

nodejs/__test__/table.test.ts

@@ -10,7 +10,13 @@ import * as arrow16 from "apache-arrow-16";
 import * as arrow17 from "apache-arrow-17";
 import * as arrow18 from "apache-arrow-18";
 
-import { MatchQuery, PhraseQuery, Table, connect } from "../lancedb";
+import {
+  Connection,
+  MatchQuery,
+  PhraseQuery,
+  Table,
+  connect,
+} from "../lancedb";
 import {
   Table as ArrowTable,
   Field,
@@ -21,6 +27,8 @@ import {
   Int64,
   List,
   Schema,
+  SchemaLike,
+  Type,
   Uint8,
   Utf8,
   makeArrowTable,
@@ -211,8 +219,7 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       },
     );
 
-    // TODO: https://github.com/lancedb/lancedb/issues/1832
-    it.skip("should be able to omit nullable fields", async () => {
+    it("should be able to omit nullable fields", async () => {
       const db = await connect(tmpDir.name);
       const schema = new arrow.Schema([
         new arrow.Field(
@@ -236,23 +243,36 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       await table.add([data3]);
 
       let res = await table.query().limit(10).toArray();
-      const resVector = res.map((r) => r.get("vector").toArray());
+      const resVector = res.map((r) =>
+        r.vector ? Array.from(r.vector) : null,
+      );
       expect(resVector).toEqual([null, data2.vector, data3.vector]);
-      const resItem = res.map((r) => r.get("item").toArray());
+      const resItem = res.map((r) => r.item);
       expect(resItem).toEqual(["foo", null, "bar"]);
-      const resPrice = res.map((r) => r.get("price").toArray());
+      const resPrice = res.map((r) => r.price);
       expect(resPrice).toEqual([10.0, 2.0, 3.0]);
 
       const data4 = { item: "foo" };
       // We can't omit a column if it's not nullable
-      await expect(table.add([data4])).rejects.toThrow("Invalid user input");
+      await expect(table.add([data4])).rejects.toThrow(
+        "Append with different schema",
+      );
 
       // But we can alter columns to make them nullable
       await table.alterColumns([{ path: "price", nullable: true }]);
       await table.add([data4]);
 
-      res = (await table.query().limit(10).toArray()).map((r) => r.toJSON());
-      expect(res).toEqual([data1, data2, data3, data4]);
+      res = (await table.query().limit(10).toArray()).map((r) => ({
+        ...r.toJSON(),
+        vector: r.vector ? Array.from(r.vector) : null,
+      }));
+      // Rust fills missing nullable fields with null
+      expect(res).toEqual([
+        { ...data1, vector: null },
+        { ...data2, item: null },
+        data3,
+        { ...data4, price: null, vector: null },
+      ]);
     });
 
     it("should be able to insert nullable data for non-nullable fields", async () => {
@@ -330,6 +350,43 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       const table = await db.createTable("my_table", data);
       expect(await table.countRows()).toEqual(2);
     });
+
+    it("should allow undefined and omitted nullable vector fields", async () => {
+      // Test for the bug: can't pass undefined or omit vector column
+      const db = await connect("memory://");
+      const schema = new arrow.Schema([
+        new arrow.Field("id", new arrow.Int32(), true),
+        new arrow.Field(
+          "vector",
+          new arrow.FixedSizeList(
+            32,
+            new arrow.Field("item", new arrow.Float32(), true),
+          ),
+          true, // nullable = true
+        ),
+      ]);
+      const table = await db.createEmptyTable("test_table", schema);
+
+      // Should not throw error for undefined value
+      await table.add([{ id: 0, vector: undefined }]);
+
+      // Should not throw error for omitted field
+      await table.add([{ id: 1 }]);
+
+      // Should still work for null
+      await table.add([{ id: 2, vector: null }]);
+
+      // Should still work for actual vector
+      const testVector = new Array(32).fill(0.5);
+      await table.add([{ id: 3, vector: testVector }]);
+      expect(await table.countRows()).toEqual(4);
+
+      const res = await table.query().limit(10).toArray();
+      const resVector = res.map((r) =>
+        r.vector ? Array.from(r.vector) : null,
+      );
+      expect(resVector).toEqual([null, null, null, testVector]);
+    });
   },
 );
 
