Bump version: 0.1.14 → 0.1.15

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.1.19
+current_version = 0.1.15
 commit = True
 message = Bump version: {current_version} → {new_version}
 tag = True

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

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

.github/workflows/node.yml

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

.github/workflows/npm-publish.yml

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

@@ -30,7 +30,7 @@ jobs:
         python-version: 3.${{ matrix.python-minor-version }}
     - name: Install lancedb
       run: |
-        pip install -e .[tests]
+        pip install -e .
         pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985
         pip install pytest pytest-mock black isort
     - name: Black
@@ -59,7 +59,7 @@ jobs:
         python-version: "3.11"
     - name: Install lancedb
       run: |
-        pip install -e .[tests]
+        pip install -e .
         pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985
         pip install pytest pytest-mock black
     - name: Black

.github/workflows/update_package_lock/action.yml

@@ -1,33 +0,0 @@
-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

@@ -1,19 +0,0 @@
-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

@@ -6,12 +6,11 @@ members = [
 resolver = "2"
 
 [workspace.dependencies]
-lance = "=0.5.9"
+lance = "=0.5.8"
 arrow-array = "42.0"
 arrow-data = "42.0"
 arrow-schema = "42.0"
 arrow-ipc = "42.0"
 half = { "version" = "=2.2.1", default-features = false }
 object_store = "0.6.1"
-snafu = "0.7.4"
 

ci/build_linux_artifacts.sh

@@ -1,19 +1,72 @@
 #!/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}
 
-# We pass down the current user so that when we later mount the local files 
+setup_dependencies() {
-# into the container, the files are accessible by the current user.
+    echo "Installing system dependencies..."
-pushd ci/manylinux_node
+    if [[ $1 == *musl ]]; then
-docker build \
+        # musllinux
-    -t lancedb-node-manylinux \
+        apk add openssl-dev
-    --build-arg="ARCH=$ARCH" \
+    else
-    --build-arg="DOCKER_USER=$(id -u)" \
+        # rust / debian
-    --progress=plain \
+        apt update
-    .
+        apt install -y libssl-dev protobuf-compiler
-popd
+    fi
+}
 
-docker run \
+install_node() {
-    -v $(pwd):/io -w /io \
+    echo "Installing node..."
-    lancedb-node-manylinux \
+    curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.34.0/install.sh | bash
-    bash ci/manylinux_node/build.sh $ARCH
+    source "$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

ci/manylinux_node/Dockerfile

@@ -1,31 +0,0 @@
-# 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

@@ -1,19 +0,0 @@
-#!/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

@@ -1,26 +0,0 @@
-#!/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

@@ -1,15 +0,0 @@
-#!/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

@@ -1,21 +0,0 @@
-#!/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/mkdocs.yml

@@ -57,14 +57,12 @@ nav:
 - Basics: basic.md
 - Embeddings: embedding.md
 - Python full-text search: fts.md
-- Integrations:
+- Python integrations:
   - Pandas and PyArrow: python/arrow.md
   - DuckDB: python/duckdb.md
   - LangChain 🦜️🔗: https://python.langchain.com/en/latest/modules/indexes/vectorstores/examples/lancedb.html
-  - LangChain JS/TS 🦜️🔗: https://js.langchain.com/docs/modules/data_connection/vectorstores/integrations/lancedb
   - LlamaIndex 🦙: https://gpt-index.readthedocs.io/en/latest/examples/vector_stores/LanceDBIndexDemo.html
   - Pydantic: python/pydantic.md
-  - Voxel51: integrations/voxel51.md
 - Python examples:
   - YouTube Transcript Search: notebooks/youtube_transcript_search.ipynb
   - Documentation QA Bot using LangChain: notebooks/code_qa_bot.ipynb
@@ -74,7 +72,6 @@ nav:
 - Javascript examples:
   - YouTube Transcript Search: examples/youtube_transcript_bot_with_nodejs.md
   - TransformersJS Embedding Search: examples/transformerjs_embedding_search_nodejs.md
-
 - References:
   - Vector Search: search.md
   - SQL filters: sql.md

docs/src/examples/transformerjs_embedding_search_nodejs.md

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

docs/src/examples/youtube_transcript_bot.md

@@ -4,10 +4,4 @@
 
 <img id="splash" width="400" alt="youtube transcript search" src="https://user-images.githubusercontent.com/917119/236965568-def7394d-171c-45f2-939d-8edfeaadd88c.png">
 
-
-<a href="https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/youtube_bot/main.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab">
-
-Scripts - [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](./examples/youtube_bot/main.py)  [![JavaScript](https://img.shields.io/badge/javascript-%23323330.svg?style=for-the-badge&logo=javascript&logoColor=%23F7DF1E)](./examples/youtube_bot/index.js)
-
-
 This example is in a [notebook](https://github.com/lancedb/lancedb/blob/main/docs/src/notebooks/youtube_transcript_search.ipynb)

docs/src/integrations/voxel51.md

@@ -1,71 +0,0 @@
-![example](/assets/voxel.gif)
-
-Basic recipe
-____________
-
-The basic workflow to use LanceDB to create a similarity index on your FiftyOne
-datasets and use this to query your data is as follows:
-
-1)  Load a dataset into FiftyOne
-
-2)  Compute embedding vectors for samples or patches in your dataset, or select
-    a model to use to generate embeddings
-
-3)  Use the `compute_similarity()`
-    method to generate a LanceDB table for the samples or object
-    patches embeddings in a dataset by setting the parameter `backend="lancedb"` and
-    specifying a `brain_key` of your choice
-
-4)  Use this LanceDB table to query your data with
-    `sort_by_similarity()`
-
-5) If desired, delete the table
-
-The example below demonstrates this workflow.
-
-!!! Note
-
-    You must install the LanceDB Python client to run this
-    ```
-    pip install lancedb
-    ```
-
-```python
-
-import fiftyone as fo
-import fiftyone.brain as fob
-import fiftyone.zoo as foz
-
-# Step 1: Load your data into FiftyOne
-dataset = foz.load_zoo_dataset("quickstart")
-
-# Steps 2 and 3: Compute embeddings and create a similarity index
-lancedb_index = fob.compute_similarity(
-    dataset, 
-    model="clip-vit-base32-torch",
-    brain_key="lancedb_index",
-    backend="lancedb",
-)
-```
-Once the similarity index has been generated, we can query our data in FiftyOne
-by specifying the `brain_key`:
-
-```python
-# Step 4: Query your data
-query = dataset.first().id  # query by sample ID
-view = dataset.sort_by_similarity(
-    query, 
-    brain_key="lancedb_index",
-    k=10,  # limit to 10 most similar samples
-)
-
-# Step 5 (optional): Cleanup
-
-# Delete the LanceDB table
-lancedb_index.cleanup()
-
-# Delete run record from FiftyOne
-dataset.delete_brain_run("lancedb_index")
-```
-
-More in depth walkthrough of the integration, visit the LanceDB guide on Voxel51 - [LaceDB x Voxel51](https://docs.voxel51.com/integrations/lancedb.html)

docs/src/notebooks/code_qa_bot.ipynb

@@ -10,11 +10,7 @@
     "\n",
     "This Q&A bot will allow you to query your own documentation easily using questions. We'll also demonstrate the use of LangChain and LanceDB using the OpenAI API. \n",
     "\n",
-    "In this example we'll use Pandas 2.0 documentation, but, this could be replaced for your own docs as well\n",
+    "In this example we'll use Pandas 2.0 documentation, but, this could be replaced for your own docs as well"
-    "\n",
-    "<a href=\"https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/Code-Documentation-QA-Bot/main.ipynb\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"></a>\n",
-    "\n",
-    "Scripts -  [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](./examples/Code-Documentation-QA-Bot/main.py)  [![JavaScript](https://img.shields.io/badge/javascript-%23323330.svg?style=for-the-badge&logo=javascript&logoColor=%23F7DF1E)](./examples/Code-Documentation-QA-Bot/index.js)"
    ]
   },
   {
@@ -185,7 +181,7 @@
    "id": "c3852dd3",
    "metadata": {},
    "source": [
-    "# Generating embeddings from our docs\n",
+    "# Generating emebeddings from our docs\n",
     "\n",
     "Now that we have our raw documents loaded, we need to pre-process them to generate embeddings:"
    ]

docs/src/notebooks/multimodal_search.ipynb

@@ -1,14 +1,5 @@
 {
  "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "![example](https://github.com/lancedb/vectordb-recipes/assets/15766192/799f94a1-a01d-4a5b-a627-2a733bbb4227)\n",
-    "\n",
-    " <a href=\"https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/multimodal_clip/main.ipynb\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"></a>| [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](./examples/multimodal_clip/main.py)  |"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": 2,
@@ -51,19 +42,6 @@
     "## First run setup: Download data and pre-process"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "### Get dataset\n",
-    "\n",
-    "!wget https://eto-public.s3.us-west-2.amazonaws.com/datasets/diffusiondb_lance.tar.gz\n",
-    "!tar -xvf diffusiondb_lance.tar.gz\n",
-    "!mv diffusiondb_test rawdata.lance\n"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": 30,
@@ -269,7 +247,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3.11.4 64-bit",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -283,12 +261,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.11.3"
-  },
-  "vscode": {
-   "interpreter": {
-    "hash": "b0fa6594d8f4cbf19f97940f81e996739fb7646882a419484c72d19e05852a7e"
-   }
   }
  },
  "nbformat": 4,

