[python] Bump version: 0.1.14 → 0.1.15

Closes #359

TODO:
 * [x] test in a sample of Linux distro docker containers

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.1.14
+current_version = 0.1.18
 commit = True
 message = Bump version: {current_version} → {new_version}
 tag = True

.github/workflows/make-release-commit.yml

@@ -25,38 +25,35 @@ jobs:
   bump-version:
     runs-on: ubuntu-latest
     steps:
-    - name: Check out main
-      uses: actions/checkout@v3
-      with:
-        ref: main
-        persist-credentials: false
-        fetch-depth: 0
-        lfs: true
-    - name: Set git configs for bumpversion
-      shell: bash
-      run: |
-        git config user.name 'Lance Release'
-        git config user.email 'lance-dev@lancedb.com'
-    - name: Set up Python 3.10
-      uses: actions/setup-python@v4
-      with:
-        python-version: "3.10"
-    - name: Bump version, create tag and commit
-      run: |
-        pip install bump2version
-        bumpversion --verbose ${{ inputs.part }}
-    - name: Update package-lock.json file
-      run: |
-        npm install
-        git add package-lock.json
-        # Add this change to the commit created by bumpversion
-        git commit --amend --no-edit
-      working-directory: node
-    - name: Push new version and tag
-      if: ${{ inputs.dry_run }} == "false"
-      uses: ad-m/github-push-action@master
-      with:
-        github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
-        branch: main
-        tags: true
+      - name: Check out main
+        uses: actions/checkout@v3
+        with:
+          ref: main
+          persist-credentials: false
+          fetch-depth: 0
+          lfs: true
+      - name: Set git configs for bumpversion
+        shell: bash
+        run: |
+          git config user.name 'Lance Release'
+          git config user.email 'lance-dev@lancedb.com'
+      - name: Set up Python 3.10
+        uses: actions/setup-python@v4
+        with:
+          python-version: "3.10"
+      - name: Bump version, create tag and commit
+        run: |
+          pip install bump2version
+          bumpversion --verbose ${{ inputs.part }}
+      - name: Push new version and tag
+        if: ${{ inputs.dry_run }} == "false"
+        uses: ad-m/github-push-action@master
+        with:
+          github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
+          branch: main
+          tags: true
+      - uses: ./.github/workflows/update_package_lock
+        if: ${{ inputs.dry_run }} == "false"
+        with:
+          github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
 

.github/workflows/node.yml

@@ -70,7 +70,7 @@ jobs:
         npm run tsc
         npm run build
         npm run pack-build
-        npm install --no-save ./dist/vectordb-*.tgz
+        npm install --no-save ./dist/lancedb-vectordb-*.tgz
         # Remove index.node to test with dependency installed
         rm index.node
     - name: Test
@@ -101,7 +101,7 @@ jobs:
         npm run tsc
         npm run build
         npm run pack-build
-        npm install --no-save ./dist/vectordb-*.tgz
+        npm install --no-save ./dist/lancedb-vectordb-*.tgz
         # Remove index.node to test with dependency installed
         rm index.node
     - name: Test

.github/workflows/npm-publish.yml

@@ -46,75 +46,51 @@ jobs:
       matrix:
         target: [x86_64-apple-darwin, aarch64-apple-darwin]
     steps:
-    - name: Checkout
-      uses: actions/checkout@v3
-    - name: Install system dependencies
-      run: brew install protobuf
-    - name: Install npm dependencies
-      run: |
-        cd node
-        npm ci
-    - name: Install rustup target
-      if: ${{ matrix.target == 'aarch64-apple-darwin' }}
-      run: rustup target add aarch64-apple-darwin
-    - name: Build MacOS native node modules
-      run: bash ci/build_macos_artifacts.sh ${{ matrix.target }}
-    - name: Upload Darwin Artifacts
-      uses: actions/upload-artifact@v3
-      with:
-        name: darwin-native
-        path: |
-          node/dist/vectordb-darwin*.tgz
+      - name: Checkout
+        uses: actions/checkout@v3
+      - name: Install system dependencies
+        run: brew install protobuf
+      - name: Install npm dependencies
+        run: |
+          cd node
+          npm ci
+      - name: Install rustup target
+        if: ${{ matrix.target == 'aarch64-apple-darwin' }}
+        run: rustup target add aarch64-apple-darwin
+      - name: Build MacOS native node modules
+        run: bash ci/build_macos_artifacts.sh ${{ matrix.target }}
+      - name: Upload Darwin Artifacts
+        uses: actions/upload-artifact@v3
+        with:
+          name: native-darwin
+          path: |
+            node/dist/lancedb-vectordb-darwin*.tgz
 
   node-linux:
-    name: node-linux (${{ matrix.arch}}-unknown-linux-${{ matrix.libc }})
-    runs-on: ubuntu-latest
+    name: node-linux (${{ matrix.config.arch}}-unknown-linux-gnu
+    runs-on: ${{ matrix.config.runner }}
     # Only runs on tags that matches the make-release action
     if: startsWith(github.ref, 'refs/tags/v')
     strategy:
       fail-fast: false
       matrix:
-        libc:
-          - gnu
-          # TODO: re-enable musl once we have refactored to pre-built containers
-          # Right now we have to build node from source which is too expensive.
-          # - musl
-        arch:
-          - x86_64
-          # Building on aarch64 is too slow for now
-          # - aarch64
+        config:
+          - arch: x86_64
+            runner: ubuntu-latest
+          - arch: aarch64
+            runner: buildjet-4vcpu-ubuntu-2204-arm
     steps:
-    - name: Checkout
-      uses: actions/checkout@v3
-    - name: Change owner to root (for npm)
-      # The docker container is run as root, so we need the files to be owned by root
-      # Otherwise npm is a nightmare: https://github.com/npm/cli/issues/3773
-      run: sudo chown -R root:root .
-    - name: Set up QEMU
-      if: ${{ matrix.arch == 'aarch64' }}
-      uses: docker/setup-qemu-action@v2
-      with:
-        platforms: arm64
-    - name: Build Linux GNU native node modules
-      if: ${{ matrix.libc == 'gnu' }}
-      run: |
-        docker run \
-          -v $(pwd):/io -w /io \
-          rust:1.70-bookworm \
-          bash ci/build_linux_artifacts.sh ${{ matrix.arch }}-unknown-linux-gnu
-    - name: Build musl Linux native node modules
-      if: ${{ matrix.libc == 'musl' }}
-      run: |
-        docker run --platform linux/arm64/v8 \
-          -v $(pwd):/io -w /io \
-          quay.io/pypa/musllinux_1_1_${{ matrix.arch }} \
-          bash ci/build_linux_artifacts.sh ${{ matrix.arch }}-unknown-linux-musl
-    - name: Upload Linux Artifacts
-      uses: actions/upload-artifact@v3
-      with:
-        name: linux-native
-        path: |
-          node/dist/vectordb-linux*.tgz
+      - name: Checkout
+        uses: actions/checkout@v3
+      - name: Build Linux Artifacts
+        run: |
+          bash ci/build_linux_artifacts.sh ${{ matrix.config.arch }}
+      - name: Upload Linux Artifacts
+        uses: actions/upload-artifact@v3
+        with:
+          name: native-linux
+          path: |
+            node/dist/lancedb-vectordb-linux*.tgz
 
   node-windows:
     runs-on: windows-2022
@@ -145,12 +121,12 @@ jobs:
       - name: Upload Windows Artifacts
         uses: actions/upload-artifact@v3
         with:
-          name: windows-native
+          name: native-windows
           path: |
-            node/dist/vectordb-win32*.tgz
+            node/dist/lancedb-vectordb-win32*.tgz
 
   release:
-    needs: [node, node-macos, node-linux]
+    needs: [node, node-macos, node-linux, node-windows]
     runs-on: ubuntu-latest
     # Only runs on tags that matches the make-release action
     if: startsWith(github.ref, 'refs/tags/v')
@@ -170,3 +146,18 @@ jobs:
           for filename in *.tgz; do
             npm publish $filename
           done
+
+  update-package-lock:
+    needs: [release]
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+        with:
+          ref: main
+          persist-credentials: false
+          fetch-depth: 0
+          lfs: true
+      - uses: ./.github/workflows/update_package_lock
+        with:
+          github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}