@@ -1454,7 +1511,9 @@ describe("when optimizing a dataset", () => {
 
   it("delete unverified", async () => {
     const version = await table.version();
-    const versionFile = `${tmpDir.name}/${table.name}.lance/_versions/${version - 1}.manifest`;
+    const versionFile = `${tmpDir.name}/${table.name}.lance/_versions/${
+      version - 1
+    }.manifest`;
     fs.rmSync(versionFile);
 
     let stats = await table.optimize({ deleteUnverified: false });
@@ -1968,3 +2027,52 @@ describe("column name options", () => {
     expect(results2.length).toBe(10);
   });
 });
+
+describe("when creating an empty table", () => {
+  let con: Connection;
+  beforeEach(async () => {
+    const tmpDir = tmp.dirSync({ unsafeCleanup: true });
+    con = await connect(tmpDir.name);
+  });
+  afterEach(() => {
+    con.close();
+  });
+
+  it("can create an empty table from an arrow Schema", async () => {
+    const schema = new Schema([
+      new Field("id", new Int64()),
+      new Field("vector", new Float64()),
+    ]);
+    const table = await con.createEmptyTable("test", schema);
+    const actualSchema = await table.schema();
+    expect(actualSchema.fields[0].type.typeId).toBe(Type.Int);
+    expect((actualSchema.fields[0].type as Int64).bitWidth).toBe(64);
+    expect(actualSchema.fields[1].type.typeId).toBe(Type.Float);
+    expect((actualSchema.fields[1].type as Float64).precision).toBe(2);
+  });
+
+  it("can create an empty table from schema that specifies field types by name", async () => {
+    const schemaLike = {
+      fields: [
+        {
+          name: "id",
+          type: "int64",
+          nullable: true,
+        },
+        {
+          name: "vector",
+          type: "float64",
+          nullable: true,
+        },
+      ],
+      metadata: new Map(),
+      names: ["id", "vector"],
+    } satisfies SchemaLike;
+    const table = await con.createEmptyTable("test", schemaLike);
+    const actualSchema = await table.schema();
+    expect(actualSchema.fields[0].type.typeId).toBe(Type.Int);
+    expect((actualSchema.fields[0].type as Int64).bitWidth).toBe(64);
+    expect(actualSchema.fields[1].type.typeId).toBe(Type.Float);
+    expect((actualSchema.fields[1].type as Float64).precision).toBe(2);
+  });
+});

nodejs/lancedb/arrow.ts

@@ -73,7 +73,7 @@ export type FieldLike =
   | {
       type: string;
       name: string;
-      nullable?: boolean;
+      nullable: boolean;
       metadata?: Map<string, string>;
     };
 
@@ -1285,19 +1285,36 @@ function validateSchemaEmbeddings(
     if (isFixedSizeList(field.type)) {
       field = sanitizeField(field);
       if (data.length !== 0 && data?.[0]?.[field.name] === undefined) {
+        // Check if there's an embedding function registered for this field
+        let hasEmbeddingFunction = false;
+
+        // Check schema metadata for embedding functions
         if (schema.metadata.has("embedding_functions")) {
           const embeddings = JSON.parse(
             schema.metadata.get("embedding_functions")!,
           );
-          if (
-            // biome-ignore lint/suspicious/noExplicitAny: we don't know the type of `f`
-            embeddings.find((f: any) => f["vectorColumn"] === field.name) ===
-            undefined
-          ) {
+          // biome-ignore lint/suspicious/noExplicitAny: we don't know the type of `f`
+          if (embeddings.find((f: any) => f["vectorColumn"] === field.name)) {
+            hasEmbeddingFunction = true;
+          }
+        }
+
+        // Check passed embedding function parameter
+        if (embeddings && embeddings.vectorColumn === field.name) {
+          hasEmbeddingFunction = true;
+        }
+
+        // If the field is nullable AND there's no embedding function, allow undefined/omitted values
+        if (field.nullable && !hasEmbeddingFunction) {
+          fields.push(field);
+        } else {
+          // Either not nullable OR has embedding function - require explicit values
+          if (hasEmbeddingFunction) {
+            // Don't add to missingEmbeddingFields since this is expected to be filled by embedding function
+            fields.push(field);
+          } else {
             missingEmbeddingFields.push(field);
           }
-        } else {
-          missingEmbeddingFields.push(field);
         }
       } else {
         fields.push(field);

nodejs/lancedb/sanitize.ts

@@ -326,6 +326,9 @@ export function sanitizeDictionary(typeLike: object) {
 
 // biome-ignore lint/suspicious/noExplicitAny: skip
 export function sanitizeType(typeLike: unknown): DataType<any> {
+  if (typeof typeLike === "string") {
+    return dataTypeFromName(typeLike);
+  }
   if (typeof typeLike !== "object" || typeLike === null) {
     throw Error("Expected a Type but object was null/undefined");
   }
@@ -447,7 +450,7 @@ export function sanitizeType(typeLike: unknown): DataType<any> {
     case Type.DurationSecond:
       return new DurationSecond();
     default:
-      throw new Error("Unrecoginized type id in schema: " + typeId);
+      throw new Error("Unrecognized type id in schema: " + typeId);
   }
 }
 