docs/src/notebooks/youtube_transcript_search.ipynb

@@ -8,12 +8,7 @@
    "source": [
     "# Youtube Transcript Search QA Bot\n",
     "\n",
-    "This Q&A bot will allow you to search through youtube transcripts using natural language! By going through this notebook, we'll introduce how you can use LanceDB to store and manage your data easily.\n",
+    "This Q&A bot will allow you to search through youtube transcripts using natural language! By going through this notebook, we'll introduce how you can use LanceDB to store and manage your data easily."
-    "\n",
-    "\n",
-    "<a href=\"https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/youtube_bot/main.ipynb\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\">\n",
-    "\n",
-    "Scripts - [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](./examples/youtube_bot/main.py)  [![JavaScript](https://img.shields.io/badge/javascript-%23323330.svg?style=for-the-badge&logo=javascript&logoColor=%23F7DF1E)](./examples/youtube_bot/index.js)\n"
    ]
   },
   {

docs/src/python/pydantic.md

@@ -1,8 +1,6 @@
 # Pydantic
 
 [Pydantic](https://docs.pydantic.dev/latest/) is a data validation library in Python.
-LanceDB integrates with Pydantic for schema inference, data ingestion, and query result casting.
-
 
 ## Schema
 

docs/test/md_testing.py

@@ -7,8 +7,7 @@ excluded_files = [
     "../src/embedding.md",
     "../src/examples/serverless_lancedb_with_s3_and_lambda.md",
     "../src/examples/serverless_qa_bot_with_modal_and_langchain.md",
-    "../src/examples/youtube_transcript_bot_with_nodejs.md",
+    "../src/examples/youtube_transcript_bot_with_nodejs.md"
-    "../src/integrations/voxel51.md",
 ]
 
 python_prefix = "py"

node/native.js

@@ -17,7 +17,7 @@ const { currentTarget } = require('@neon-rs/load');
 let nativeLib;
 
 try {
-    nativeLib = require(`@lancedb/vectordb-${currentTarget()}`);
+    nativeLib = require(`vectordb-${currentTarget()}`);
 } catch (e) {
     try {
         // Might be developing locally, so try that. But don't expose that error
@@ -25,12 +25,12 @@ try {
         nativeLib = require("./index.node");
     } catch {
         throw new Error(`vectordb: failed to load native library.
-  You may need to run \`npm install @lancedb/vectordb-${currentTarget()}\`.
+  You may need to run \`npm install vectordb-${currentTarget()}\`.
-
+      
   If that does not work, please file a bug report at https://github.com/lancedb/lancedb/issues
       
   Source error: ${e}`);
-    }
+    }   
 }
 
 // Dynamic require for runtime.

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.1.19",
+  "version": "0.1.14",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.1.19",
+      "version": "0.1.14",
       "cpu": [
         "x64",
         "arm64"
@@ -24,7 +24,7 @@
         "axios": "^1.4.0"
       },
       "devDependencies": {
-        "@neon-rs/cli": "^0.0.160",
+        "@neon-rs/cli": "^0.0.74",
         "@types/chai": "^4.3.4",
         "@types/chai-as-promised": "^7.1.5",
         "@types/mocha": "^10.0.1",
@@ -51,11 +51,11 @@
         "typescript": "*"
       },
       "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.1.19",
+        "vectordb-darwin-arm64": "0.1.14",
-        "@lancedb/vectordb-darwin-x64": "0.1.19",
+        "vectordb-darwin-x64": "0.1.14",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.1.19",
+        "vectordb-linux-arm64-gnu": "0.1.14",
-        "@lancedb/vectordb-linux-x64-gnu": "0.1.19",
+        "vectordb-linux-x64-gnu": "0.1.14",
-        "@lancedb/vectordb-win32-x64-msvc": "0.1.19"
+        "vectordb-win32-x64-msvc": "0.1.14"
       }
     },
     "node_modules/@apache-arrow/ts": {
@@ -85,97 +85,6 @@
       "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",
@@ -314,82 +223,13 @@
         "@jridgewell/sourcemap-codec": "^1.4.10"
       }
     },
-    "node_modules/@lancedb/vectordb-darwin-arm64": {
-      "version": "0.1.19",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.1.19.tgz",
-      "integrity": "sha512-efQhJkBKvMNhjFq3Sw3/qHo9D9gb9UqiIr98n3STsbNxBQjMnWemXn91Ckl40siRG1O8qXcINW7Qs/EGmus+kg==",
-      "cpu": [
-        "arm64"
-      ],
-      "optional": true,
-      "os": [
-        "darwin"
-      ]
-    },
-    "node_modules/@lancedb/vectordb-darwin-x64": {
-      "version": "0.1.19",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.1.19.tgz",
-      "integrity": "sha512-r6OZNVyemAssABz2w7CRhe7dyREwBEfTytn+ux1zzTnzsgMgDovCQ0rQ3WZcxWvcy7SFCxiemA9IP1b/lsb4tQ==",
-      "cpu": [
-        "x64"
-      ],
-      "optional": true,
-      "os": [
-        "darwin"
-      ]
-    },
-    "node_modules/@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.1.19",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.1.19.tgz",
-      "integrity": "sha512-mL/hRmZp6Kw7hmGJBdOZfp/tTYiCdlOcs8DA/+nr2eiXERv0gIhyiKvr2P5DwbBmut3qXEkDalMHTo95BSdL2A==",
-      "cpu": [
-        "arm64"
-      ],
-      "optional": true,
-      "os": [
-        "linux"
-      ]
-    },
-    "node_modules/@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.1.19",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.1.19.tgz",
-      "integrity": "sha512-AG0FHksbbr+cHVKPi4B8cmBtqb6T9E0uaK4kyZkXrX52/xtv9RYVZcykaB/tSSm0XNFPWWRnx9R8UqNZV/hxMA==",
-      "cpu": [
-        "x64"
-      ],
-      "optional": true,
-      "os": [
-        "linux"
-      ]
-    },
-    "node_modules/@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.1.19",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.1.19.tgz",
-      "integrity": "sha512-PDWZ2hvLVXH4Z4WIO1rsWY8ev3NpNm7aXlaey32P+l1Iz9Hia9+F2GBpp2UiEQKfvbk82ucAvBLRmpSsHY8Tlw==",
-      "cpu": [
-        "x64"
-      ],
-      "optional": true,
-      "os": [
-        "win32"
-      ]
-    },
     "node_modules/@neon-rs/cli": {
-      "version": "0.0.160",
+      "version": "0.0.74",
-      "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.160.tgz",
+      "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.74.tgz",
-      "integrity": "sha512-GQjzHPJVTOARbX3nP/fAWqBq7JlQ8XgfYlCa+iwzIXf0LC1EyfJTX+vqGD/36b9lKoyY01Z/aDUB9o/qF6ztHA==",
+      "integrity": "sha512-9lPmNmjej5iKKOTMPryOMubwkgMRyTWRuaq1yokASvI5mPhr2kzPN7UVjdCOjQvpunNPngR9yAHoirpjiWhUHw==",
       "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": {
@@ -4457,6 +4297,42 @@
       "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",
@@ -4702,55 +4578,6 @@
         }
       }
     },
-    "@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",
@@ -4851,50 +4678,11 @@
         "@jridgewell/sourcemap-codec": "^1.4.10"
       }
     },
-    "@lancedb/vectordb-darwin-arm64": {
-      "version": "0.1.19",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.1.19.tgz",
-      "integrity": "sha512-efQhJkBKvMNhjFq3Sw3/qHo9D9gb9UqiIr98n3STsbNxBQjMnWemXn91Ckl40siRG1O8qXcINW7Qs/EGmus+kg==",
-      "optional": true
-    },
-    "@lancedb/vectordb-darwin-x64": {
-      "version": "0.1.19",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.1.19.tgz",
-      "integrity": "sha512-r6OZNVyemAssABz2w7CRhe7dyREwBEfTytn+ux1zzTnzsgMgDovCQ0rQ3WZcxWvcy7SFCxiemA9IP1b/lsb4tQ==",
-      "optional": true
-    },
-    "@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.1.19",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.1.19.tgz",
-      "integrity": "sha512-mL/hRmZp6Kw7hmGJBdOZfp/tTYiCdlOcs8DA/+nr2eiXERv0gIhyiKvr2P5DwbBmut3qXEkDalMHTo95BSdL2A==",
-      "optional": true
-    },
-    "@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.1.19",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.1.19.tgz",
-      "integrity": "sha512-AG0FHksbbr+cHVKPi4B8cmBtqb6T9E0uaK4kyZkXrX52/xtv9RYVZcykaB/tSSm0XNFPWWRnx9R8UqNZV/hxMA==",
-      "optional": true
-    },
-    "@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.1.19",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.1.19.tgz",
-      "integrity": "sha512-PDWZ2hvLVXH4Z4WIO1rsWY8ev3NpNm7aXlaey32P+l1Iz9Hia9+F2GBpp2UiEQKfvbk82ucAvBLRmpSsHY8Tlw==",
-      "optional": true
-    },
     "@neon-rs/cli": {
-      "version": "0.0.160",
+      "version": "0.0.74",
-      "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.160.tgz",
+      "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.74.tgz",
-      "integrity": "sha512-GQjzHPJVTOARbX3nP/fAWqBq7JlQ8XgfYlCa+iwzIXf0LC1EyfJTX+vqGD/36b9lKoyY01Z/aDUB9o/qF6ztHA==",
+      "integrity": "sha512-9lPmNmjej5iKKOTMPryOMubwkgMRyTWRuaq1yokASvI5mPhr2kzPN7UVjdCOjQvpunNPngR9yAHoirpjiWhUHw==",
-      "dev": true,
+      "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",
@@ -7850,6 +7638,24 @@
       "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": "0.1.19",
+  "version": "0.1.15",
   "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": "^0.0.160",
+    "@neon-rs/cli": "^0.0.74",
     "@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": "@lancedb/vectordb-darwin-x64",
+      "x86_64-apple-darwin": "vectordb-darwin-x64",
-      "aarch64-apple-darwin": "@lancedb/vectordb-darwin-arm64",
+      "aarch64-apple-darwin": "vectordb-darwin-arm64",
-      "x86_64-unknown-linux-gnu": "@lancedb/vectordb-linux-x64-gnu",
+      "x86_64-unknown-linux-gnu": "vectordb-linux-x64-gnu",
-      "aarch64-unknown-linux-gnu": "@lancedb/vectordb-linux-arm64-gnu",
+      "aarch64-unknown-linux-gnu": "vectordb-linux-arm64-gnu",
-      "x86_64-pc-windows-msvc": "@lancedb/vectordb-win32-x64-msvc"
+      "x86_64-pc-windows-msvc": "vectordb-win32-x64-msvc"
     }
   },
   "optionalDependencies": {
-    "@lancedb/vectordb-darwin-arm64": "0.1.19",
+    "vectordb-darwin-arm64": "0.1.15",
-    "@lancedb/vectordb-darwin-x64": "0.1.19",
+    "vectordb-darwin-x64": "0.1.15",
-    "@lancedb/vectordb-linux-arm64-gnu": "0.1.19",
+    "vectordb-linux-arm64-gnu": "0.1.15",
-    "@lancedb/vectordb-linux-x64-gnu": "0.1.19",
+    "vectordb-linux-x64-gnu": "0.1.15",
-    "@lancedb/vectordb-win32-x64-msvc": "0.1.19"
+    "vectordb-win32-x64-msvc": "0.1.15"
   }
 }