.github/workflows/update_package_lock/action.yml

@@ -0,0 +1,33 @@
+name: update_package_lock
+description: "Update node's package.lock"
+
+inputs:
+  github_token:
+    required: true
+    description: "github token for the repo"
+
+runs:
+  using: "composite"
+  steps:
+    - uses: actions/setup-node@v3
+      with:
+        node-version: 20
+    - name: Set git configs
+      shell: bash
+      run: |
+        git config user.name 'Lance Release'
+        git config user.email 'lance-dev@lancedb.com'
+    - name: Update package-lock.json file
+      working-directory: ./node
+      run: |
+        npm install
+        git add package-lock.json
+        git commit -m "Updating package-lock.json"
+      shell: bash
+    - name: Push changes
+      if: ${{ inputs.dry_run }} == "false"
+      uses: ad-m/github-push-action@master
+      with:
+        github_token: ${{ inputs.github_token }}
+        branch: main
+        tags: true

.github/workflows/update_package_lock_run.yml

@@ -0,0 +1,19 @@
+name: Update package-lock.json
+
+on:
+  workflow_dispatch:
+
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+        with:
+          ref: main
+          persist-credentials: false
+          fetch-depth: 0
+          lfs: true
+      - uses: ./.github/workflows/update_package_lock
+        with:
+          github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}

Cargo.toml

@@ -6,11 +6,11 @@ members = [
 resolver = "2"
 
 [workspace.dependencies]
-lance = "=0.5.8"
+lance = "=0.5.9"
 arrow-array = "42.0"
 arrow-data = "42.0"
 arrow-schema = "42.0"
 arrow-ipc = "42.0"
-half = { "version" = "2.2.1", default-features = false }
+half = { "version" = "=2.2.1", default-features = false }
 object_store = "0.6.1"
 

ci/build_linux_artifacts.sh

@@ -1,72 +1,19 @@
 #!/bin/bash
-# Builds the Linux artifacts (node binaries).
-# Usage: ./build_linux_artifacts.sh [target]
-# Targets supported: 
-# - x86_64-unknown-linux-gnu:centos
-# - aarch64-unknown-linux-gnu:centos
-# - aarch64-unknown-linux-musl
-# - x86_64-unknown-linux-musl
-
-# TODO: refactor this into a Docker container we can pull
-
 set -e
+ARCH=${1:-x86_64}
 
-setup_dependencies() {
-    echo "Installing system dependencies..."
-    if [[ $1 == *musl ]]; then
-        # musllinux
-        apk add openssl-dev
-    else
-        # rust / debian
-        apt update
-        apt install -y libssl-dev protobuf-compiler
-    fi
-}
+# We pass down the current user so that when we later mount the local files 
+# into the container, the files are accessible by the current user.
+pushd ci/manylinux_node
+docker build \
+    -t lancedb-node-manylinux \
+    --build-arg="ARCH=$ARCH" \
+    --build-arg="DOCKER_USER=$(id -u)" \
+    --progress=plain \
+    .
+popd
 
-install_node() {
-    echo "Installing node..."
-    curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.34.0/install.sh | bash
-    source "$HOME"/.bashrc
-
-    if [[ $1 == *musl ]]; then
-        # This node version is 15, we need 16 or higher:
-        # apk add nodejs-current npm 
-        # So instead we install from source (nvm doesn't provide binaries for musl):
-        nvm install -s --no-progress 17
-    else
-        nvm install --no-progress 17 # latest that supports glibc 2.17
-    fi
-}
-
-build_node_binary() {
-    echo "Building node library for $1..."
-    pushd node
-
-    npm ci
-    
-    if [[ $1 == *musl ]]; then
-        # This is needed for cargo to allow build cdylibs with musl
-        export RUSTFLAGS="-C target-feature=-crt-static"
-    fi
-
-    # Cargo can run out of memory while pulling dependencies, especially when running
-    # in QEMU. This is a workaround for that.
-    export CARGO_NET_GIT_FETCH_WITH_CLI=true
-
-    # We don't pass in target, since the native target here already matches
-    # We need to pass OPENSSL_LIB_DIR and OPENSSL_INCLUDE_DIR for static build to work https://github.com/sfackler/rust-openssl/issues/877
-    OPENSSL_STATIC=1 OPENSSL_LIB_DIR=/usr/lib/x86_64-linux-gnu OPENSSL_INCLUDE_DIR=/usr/include/openssl/ npm run build-release
-    npm run pack-build
-
-    popd
-}
-
-TARGET=${1:-x86_64-unknown-linux-gnu}
-# Others:
-# aarch64-unknown-linux-gnu
-# x86_64-unknown-linux-musl
-# aarch64-unknown-linux-musl
-
-setup_dependencies $TARGET
-install_node $TARGET
-build_node_binary $TARGET
+docker run \
+    -v $(pwd):/io -w /io \
+    lancedb-node-manylinux \
+    bash ci/manylinux_node/build.sh $ARCH

ci/manylinux_node/Dockerfile

@@ -0,0 +1,31 @@
+# Many linux dockerfile with Rust, Node, and Lance dependencies installed.
+# This container allows building the node modules native libraries in an 
+# environment with a very old glibc, so that we are compatible with a wide
+# range of linux distributions.
+ARG ARCH=x86_64
+
+FROM quay.io/pypa/manylinux2014_${ARCH}
+
+ARG ARCH=x86_64
+ARG DOCKER_USER=default_user
+
+# Install static openssl
+COPY install_openssl.sh install_openssl.sh
+RUN ./install_openssl.sh ${ARCH} > /dev/null
+
+# Protobuf is also installed as root.
+COPY install_protobuf.sh install_protobuf.sh
+RUN ./install_protobuf.sh ${ARCH}
+
+ENV DOCKER_USER=${DOCKER_USER}
+# Create a group and user
+RUN echo ${ARCH} && adduser --user-group --create-home --uid ${DOCKER_USER} build_user
+
+# We switch to the user to install Rust and Node, since those like to be 
+# installed at the user level.
+USER ${DOCKER_USER}
+
+COPY prepare_manylinux_node.sh prepare_manylinux_node.sh
+RUN cp /prepare_manylinux_node.sh $HOME/ && \
+    cd $HOME && \
+    ./prepare_manylinux_node.sh ${ARCH}

ci/manylinux_node/build.sh

@@ -0,0 +1,19 @@
+#!/bin/bash
+# Builds the node module for manylinux. Invoked by ci/build_linux_artifacts.sh.
+set -e
+ARCH=${1:-x86_64}
+
+if [ "$ARCH" = "x86_64" ]; then
+    export OPENSSL_LIB_DIR=/usr/local/lib64/
+else 
+    export OPENSSL_LIB_DIR=/usr/local/lib/
+fi
+export OPENSSL_STATIC=1
+export OPENSSL_INCLUDE_DIR=/usr/local/include/openssl
+
+source $HOME/.bashrc
+
+cd node
+npm ci
+npm run build-release
+npm run pack-build

ci/manylinux_node/install_openssl.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+# Builds openssl from source so we can statically link to it
+
+# this is to avoid the error we get with the system installation:
+# /usr/bin/ld: <library>: version node not found for symbol SSLeay@@OPENSSL_1.0.1
+# /usr/bin/ld: failed to set dynamic section sizes: Bad value
+set -e
+
+git clone -b OpenSSL_1_1_1u \
+    --single-branch \
+    https://github.com/openssl/openssl.git
+
+pushd openssl
+
+if [[ $1 == x86_64* ]]; then
+    ARCH=linux-x86_64
+else
+    # gnu target
+    ARCH=linux-aarch64
+fi
+
+./Configure no-shared $ARCH
+
+make
+
+make install

ci/manylinux_node/install_protobuf.sh

@@ -0,0 +1,15 @@
+#!/bin/bash
+# Installs protobuf compiler. Should be run as root.
+set -e
+
+if [[ $1 == x86_64* ]]; then
+    ARCH=x86_64
+else
+    # gnu target
+    ARCH=aarch_64
+fi
+
+PB_REL=https://github.com/protocolbuffers/protobuf/releases
+PB_VERSION=23.1
+curl -LO $PB_REL/download/v$PB_VERSION/protoc-$PB_VERSION-linux-$ARCH.zip
+unzip protoc-$PB_VERSION-linux-$ARCH.zip -d /usr/local

ci/manylinux_node/prepare_manylinux_node.sh

@@ -0,0 +1,21 @@
+#!/bin/bash
+set -e
+
+install_node() {
+    echo "Installing node..."
+
+    curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.34.0/install.sh | bash
+
+    source "$HOME"/.bashrc
+
+    nvm install --no-progress 16
+}
+
+install_rust() {
+    echo "Installing rust..."
+    curl https://sh.rustup.rs -sSf | bash -s -- -y
+    export PATH="$PATH:/root/.cargo/bin"
+}
+
+install_node
+install_rust

docs/src/examples/transformerjs_embedding_search_nodejs.md

@@ -1,6 +1,6 @@
 # Vector embedding search using TransformersJS
 
-## Embed and query data from LacneDB using TransformersJS
+## Embed and query data from LanceDB using TransformersJS
 