@@ -467,7 +470,15 @@ export function sanitizeField(fieldLike: unknown): Field {
       "The field passed in is missing a `type`/`name`/`nullable` property",
     );
   }
-  const type = sanitizeType(fieldLike.type);
+  let type: DataType;
+  try {
+    type = sanitizeType(fieldLike.type);
+  } catch (error: unknown) {
+    throw Error(
+      `Unable to sanitize type for field: ${fieldLike.name} due to error: ${error}`,
+      { cause: error },
+    );
+  }
   const name = fieldLike.name;
   if (!(typeof name === "string")) {
     throw Error("The field passed in had a non-string `name` property");
@@ -581,3 +592,46 @@ function sanitizeData(
     },
   );
 }
+
+const constructorsByTypeName = {
+  null: () => new Null(),
+  binary: () => new Binary(),
+  utf8: () => new Utf8(),
+  bool: () => new Bool(),
+  int8: () => new Int8(),
+  int16: () => new Int16(),
+  int32: () => new Int32(),
+  int64: () => new Int64(),
+  uint8: () => new Uint8(),
+  uint16: () => new Uint16(),
+  uint32: () => new Uint32(),
+  uint64: () => new Uint64(),
+  float16: () => new Float16(),
+  float32: () => new Float32(),
+  float64: () => new Float64(),
+  datemillisecond: () => new DateMillisecond(),
+  dateday: () => new DateDay(),
+  timenanosecond: () => new TimeNanosecond(),
+  timemicrosecond: () => new TimeMicrosecond(),
+  timemillisecond: () => new TimeMillisecond(),
+  timesecond: () => new TimeSecond(),
+  intervaldaytime: () => new IntervalDayTime(),
+  intervalyearmonth: () => new IntervalYearMonth(),
+  durationnanosecond: () => new DurationNanosecond(),
+  durationmicrosecond: () => new DurationMicrosecond(),
+  durationmillisecond: () => new DurationMillisecond(),
+  durationsecond: () => new DurationSecond(),
+} as const;
+
+type MappableTypeName = keyof typeof constructorsByTypeName;
+
+export function dataTypeFromName(typeName: string): DataType {
+  const normalizedTypeName = typeName.toLowerCase() as MappableTypeName;
+  const _constructor = constructorsByTypeName[normalizedTypeName];
+
+  if (!_constructor) {
+    throw new Error("Unrecognized type name in schema: " + typeName);
+  }
+
+  return _constructor();
+}

nodejs/npm/darwin-arm64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-arm64",
-	"version": "0.22.1",
+	"version": "0.22.2-beta.2",
 	"os": ["darwin"],
 	"cpu": ["arm64"],
 	"main": "lancedb.darwin-arm64.node",

nodejs/npm/darwin-x64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-x64",
-	"version": "0.22.1",
+	"version": "0.22.2-beta.2",
 	"os": ["darwin"],
 	"cpu": ["x64"],
 	"main": "lancedb.darwin-x64.node",

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-gnu",
-	"version": "0.22.1",
+	"version": "0.22.2-beta.2",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-gnu.node",

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-musl",
-	"version": "0.22.1",
+	"version": "0.22.2-beta.2",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-musl.node",

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-gnu",
-	"version": "0.22.1",
+	"version": "0.22.2-beta.2",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-gnu.node",

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-musl",
-	"version": "0.22.1",
+	"version": "0.22.2-beta.2",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-musl.node",

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

@@ -1,6 +1,6 @@
 {
   "name": "@lancedb/lancedb-win32-arm64-msvc",
-  "version": "0.22.1",
+  "version": "0.22.2-beta.2",
   "os": [
     "win32"
   ],

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-win32-x64-msvc",
-	"version": "0.22.1",
+	"version": "0.22.2-beta.2",
 	"os": ["win32"],
 	"cpu": ["x64"],
 	"main": "lancedb.win32-x64-msvc.node",

nodejs/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "@lancedb/lancedb",
-  "version": "0.22.1",
+  "version": "0.22.2-beta.2",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "@lancedb/lancedb",
-      "version": "0.22.1",
+      "version": "0.22.2-beta.2",
       "cpu": [
         "x64",
         "arm64"

nodejs/package.json

@@ -11,7 +11,7 @@
     "ann"
   ],
   "private": false,
-  "version": "0.22.1",
+  "version": "0.22.2-beta.2",
   "main": "dist/index.js",
   "exports": {
     ".": "./dist/index.js",

python/.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.25.2-beta.0"
+current_version = "0.25.2"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

python/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-python"
-version = "0.25.2-beta.0"
+version = "0.25.2"
 edition.workspace = true
 description = "Python bindings for LanceDB"
 license.workspace = true
@@ -14,12 +14,12 @@ name = "_lancedb"
 crate-type = ["cdylib"]
 
 [dependencies]
-arrow = { version = "55.1", features = ["pyarrow"] }
+arrow = { version = "56.2", features = ["pyarrow"] }
 async-trait = "0.1"
 lancedb = { path = "../rust/lancedb", default-features = false }
 env_logger.workspace = true
-pyo3 = { version = "0.24", features = ["extension-module", "abi3-py39"] }
-pyo3-async-runtimes = { version = "0.24", features = [
+pyo3 = { version = "0.25", features = ["extension-module", "abi3-py39"] }
+pyo3-async-runtimes = { version = "0.25", features = [
     "attributes",
     "tokio-runtime",
 ] }
@@ -28,7 +28,7 @@ futures.workspace = true
 tokio = { version = "1.40", features = ["sync"] }
 