node/src/embedding/embedding_function.ts

@@ -26,8 +26,3 @@ export interface EmbeddingFunction<T> {
      */
   embed: (data: T[]) => Promise<number[][]>
 }
-
-export function isEmbeddingFunction<T> (value: any): value is EmbeddingFunction<T> {
-  return typeof value.sourceColumn === 'string' &&
-      typeof value.embed === 'function'
-}

node/src/index.ts

@@ -20,12 +20,10 @@ 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'
 
@@ -102,35 +100,10 @@ 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>>, embeddings: EmbeddingFunction<T>): Promise<Table<T>>
+  createTable<T>(name: string, data: Array<Record<string, unknown>>, mode?: WriteMode, 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>
 
@@ -264,19 +237,32 @@ export class LocalConnection implements Connection {
     }
   }
 
-  async createTable<T> (name: string, data: Array<Record<string, unknown>>, optsOrEmbedding?: WriteOptions | EmbeddingFunction<T>, opt?: WriteOptions): Promise<Table<T>> {
+  /**
-    let writeOptions: WriteOptions = new DefaultWriteOptions()
+   * Creates a new Table and initialize it with new data.
-    if (opt !== undefined && isWriteOptions(opt)) {
+   *
-      writeOptions = opt
+   * @param name The name of the table.
-    } else if (optsOrEmbedding !== undefined && isWriteOptions(optsOrEmbedding)) {
+   * @param data Non-empty Array of Records to be inserted into the Table
-      writeOptions = optsOrEmbedding
+   * @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
     }
 
-    let embeddings: undefined | EmbeddingFunction<T>
+    const createArgs = [this._db, name, await fromRecordsToBuffer(data, embeddings), mode.toLowerCase()]
-    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)
@@ -473,23 +459,6 @@ 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

@@ -18,15 +18,13 @@ import { tableFromIPC, type Table as ArrowTable } from 'apache-arrow'
 
 export class HttpLancedbClient {
   private readonly _url: string
-  private readonly _apiKey: () => string
 
   public constructor (
     url: string,
-    apiKey: string,
+    private readonly _apiKey: string,
     private readonly _dbName?: string
   ) {
     this._url = url
-    this._apiKey = () => apiKey
   }
 
   get uri (): string {
@@ -43,7 +41,7 @@ export class HttpLancedbClient {
     filter?: string
   ): Promise<ArrowTable<any>> {
     const response = await axios.post(
-              `${this._url}/v1/table/${tableName}/query/`,
+              `${this._url}/v1/table/${tableName}`,
               {
                 vector,
                 k,
@@ -55,7 +53,7 @@ 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',
@@ -86,7 +84,7 @@ export class HttpLancedbClient {
       {
         headers: {
           'Content-Type': 'application/json',
-          'x-api-key': this._apiKey()
+          'x-api-key': this._apiKey
         },
         params,
         timeout: 10000

node/src/test/embedding/openai.ts

@@ -16,7 +16,6 @@ import { describe } from 'mocha'
 import { assert } from 'chai'
 
 import { OpenAIEmbeddingFunction } from '../../embedding/openai'
-import { isEmbeddingFunction } from '../../embedding/embedding_function'
 
 // eslint-disable-next-line @typescript-eslint/no-var-requires
 const { OpenAIApi } = require('openai')
@@ -48,10 +47,4 @@ describe('OpenAPIEmbeddings', function () {
       assert.deepEqual(vectors[1], stubValue.data.data[1].embedding)
     })
   })
-
-  describe('isEmbeddingFunction', function () {
-    it('should match the isEmbeddingFunction guard', function () {
-      assert.isTrue(isEmbeddingFunction(new OpenAIEmbeddingFunction('text', 'sk-key')))
-    })
-  })
 })

node/src/test/test.ts

@@ -18,7 +18,8 @@ import * as chai from 'chai'
 import * as chaiAsPromised from 'chai-as-promised'
 
 import * as lancedb from '../index'
-import { type AwsCredentials, type EmbeddingFunction, MetricType, Query, WriteMode, DefaultWriteOptions, isWriteOptions } from '../index'
+import { type AwsCredentials, type EmbeddingFunction, MetricType, WriteMode } from '../index'
+import { Query } from '../query'
 
 const expect = chai.expect
 const assert = chai.assert
@@ -134,18 +135,6 @@ describe('LanceDB client', function () {
       assert.equal(await table.countRows(), 2)
     })
 
-    it('fails to create a new table when the vector column is missing', async function () {
-      const dir = await track().mkdir('lancejs')
-      const con = await lancedb.connect(dir)
-
-      const data = [
-        { id: 1, price: 10 }
-      ]
-
-      const create = con.createTable('missing_vector', data)
-      await expect(create).to.be.rejectedWith(Error, 'column \'vector\' is missing')
-    })
-
     it('use overwrite flag to overwrite existing table', async function () {
       const dir = await track().mkdir('lancejs')
       const con = await lancedb.connect(dir)
@@ -156,7 +145,7 @@ describe('LanceDB client', function () {
       ]
 
       const tableName = 'overwrite'
-      await con.createTable(tableName, data, { writeMode: WriteMode.Create })
+      await con.createTable(tableName, data, WriteMode.Create)
 
       const newData = [
         { id: 1, vector: [0.1, 0.2], price: 10 },
@@ -166,7 +155,7 @@ describe('LanceDB client', function () {
 
       await expect(con.createTable(tableName, newData)).to.be.rejectedWith(Error, 'already exists')
 
