[python] Bump version: 0.4.2 → 0.4.3

issue separate requests under the hood and concatenate results

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.3.8
+current_version = 0.4.1
 commit = True
 message = Bump version: {current_version} → {new_version}
 tag = True

.github/ISSUE_TEMPLATE/bug-node.yml

@@ -0,0 +1,33 @@
+name: Bug Report - Node / Typescript
+description: File a bug report
+title: "bug(node): "
+labels: [bug, typescript]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for taking the time to fill out this bug report!
+  - type: input
+    id: version
+    attributes:
+      label: LanceDB version
+      description: What version of LanceDB are you using? `npm list | grep vectordb`.
+      placeholder: v0.3.2
+    validations:
+      required: false
+  - type: textarea
+    id: what-happened
+    attributes:
+      label: What happened?
+      description: Also tell us, what did you expect to happen?
+    validations:
+      required: true
+  - type: textarea
+    id: reproduction
+    attributes:
+      label: Are there known steps to reproduce?
+      description: |
+        Let us know how to reproduce the bug and we may be able to fix it more
+        quickly. This is not required, but it is helpful.
+    validations:
+      required: false

.github/ISSUE_TEMPLATE/bug-python.yml

@@ -0,0 +1,33 @@
+name: Bug Report - Python
+description: File a bug report
+title: "bug(python): "
+labels: [bug, python]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for taking the time to fill out this bug report!
+  - type: input
+    id: version
+    attributes:
+      label: LanceDB version
+      description: What version of LanceDB are you using? `python -c "import lancedb; print(lancedb.__version__)"`.
+      placeholder: v0.3.2
+    validations:
+      required: false
+  - type: textarea
+    id: what-happened
+    attributes:
+      label: What happened?
+      description: Also tell us, what did you expect to happen?
+    validations:
+      required: true
+  - type: textarea
+    id: reproduction
+    attributes:
+      label: Are there known steps to reproduce?
+      description: |
+        Let us know how to reproduce the bug and we may be able to fix it more
+        quickly. This is not required, but it is helpful.
+    validations:
+      required: false

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,5 @@
+blank_issues_enabled: true
+contact_links:
+  - name: Discord Community Support
+    url: https://discord.com/invite/zMM32dvNtd
+    about: Please ask and answer questions here.

.github/ISSUE_TEMPLATE/documentation.yml

@@ -0,0 +1,23 @@
+name: 'Documentation improvement'
+description: Report an issue with the documentation.
+labels: [documentation]
+
+body:
+  - type: textarea
+    id: description
+    attributes:
+      label: Description
+      description: >
+        Describe the issue with the documentation and how it can be fixed or improved.
+    validations:
+      required: true
+
+  - type: input
+    id: link
+    attributes:
+      label: Link
+      description: >
+        Provide a link to the existing documentation, if applicable.
+      placeholder: ex. https://lancedb.github.io/lancedb/guides/tables/...
+    validations:
+      required: false

.github/ISSUE_TEMPLATE/feature.yml

@@ -0,0 +1,31 @@
+name: Feature suggestion
+description: Suggestion a new feature for LanceDB
+title: "Feature: "
+labels: [enhancement]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Share a new idea for a feature or improvement. Be sure to search existing
+        issues first to avoid duplicates.
+  - type: dropdown
+    id: sdk
+    attributes:
+      label: SDK
+      description: Which SDK are you using? This helps us prioritize.
+      options:
+        - Python
+        - Node
+        - Rust
+      default: 0
+    validations:
+      required: false
+  - type: textarea
+    id: description
+    attributes:
+      label: Description
+      description: |
+        Describe the feature and why it would be useful. If applicable, consider
+        providing a code example of what it might be like to use the feature.
+    validations:
+      required: true