 [build-dependencies]
-pyo3-build-config = { version = "0.24", features = [
+pyo3-build-config = { version = "0.25", features = [
     "extension-module",
     "abi3-py39",
 ] }

python/pyproject.toml

@@ -5,12 +5,12 @@ dynamic = ["version"]
 dependencies = [
     "deprecation",
     "numpy",
-    "overrides>=0.7",
+    "overrides>=0.7; python_version<'3.12'",
     "packaging",
     "pyarrow>=16",
     "pydantic>=1.10",
     "tqdm>=4.27.0",
-    "lance-namespace==0.0.6"
+    "lance-namespace>=0.0.16"
 ]
 description = "lancedb"
 authors = [{ name = "LanceDB Devs", email = "dev@lancedb.com" }]

python/python/lancedb/_lancedb.pyi

@@ -133,6 +133,7 @@ class Tags:
     async def update(self, tag: str, version: int): ...
 
 class IndexConfig:
+    name: str
     index_type: str
     columns: List[str]
 

python/python/lancedb/db.py

@@ -6,10 +6,18 @@ from __future__ import annotations
 
 from abc import abstractmethod
 from pathlib import Path
+import sys
 from typing import TYPE_CHECKING, Dict, Iterable, List, Literal, Optional, Union
 
+if sys.version_info >= (3, 12):
+    from typing import override
+
+    class EnforceOverrides:
+        pass
+else:
+    from overrides import EnforceOverrides, override  # type: ignore
+
 from lancedb.embeddings.registry import EmbeddingFunctionRegistry
-from overrides import EnforceOverrides, override  # type: ignore
 
 from lancedb.common import data_to_reader, sanitize_uri, validate_schema
 from lancedb.background_loop import LOOP

python/python/lancedb/namespace.py

@@ -12,13 +12,18 @@ from __future__ import annotations
 
 from typing import Dict, Iterable, List, Optional, Union
 import os
+import sys
+
+if sys.version_info >= (3, 12):
+    from typing import override
+else:
+    from overrides import override
 
 from lancedb.db import DBConnection
 from lancedb.table import LanceTable, Table
 from lancedb.util import validate_table_name
 from lancedb.common import validate_schema
 from lancedb.table import sanitize_create_table
-from overrides import override
 
 from lance_namespace import LanceNamespace, connect as namespace_connect
 from lance_namespace_urllib3_client.models import (

python/python/lancedb/remote/db.py

@@ -5,15 +5,20 @@
 from datetime import timedelta
 import logging
 from concurrent.futures import ThreadPoolExecutor
+import sys
 from typing import Any, Dict, Iterable, List, Optional, Union
 from urllib.parse import urlparse
 import warnings
 
+if sys.version_info >= (3, 12):
+    from typing import override
+else:
+    from overrides import override
+
 # Remove this import to fix circular dependency
 # from lancedb import connect_async
 from lancedb.remote import ClientConfig
 import pyarrow as pa
-from overrides import override
 
 from ..common import DATA
 from ..db import DBConnection, LOOP

python/python/lancedb/remote/table.py

@@ -114,7 +114,7 @@ class RemoteTable(Table):
         index_type: Literal["BTREE", "BITMAP", "LABEL_LIST", "scalar"] = "scalar",
         *,
         replace: bool = False,
-        wait_timeout: timedelta = None,
+        wait_timeout: Optional[timedelta] = None,
         name: Optional[str] = None,
     ):
         """Creates a scalar index
@@ -153,7 +153,7 @@ class RemoteTable(Table):
         column: str,
         *,
         replace: bool = False,
-        wait_timeout: timedelta = None,
+        wait_timeout: Optional[timedelta] = None,
         with_position: bool = False,
         # tokenizer configs:
         base_tokenizer: str = "simple",

python/python/tests/test_index.py

@@ -35,6 +35,8 @@ async def some_table(db_async):
             "tags": [
                 [f"tag{random.randint(0, 8)}" for _ in range(2)] for _ in range(NROWS)
             ],
+            "is_active": [random.choice([True, False]) for _ in range(NROWS)],
+            "data": [random.randbytes(random.randint(0, 128)) for _ in range(NROWS)],
         }
     )
     return await db_async.create_table(
@@ -99,10 +101,17 @@ async def test_create_fixed_size_binary_index(some_table: AsyncTable):
 @pytest.mark.asyncio
 async def test_create_bitmap_index(some_table: AsyncTable):
     await some_table.create_index("id", config=Bitmap())
+    await some_table.create_index("is_active", config=Bitmap())
+    await some_table.create_index("data", config=Bitmap())
     indices = await some_table.list_indices()
-    assert str(indices) == '[Index(Bitmap, columns=["id"], name="id_idx")]'
-    indices = await some_table.list_indices()
-    assert len(indices) == 1
+    assert len(indices) == 3
+    assert indices[0].index_type == "Bitmap"
+    assert indices[0].columns == ["id"]
+    assert indices[1].index_type == "Bitmap"
+    assert indices[1].columns == ["is_active"]
+    assert indices[2].index_type == "Bitmap"
+    assert indices[2].columns == ["data"]
+
     index_name = indices[0].name
     stats = await some_table.index_stats(index_name)
     assert stats.index_type == "BITMAP"
@@ -111,6 +120,11 @@ async def test_create_bitmap_index(some_table: AsyncTable):
     assert stats.num_unindexed_rows == 0
     assert stats.num_indices == 1
 