-      const table = await con.createTable(tableName, newData, { writeMode: WriteMode.Overwrite })
+      const table = await con.createTable(tableName, newData, WriteMode.Overwrite)
       assert.equal(table.name, tableName)
       assert.equal(await table.countRows(), 3)
     })
@@ -242,22 +231,6 @@ describe('LanceDB client', function () {
       // Default replace = true
       await table.createIndex({ type: 'ivf_pq', column: 'vector', num_partitions: 2, max_iters: 2, num_sub_vectors: 2 })
     }).timeout(50_000)
-
-    it('it should fail when the column is not a vector', async function () {
-      const uri = await createTestDB(32, 300)
-      const con = await lancedb.connect(uri)
-      const table = await con.openTable('vectors')
-      const createIndex = table.createIndex({ type: 'ivf_pq', column: 'name', num_partitions: 2, max_iters: 2, num_sub_vectors: 2 })
-      await expect(createIndex).to.be.rejectedWith(/VectorIndex requires the column data type to be fixed size list of float32s/)
-    })
-
-    it('it should fail when the column is not a vector', async function () {
-      const uri = await createTestDB(32, 300)
-      const con = await lancedb.connect(uri)
-      const table = await con.openTable('vectors')
-      const createIndex = table.createIndex({ type: 'ivf_pq', column: 'name', num_partitions: -1, max_iters: 2, num_sub_vectors: 2 })
-      await expect(createIndex).to.be.rejectedWith('num_partitions: must be > 0')
-    })
   })
 
   describe('when using a custom embedding function', function () {
@@ -287,7 +260,7 @@ describe('LanceDB client', function () {
         { price: 10, name: 'foo' },
         { price: 50, name: 'bar' }
       ]
-      const table = await con.createTable('vectors', data, embeddings, { writeMode: WriteMode.Create })
+      const table = await con.createTable('vectors', data, WriteMode.Create, embeddings)
       const results = await table.search('foo').execute()
       assert.equal(results.length, 2)
     })
@@ -345,20 +318,3 @@ 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 = 0.1.16
+current_version = 0.1.12
 commit = True
 message = [python] Bump version: {current_version} → {new_version}
 tag = True

python/lancedb/__init__.py

@@ -19,11 +19,7 @@ from .schema import vector
 
 
 def connect(
-    uri: URI,
+    uri: URI, *, api_key: Optional[str] = None, region: str = "us-west-2"
-    *,
-    api_key: Optional[str] = None,
-    region: str = "us-west-2",
-    host_override: Optional[str] = None,
 ) -> DBConnection:
     """Connect to a LanceDB database.
 
@@ -59,5 +55,5 @@ 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, host_override)
+        return RemoteDBConnection(uri, api_key, region)
     return LanceDBConnection(uri)

python/lancedb/common.py

@@ -11,18 +11,17 @@
 #  See the License for the specific language governing permissions and
 #  limitations under the License.
 from pathlib import Path
-from typing import Iterable, List, Union
+from typing import List, Union
 
 import numpy as np
+import pandas as pd
 import pyarrow as pa
 
-from .util import safe_import_pandas
-
-pd = safe_import_pandas()
-
-DATA = Union[List[dict], dict, "pd.DataFrame", pa.Table, Iterable[pa.RecordBatch]]
 VEC = Union[list, np.ndarray, pa.Array, pa.ChunkedArray]
 URI = Union[str, Path]
+
+# TODO support generator
+DATA = Union[List[dict], dict, pd.DataFrame]
 VECTOR_COLUMN_NAME = "vector"
 
 

python/lancedb/context.py

@@ -12,13 +12,12 @@
 #  limitations under the License.
 from __future__ import annotations
 
+import pandas as pd
+
 from .exceptions import MissingColumnError, MissingValueError
-from .util import safe_import_pandas
-
-pd = safe_import_pandas()
 
 
-def contextualize(raw_df: "pd.DataFrame") -> Contextualizer:
+def contextualize(raw_df: pd.DataFrame) -> Contextualizer:
     """Create a Contextualizer object for the given DataFrame.
 
     Used to create context windows. Context windows are rolling subsets of text
@@ -176,12 +175,8 @@ class Contextualizer:
         self._min_window_size = min_window_size
         return self
 
-    def to_df(self) -> "pd.DataFrame":
+    def to_df(self) -> pd.DataFrame:
         """Create the context windows and return a DataFrame."""
-        if pd is None:
-            raise ImportError(
-                "pandas is required to create context windows using lancedb"
-            )
 
         if self._text_col not in self._raw_df.columns.tolist():
             raise MissingColumnError(self._text_col)

python/lancedb/db.py

@@ -16,8 +16,9 @@ from __future__ import annotations
 import os
 from abc import ABC, abstractmethod
 from pathlib import Path
-from typing import Optional
+from typing import Dict, Iterable, List, Optional, Tuple, Union
 
+import pandas as pd
 import pyarrow as pa
 from pyarrow import fs
 
@@ -38,7 +39,9 @@ class DBConnection(ABC):
     def create_table(
         self,
         name: str,
-        data: Optional[DATA] = None,
+        data: Optional[
+            Union[List[dict], dict, pd.DataFrame, pa.Table, Iterable[pa.RecordBatch]],
+        ] = None,
         schema: Optional[pa.Schema] = None,
         mode: str = "create",
         on_bad_vectors: str = "error",
@@ -276,7 +279,7 @@ class LanceDBConnection(DBConnection):
     def create_table(
         self,
         name: str,
-        data: Optional[DATA] = None,
+        data: Optional[Union[List[dict], dict, pd.DataFrame]] = None,
         schema: pa.Schema = None,
         mode: str = "create",
         on_bad_vectors: str = "error",
@@ -316,20 +319,14 @@ class LanceDBConnection(DBConnection):
         """
         return LanceTable.open(self, name)
 
-    def drop_table(self, name: str, ignore_missing: bool = False):
+    def drop_table(self, name: str):
         """Drop a table from the database.
 
         Parameters
         ----------
         name: str
             The name of the table.
-        ignore_missing: bool, default False
-            If True, ignore if the table does not exist.
         """
-        try:
+        filesystem, path = pa.fs.FileSystem.from_uri(self.uri)
-            filesystem, path = fs_from_uri(self.uri)
+        table_path = os.path.join(path, name + ".lance")
-            table_path = os.path.join(path, name + ".lance")
+        filesystem.delete_dir(table_path)
-            filesystem.delete_dir(table_path)
-        except FileNotFoundError:
-            if not ignore_missing:
-                raise

python/lancedb/embeddings.py

@@ -16,19 +16,15 @@ import sys
 from typing import Callable, Union
 
 import numpy as np
+import pandas as pd
 import pyarrow as pa
 from lance.vector import vec_to_table
 from retry import retry
 
-from .util import safe_import_pandas
-
-pd = safe_import_pandas()
-DATA = Union[pa.Table, "pd.DataFrame"]
-
 
 def with_embeddings(
     func: Callable,
-    data: DATA,
+    data: Union[pa.Table, pd.DataFrame],
     column: str = "text",
     wrap_api: bool = True,
     show_progress: bool = False,
@@ -64,7 +60,7 @@ def with_embeddings(
     func = func.batch_size(batch_size)
     if show_progress:
         func = func.show_progress()
-    if pd is not None and isinstance(data, pd.DataFrame):
+    if isinstance(data, pd.DataFrame):
         data = pa.Table.from_pandas(data, preserve_index=False)
     embeddings = func(data[column].to_numpy())
     table = vec_to_table(np.array(embeddings))

python/lancedb/pydantic.py

@@ -249,36 +249,3 @@ def pydantic_to_schema(model: Type[pydantic.BaseModel]) -> pa.Schema:
     """
     fields = _pydantic_model_to_fields(model)
     return pa.schema(fields)
-
-
-class LanceModel(pydantic.BaseModel):
-    """
-    A Pydantic Model base class that can be converted to a LanceDB Table.
-
-    Examples
-    --------
-    >>> import lancedb
-    >>> from lancedb.pydantic import LanceModel, vector
-    >>>
-    >>> class TestModel(LanceModel):
-    ...     name: str
-    ...     vector: vector(2)
-    ...
-    >>> db = lancedb.connect("/tmp")
-    >>> table = db.create_table("test", schema=TestModel.to_arrow_schema())
-    >>> table.add([
-    ...     TestModel(name="test", vector=[1.0, 2.0])
-    ... ])
-    >>> table.search([0., 0.]).limit(1).to_pydantic(TestModel)
-    [TestModel(name='test', vector=FixedSizeList(dim=2))]
-    """
-
-    @classmethod
-    def to_arrow_schema(cls):
-        return pydantic_to_schema(cls)
-
-    @classmethod
-    def field_names(cls) -> List[str]:
-        if PYDANTIC_VERSION.major < 2:
-            return list(cls.__fields__.keys())
-        return list(cls.model_fields.keys())

python/lancedb/query.py

@@ -13,20 +13,17 @@
 
 from __future__ import annotations
 
-from typing import List, Literal, Optional, Type, Union
+from typing import List, Literal, Optional, Union
 
 import numpy as np
+import pandas as pd
 import pyarrow as pa
-import pydantic
+from pydantic import BaseModel
 
 from .common import VECTOR_COLUMN_NAME
-from .pydantic import LanceModel
-from .util import safe_import_pandas
-
-pd = safe_import_pandas()
 
 
-class Query(pydantic.BaseModel):
+class Query(BaseModel):
     """A Query"""
 
     vector_column: str = VECTOR_COLUMN_NAME
@@ -201,7 +198,7 @@ class LanceQueryBuilder:
         self._refine_factor = refine_factor
         return self
 
-    def to_df(self) -> "pd.DataFrame":
+    def to_df(self) -> pd.DataFrame:
         """
         Execute the query and return the results as a pandas DataFrame.
         In addition to the selected columns, LanceDB also returns a vector
@@ -233,26 +230,9 @@ class LanceQueryBuilder:
         )
         return self._table._execute_query(query)
 