 <img id="splash" width="400" alt="transformersjs" src="https://github.com/lancedb/lancedb/assets/43097991/88a31e30-3d6f-4eef-9216-4b7c688f1b4f">
 

docs/src/notebooks/code_qa_bot.ipynb

@@ -181,7 +181,7 @@
    "id": "c3852dd3",
    "metadata": {},
    "source": [
-    "# Generating emebeddings from our docs\n",
+    "# Generating embeddings from our docs\n",
     "\n",
     "Now that we have our raw documents loaded, we need to pre-process them to generate embeddings:"
    ]

node/native.js

@@ -17,7 +17,7 @@ const { currentTarget } = require('@neon-rs/load');
 let nativeLib;
 
 try {
-    nativeLib = require(`vectordb-${currentTarget()}`);
+    nativeLib = require(`@lancedb/vectordb-${currentTarget()}`);
 } catch (e) {
     try {
         // Might be developing locally, so try that. But don't expose that error
@@ -25,7 +25,7 @@ try {
         nativeLib = require("./index.node");
     } catch {
         throw new Error(`vectordb: failed to load native library.
-  You may need to run \`npm install vectordb-${currentTarget()}\`.
+  You may need to run \`npm install @lancedb/vectordb-${currentTarget()}\`.
 
   If that does not work, please file a bug report at https://github.com/lancedb/lancedb/issues
       