+    assert (
+        "ScalarIndexQuery"
+        in await some_table.query().where("is_active = TRUE").explain_plan()
+    )
+
 
 @pytest.mark.asyncio
 async def test_create_label_list_index(some_table: AsyncTable):

rust/lancedb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb"
-version = "0.22.1"
+version = "0.22.2-beta.2"
 edition.workspace = true
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license.workspace = true
@@ -82,6 +82,7 @@ crunchy.workspace = true
 bytemuck_derive.workspace = true
 
 [dev-dependencies]
+anyhow = "1"
 tempfile = "3.5.0"
 rand = { version = "0.9", features = ["small_rng"] }
 random_word = { version = "0.4.3", features = ["en"] }

rust/lancedb/Makefile

@@ -0,0 +1,19 @@
+#
+# Makefile for running tests.
+#
+
+# Run all tests.
+all-tests: feature-tests remote-tests
+
+# Run tests for every feature. This requires using docker compose to set up
+# the environment.
+feature-tests:
+	../../ci/run_with_docker_compose.sh \
+		cargo test --all-features --tests --locked --examples
+.PHONY: feature-tests
+
+# Run tests against remote endpoints.
+remote-tests:
+	../../ci/run_with_test_connection.sh \
+		cargo test --features remote --locked
+.PHONY: remote-tests

rust/lancedb/src/connection.rs

@@ -1170,6 +1170,7 @@ mod tests {
     use crate::database::listing::{ListingDatabaseOptions, NewTableConfig};
     use crate::query::QueryBase;
     use crate::query::{ExecutableQuery, QueryExecutionOptions};
+    use crate::test_connection::test_utils::new_test_connection;
     use arrow::compute::concat_batches;
     use arrow_array::RecordBatchReader;
     use arrow_schema::{DataType, Field, Schema};
@@ -1185,11 +1186,8 @@ mod tests {
 
     #[tokio::test]
     async fn test_connect() {
-        let tmp_dir = tempdir().unwrap();
-        let uri = tmp_dir.path().to_str().unwrap();
-        let db = connect(uri).execute().await.unwrap();
-
-        assert_eq!(db.uri, uri);
+        let tc = new_test_connection().await.unwrap();
+        assert_eq!(tc.connection.uri, tc.uri);
     }
 
     #[cfg(not(windows))]
@@ -1255,16 +1253,10 @@ mod tests {
         assert_eq!(tables, names[..7]);
     }
 
-    #[tokio::test]
-    async fn test_connect_s3() {
-        // let db = Database::connect("s3://bucket/path/to/database").await.unwrap();
-    }
-
     #[tokio::test]
     async fn test_open_table() {
-        let tmp_dir = tempdir().unwrap();
-        let uri = tmp_dir.path().to_str().unwrap();
-        let db = connect(uri).execute().await.unwrap();
+        let tc = new_test_connection().await.unwrap();
+        let db = tc.connection;
 
         assert_eq!(db.table_names().execute().await.unwrap().len(), 0);
         // open non-exist table

rust/lancedb/src/data/inspect.rs

@@ -52,13 +52,13 @@ pub fn infer_vector_columns(
     for field in reader.schema().fields() {
         match field.data_type() {
             DataType::FixedSizeList(sub_field, _) if sub_field.data_type().is_floating() => {
-                columns.push(field.name().to_string());
+                columns.push(field.name().clone());
             }
             DataType::List(sub_field) if sub_field.data_type().is_floating() && !strict => {
-                columns_to_infer.insert(field.name().to_string(), None);
+                columns_to_infer.insert(field.name().clone(), None);
             }
             DataType::LargeList(sub_field) if sub_field.data_type().is_floating() && !strict => {
-                columns_to_infer.insert(field.name().to_string(), None);
+                columns_to_infer.insert(field.name().clone(), None);
             }
             _ => {}
         }

rust/lancedb/src/database/listing.rs

@@ -718,9 +718,9 @@ impl Database for ListingDatabase {
             .map_err(|e| Error::Lance { source: e })?;
 
         let version_ref = match (request.source_version, request.source_tag) {
-            (Some(v), None) => Ok(Ref::Version(v)),
+            (Some(v), None) => Ok(Ref::Version(None, Some(v))),
             (None, Some(tag)) => Ok(Ref::Tag(tag)),
-            (None, None) => Ok(Ref::Version(source_dataset.version().version)),
+            (None, None) => Ok(Ref::Version(None, Some(source_dataset.version().version))),
             _ => Err(Error::InvalidInput {
                 message: "Cannot specify both source_version and source_tag".to_string(),
             }),
@@ -728,7 +728,7 @@ impl Database for ListingDatabase {
 