-    def to_pydantic(self, model: Type[LanceModel]) -> List[LanceModel]:
-        """Return the table as a list of pydantic models.
-
-        Parameters
-        ----------
-        model: Type[LanceModel]
-            The pydantic model to use.
-
-        Returns
-        -------
-        List[LanceModel]
-        """
-        return [
-            model(**{k: v for k, v in row.items() if k in model.field_names()})
-            for row in self.to_arrow().to_pylist()
-        ]
-
 
 class LanceFtsQueryBuilder(LanceQueryBuilder):
-    def to_arrow(self) -> pa.Table:
+    def to_arrow(self) -> pd.Table:
         try:
             import tantivy
         except ImportError:

python/lancedb/remote/client.py

@@ -48,16 +48,11 @@ 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 = (
+        url = f"https://{self.db_name}.{self.region}.api.lancedb.com"
-            self.host_override
-            or f"https://{self.db_name}.{self.region}.api.lancedb.com"
-        )
         return aiohttp.ClientSession(url)
 
     async def close(self):
@@ -71,8 +66,6 @@ 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
@@ -105,7 +98,7 @@ class RestfulLanceDBClient:
     async def post(
         self,
         uri: str,
-        data: Optional[Union[Dict[str, Any], BaseModel, bytes]] = None,
+        data: Union[Dict[str, Any], BaseModel, bytes],
         params: Optional[Dict[str, Any]] = None,
         content_type: Optional[str] = None,
         deserialize: Callable = lambda resp: resp.json(),
@@ -148,7 +141,5 @@ class RestfulLanceDBClient:
     @_check_not_closed
     async def query(self, table_name: str, query: VectorQuery) -> VectorQueryResult:
         """Query a table."""
-        tbl = await self.post(
+        tbl = await self.post(f"/v1/table/{table_name}/", query, deserialize=_read_ipc)
-            f"/v1/table/{table_name}/query/", query, deserialize=_read_ipc
-        )
         return VectorQueryResult(tbl)

python/lancedb/remote/db.py

@@ -13,13 +13,14 @@
 
 import asyncio
 import uuid
-from typing import List, Optional
+from typing import List
 from urllib.parse import urlparse
 
 import pyarrow as pa
 
 from lancedb.common import DATA
 from lancedb.db import DBConnection
+from lancedb.schema import schema_to_json
 from lancedb.table import Table, _sanitize_data
 
 from .arrow import to_ipc_binary
@@ -29,22 +30,14 @@ from .client import ARROW_STREAM_CONTENT_TYPE, RestfulLanceDBClient
 class RemoteDBConnection(DBConnection):
     """A connection to a remote LanceDB database."""
 
-    def __init__(
+    def __init__(self, db_url: str, api_key: str, region: str):
-        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._client = RestfulLanceDBClient(self.db_name, region, api_key)
-            self.db_name, region, api_key, host_override
-        )
         try:
             self._loop = asyncio.get_running_loop()
         except RuntimeError:
@@ -102,7 +95,7 @@ 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

@@ -16,11 +16,11 @@ from functools import cached_property
 from typing import Union
 
 import pyarrow as pa
-from lance import json_to_schema
 
 from lancedb.common import DATA, VEC, VECTOR_COLUMN_NAME
 
-from ..query import LanceQueryBuilder
+from ..query import LanceQueryBuilder, Query
+from ..schema import json_to_schema
 from ..table import Query, Table, _sanitize_data
 from .arrow import to_ipc_binary
 from .client import ARROW_STREAM_CONTENT_TYPE
@@ -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.post(f"/v1/table/{self._name}/describe/")
+            self._conn._client.get(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/schema.py

@@ -12,7 +12,11 @@
 #  limitations under the License.
 
 """Schema related utilities."""
+
+from typing import Any, Dict, Type
+
 import pyarrow as pa
+from lance import json_to_schema, schema_to_json
 
 
 def vector(dimension: int, value_type: pa.DataType = pa.float32()) -> pa.DataType:

python/lancedb/table.py

@@ -20,32 +20,26 @@ from typing import Iterable, List, Union
 
 import lance
 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 .pydantic import LanceModel
 from .query import LanceFtsQueryBuilder, LanceQueryBuilder, Query
-from .util import fs_from_uri, safe_import_pandas
-
-pd = safe_import_pandas()
 
 
 def _sanitize_data(data, schema, on_bad_vectors, fill_value):
     if isinstance(data, list):
-        # convert to list of dict if data is a bunch of LanceModels
-        if isinstance(data[0], LanceModel):
-            schema = data[0].__class__.to_arrow_schema()
-            data = [dict(d) for d in data]
         data = pa.Table.from_pylist(data)
         data = _sanitize_schema(
             data, schema=schema, on_bad_vectors=on_bad_vectors, fill_value=fill_value
         )
     if isinstance(data, dict):
         data = vec_to_table(data)
-    if pd is not None and isinstance(data, pd.DataFrame):
+    if isinstance(data, pd.DataFrame):
         data = pa.Table.from_pandas(data)
         data = _sanitize_schema(
             data, schema=schema, on_bad_vectors=on_bad_vectors, fill_value=fill_value
@@ -100,7 +94,7 @@ class Table(ABC):
         """
         raise NotImplementedError
 
-    def to_pandas(self):
+    def to_pandas(self) -> pd.DataFrame:
         """Return the table as a pandas DataFrame.
 
         Returns
@@ -334,7 +328,7 @@ class LanceTable(Table):
         """Return the first n rows of the table."""
         return self._dataset.head(n)
 
-    def to_pandas(self) -> "pd.DataFrame":
+    def to_pandas(self) -> pd.DataFrame:
         """Return the table as a pandas DataFrame.
 
         Returns
@@ -533,7 +527,7 @@ class LanceTable(Table):
     @classmethod
     def open(cls, db, name):
         tbl = cls(db, name)
-        fs, path = fs_from_uri(tbl._dataset_uri)
+        fs, path = pa.fs.FileSystem.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

@@ -15,6 +15,7 @@ import os
 from typing import Tuple
 from urllib.parse import urlparse
 
+import pyarrow as pa
 import pyarrow.fs as pa_fs
 
 
@@ -70,17 +71,7 @@ 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":
-        fs = pa_fs.S3FileSystem(endpoint_override=os.environ.get("AWS_ENDPOINT"))
+        if os.environ["AWS_ENDPOINT"]:
-        path = get_uri_location(uri)
+            uri += "?endpoint_override=" + os.environ["AWS_ENDPOINT"]
-        return fs, path
 
     return pa_fs.FileSystem.from_uri(uri)
-
-
-def safe_import_pandas():
-    try:
-        import pandas as pd
-
-        return pd
-    except ImportError:
-        return None

python/pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "lancedb"
-version = "0.1.16"
+version = "0.1.12"
-dependencies = ["pylance==0.5.10", "ratelimiter", "retry", "tqdm", "aiohttp", "pydantic", "attr", "semver"]
+dependencies = ["pylance~=0.5.8", "ratelimiter", "retry", "tqdm", "aiohttp", "pydantic", "attr", "semver"]
 description = "lancedb"
 authors = [
     { name = "LanceDB Devs", email = "dev@lancedb.com" },
@@ -37,7 +37,7 @@ repository = "https://github.com/lancedb/lancedb"
 
 [project.optional-dependencies]
 tests = [
-    "pandas>=1.4", "pytest", "pytest-mock", "pytest-asyncio"
+    "pytest", "pytest-mock", "pytest-asyncio"
 ]
 dev = [
     "ruff", "pre-commit", "black"

python/tests/test_db.py

@@ -149,10 +149,6 @@ def test_delete_table(tmp_path):
     db.create_table("test", data=data)
     assert db.table_names() == ["test"]
 
-    # dropping a table that does not exist should pass
-    # if ignore_missing=True
-    db.drop_table("does_not_exist", ignore_missing=True)
-
 
 def test_empty_or_nonexistent_table(tmp_path):
     db = lancedb.connect(tmp_path)

python/tests/test_pydantic.py

@@ -20,7 +20,7 @@ import pyarrow as pa
 import pydantic
 import pytest
 
-from lancedb.pydantic import PYDANTIC_VERSION, LanceModel, pydantic_to_schema, vector
+from lancedb.pydantic import PYDANTIC_VERSION, pydantic_to_schema, vector
 
 
 @pytest.mark.skipif(
@@ -163,13 +163,3 @@ def test_fixed_size_list_validation():
         TestModel(vec=range(7))
 
     TestModel(vec=range(8))
-
-
-def test_lance_model():
-    class TestModel(LanceModel):
-        vec: vector(16)
-        li: List[int]
-
-    schema = pydantic_to_schema(TestModel)
-    assert schema == TestModel.to_arrow_schema()
-    assert TestModel.field_names() == ["vec", "li"]

python/tests/test_query.py

@@ -20,7 +20,6 @@ import pyarrow as pa
 import pytest
 
 from lancedb.db import LanceDBConnection
-from lancedb.pydantic import LanceModel, vector
 from lancedb.query import LanceQueryBuilder, Query
 from lancedb.table import LanceTable
 
@@ -65,24 +64,6 @@ def table(tmp_path) -> MockTable:
     return MockTable(tmp_path)
 