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.1.14",
+  "version": "0.1.18",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.1.14",
+      "version": "0.1.18",
       "cpu": [
         "x64",
         "arm64"
@@ -24,7 +24,7 @@
         "axios": "^1.4.0"
       },
       "devDependencies": {
-        "@neon-rs/cli": "^0.0.74",
+        "@neon-rs/cli": "^0.0.160",
         "@types/chai": "^4.3.4",
         "@types/chai-as-promised": "^7.1.5",
         "@types/mocha": "^10.0.1",
@@ -51,11 +51,11 @@
         "typescript": "*"
       },
       "optionalDependencies": {
-        "vectordb-darwin-arm64": "0.1.14",
-        "vectordb-darwin-x64": "0.1.14",
-        "vectordb-linux-arm64-gnu": "0.1.14",
-        "vectordb-linux-x64-gnu": "0.1.14",
-        "vectordb-win32-x64-msvc": "0.1.14"
+        "@lancedb/vectordb-darwin-arm64": "0.1.18",
+        "@lancedb/vectordb-darwin-x64": "0.1.18",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.1.18",
+        "@lancedb/vectordb-linux-x64-gnu": "0.1.18",
+        "@lancedb/vectordb-win32-x64-msvc": "0.1.18"
       }
     },
     "node_modules/@apache-arrow/ts": {
@@ -85,6 +85,97 @@
       "resolved": "https://registry.npmjs.org/tslib/-/tslib-2.5.0.tgz",
       "integrity": "sha512-336iVw3rtn2BUK7ORdIAHTyxHGRIHVReokCR3XjbckJMK7ms8FysBfhLR8IXnAgy7T0PTPNBWKiH514FOW/WSg=="
     },
+    "node_modules/@cargo-messages/android-arm-eabi": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/android-arm-eabi/-/android-arm-eabi-0.0.160.tgz",
+      "integrity": "sha512-PTgCEmBHEPKJbxwlHVXB3aGES+NqpeBvn6hJNYWIkET3ZQCSJnScMlIDQXEkWndK7J+hW3Or3H32a93B/MbbfQ==",
+      "cpu": [
+        "arm"
+      ],
+      "dev": true,
+      "optional": true,
+      "os": [
+        "android"
+      ]
+    },
+    "node_modules/@cargo-messages/darwin-arm64": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/darwin-arm64/-/darwin-arm64-0.0.160.tgz",
+      "integrity": "sha512-YSVUuc8TUTi/XmZVg9KrH0bDywKLqC1zeTyZYAYDDmqVDZW9KeTnbBUECKRs56iyHeO+kuEkVW7MKf7j2zb/FA==",
+      "cpu": [
+        "arm64"
+      ],
+      "dev": true,
+      "optional": true,
+      "os": [
+        "darwin"
+      ]
+    },
+    "node_modules/@cargo-messages/darwin-x64": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/darwin-x64/-/darwin-x64-0.0.160.tgz",
+      "integrity": "sha512-U+YlAR+9tKpBljnNPWMop5YhvtwfIPQSAaUYN2llteC7ZNU5/cv8CGT1vm7uFNxr2LeGuAtRbzIh2gUmTV8mng==",
+      "cpu": [
+        "x64"
+      ],
+      "dev": true,
+      "optional": true,
+      "os": [
+        "darwin"
+      ]
+    },
+    "node_modules/@cargo-messages/linux-arm-gnueabihf": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/linux-arm-gnueabihf/-/linux-arm-gnueabihf-0.0.160.tgz",
+      "integrity": "sha512-wqAelTzVv1E7Ls4aviqUbem5xjzCaJQxQtVnLhv6pf1k0UyEHCS2WdufFFmWcojGe7QglI4uve3KTe01MKYj0A==",
+      "cpu": [
+        "arm"
+      ],
+      "dev": true,
+      "optional": true,
+      "os": [
+        "linux"
+      ]
+    },
+    "node_modules/@cargo-messages/linux-x64-gnu": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/linux-x64-gnu/-/linux-x64-gnu-0.0.160.tgz",
+      "integrity": "sha512-LQ6e7O7YYkWfDNIi/53q2QG/+lZok72LOG+NKDVCrrY4TYUcrTqWAybOV6IlkVntKPnpx8YB95umSQGeVuvhpQ==",
+      "cpu": [
+        "x64"
+      ],
+      "dev": true,
+      "optional": true,
+      "os": [
+        "linux"
+      ]
+    },
+    "node_modules/@cargo-messages/win32-arm64-msvc": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/win32-arm64-msvc/-/win32-arm64-msvc-0.0.160.tgz",
+      "integrity": "sha512-VDMBhyun02gIDwmEhkYP1W9Z0tYqn4drgY5Iua1qV2tYOU58RVkWhzUYxM9rzYbnwKZlltgM46J/j5QZ3VaFrA==",
+      "cpu": [
+        "arm64"
+      ],
+      "dev": true,
+      "optional": true,
+      "os": [
+        "win32"
+      ]
+    },
+    "node_modules/@cargo-messages/win32-x64-msvc": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/win32-x64-msvc/-/win32-x64-msvc-0.0.160.tgz",
+      "integrity": "sha512-vnoglDxF6zj0W/Co9D0H/bgnrhUuO5EumIf9v3ujLtBH94rAX11JsXh/FgC/8wQnQSsLyWSq70YxNS2wdETxjA==",
+      "cpu": [
+        "x64"
+      ],
+      "dev": true,
+      "optional": true,
+      "os": [
+        "win32"
+      ]
+    },
     "node_modules/@cspotcode/source-map-support": {
       "version": "0.8.1",
       "resolved": "https://registry.npmjs.org/@cspotcode/source-map-support/-/source-map-support-0.8.1.tgz",
@@ -223,13 +314,82 @@
         "@jridgewell/sourcemap-codec": "^1.4.10"
       }
     },
+    "node_modules/@lancedb/vectordb-darwin-arm64": {
+      "version": "0.1.18",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.1.18.tgz",
+      "integrity": "sha512-vu8MCFgaAAGmTJF+4RaoApROMpRVVgrCk+V9my4adAfWkkXbSmtxiDgiIwwL1VqdGb8UwzGn3kVbNW7idE1ojA==",
+      "cpu": [
+        "arm64"
+      ],
+      "optional": true,
+      "os": [
+        "darwin"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-darwin-x64": {
+      "version": "0.1.18",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.1.18.tgz",
+      "integrity": "sha512-ZU30bd6frRyKJ515ow972PlqO2wIiNT4Ohor9+KbUwl/VKDyAwKOKG8cWhRJXTxk0k1oqpiJ6+Q28TcYJ0sSAw==",
+      "cpu": [
+        "x64"
+      ],
+      "optional": true,
+      "os": [
+        "darwin"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-linux-arm64-gnu": {
+      "version": "0.1.18",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.1.18.tgz",
+      "integrity": "sha512-2UroC026bUYwyciSRonYlXei0SoYbKgfWpozxYOu7GgBAV2CQQtaAPgWJTEl6ZiCNeBmBTx+j0h3+ydUfZA73Q==",
+      "cpu": [
+        "arm64"
+      ],
+      "optional": true,
+      "os": [
+        "linux"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-linux-x64-gnu": {
+      "version": "0.1.18",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.1.18.tgz",
+      "integrity": "sha512-DoQBskl22JAJFZh219ZOJ6o+f1niTZp0qRYngHa/kTIpLKzHWQ0OTtMCz32VBAjAsKjSLNxHE8rrT/S6tvS7KQ==",
+      "cpu": [
+        "x64"
+      ],
+      "optional": true,
+      "os": [
+        "linux"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-win32-x64-msvc": {
+      "version": "0.1.18",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.1.18.tgz",
+      "integrity": "sha512-a/kUM3V6rWuXS80pPECYxKfCUAnq56Of/GPCvnAkpk9C9ldyX10iff4aA6DiPHjEk9V2ytqDfJKl9N3QcMLKLA==",
+      "cpu": [
+        "x64"
+      ],
+      "optional": true,
+      "os": [
+        "win32"
+      ]
+    },
     "node_modules/@neon-rs/cli": {
-      "version": "0.0.74",
-      "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.74.tgz",
-      "integrity": "sha512-9lPmNmjej5iKKOTMPryOMubwkgMRyTWRuaq1yokASvI5mPhr2kzPN7UVjdCOjQvpunNPngR9yAHoirpjiWhUHw==",
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.160.tgz",
+      "integrity": "sha512-GQjzHPJVTOARbX3nP/fAWqBq7JlQ8XgfYlCa+iwzIXf0LC1EyfJTX+vqGD/36b9lKoyY01Z/aDUB9o/qF6ztHA==",
       "dev": true,
       "bin": {
         "neon": "index.js"
+      },
+      "optionalDependencies": {
+        "@cargo-messages/android-arm-eabi": "0.0.160",
+        "@cargo-messages/darwin-arm64": "0.0.160",
+        "@cargo-messages/darwin-x64": "0.0.160",
+        "@cargo-messages/linux-arm-gnueabihf": "0.0.160",
+        "@cargo-messages/linux-x64-gnu": "0.0.160",
+        "@cargo-messages/win32-arm64-msvc": "0.0.160",
+        "@cargo-messages/win32-x64-msvc": "0.0.160"
       }
     },
     "node_modules/@neon-rs/load": {
@@ -4297,42 +4457,6 @@
       "integrity": "sha512-wa7YjyUGfNZngI/vtK0UHAN+lgDCxBPCylVXGp0zu59Fz5aiGtNXaq3DhIov063MorB+VfufLh3JlF2KdTK3xg==",
       "dev": true
     },
-    "node_modules/vectordb-darwin-arm64": {
-      "version": "0.1.14",
-      "resolved": "https://registry.npmjs.org/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.1.14.tgz",
-      "integrity": "sha512-5doSFMUR4scxseo73thCxScmO3Wpb+cqPsIa7+2uneTEtBSViMbkw/1mGTC+rV4NTCnxhoiqHk9pJzZVeDMkPg==",
-      "cpu": [
-        "arm64"
-      ],
-      "optional": true,
-      "os": [
-        "darwin"
-      ]
-    },
-    "node_modules/vectordb-darwin-x64": {
-      "version": "0.1.14",
-      "resolved": "https://registry.npmjs.org/vectordb-darwin-x64/-/vectordb-darwin-x64-0.1.14.tgz",
-      "integrity": "sha512-x+qVaKNhAG65HdENL6GRJjxl1hZ7erRm3a2rhplyYoQyzuRPPBILeWzxkE01G1fb0+47dehe7Q4f/8BDaghcCQ==",
-      "cpu": [
-        "x64"
-      ],
-      "optional": true,
-      "os": [
-        "darwin"
-      ]
-    },
-    "node_modules/vectordb-linux-x64-gnu": {
-      "version": "0.1.14",
-      "resolved": "https://registry.npmjs.org/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.1.14.tgz",
-      "integrity": "sha512-hvA2YYwTZK92k6nPH99Jn5N0CwagDOdnwMmjtCpzFOEYK7dY/2kcTOoQNlBwwNP9MYvgN6jdFD/Cwkih1X/qjA==",
-      "cpu": [
-        "x64"
-      ],
-      "optional": true,
-      "os": [
-        "linux"
-      ]
-    },
     "node_modules/vscode-oniguruma": {
       "version": "1.7.0",
       "resolved": "https://registry.npmjs.org/vscode-oniguruma/-/vscode-oniguruma-1.7.0.tgz",
@@ -4578,6 +4702,55 @@
         }
       }
     },
+    "@cargo-messages/android-arm-eabi": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/android-arm-eabi/-/android-arm-eabi-0.0.160.tgz",
+      "integrity": "sha512-PTgCEmBHEPKJbxwlHVXB3aGES+NqpeBvn6hJNYWIkET3ZQCSJnScMlIDQXEkWndK7J+hW3Or3H32a93B/MbbfQ==",
+      "dev": true,
+      "optional": true
+    },
+    "@cargo-messages/darwin-arm64": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/darwin-arm64/-/darwin-arm64-0.0.160.tgz",
+      "integrity": "sha512-YSVUuc8TUTi/XmZVg9KrH0bDywKLqC1zeTyZYAYDDmqVDZW9KeTnbBUECKRs56iyHeO+kuEkVW7MKf7j2zb/FA==",
+      "dev": true,
+      "optional": true
+    },
+    "@cargo-messages/darwin-x64": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/darwin-x64/-/darwin-x64-0.0.160.tgz",
+      "integrity": "sha512-U+YlAR+9tKpBljnNPWMop5YhvtwfIPQSAaUYN2llteC7ZNU5/cv8CGT1vm7uFNxr2LeGuAtRbzIh2gUmTV8mng==",
+      "dev": true,
+      "optional": true
+    },
+    "@cargo-messages/linux-arm-gnueabihf": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/linux-arm-gnueabihf/-/linux-arm-gnueabihf-0.0.160.tgz",
+      "integrity": "sha512-wqAelTzVv1E7Ls4aviqUbem5xjzCaJQxQtVnLhv6pf1k0UyEHCS2WdufFFmWcojGe7QglI4uve3KTe01MKYj0A==",
+      "dev": true,
+      "optional": true
+    },
+    "@cargo-messages/linux-x64-gnu": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/linux-x64-gnu/-/linux-x64-gnu-0.0.160.tgz",
+      "integrity": "sha512-LQ6e7O7YYkWfDNIi/53q2QG/+lZok72LOG+NKDVCrrY4TYUcrTqWAybOV6IlkVntKPnpx8YB95umSQGeVuvhpQ==",
+      "dev": true,
+      "optional": true
+    },
+    "@cargo-messages/win32-arm64-msvc": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/win32-arm64-msvc/-/win32-arm64-msvc-0.0.160.tgz",
+      "integrity": "sha512-VDMBhyun02gIDwmEhkYP1W9Z0tYqn4drgY5Iua1qV2tYOU58RVkWhzUYxM9rzYbnwKZlltgM46J/j5QZ3VaFrA==",
+      "dev": true,
+      "optional": true
+    },
+    "@cargo-messages/win32-x64-msvc": {
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@cargo-messages/win32-x64-msvc/-/win32-x64-msvc-0.0.160.tgz",
+      "integrity": "sha512-vnoglDxF6zj0W/Co9D0H/bgnrhUuO5EumIf9v3ujLtBH94rAX11JsXh/FgC/8wQnQSsLyWSq70YxNS2wdETxjA==",
+      "dev": true,
+      "optional": true
+    },
     "@cspotcode/source-map-support": {
       "version": "0.8.1",
       "resolved": "https://registry.npmjs.org/@cspotcode/source-map-support/-/source-map-support-0.8.1.tgz",
@@ -4678,11 +4851,50 @@
         "@jridgewell/sourcemap-codec": "^1.4.10"
       }
     },
+    "@lancedb/vectordb-darwin-arm64": {
+      "version": "0.1.18",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.1.18.tgz",
+      "integrity": "sha512-vu8MCFgaAAGmTJF+4RaoApROMpRVVgrCk+V9my4adAfWkkXbSmtxiDgiIwwL1VqdGb8UwzGn3kVbNW7idE1ojA==",
+      "optional": true
+    },
+    "@lancedb/vectordb-darwin-x64": {
+      "version": "0.1.18",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.1.18.tgz",
+      "integrity": "sha512-ZU30bd6frRyKJ515ow972PlqO2wIiNT4Ohor9+KbUwl/VKDyAwKOKG8cWhRJXTxk0k1oqpiJ6+Q28TcYJ0sSAw==",
+      "optional": true
+    },
+    "@lancedb/vectordb-linux-arm64-gnu": {
+      "version": "0.1.18",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.1.18.tgz",
+      "integrity": "sha512-2UroC026bUYwyciSRonYlXei0SoYbKgfWpozxYOu7GgBAV2CQQtaAPgWJTEl6ZiCNeBmBTx+j0h3+ydUfZA73Q==",
+      "optional": true
+    },
+    "@lancedb/vectordb-linux-x64-gnu": {
+      "version": "0.1.18",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.1.18.tgz",
+      "integrity": "sha512-DoQBskl22JAJFZh219ZOJ6o+f1niTZp0qRYngHa/kTIpLKzHWQ0OTtMCz32VBAjAsKjSLNxHE8rrT/S6tvS7KQ==",
+      "optional": true
+    },
+    "@lancedb/vectordb-win32-x64-msvc": {
+      "version": "0.1.18",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.1.18.tgz",
+      "integrity": "sha512-a/kUM3V6rWuXS80pPECYxKfCUAnq56Of/GPCvnAkpk9C9ldyX10iff4aA6DiPHjEk9V2ytqDfJKl9N3QcMLKLA==",
+      "optional": true
+    },
     "@neon-rs/cli": {
-      "version": "0.0.74",
-      "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.74.tgz",
-      "integrity": "sha512-9lPmNmjej5iKKOTMPryOMubwkgMRyTWRuaq1yokASvI5mPhr2kzPN7UVjdCOjQvpunNPngR9yAHoirpjiWhUHw==",
-      "dev": true
+      "version": "0.0.160",
+      "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.160.tgz",
+      "integrity": "sha512-GQjzHPJVTOARbX3nP/fAWqBq7JlQ8XgfYlCa+iwzIXf0LC1EyfJTX+vqGD/36b9lKoyY01Z/aDUB9o/qF6ztHA==",
+      "dev": true,
+      "requires": {
+        "@cargo-messages/android-arm-eabi": "0.0.160",
+        "@cargo-messages/darwin-arm64": "0.0.160",
+        "@cargo-messages/darwin-x64": "0.0.160",
+        "@cargo-messages/linux-arm-gnueabihf": "0.0.160",
+        "@cargo-messages/linux-x64-gnu": "0.0.160",
+        "@cargo-messages/win32-arm64-msvc": "0.0.160",
+        "@cargo-messages/win32-x64-msvc": "0.0.160"
+      }
     },
     "@neon-rs/load": {
       "version": "0.0.74",
@@ -7638,24 +7850,6 @@
       "integrity": "sha512-wa7YjyUGfNZngI/vtK0UHAN+lgDCxBPCylVXGp0zu59Fz5aiGtNXaq3DhIov063MorB+VfufLh3JlF2KdTK3xg==",
       "dev": true
     },
-    "vectordb-darwin-arm64": {
-      "version": "0.1.14",
-      "resolved": "https://registry.npmjs.org/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.1.14.tgz",
-      "integrity": "sha512-5doSFMUR4scxseo73thCxScmO3Wpb+cqPsIa7+2uneTEtBSViMbkw/1mGTC+rV4NTCnxhoiqHk9pJzZVeDMkPg==",
-      "optional": true
-    },
-    "vectordb-darwin-x64": {
-      "version": "0.1.14",
-      "resolved": "https://registry.npmjs.org/vectordb-darwin-x64/-/vectordb-darwin-x64-0.1.14.tgz",
-      "integrity": "sha512-x+qVaKNhAG65HdENL6GRJjxl1hZ7erRm3a2rhplyYoQyzuRPPBILeWzxkE01G1fb0+47dehe7Q4f/8BDaghcCQ==",
-      "optional": true
-    },
-    "vectordb-linux-x64-gnu": {
-      "version": "0.1.14",
-      "resolved": "https://registry.npmjs.org/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.1.14.tgz",
-      "integrity": "sha512-hvA2YYwTZK92k6nPH99Jn5N0CwagDOdnwMmjtCpzFOEYK7dY/2kcTOoQNlBwwNP9MYvgN6jdFD/Cwkih1X/qjA==",
-      "optional": true
-    },
     "vscode-oniguruma": {
       "version": "1.7.0",
       "resolved": "https://registry.npmjs.org/vscode-oniguruma/-/vscode-oniguruma-1.7.0.tgz",