         let target_uri = self.table_uri(&request.target_table_name)?;
         source_dataset
-            .shallow_clone(&target_uri, version_ref, storage_params)
+            .shallow_clone(&target_uri, version_ref, Some(storage_params))
             .await
             .map_err(|e| Error::Lance { source: e })?;
 

rust/lancedb/src/database/namespace.rs

@@ -261,7 +261,7 @@ impl Database for LanceNamespaceDatabase {
                     return listing_db
                         .open_table(OpenTableRequest {
                             name: request.name.clone(),
-                            namespace: request.namespace.clone(),
+                            namespace: vec![],
                             index_cache_size: None,
                             lance_read_params: None,
                         })
@@ -305,7 +305,14 @@ impl Database for LanceNamespaceDatabase {
             )
             .await?;
 
-        listing_db.create_table(request).await
+        let create_request = DbCreateTableRequest {
+            name: request.name,
+            namespace: vec![],
+            data: request.data,
+            mode: request.mode,
+            write_options: request.write_options,
+        };
+        listing_db.create_table(create_request).await
     }
 
     async fn open_table(&self, request: OpenTableRequest) -> Result<Arc<dyn BaseTable>> {
@@ -332,7 +339,13 @@ impl Database for LanceNamespaceDatabase {
             .create_listing_database(&request.name, &location, response.storage_options)
             .await?;
 
-        listing_db.open_table(request).await
+        let open_request = OpenTableRequest {
+            name: request.name.clone(),
+            namespace: vec![],
+            index_cache_size: request.index_cache_size,
+            lance_read_params: request.lance_read_params,
+        };
+        listing_db.open_table(open_request).await
     }
 
     async fn clone_table(&self, _request: CloneTableRequest) -> Result<Arc<dyn BaseTable>> {

rust/lancedb/src/lib.rs

@@ -206,6 +206,7 @@ pub mod query;
 pub mod remote;
 pub mod rerankers;
 pub mod table;
+pub mod test_connection;
 pub mod utils;
 
 use std::fmt::Display;

rust/lancedb/src/remote/db.rs

@@ -647,7 +647,7 @@ impl From<StorageOptions> for RemoteOptions {
         let mut filtered = HashMap::new();
         for opt in supported_opts {
             if let Some(v) = options.0.get(opt) {
-                filtered.insert(opt.to_string(), v.to_string());
+                filtered.insert(opt.to_string(), v.clone());
             }
         }
         Self::new(filtered)

rust/lancedb/src/table.rs

@@ -1383,30 +1383,35 @@ impl Table {
 }
 
 pub struct NativeTags {
-    inner: LanceTags,
+    dataset: dataset::DatasetConsistencyWrapper,
 }
 #[async_trait]
 impl Tags for NativeTags {
     async fn list(&self) -> Result<HashMap<String, TagContents>> {
-        Ok(self.inner.list().await?)
+        let dataset = self.dataset.get().await?;
+        Ok(dataset.tags().list().await?)
     }
 
     async fn get_version(&self, tag: &str) -> Result<u64> {
-        Ok(self.inner.get_version(tag).await?)
+        let dataset = self.dataset.get().await?;
+        Ok(dataset.tags().get_version(tag).await?)
     }
 
     async fn create(&mut self, tag: &str, version: u64) -> Result<()> {
-        self.inner.create(tag, version).await?;
+        let dataset = self.dataset.get().await?;
+        dataset.tags().create(tag, version).await?;
         Ok(())
     }
 
     async fn delete(&mut self, tag: &str) -> Result<()> {
-        self.inner.delete(tag).await?;
+        let dataset = self.dataset.get().await?;
+        dataset.tags().delete(tag).await?;
         Ok(())
     }
 
     async fn update(&mut self, tag: &str, version: u64) -> Result<()> {
-        self.inner.update(tag, version).await?;
+        let dataset = self.dataset.get().await?;
+        dataset.tags().update(tag, version).await?;
         Ok(())
     }
 }
@@ -1780,13 +1785,13 @@ impl NativeTable {
                         BuiltinIndexType::BTree,
                     )))
                 } else {
-                    return Err(Error::InvalidInput {
+                    Err(Error::InvalidInput {
                         message: format!(
                             "there are no indices supported for the field `{}` with the data type {}",
                             field.name(),
                             field.data_type()
                         ),
-                    });
+                    })?
                 }
             }
             Index::BTree(_) => {
@@ -2440,10 +2445,8 @@ impl BaseTable for NativeTable {
     }
 
     async fn tags(&self) -> Result<Box<dyn Tags + '_>> {
-        let dataset = self.dataset.get().await?;
-
         Ok(Box::new(NativeTags {
-            inner: dataset.tags.clone(),
+            dataset: self.dataset.clone(),
         }))
     }
 