 
-def test_cast(table):
-    class TestModel(LanceModel):
-        vector: vector(2)
-        id: int
-        str_field: str
-        float_field: float
-
-    q = LanceQueryBuilder(table, [0, 0], "vector").limit(1)
-    results = q.to_pydantic(TestModel)
-    assert len(results) == 1
-    r0 = results[0]
-    assert isinstance(r0, TestModel)
-    assert r0.id == 1
-    assert r0.vector == [1, 2]
-    assert r0.str_field == "a"
-    assert r0.float_field == 1.0
-
-
 def test_query_builder(table):
     df = LanceQueryBuilder(table, [0, 0], "vector").limit(1).select(["id"]).to_df()
     assert df["id"].values[0] == 1

python/tests/test_table.py

@@ -13,16 +13,15 @@
 
 import functools
 from pathlib import Path
-from typing import List
 from unittest.mock import PropertyMock, patch
 
 import numpy as np
 import pandas as pd
 import pyarrow as pa
 import pytest
+from lance.vector import vec_to_table
 
 from lancedb.db import LanceDBConnection
-from lancedb.pydantic import LanceModel, vector
 from lancedb.table import LanceTable
 
 
@@ -136,17 +135,6 @@ def test_add(db):
     _add(table, schema)
 
 
-def test_add_pydantic_model(db):
-    class TestModel(LanceModel):
-        vector: vector(16)
-        li: List[int]
-
-    data = TestModel(vector=list(range(16)), li=[1, 2, 3])
-    table = LanceTable.create(db, "test", data=[data])
-    assert len(table) == 1
-    assert table.schema == TestModel.to_arrow_schema()
-
-
 def _add(table, schema):
     # table = LanceTable(db, "test")
     assert len(table) == 2

rust/ffi/node/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "vectordb-node"
-version = "0.1.19"
+version = "0.1.15"
 description = "Serverless, low-latency vector database for AI applications"
 license = "Apache-2.0"
 edition = "2018"
@@ -13,7 +13,6 @@ crate-type = ["cdylib"]
 arrow-array = { workspace = true }
 arrow-ipc = { workspace = true }
 arrow-schema = { workspace = true }
-conv = "0.3.3"
 once_cell = "1"
 futures = "0.3"
 half = { workspace = true }
@@ -22,6 +21,5 @@ vectordb = { path = "../../vectordb" }
 tokio = { version = "1.23", features = ["rt-multi-thread"] }
 neon = {version = "0.10.1", default-features = false, features = ["channel-api", "napi-6", "promise-api", "task-api"] }
 object_store = { workspace = true, features = ["aws"] }
-snafu = { workspace = true }
 async-trait = "0"
 env_logger = "0"

rust/ffi/node/src/arrow.rs

@@ -13,30 +13,27 @@
 // limitations under the License.
 
 use std::io::Cursor;
-use std::ops::Deref;
 use std::sync::Arc;
 
 use arrow_array::cast::as_list_array;
-use arrow_array::{Array, ArrayRef, FixedSizeListArray, RecordBatch};
+use arrow_array::{Array, FixedSizeListArray, RecordBatch};
 use arrow_ipc::reader::FileReader;
-use arrow_ipc::writer::FileWriter;
 use arrow_schema::{DataType, Field, Schema};
 use lance::arrow::{FixedSizeListArrayExt, RecordBatchExt};
-use vectordb::table::VECTOR_COLUMN_NAME;
 
-use crate::error::{MissingColumnSnafu, Result};
+pub(crate) fn convert_record_batch(record_batch: RecordBatch) -> RecordBatch {
-use snafu::prelude::*;
+    let column = record_batch
-
+        .column_by_name("vector")
-pub(crate) fn convert_record_batch(record_batch: RecordBatch) -> Result<RecordBatch> {
+        .cloned()
-    let column = get_column(VECTOR_COLUMN_NAME, &record_batch)?;
+        .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 = FixedSizeListArray::try_new_from_values(arr.values().to_owned(), list_size as i32)?;
+    let r =
+        FixedSizeListArray::try_new_from_values(arr.values().to_owned(), list_size as i32).unwrap();
 
     let schema = Arc::new(Schema::new(vec![Field::new(
-        VECTOR_COLUMN_NAME,
+        "vector",
         DataType::FixedSizeList(
             Arc::new(Field::new("item", DataType::Float32, true)),
             list_size as i32,
@@ -44,42 +41,22 @@ pub(crate) fn convert_record_batch(record_batch: RecordBatch) -> Result<RecordBa
         true,
     )]));
 
-    let mut new_batch = RecordBatch::try_new(schema.clone(), vec![Arc::new(r)])?;
+    let mut new_batch = RecordBatch::try_new(schema.clone(), vec![Arc::new(r)]).unwrap();
 
     if record_batch.num_columns() > 1 {
-        let rb = record_batch.drop_column(VECTOR_COLUMN_NAME)?;
+        let rb = record_batch.drop_column("vector").unwrap();
-        new_batch = new_batch.merge(&rb)?;
+        new_batch = new_batch.merge(&rb).unwrap();
     }
-    Ok(new_batch)
+    new_batch
 }
 
-fn get_column(column_name: &str, record_batch: &RecordBatch) -> Result<ArrayRef> {
+pub(crate) fn arrow_buffer_to_record_batch(slice: &[u8]) -> Vec<RecordBatch> {
-    record_batch
-        .column_by_name(column_name)
-        .cloned()
-        .context(MissingColumnSnafu { name: column_name })
-}
-
-pub(crate) fn arrow_buffer_to_record_batch(slice: &[u8]) -> Result<Vec<RecordBatch>> {
     let mut batches: Vec<RecordBatch> = Vec::new();
-    let file_reader = FileReader::try_new(Cursor::new(slice), None)?;
+    let fr = FileReader::try_new(Cursor::new(slice), None);
+    let file_reader = fr.unwrap();
     for b in file_reader {
-        let record_batch = convert_record_batch(b?)?;
+        let record_batch = convert_record_batch(b.unwrap());
         batches.push(record_batch);
     }
-    Ok(batches)
+    batches
-}
-
-pub(crate) fn record_batch_to_buffer(batches: Vec<RecordBatch>) -> Result<Vec<u8>> {
-    if batches.is_empty() {
-        return Ok(Vec::new());
-    }
-
-    let schema = batches.get(0).unwrap().schema();
-    let mut fr = FileWriter::try_new(Vec::new(), schema.deref())?;
-    for batch in batches.iter() {
-        fr.write(batch)?
-    }
-    fr.finish()?;
-    Ok(fr.into_inner()?)
 }

rust/ffi/node/src/error.rs

@@ -1,88 +0,0 @@
-// Copyright 2023 Lance Developers.
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-use arrow_schema::ArrowError;
-use neon::context::Context;
-use neon::prelude::NeonResult;
-use snafu::Snafu;
-
-#[derive(Debug, Snafu)]
-#[snafu(visibility(pub(crate)))]
-pub enum Error {
-    #[snafu(display("column '{name}' is missing"))]
-    MissingColumn { name: String },
-    #[snafu(display("{name}: {message}"))]
-    RangeError { name: String, message: String },
-    #[snafu(display("{index_type} is not a valid index type"))]
-    InvalidIndexType { index_type: String },
-
-    #[snafu(display("{message}"))]
-    LanceDB { message: String },
-    #[snafu(display("{message}"))]
-    Neon { message: String },
-}
-
-pub type Result<T> = std::result::Result<T, Error>;
-
-impl From<vectordb::error::Error> for Error {
-    fn from(e: vectordb::error::Error) -> Self {
-        Self::LanceDB {
-            message: e.to_string(),
-        }
-    }
-}
-
-impl From<lance::Error> for Error {
-    fn from(e: lance::Error) -> Self {
-        Self::LanceDB {
-            message: e.to_string(),
-        }
-    }
-}
-
-impl From<ArrowError> for Error {
-    fn from(value: ArrowError) -> Self {
-        Self::LanceDB {
-            message: value.to_string(),
-        }
-    }
-}
-
-impl From<neon::result::Throw> for Error {
-    fn from(value: neon::result::Throw) -> Self {
-        Self::Neon {
-            message: value.to_string(),
-        }
-    }
-}
-
-/// ResultExt is used to transform a [`Result`] into a [`NeonResult`],
-/// so it can be returned as a JavaScript error
-/// Copied from [Neon](https://github.com/neon-bindings/neon/blob/4c2e455a9e6814f1ba0178616d63caec7f4df317/crates/neon/src/result/mod.rs#L88)
-pub trait ResultExt<T> {
-    fn or_throw<'a, C: Context<'a>>(self, cx: &mut C) -> NeonResult<T>;
-}
-
-/// Implement ResultExt for the std Result so it can be used any Result type
-impl<T, E> ResultExt<T> for std::result::Result<T, E>
-where
-    E: std::fmt::Display,
-{
-    fn or_throw<'a, C: Context<'a>>(self, cx: &mut C) -> NeonResult<T> {
-        match self {
-            Ok(value) => Ok(value),
-            Err(error) => cx.throw_error(error.to_string()),
-        }
-    }
-}

rust/ffi/node/src/index/vector.rs

@@ -22,15 +22,12 @@ use neon::prelude::*;
 
 use vectordb::index::vector::{IvfPQIndexBuilder, VectorIndexBuilder};
 