node/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vectordb",
-  "version": "0.1.14",
+  "version": "0.1.18",
   "description": " Serverless, low-latency vector database for AI applications",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",
@@ -27,7 +27,7 @@
   "author": "Lance Devs",
   "license": "Apache-2.0",
   "devDependencies": {
-    "@neon-rs/cli": "^0.0.74",
+    "@neon-rs/cli": "^0.0.160",
     "@types/chai": "^4.3.4",
     "@types/chai-as-promised": "^7.1.5",
     "@types/mocha": "^10.0.1",
@@ -70,18 +70,18 @@
   ],
   "neon": {
     "targets": {
-      "x86_64-apple-darwin": "vectordb-darwin-x64",
-      "aarch64-apple-darwin": "vectordb-darwin-arm64",
-      "x86_64-unknown-linux-gnu": "vectordb-linux-x64-gnu",
-      "aarch64-unknown-linux-gnu": "vectordb-linux-arm64-gnu",
-      "x86_64-pc-windows-msvc": "vectordb-win32-x64-msvc"
+      "x86_64-apple-darwin": "@lancedb/vectordb-darwin-x64",
+      "aarch64-apple-darwin": "@lancedb/vectordb-darwin-arm64",
+      "x86_64-unknown-linux-gnu": "@lancedb/vectordb-linux-x64-gnu",
+      "aarch64-unknown-linux-gnu": "@lancedb/vectordb-linux-arm64-gnu",
+      "x86_64-pc-windows-msvc": "@lancedb/vectordb-win32-x64-msvc"
     }
   },
   "optionalDependencies": {
-    "vectordb-darwin-arm64": "0.1.14",
-    "vectordb-darwin-x64": "0.1.14",
-    "vectordb-linux-arm64-gnu": "0.1.14",
-    "vectordb-linux-x64-gnu": "0.1.14",
-    "vectordb-win32-x64-msvc": "0.1.14"
+    "@lancedb/vectordb-darwin-arm64": "0.1.18",
+    "@lancedb/vectordb-darwin-x64": "0.1.18",
+    "@lancedb/vectordb-linux-arm64-gnu": "0.1.18",
+    "@lancedb/vectordb-linux-x64-gnu": "0.1.18",
+    "@lancedb/vectordb-win32-x64-msvc": "0.1.18"
   }
 }

node/src/embedding/embedding_function.ts

@@ -26,3 +26,9 @@ export interface EmbeddingFunction<T> {
      */
   embed: (data: T[]) => Promise<number[][]>
 }
+
+export function isEmbeddingFunction<T> (value: any): value is EmbeddingFunction<T> {
+  return Object.keys(value).length === 2 &&
+      typeof value.sourceColumn === 'string' &&
+      typeof value.embed === 'function'
+}

node/src/index.ts

@@ -20,10 +20,12 @@ import { fromRecordsToBuffer } from './arrow'
 import type { EmbeddingFunction } from './embedding/embedding_function'
 import { RemoteConnection } from './remote'
 import { Query } from './query'
+import { isEmbeddingFunction } from './embedding/embedding_function'
 