@@ -2760,6 +2763,7 @@ mod tests {
         RecordBatchReader, StringArray, TimestampMillisecondArray, TimestampNanosecondArray,
         UInt32Array,
     };
+    use arrow_array::{BinaryArray, LargeBinaryArray};
     use arrow_data::ArrayDataBuilder;
     use arrow_schema::{DataType, Field, Schema, TimeUnit};
     use futures::TryStreamExt;
@@ -3725,6 +3729,10 @@ mod tests {
         let schema = Arc::new(Schema::new(vec![
             Field::new("id", DataType::Int32, false),
             Field::new("category", DataType::Utf8, true),
+            Field::new("large_category", DataType::LargeUtf8, true),
+            Field::new("is_active", DataType::Boolean, true),
+            Field::new("data", DataType::Binary, true),
+            Field::new("large_data", DataType::LargeBinary, true),
         ]));
 
         let batch = RecordBatch::try_new(
@@ -3734,6 +3742,16 @@ mod tests {
                 Arc::new(StringArray::from_iter_values(
                     (0..100).map(|i| format!("category_{}", i % 5)),
                 )),
+                Arc::new(LargeStringArray::from_iter_values(
+                    (0..100).map(|i| format!("large_category_{}", i % 5)),
+                )),
+                Arc::new(BooleanArray::from_iter((0..100).map(|i| Some(i % 2 == 0)))),
+                Arc::new(BinaryArray::from_iter_values(
+                    (0_u32..100).map(|i| i.to_le_bytes()),
+                )),
+                Arc::new(LargeBinaryArray::from_iter_values(
+                    (0_u32..100).map(|i| i.to_le_bytes()),
+                )),
             ],
         )
         .unwrap();
@@ -3754,12 +3772,58 @@ mod tests {
             .await
             .unwrap();
 
+        // Create bitmap index on the "is_active" column
+        table
+            .create_index(&["is_active"], Index::Bitmap(Default::default()))
+            .execute()
+            .await
+            .unwrap();
+
+        // Create bitmap index on the "data" column
+        table
+            .create_index(&["data"], Index::Bitmap(Default::default()))
+            .execute()
+            .await
+            .unwrap();
+
+        // Create bitmap index on the "large_data" column
+        table
+            .create_index(&["large_data"], Index::Bitmap(Default::default()))
+            .execute()
+            .await
+            .unwrap();
+
+        // Create bitmap index on the "large_category" column
+        table
+            .create_index(&["large_category"], Index::Bitmap(Default::default()))
+            .execute()
+            .await
+            .unwrap();
+
         // Verify the index was created
         let index_configs = table.list_indices().await.unwrap();
-        assert_eq!(index_configs.len(), 1);
-        let index = index_configs.into_iter().next().unwrap();
+        assert_eq!(index_configs.len(), 5);
+
+        let mut configs_iter = index_configs.into_iter();
+        let index = configs_iter.next().unwrap();
         assert_eq!(index.index_type, crate::index::IndexType::Bitmap);
         assert_eq!(index.columns, vec!["category".to_string()]);
+
+        let index = configs_iter.next().unwrap();
+        assert_eq!(index.index_type, crate::index::IndexType::Bitmap);
+        assert_eq!(index.columns, vec!["is_active".to_string()]);
+
+        let index = configs_iter.next().unwrap();
+        assert_eq!(index.index_type, crate::index::IndexType::Bitmap);
+        assert_eq!(index.columns, vec!["data".to_string()]);
+
+        let index = configs_iter.next().unwrap();
+        assert_eq!(index.index_type, crate::index::IndexType::Bitmap);
+        assert_eq!(index.columns, vec!["large_data".to_string()]);
+
+        let index = configs_iter.next().unwrap();
+        assert_eq!(index.index_type, crate::index::IndexType::Bitmap);
+        assert_eq!(index.columns, vec!["large_category".to_string()]);
     }
 
     #[tokio::test]

rust/lancedb/src/table/datafusion.rs

@@ -172,7 +172,7 @@ impl TableProvider for BaseTableAdapter {
         if let Some(projection) = projection {
             let field_names = projection
                 .iter()
-                .map(|i| self.schema.field(*i).name().to_string())
+                .map(|i| self.schema.field(*i).name().clone())
                 .collect();
             query.select = Select::Columns(field_names);
         }

rust/lancedb/src/table/dataset.rs

@@ -98,8 +98,9 @@ impl DatasetRef {
             }
             Self::TimeTravel { dataset, version } => {
                 let should_checkout = match &target_ref {
-                    refs::Ref::Version(target_ver) => version != target_ver,
-                    refs::Ref::Tag(_) => true, // Always checkout for tags
+                    refs::Ref::Version(_, Some(target_ver)) => version != target_ver,
+                    refs::Ref::Version(_, None) => true, // No specific version, always checkout
+                    refs::Ref::Tag(_) => true,           // Always checkout for tags
                 };
 