-use crate::error::Error::InvalidIndexType;
-use crate::error::ResultExt;
-use crate::neon_ext::js_object_ext::JsObjectExt;
 use crate::{runtime, JsTable};
 
 pub(crate) fn table_create_vector_index(mut cx: FunctionContext) -> JsResult<JsPromise> {
     let js_table = cx.this().downcast_or_throw::<JsBox<JsTable>, _>(&mut cx)?;
     let index_params = cx.argument::<JsObject>(0)?;
-    let index_params_builder = get_index_params_builder(&mut cx, index_params).or_throw(&mut cx)?;
+    let index_params_builder = get_index_params_builder(&mut cx, index_params).unwrap();
 
     let rt = runtime(&mut cx)?;
     let channel = cx.channel();
@@ -57,21 +54,27 @@ pub(crate) fn table_create_vector_index(mut cx: FunctionContext) -> JsResult<JsP
 fn get_index_params_builder(
     cx: &mut FunctionContext,
     obj: Handle<JsObject>,
-) -> crate::error::Result<impl VectorIndexBuilder> {
+) -> Result<impl VectorIndexBuilder, String> {
-    let idx_type = obj.get::<JsString, _, _>(cx, "type")?.value(cx);
+    let idx_type = obj
+        .get::<JsString, _, _>(cx, "type")
+        .map_err(|t| t.to_string())?
+        .value(cx);
 
     match idx_type.as_str() {
         "ivf_pq" => {
             let mut index_builder: IvfPQIndexBuilder = IvfPQIndexBuilder::new();
             let mut pq_params = PQBuildParams::default();
 
-            obj.get_opt::<JsString, _, _>(cx, "column")?
+            obj.get_opt::<JsString, _, _>(cx, "column")
+                .map_err(|t| t.to_string())?
                 .map(|s| index_builder.column(s.value(cx)));
 
-            obj.get_opt::<JsString, _, _>(cx, "index_name")?
+            obj.get_opt::<JsString, _, _>(cx, "index_name")
+                .map_err(|t| t.to_string())?
                 .map(|s| index_builder.index_name(s.value(cx)));
 
-            obj.get_opt::<JsString, _, _>(cx, "metric_type")?
+            obj.get_opt::<JsString, _, _>(cx, "metric_type")
+                .map_err(|t| t.to_string())?
                 .map(|s| MetricType::try_from(s.value(cx).as_str()))
                 .map(|mt| {
                     let metric_type = mt.unwrap();
@@ -79,8 +82,15 @@ fn get_index_params_builder(
                     pq_params.metric_type = metric_type;
                 });
 
-            let num_partitions = obj.get_opt_usize(cx, "num_partitions")?;
+            let num_partitions = obj
-            let max_iters = obj.get_opt_usize(cx, "max_iters")?;
+                .get_opt::<JsNumber, _, _>(cx, "num_partitions")
+                .map_err(|t| t.to_string())?
+                .map(|s| s.value(cx) as usize);
+
+            let max_iters = obj
+                .get_opt::<JsNumber, _, _>(cx, "max_iters")
+                .map_err(|t| t.to_string())?
+                .map(|s| s.value(cx) as usize);
 
             num_partitions.map(|np| {
                 let max_iters = max_iters.unwrap_or(50);
@@ -92,28 +102,32 @@ fn get_index_params_builder(
                 index_builder.ivf_params(ivf_params)
             });
 
-            obj.get_opt::<JsBoolean, _, _>(cx, "use_opq")?
+            obj.get_opt::<JsBoolean, _, _>(cx, "use_opq")
+                .map_err(|t| t.to_string())?
                 .map(|s| pq_params.use_opq = s.value(cx));
 
-            obj.get_opt_usize(cx, "num_sub_vectors")?
+            obj.get_opt::<JsNumber, _, _>(cx, "num_sub_vectors")
-                .map(|s| pq_params.num_sub_vectors = s);
+                .map_err(|t| t.to_string())?
+                .map(|s| pq_params.num_sub_vectors = s.value(cx) as usize);
 
-            obj.get_opt_usize(cx, "num_bits")?
+            obj.get_opt::<JsNumber, _, _>(cx, "num_bits")
-                .map(|s| pq_params.num_bits = s);
+                .map_err(|t| t.to_string())?
+                .map(|s| pq_params.num_bits = s.value(cx) as usize);
 
-            obj.get_opt_usize(cx, "max_iters")?
+            obj.get_opt::<JsNumber, _, _>(cx, "max_iters")
-                .map(|s| pq_params.max_iters = s);
+                .map_err(|t| t.to_string())?
+                .map(|s| pq_params.max_iters = s.value(cx) as usize);
 
-            obj.get_opt_usize(cx, "max_opq_iters")?
+            obj.get_opt::<JsNumber, _, _>(cx, "max_opq_iters")
-                .map(|s| pq_params.max_opq_iters = s);
+                .map_err(|t| t.to_string())?
+                .map(|s| pq_params.max_opq_iters = s.value(cx) as usize);
 
-            obj.get_opt::<JsBoolean, _, _>(cx, "replace")?
+            obj.get_opt::<JsBoolean, _, _>(cx, "replace")
+                .map_err(|t| t.to_string())?
                 .map(|s| index_builder.replace(s.value(cx)));
 
             Ok(index_builder)
         }
-        index_type => Err(InvalidIndexType {
+        t => Err(format!("{} is not a valid index type", t).to_string()),
-            index_type: index_type.into(),
-        }),
     }
 }

rust/ffi/node/src/lib.rs

@@ -18,6 +18,7 @@ use std::ops::Deref;
 use std::sync::{Arc, Mutex};
 
 use arrow_array::{Float32Array, RecordBatchIterator};
+use arrow_ipc::writer::FileWriter;
 use async_trait::async_trait;
 use futures::{TryFutureExt, TryStreamExt};
 use lance::dataset::{WriteMode, WriteParams};
@@ -31,17 +32,14 @@ use once_cell::sync::OnceCell;
 use tokio::runtime::Runtime;
 
 use vectordb::database::Database;
+use vectordb::error::Error;
 use vectordb::table::{ReadParams, Table};
 
-use crate::arrow::{arrow_buffer_to_record_batch, record_batch_to_buffer};
+use crate::arrow::arrow_buffer_to_record_batch;
-use crate::error::ResultExt;
-use crate::neon_ext::js_object_ext::JsObjectExt;
 
 mod arrow;
 mod convert;
-mod error;
 mod index;
-mod neon_ext;
 
 struct JsDatabase {
     database: Arc<Database>,
@@ -56,7 +54,7 @@ struct JsTable {
 impl Finalize for JsTable {}
 
 // TODO: object_store didn't export this type so I copied it.
-// Make a request to object_store to export this type
+// Make a requiest to object_store to export this type
 #[derive(Debug)]
 pub struct StaticCredentialProvider<T> {
     credential: Arc<T>,
@@ -88,7 +86,7 @@ fn runtime<'a, C: Context<'a>>(cx: &mut C) -> NeonResult<&'static Runtime> {
 
     LOG.get_or_init(|| env_logger::init());
 
-    RUNTIME.get_or_try_init(|| Runtime::new().or_throw(cx))
+    RUNTIME.get_or_try_init(|| Runtime::new().or_else(|err| cx.throw_error(err.to_string())))
 }
 
 fn database_new(mut cx: FunctionContext) -> JsResult<JsPromise> {
@@ -103,7 +101,7 @@ fn database_new(mut cx: FunctionContext) -> JsResult<JsPromise> {
 
         deferred.settle_with(&channel, move |mut cx| {
             let db = JsDatabase {
-                database: Arc::new(database.or_throw(&mut cx)?),
+                database: Arc::new(database.or_else(|err| cx.throw_error(err.to_string()))?),
             };
             Ok(cx.boxed(db))
         });
@@ -125,7 +123,7 @@ fn database_table_names(mut cx: FunctionContext) -> JsResult<JsPromise> {
         let tables_rst = database.table_names().await;
 
         deferred.settle_with(&channel, move |mut cx| {
-            let tables = tables_rst.or_throw(&mut cx)?;
+            let tables = tables_rst.or_else(|err| cx.throw_error(err.to_string()))?;
             let table_names = convert::vec_str_to_array(&tables, &mut cx);
             table_names
         });
@@ -196,7 +194,9 @@ fn database_open_table(mut cx: FunctionContext) -> JsResult<JsPromise> {
         let table_rst = database.open_table_with_params(&table_name, &params).await;
 