 // eslint-disable-next-line @typescript-eslint/no-var-requires
 const { databaseNew, databaseTableNames, databaseOpenTable, databaseDropTable, tableCreate, tableAdd, tableCreateVectorIndex, tableCountRows, tableDelete } = require('../native.js')
 
+export { Query }
 export type { EmbeddingFunction }
 export { OpenAIEmbeddingFunction } from './embedding/openai'
 
@@ -44,6 +46,9 @@ export interface ConnectionOptions {
   apiKey?: string
   // Region to connect
   region?: string
+
+  // override the host for the remote connections
+  hostOverride?: string
 }
 
 /**
@@ -97,10 +102,35 @@ export interface Connection {
    *
    * @param {string} name - The name of the table.
    * @param data - Non-empty Array of Records to be inserted into the table
-   * @param {WriteMode} mode - The write mode to use when creating the table.
+   */
+  createTable (name: string, data: Array<Record<string, unknown>>): Promise<Table>
+
+  /**
+   * Creates a new Table and initialize it with new data.
+   *
+   * @param {string} name - The name of the table.
+   * @param data - Non-empty Array of Records to be inserted into the table
+   * @param {WriteOptions} options - The write options to use when creating the table.
+   */
+  createTable (name: string, data: Array<Record<string, unknown>>, options: WriteOptions): Promise<Table>
+
+  /**
+   * Creates a new Table and initialize it with new data.
+   *
+   * @param {string} name - The name of the table.
+   * @param data - Non-empty Array of Records to be inserted into the table
    * @param {EmbeddingFunction} embeddings - An embedding function to use on this table
    */
-  createTable<T>(name: string, data: Array<Record<string, unknown>>, mode?: WriteMode, embeddings?: EmbeddingFunction<T>): Promise<Table<T>>
+  createTable<T> (name: string, data: Array<Record<string, unknown>>, embeddings: EmbeddingFunction<T>): Promise<Table<T>>
+  /**
+   * Creates a new Table and initialize it with new data.
+   *
+   * @param {string} name - The name of the table.
+   * @param data - Non-empty Array of Records to be inserted into the table
+   * @param {EmbeddingFunction} embeddings - An embedding function to use on this table
+   * @param {WriteOptions} options - The write options to use when creating the table.
+   */
+  createTable<T> (name: string, data: Array<Record<string, unknown>>, embeddings: EmbeddingFunction<T>, options: WriteOptions): Promise<Table<T>>
 
   createTableArrow(name: string, table: ArrowTable): Promise<Table>
 
@@ -234,32 +264,19 @@ export class LocalConnection implements Connection {
     }
   }
 
-  /**
-   * Creates a new Table and initialize it with new data.
-   *
-   * @param name The name of the table.
-   * @param data Non-empty Array of Records to be inserted into the Table
-   * @param mode The write mode to use when creating the table.
-   */
-  async createTable (name: string, data: Array<Record<string, unknown>>, mode?: WriteMode): Promise<Table>
-  async createTable (name: string, data: Array<Record<string, unknown>>, mode: WriteMode): Promise<Table>
-
-  /**
-   * Creates a new Table and initialize it with new data.
-   *
-   * @param name The name of the table.
-   * @param data Non-empty Array of Records to be inserted into the Table
-   * @param mode The write mode to use when creating the table.
-   * @param embeddings An embedding function to use on this Table
-   */
-  async createTable<T> (name: string, data: Array<Record<string, unknown>>, mode: WriteMode, embeddings: EmbeddingFunction<T>): Promise<Table<T>>
-  async createTable<T> (name: string, data: Array<Record<string, unknown>>, mode: WriteMode, embeddings?: EmbeddingFunction<T>): Promise<Table<T>>
-  async createTable<T> (name: string, data: Array<Record<string, unknown>>, mode: WriteMode, embeddings?: EmbeddingFunction<T>): Promise<Table<T>> {
-    if (mode === undefined) {
-      mode = WriteMode.Create
+  async createTable<T> (name: string, data: Array<Record<string, unknown>>, optsOrEmbedding?: WriteOptions | EmbeddingFunction<T>, opt?: WriteOptions): Promise<Table<T>> {
+    let writeOptions: WriteOptions = new DefaultWriteOptions()
+    if (opt !== undefined && isWriteOptions(opt)) {
+      writeOptions = opt
+    } else if (optsOrEmbedding !== undefined && isWriteOptions(optsOrEmbedding)) {
+      writeOptions = optsOrEmbedding
     }
 
-    const createArgs = [this._db, name, await fromRecordsToBuffer(data, embeddings), mode.toLowerCase()]
+    let embeddings: undefined | EmbeddingFunction<T>
+    if (optsOrEmbedding !== undefined && isEmbeddingFunction(optsOrEmbedding)) {
+      embeddings = optsOrEmbedding
+    }
+    const createArgs = [this._db, name, await fromRecordsToBuffer(data, embeddings), writeOptions.writeMode?.toString()]
     if (this._options.awsCredentials !== undefined) {
       createArgs.push(this._options.awsCredentials.accessKeyId)
       createArgs.push(this._options.awsCredentials.secretKey)
@@ -456,6 +473,23 @@ export enum WriteMode {
   Append = 'append'
 }
 
+/**
+ * Write options when creating a Table.
+ */
+export interface WriteOptions {
+  /** A {@link WriteMode} to use on this operation */
+  writeMode?: WriteMode
+}
+
+export class DefaultWriteOptions implements WriteOptions {
+  writeMode = WriteMode.Create
+}
+
+export function isWriteOptions (value: any): value is WriteOptions {
+  return Object.keys(value).length === 1 &&
+      (value.writeMode === undefined || typeof value.writeMode === 'string')
+}
+
 /**
  * Distance metrics type.
  */

node/src/remote/client.ts

@@ -18,9 +18,15 @@ import { tableFromIPC, type Table as ArrowTable } from 'apache-arrow'
 
 export class HttpLancedbClient {
   private readonly _url: string
+  private readonly _apiKey: () => string
 
-  public constructor (url: string, private readonly _apiKey: string) {
+  public constructor (
+    url: string,
+    apiKey: string,
+    private readonly _dbName?: string
+  ) {
     this._url = url
+    this._apiKey = () => apiKey
   }
 
   get uri (): string {
@@ -37,7 +43,7 @@ export class HttpLancedbClient {
     filter?: string
   ): Promise<ArrowTable<any>> {
     const response = await axios.post(
-              `${this._url}/v1/table/${tableName}`,
+              `${this._url}/v1/table/${tableName}/query/`,
               {
                 vector,
                 k,
@@ -49,7 +55,8 @@ export class HttpLancedbClient {
               {
                 headers: {
                   'Content-Type': 'application/json',
-                  'x-api-key': this._apiKey
+                  'x-api-key': this._apiKey(),
+                  ...(this._dbName !== undefined ? { 'x-lancedb-database': this._dbName } : {})
                 },
                 responseType: 'arraybuffer',
                 timeout: 10000
@@ -79,7 +86,7 @@ export class HttpLancedbClient {
       {
         headers: {
           'Content-Type': 'application/json',
-          'x-api-key': this._apiKey
+          'x-api-key': this._apiKey()
         },
         params,
         timeout: 10000

node/src/remote/index.ts

@@ -37,8 +37,13 @@ export class RemoteConnection implements Connection {
     }
 
     this._dbName = opts.uri.slice('db://'.length)
-    const server = `https://${this._dbName}.${opts.region}.api.lancedb.com`
-    this._client = new HttpLancedbClient(server, opts.apiKey)
+    let server: string
+    if (opts.hostOverride === undefined) {
+      server = `https://${this._dbName}.${opts.region}.api.lancedb.com`
+    } else {
+      server = opts.hostOverride
+    }
+    this._client = new HttpLancedbClient(server, opts.apiKey, opts.hostOverride === undefined ? undefined : this._dbName)
   }
 
   get uri (): string {

node/src/test/test.ts

@@ -18,8 +18,7 @@ import * as chai from 'chai'
 import * as chaiAsPromised from 'chai-as-promised'
 
 import * as lancedb from '../index'
-import { type AwsCredentials, type EmbeddingFunction, MetricType, WriteMode } from '../index'
-import { Query } from '../query'
+import { type AwsCredentials, type EmbeddingFunction, MetricType, Query, WriteMode, DefaultWriteOptions, isWriteOptions } from '../index'
 
 const expect = chai.expect
 const assert = chai.assert
@@ -145,7 +144,7 @@ describe('LanceDB client', function () {
       ]
 
       const tableName = 'overwrite'
-      await con.createTable(tableName, data, WriteMode.Create)
+      await con.createTable(tableName, data, { writeMode: WriteMode.Create })
 
       const newData = [
         { id: 1, vector: [0.1, 0.2], price: 10 },
@@ -155,7 +154,7 @@ describe('LanceDB client', function () {
 
       await expect(con.createTable(tableName, newData)).to.be.rejectedWith(Error, 'already exists')
 