                 if should_checkout {

rust/lancedb/src/test_connection.rs

@@ -0,0 +1,126 @@
+// SPDX-License-Identifier: Apache-2.0
+// SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+//! Functions for testing connections.
+
+#[cfg(test)]
+pub mod test_utils {
+    use regex::Regex;
+    use std::env;
+    use std::io::{BufRead, BufReader};
+    use std::process::{Child, ChildStdout, Command, Stdio};
+
+    use crate::{connect, Connection};
+    use anyhow::{bail, Result};
+    use tempfile::{tempdir, TempDir};
+
+    pub struct TestConnection {
+        pub uri: String,
+        pub connection: Connection,
+        _temp_dir: Option<TempDir>,
+        _process: Option<TestProcess>,
+    }
+
+    struct TestProcess {
+        child: Child,
+    }
+
+    impl Drop for TestProcess {
+        #[allow(unused_must_use)]
+        fn drop(&mut self) {
+            self.child.kill();
+        }
+    }
+
+    pub async fn new_test_connection() -> Result<TestConnection> {
+        match env::var("CREATE_LANCEDB_TEST_CONNECTION_SCRIPT") {
+            Ok(script_path) => new_remote_connection(&script_path).await,
+            Err(_e) => new_local_connection().await,
+        }
+    }
+
+    async fn new_remote_connection(script_path: &str) -> Result<TestConnection> {
+        let temp_dir = tempdir()?;
+        let data_path = temp_dir.path().to_str().unwrap().to_string();
+        let child_result = Command::new(script_path)
+            .stdin(Stdio::null())
+            .stdout(Stdio::piped())
+            .stderr(Stdio::piped())
+            .arg(data_path.clone())
+            .spawn();
+        if child_result.is_err() {
+            bail!(format!(
+                "Unable to run {}: {:?}",
+                script_path,
+                child_result.err()
+            ));
+        }
+        let mut process = TestProcess {
+            child: child_result.unwrap(),
+        };
+        let stdout = BufReader::new(process.child.stdout.take().unwrap());
+        let port = read_process_port(stdout)?;
+        let uri = "db://test";
+        let host_override = format!("http://localhost:{}", port);
+        let connection = create_new_connection(uri, &host_override).await?;
+        Ok(TestConnection {
+            uri: uri.to_string(),
+            connection,
+            _temp_dir: Some(temp_dir),
+            _process: Some(process),
+        })
+    }
+
+    fn read_process_port(mut stdout: BufReader<ChildStdout>) -> Result<String> {
+        let mut line = String::new();
+        let re = Regex::new(r"Query node now listening on 0.0.0.0:(.*)").unwrap();
+        loop {
+            let result = stdout.read_line(&mut line);
+            if let Err(err) = result {
+                bail!(format!(
+                    "read_process_port: error while reading from process output: {}",
+                    err
+                ));
+            } else if result.unwrap() == 0 {
+                bail!("read_process_port: hit EOF before reading port from process output.");
+            }
+            if re.is_match(&line) {
+                let caps = re.captures(&line).unwrap();
+                return Ok(caps[1].to_string());
+            }
+        }
+    }
+
+    #[cfg(feature = "remote")]
+    async fn create_new_connection(
+        uri: &str,
+        host_override: &str,
+    ) -> crate::error::Result<Connection> {
+        connect(uri)
+            .region("us-east-1")
+            .api_key("sk_localtest")
+            .host_override(host_override)
+            .execute()
+            .await
+    }
+
+    #[cfg(not(feature = "remote"))]
+    async fn create_new_connection(
+        _uri: &str,
+        _host_override: &str,
+    ) -> crate::error::Result<Connection> {
+        panic!("remote feature not supported");
+    }
+
+    async fn new_local_connection() -> Result<TestConnection> {
+        let temp_dir = tempdir()?;
+        let uri = temp_dir.path().to_str().unwrap();
+        let connection = connect(uri).execute().await?;
+        Ok(TestConnection {
+            uri: uri.to_string(),
+            connection,
+            _temp_dir: Some(temp_dir),
+            _process: None,
+        })
+    }
+}

rust/lancedb/src/utils.rs

@@ -174,7 +174,7 @@ pub(crate) fn default_vector_column(schema: &Schema, dim: Option<i32>) -> Result
             ),
         })
     } else {
-        Ok(candidates[0].to_string())
+        Ok(candidates[0].clone())
     }
 }
 
@@ -195,7 +195,15 @@ pub fn supported_btree_data_type(dtype: &DataType) -> bool {
 }
 
 pub fn supported_bitmap_data_type(dtype: &DataType) -> bool {
-    dtype.is_integer() || matches!(dtype, DataType::Utf8)
+    dtype.is_integer()
+        || matches!(
+            dtype,
+            DataType::Utf8
+                | DataType::LargeUtf8
+                | DataType::Binary
+                | DataType::LargeBinary
+                | DataType::Boolean
+        )
 }
 
 pub fn supported_label_list_data_type(dtype: &DataType) -> bool {