         deferred.settle_with(&channel, move |mut cx| {
-            let table = Arc::new(Mutex::new(table_rst.or_throw(&mut cx)?));
+            let table = Arc::new(Mutex::new(
+                table_rst.or_else(|err| cx.throw_error(err.to_string()))?,
+            ));
             Ok(cx.boxed(JsTable { table }))
         });
     });
@@ -217,7 +217,7 @@ fn database_drop_table(mut cx: FunctionContext) -> JsResult<JsPromise> {
     rt.spawn(async move {
         let result = database.drop_table(&table_name).await;
         deferred.settle_with(&channel, move |mut cx| {
-            result.or_throw(&mut cx)?;
+            result.or_else(|err| cx.throw_error(err.to_string()))?;
             Ok(cx.null())
         });
     });
@@ -246,9 +246,12 @@ fn table_search(mut cx: FunctionContext) -> JsResult<JsPromise> {
         .get_opt::<JsString, _, _>(&mut cx, "_filter")?
         .map(|s| s.value(&mut cx));
     let refine_factor = query_obj
-        .get_opt_u32(&mut cx, "_refineFactor")
+        .get_opt::<JsNumber, _, _>(&mut cx, "_refineFactor")?
-        .or_throw(&mut cx)?;
+        .map(|s| s.value(&mut cx))
-    let nprobes = query_obj.get_usize(&mut cx, "_nprobes").or_throw(&mut cx)?;
+        .map(|i| i as u32);
+    let nprobes = query_obj
+        .get::<JsNumber, _, _>(&mut cx, "_nprobes")?
+        .value(&mut cx) as usize;
     let metric_type = query_obj
         .get_opt::<JsString, _, _>(&mut cx, "_metricType")?
         .map(|s| s.value(&mut cx))
@@ -275,17 +278,30 @@ fn table_search(mut cx: FunctionContext) -> JsResult<JsPromise> {
             .select(select);
         let record_batch_stream = builder.execute();
         let results = record_batch_stream
-            .and_then(|stream| {
+            .and_then(|stream| stream.try_collect::<Vec<_>>().map_err(Error::from))
-                stream
-                    .try_collect::<Vec<_>>()
-                    .map_err(vectordb::error::Error::from)
-            })
             .await;
 
         deferred.settle_with(&channel, move |mut cx| {
-            let results = results.or_throw(&mut cx)?;
+            let results = results.or_else(|err| cx.throw_error(err.to_string()))?;
-            let buffer = record_batch_to_buffer(results).or_throw(&mut cx)?;
+            let vector: Vec<u8> = Vec::new();
-            Ok(JsBuffer::external(&mut cx, buffer))
+
+            if results.is_empty() {
+                return cx.buffer(0);
+            }
+
+            let schema = results.get(0).unwrap().schema();
+            let mut fr = FileWriter::try_new(vector, schema.deref())
+                .or_else(|err| cx.throw_error(err.to_string()))?;
+
+            for batch in results.iter() {
+                fr.write(batch)
+                    .or_else(|err| cx.throw_error(err.to_string()))?;
+            }
+            fr.finish().or_else(|err| cx.throw_error(err.to_string()))?;
+            let buf = fr
+                .into_inner()
+                .or_else(|err| cx.throw_error(err.to_string()))?;
+            Ok(JsBuffer::external(&mut cx, buf))
         });
     });
     Ok(promise)
@@ -297,7 +313,7 @@ fn table_create(mut cx: FunctionContext) -> JsResult<JsPromise> {
         .downcast_or_throw::<JsBox<JsDatabase>, _>(&mut cx)?;
     let table_name = cx.argument::<JsString>(0)?.value(&mut cx);
     let buffer = cx.argument::<JsBuffer>(1)?;
-    let batches = arrow_buffer_to_record_batch(buffer.as_slice(&mut cx)).or_throw(&mut cx)?;
+    let batches = arrow_buffer_to_record_batch(buffer.as_slice(&mut cx));
     let schema = batches[0].schema();
 
     // Write mode
@@ -335,7 +351,9 @@ fn table_create(mut cx: FunctionContext) -> JsResult<JsPromise> {
             .await;
 
         deferred.settle_with(&channel, move |mut cx| {
-            let table = Arc::new(Mutex::new(table_rst.or_throw(&mut cx)?));
+            let table = Arc::new(Mutex::new(
+                table_rst.or_else(|err| cx.throw_error(err.to_string()))?,
+            ));
             Ok(cx.boxed(JsTable { table }))
         });
     });
@@ -352,8 +370,7 @@ fn table_add(mut cx: FunctionContext) -> JsResult<JsPromise> {
     let js_table = cx.this().downcast_or_throw::<JsBox<JsTable>, _>(&mut cx)?;
     let buffer = cx.argument::<JsBuffer>(0)?;
     let write_mode = cx.argument::<JsString>(1)?.value(&mut cx);
-
+    let batches = arrow_buffer_to_record_batch(buffer.as_slice(&mut cx));
-    let batches = arrow_buffer_to_record_batch(buffer.as_slice(&mut cx)).or_throw(&mut cx)?;
     let schema = batches[0].schema();
 
     let rt = runtime(&mut cx)?;
@@ -382,7 +399,7 @@ fn table_add(mut cx: FunctionContext) -> JsResult<JsPromise> {
         let add_result = table.lock().unwrap().add(batch_reader, Some(params)).await;
 
         deferred.settle_with(&channel, move |mut cx| {
-            let _added = add_result.or_throw(&mut cx)?;
+            let _added = add_result.or_else(|err| cx.throw_error(err.to_string()))?;
             Ok(cx.boolean(true))
         });
     });
@@ -401,7 +418,7 @@ fn table_count_rows(mut cx: FunctionContext) -> JsResult<JsPromise> {
         let num_rows_result = table.lock().unwrap().count_rows().await;
 
         deferred.settle_with(&channel, move |mut cx| {
-            let num_rows = num_rows_result.or_throw(&mut cx)?;
+            let num_rows = num_rows_result.or_else(|err| cx.throw_error(err.to_string()))?;
             Ok(cx.number(num_rows as f64))
         });
     });
@@ -421,7 +438,7 @@ fn table_delete(mut cx: FunctionContext) -> JsResult<JsPromise> {
     let delete_result = rt.block_on(async move { table.lock().unwrap().delete(&predicate).await });
 
     deferred.settle_with(&channel, move |mut cx| {
-        delete_result.or_throw(&mut cx)?;
+        delete_result.or_else(|err| cx.throw_error(err.to_string()))?;
         Ok(cx.undefined())
     });
 

rust/ffi/node/src/neon_ext.rs

@@ -1,15 +0,0 @@
-// Copyright 2023 Lance Developers.
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-pub mod js_object_ext;

rust/ffi/node/src/neon_ext/js_object_ext.rs

@@ -1,82 +0,0 @@
-// Copyright 2023 Lance Developers.
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-use crate::error::{Error, Result};
-use neon::prelude::*;
-
-// extends neon's [JsObject] with helper functions to extract properties
-pub trait JsObjectExt {
-    fn get_opt_u32(&self, cx: &mut FunctionContext, key: &str) -> Result<Option<u32>>;
-    fn get_usize(&self, cx: &mut FunctionContext, key: &str) -> Result<usize>;
-    fn get_opt_usize(&self, cx: &mut FunctionContext, key: &str) -> Result<Option<usize>>;
-}
-
-impl JsObjectExt for JsObject {
-    fn get_opt_u32(&self, cx: &mut FunctionContext, key: &str) -> Result<Option<u32>> {
-        let val_opt = self
-            .get_opt::<JsNumber, _, _>(cx, key)?
-            .map(|s| f64_to_u32_safe(s.value(cx), key));
-        val_opt.transpose()
-    }
-
-    fn get_usize(&self, cx: &mut FunctionContext, key: &str) -> Result<usize> {
-        let val = self.get::<JsNumber, _, _>(cx, key)?.value(cx);
-        f64_to_usize_safe(val, key)
-    }
-
-    fn get_opt_usize(&self, cx: &mut FunctionContext, key: &str) -> Result<Option<usize>> {
-        let val_opt = self
-            .get_opt::<JsNumber, _, _>(cx, key)?
-            .map(|s| f64_to_usize_safe(s.value(cx), key));
-        val_opt.transpose()
-    }
-}
-
-fn f64_to_u32_safe(n: f64, key: &str) -> Result<u32> {
-    use conv::*;
-
-    n.approx_as::<u32>().map_err(|e| match e {
-        FloatError::NegOverflow(_) => Error::RangeError {
-            name: key.into(),
-            message: "must be > 0".to_string(),
-        },
-        FloatError::PosOverflow(_) => Error::RangeError {
-            name: key.into(),
-            message: format!("must be < {}", u32::MAX),
-        },
-        FloatError::NotANumber(_) => Error::RangeError {
-            name: key.into(),
-            message: "not a valid number".to_string(),
-        },
-    })
-}
-
-fn f64_to_usize_safe(n: f64, key: &str) -> Result<usize> {
-    use conv::*;
-
-    n.approx_as::<usize>().map_err(|e| match e {
-        FloatError::NegOverflow(_) => Error::RangeError {
-            name: key.into(),
-            message: "must be > 0".to_string(),
-        },
-        FloatError::PosOverflow(_) => Error::RangeError {
-            name: key.into(),
-            message: format!("must be < {}", usize::MAX),
-        },
-        FloatError::NotANumber(_) => Error::RangeError {
-            name: key.into(),
-            message: "not a valid number".to_string(),
-        },
-    })
-}

rust/vectordb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "vectordb"
-version = "0.1.19"
+version = "0.1.15"
 edition = "2021"
 description = "Serverless, low-latency vector database for AI applications"
 license = "Apache-2.0"
@@ -12,7 +12,7 @@ arrow-array = { workspace = true }
 arrow-data = { workspace = true }
 arrow-schema = { workspace = true }
 object_store = { workspace = true }
-snafu = { workspace = true }
+snafu = "0.7.4"
 half = { workspace = true }
 lance = { workspace = true }
 tokio = { version = "1.23", features = ["rt-multi-thread"] }