-      const table = await con.createTable(tableName, newData, WriteMode.Overwrite)
+      const table = await con.createTable(tableName, newData, { writeMode: WriteMode.Overwrite })
       assert.equal(table.name, tableName)
       assert.equal(await table.countRows(), 3)
     })
@@ -260,7 +259,7 @@ describe('LanceDB client', function () {
         { price: 10, name: 'foo' },
         { price: 50, name: 'bar' }
       ]
-      const table = await con.createTable('vectors', data, WriteMode.Create, embeddings)
+      const table = await con.createTable('vectors', data, embeddings, { writeMode: WriteMode.Create })
       const results = await table.search('foo').execute()
       assert.equal(results.length, 2)
     })
@@ -318,3 +317,20 @@ describe('Drop table', function () {
     assert.deepEqual(await con.tableNames(), ['t2'])
   })
 })
+
+describe('WriteOptions', function () {
+  context('#isWriteOptions', function () {
+    it('should not match empty object', function () {
+      assert.equal(isWriteOptions({}), false)
+    })
+    it('should match write options', function () {
+      assert.equal(isWriteOptions({ writeMode: WriteMode.Create }), true)
+    })
+    it('should match undefined write mode', function () {
+      assert.equal(isWriteOptions({ writeMode: undefined }), true)
+    })
+    it('should match default write options', function () {
+      assert.equal(isWriteOptions(new DefaultWriteOptions()), true)
+    })
+  })
+})

python/.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.1.12
+current_version = 0.1.15
 commit = True
 message = [python] Bump version: {current_version} → {new_version}
 tag = True

python/lancedb/__init__.py

@@ -19,7 +19,11 @@ from .schema import vector
 
 
 def connect(
-    uri: URI, *, api_key: Optional[str] = None, region: str = "us-west-2"
+    uri: URI,
+    *,
+    api_key: Optional[str] = None,
+    region: str = "us-west-2",
+    host_override: Optional[str] = None,
 ) -> DBConnection:
     """Connect to a LanceDB database.
 
@@ -55,5 +59,5 @@ def connect(
     if isinstance(uri, str) and uri.startswith("db://"):
         if api_key is None:
             raise ValueError(f"api_key is required to connected LanceDB cloud: {uri}")
-        return RemoteDBConnection(uri, api_key, region)
+        return RemoteDBConnection(uri, api_key, region, host_override)
     return LanceDBConnection(uri)

python/lancedb/db.py

@@ -327,6 +327,6 @@ class LanceDBConnection(DBConnection):
         name: str
             The name of the table.
         """
-        filesystem, path = pa.fs.FileSystem.from_uri(self.uri)
+        filesystem, path = fs_from_uri(self.uri)
         table_path = os.path.join(path, name + ".lance")
         filesystem.delete_dir(table_path)

python/lancedb/pydantic.py

@@ -11,7 +11,7 @@
 #  See the License for the specific language governing permissions and
 #  limitations under the License.
 
-"""Pydantic adapter for LanceDB"""
+"""Pydantic (v1 / v2) adapter for LanceDB"""
 
 from __future__ import annotations
 
@@ -19,11 +19,19 @@ import inspect
 import sys
 import types
 from abc import ABC, abstractmethod
-from typing import Any, List, Type, Union, _GenericAlias
+from typing import Any, Callable, Dict, Generator, List, Type, Union, _GenericAlias
 
+import numpy as np
 import pyarrow as pa
 import pydantic
-from pydantic_core import CoreSchema, core_schema
+import semver
+
+PYDANTIC_VERSION = semver.Version.parse(pydantic.__version__)
+try:
+    from pydantic_core import CoreSchema, core_schema
+except ImportError:
+    if PYDANTIC_VERSION >= (2,):
+        raise
 
 
 class FixedSizeListMixin(ABC):
@@ -73,6 +81,9 @@ def vector(
 
     # TODO: make a public parameterized type.
     class FixedSizeList(list, FixedSizeListMixin):
+        def __repr__(self):
+            return f"FixedSizeList(dim={dim})"
+
         @staticmethod
         def dim() -> int:
             return dim
@@ -94,6 +105,25 @@ def vector(
                 ),
             )
 
+        @classmethod
+        def __get_validators__(cls) -> Generator[Callable, None, None]:
+            yield cls.validate
+
+        # For pydantic v1
+        @classmethod
+        def validate(cls, v):
+            if not isinstance(v, (list, range, np.ndarray)) or len(v) != dim:
+                raise TypeError("A list of numbers or numpy.ndarray is needed")
+            return v
+
+        if PYDANTIC_VERSION < (2, 0):
+
+            @classmethod
+            def __modify_schema__(cls, field_schema: Dict[str, Any]):
+                field_schema["items"] = {"type": "number"}
+                field_schema["maxItems"] = dim
+                field_schema["minItems"] = dim
+
     return FixedSizeList
 
 
@@ -120,11 +150,20 @@ def _py_type_to_arrow_type(py_type: Type[Any]) -> pa.DataType:
     )
 
 
-def _pydantic_model_to_fields(model: pydantic.BaseModel) -> List[pa.Field]:
-    fields = []
-    for name, field in model.model_fields.items():
-        fields.append(_pydantic_to_field(name, field))
-    return fields
+if PYDANTIC_VERSION.major < 2:
+
+    def _pydantic_model_to_fields(model: pydantic.BaseModel) -> List[pa.Field]:
+        return [
+            _pydantic_to_field(name, field) for name, field in model.__fields__.items()
+        ]
+
+else:
+
+    def _pydantic_model_to_fields(model: pydantic.BaseModel) -> List[pa.Field]:
+        return [
+            _pydantic_to_field(name, field)
+            for name, field in model.model_fields.items()
+        ]
 
 
 def _pydantic_to_arrow_type(field: pydantic.fields.FieldInfo) -> pa.DataType:

python/lancedb/remote/client.py

@@ -48,11 +48,16 @@ class RestfulLanceDBClient:
     db_name: str
     region: str
     api_key: Credential
+    host_override: Optional[str] = attr.field(default=None)
+
     closed: bool = attr.field(default=False, init=False)
 
     @functools.cached_property
     def session(self) -> aiohttp.ClientSession:
-        url = f"https://{self.db_name}.{self.region}.api.lancedb.com"
+        url = (
+            self.host_override
+            or f"https://{self.db_name}.{self.region}.api.lancedb.com"
+        )
         return aiohttp.ClientSession(url)
 
     async def close(self):
@@ -66,6 +71,8 @@ class RestfulLanceDBClient:
         }
         if self.region == "local":  # Local test mode
             headers["Host"] = f"{self.db_name}.{self.region}.api.lancedb.com"
+        if self.host_override:
+            headers["x-lancedb-database"] = self.db_name
         return headers
 
     @staticmethod
@@ -98,7 +105,7 @@ class RestfulLanceDBClient:
     async def post(
         self,
         uri: str,
-        data: Union[Dict[str, Any], BaseModel, bytes],
+        data: Optional[Union[Dict[str, Any], BaseModel, bytes]] = None,
         params: Optional[Dict[str, Any]] = None,
         content_type: Optional[str] = None,
         deserialize: Callable = lambda resp: resp.json(),
@@ -141,5 +148,7 @@ class RestfulLanceDBClient:
     @_check_not_closed
     async def query(self, table_name: str, query: VectorQuery) -> VectorQueryResult:
         """Query a table."""
-        tbl = await self.post(f"/v1/table/{table_name}/", query, deserialize=_read_ipc)
+        tbl = await self.post(
+            f"/v1/table/{table_name}/query/", query, deserialize=_read_ipc
+        )
         return VectorQueryResult(tbl)

python/lancedb/remote/db.py

@@ -13,7 +13,7 @@
 
 import asyncio
 import uuid
-from typing import List
+from typing import List, Optional
 from urllib.parse import urlparse
 
 import pyarrow as pa
@@ -30,14 +30,22 @@ from .client import ARROW_STREAM_CONTENT_TYPE, RestfulLanceDBClient
 class RemoteDBConnection(DBConnection):
     """A connection to a remote LanceDB database."""
 
-    def __init__(self, db_url: str, api_key: str, region: str):
+    def __init__(
+        self,
+        db_url: str,
+        api_key: str,
+        region: str,
+        host_override: Optional[str] = None,
+    ):
         """Connect to a remote LanceDB database."""
         parsed = urlparse(db_url)
         if parsed.scheme != "db":
             raise ValueError(f"Invalid scheme: {parsed.scheme}, only accepts db://")
         self.db_name = parsed.netloc
         self.api_key = api_key
-        self._client = RestfulLanceDBClient(self.db_name, region, api_key)
+        self._client = RestfulLanceDBClient(
+            self.db_name, region, api_key, host_override
+        )
         try:
             self._loop = asyncio.get_running_loop()
         except RuntimeError:
@@ -95,7 +103,7 @@ class RemoteDBConnection(DBConnection):
 
         self._loop.run_until_complete(
             self._client.post(
-                f"/v1/table/{name}/create",
+                f"/v1/table/{name}/create/",
                 data=data,
                 params={"request_id": request_id},
                 content_type=ARROW_STREAM_CONTENT_TYPE,

python/lancedb/remote/table.py

@@ -33,13 +33,13 @@ class RemoteTable(Table):
         self._name = name
 
     def __repr__(self) -> str:
-        return f"RemoteTable({self._conn.db_name}.{self.name})"
+        return f"RemoteTable({self._conn.db_name}.{self._name})"
 
     @cached_property
     def schema(self) -> pa.Schema:
         """Return the schema of the table."""
         resp = self._conn._loop.run_until_complete(
-            self._conn._client.get(f"/v1/table/{self._name}/describe")
+            self._conn._client.post(f"/v1/table/{self._name}/describe/")
         )
         schema = json_to_schema(resp["schema"])
         return schema
@@ -73,7 +73,7 @@ class RemoteTable(Table):
 
         self._conn._loop.run_until_complete(
             self._conn._client.post(
-                f"/v1/table/{self._name}/insert",
+                f"/v1/table/{self._name}/insert/",
                 data=payload,
                 params={"request_id": request_id, "mode": mode},
                 content_type=ARROW_STREAM_CONTENT_TYPE,

python/lancedb/table.py

@@ -23,12 +23,12 @@ import numpy as np
 import pandas as pd
 import pyarrow as pa
 import pyarrow.compute as pc
-import pyarrow.fs
 from lance import LanceDataset
 from lance.vector import vec_to_table
 
 from .common import DATA, VEC, VECTOR_COLUMN_NAME
 from .query import LanceFtsQueryBuilder, LanceQueryBuilder, Query
+from .util import fs_from_uri
 
 
 def _sanitize_data(data, schema, on_bad_vectors, fill_value):
@@ -527,7 +527,7 @@ class LanceTable(Table):
     @classmethod
     def open(cls, db, name):
         tbl = cls(db, name)
-        fs, path = pa.fs.FileSystem.from_uri(tbl._dataset_uri)
+        fs, path = fs_from_uri(tbl._dataset_uri)
         file_info = fs.get_file_info(path)
         if file_info.type != pa.fs.FileType.Directory:
             raise FileNotFoundError(

python/lancedb/util.py

@@ -71,7 +71,8 @@ def fs_from_uri(uri: str) -> Tuple[pa_fs.FileSystem, str]:
     Get a PyArrow FileSystem from a URI, handling extra environment variables.
     """
     if get_uri_scheme(uri) == "s3":
-        if os.environ["AWS_ENDPOINT"]:
-            uri += "?endpoint_override=" + os.environ["AWS_ENDPOINT"]
+        fs = pa_fs.S3FileSystem(endpoint_override=os.environ.get("AWS_ENDPOINT"))
+        path = get_uri_location(uri)
+        return fs, path
 
     return pa_fs.FileSystem.from_uri(uri)

python/pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "lancedb"
-version = "0.1.12"
-dependencies = ["pylance~=0.5.8", "ratelimiter", "retry", "tqdm", "aiohttp", "pydantic>=2", "attr"]
+version = "0.1.15"
+dependencies = ["pylance~=0.5.8", "ratelimiter", "retry", "tqdm", "aiohttp", "pydantic", "attr", "semver"]
 description = "lancedb"
 authors = [
     { name = "LanceDB Devs", email = "dev@lancedb.com" },
@@ -52,3 +52,6 @@ requires = [
   "wheel",
 ]
 build-backend = "setuptools.build_meta"
+
+[tool.isort]
+profile = "black"

python/tests/test_pydantic.py

@@ -20,7 +20,7 @@ import pyarrow as pa
 import pydantic
 import pytest
 
-from lancedb.pydantic import pydantic_to_schema, vector
+from lancedb.pydantic import PYDANTIC_VERSION, pydantic_to_schema, vector
 
 
 @pytest.mark.skipif(
@@ -111,10 +111,16 @@ def test_fixed_size_list_field():
         li: List[int]
 
     data = TestModel(vec=list(range(16)), li=[1, 2, 3])
-    assert json.loads(data.model_dump_json()) == {
-        "vec": list(range(16)),
-        "li": [1, 2, 3],
-    }
+    if PYDANTIC_VERSION >= (2,):
+        assert json.loads(data.model_dump_json()) == {
+            "vec": list(range(16)),
+            "li": [1, 2, 3],
+        }
+    else:
+        assert data.dict() == {
+            "vec": list(range(16)),
+            "li": [1, 2, 3],
+        }
 
     schema = pydantic_to_schema(TestModel)
     assert schema == pa.schema(
@@ -124,7 +130,11 @@ def test_fixed_size_list_field():
         ]
     )
 
-    json_schema = TestModel.model_json_schema()
+    if PYDANTIC_VERSION >= (2,):
+        json_schema = TestModel.model_json_schema()
+    else:
+        json_schema = TestModel.schema()
+
     assert json_schema == {
         "properties": {
             "vec": {

rust/ffi/node/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "vectordb-node"
-version = "0.1.14"
+version = "0.1.18"
 description = "Serverless, low-latency vector database for AI applications"
 license = "Apache-2.0"
 edition = "2018"

rust/ffi/node/src/arrow.rs

@@ -26,7 +26,7 @@ pub(crate) fn convert_record_batch(record_batch: RecordBatch) -> RecordBatch {
         .column_by_name("vector")
         .cloned()
         .expect("vector column is missing");
-    // TODO: we should just consume the underlaying js buffer in the future instead of this arrow around a bunch of times
+    // TODO: we should just consume the underlying js buffer in the future instead of this arrow around a bunch of times
     let arr = as_list_array(column.as_ref());
     let list_size = arr.values().len() / record_batch.num_rows();
     let r =

rust/ffi/node/src/lib.rs

@@ -54,7 +54,7 @@ struct JsTable {
 impl Finalize for JsTable {}
 
 // TODO: object_store didn't export this type so I copied it.
-// Make a requiest to object_store to export this type
+// Make a request to object_store to export this type
 #[derive(Debug)]
 pub struct StaticCredentialProvider<T> {
     credential: Arc<T>,

rust/vectordb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "vectordb"
-version = "0.1.14"
+version = "0.1.18"
 edition = "2021"
 description = "Serverless, low-latency vector database for AI applications"
 license = "Apache-2.0"