.github/workflows/npm-publish.yml

@@ -38,13 +38,17 @@ jobs:
             node/vectordb-*.tgz
 
   node-macos:
-    runs-on: macos-13
+    strategy:
+      matrix:
+        config:
+          - arch: x86_64-apple-darwin
+            runner: macos-13
+          - arch: aarch64-apple-darwin
+            # xlarge is implicitly arm64.
+            runner: macos-13-xlarge
+    runs-on: ${{ matrix.config.runner }}
     # Only runs on tags that matches the make-release action
     if: startsWith(github.ref, 'refs/tags/v')
-    strategy:
-      fail-fast: false
-      matrix:
-        target: [x86_64-apple-darwin, aarch64-apple-darwin]
     steps:
       - name: Checkout
         uses: actions/checkout@v3
@@ -54,17 +58,15 @@ jobs:
         run: |
           cd node
           npm ci
-      - name: Install rustup target
-        if: ${{ matrix.target == 'aarch64-apple-darwin' }}
-        run: rustup target add aarch64-apple-darwin
       - name: Build MacOS native node modules
-        run: bash ci/build_macos_artifacts.sh ${{ matrix.target }}
+        run: bash ci/build_macos_artifacts.sh ${{ matrix.config.arch }}
       - name: Upload Darwin Artifacts
         uses: actions/upload-artifact@v3
         with:
           name: native-darwin
           path: |
             node/dist/lancedb-vectordb-darwin*.tgz
+         
 
   node-linux:
     name: node-linux (${{ matrix.config.arch}}-unknown-linux-gnu

.github/workflows/python.yml

@@ -44,12 +44,19 @@ jobs:
       run: pytest -m "not slow" -x -v --durations=30 tests
     - name: doctest
       run: pytest --doctest-modules lancedb
-  mac:
+  platform:
+    name: "Platform: ${{ matrix.config.name }}"
     timeout-minutes: 30
     strategy:
       matrix:
-        mac-runner: [ "macos-13", "macos-13-xlarge" ]
-    runs-on: "${{ matrix.mac-runner }}"
+        config: 
+          - name: x86 Mac
+            runner: macos-13
+          - name: Arm Mac
+            runner: macos-13-xlarge
+          - name: x86 Windows
+            runner: windows-latest
+    runs-on: "${{ matrix.config.runner }}"
     defaults:
       run:
         shell: bash
@@ -91,11 +98,7 @@ jobs:
         pip install "pydantic<2"
         pip install -e .[tests]
         pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985
-        pip install pytest pytest-mock black isort
-    - name: Black
-      run: black --check --diff --no-color --quiet .
-    - name: isort
-      run: isort --check --diff --quiet .
+        pip install pytest pytest-mock
     - name: Run tests
       run: pytest -m "not slow" -x -v --durations=30 tests
     - name: doctest

.github/workflows/rust.yml

@@ -24,6 +24,29 @@ env:
   RUST_BACKTRACE: "1"
 
 jobs:
+  lint:
+    timeout-minutes: 30
+    runs-on: ubuntu-22.04
+    defaults:
+      run:
+        shell: bash
+        working-directory: rust
+    steps:
+    - uses: actions/checkout@v3
+      with:
+          fetch-depth: 0
+          lfs: true
+    - uses: Swatinem/rust-cache@v2
+      with:
+        workspaces: rust
+    - name: Install dependencies
+      run: |
+          sudo apt update
+          sudo apt install -y protobuf-compiler libssl-dev
+    - name: Run format
+      run: cargo fmt --all -- --check
+    - name: Run clippy
+      run: cargo clippy --all --all-features -- -D warnings
   linux:
     timeout-minutes: 30
     runs-on: ubuntu-22.04

Cargo.toml

@@ -5,24 +5,24 @@ exclude = ["python"]
 resolver = "2"
 
 [workspace.dependencies]
-lance = { "version" = "=0.8.17", "features" = ["dynamodb"] }
-lance-index = { "version" = "=0.8.17" }
-lance-linalg = { "version" = "=0.8.17" }
-lance-testing = { "version" = "=0.8.17" }
+lance = { "version" = "=0.9.2", "features" = ["dynamodb"] }
+lance-index = { "version" = "=0.9.2" }
+lance-linalg = { "version" = "=0.9.2" }
+lance-testing = { "version" = "=0.9.2" }
 # Note that this one does not include pyarrow
-arrow = { version = "47.0.0", optional = false }
-arrow-array = "47.0"
-arrow-data = "47.0"
-arrow-ipc = "47.0"
-arrow-ord = "47.0"
-arrow-schema = "47.0"
-arrow-arith = "47.0"
-arrow-cast = "47.0"
+arrow = { version = "49.0.0", optional = false }
+arrow-array = "49.0"
+arrow-data = "49.0"
+arrow-ipc = "49.0"
+arrow-ord = "49.0"
+arrow-schema = "49.0"
+arrow-arith = "49.0"
+arrow-cast = "49.0"
 chrono = "0.4.23"
 half = { "version" = "=2.3.1", default-features = false, features = [
     "num-traits",
 ] }
 log = "0.4"
-object_store = "0.7.1"
+object_store = "0.8.0"
 snafu = "0.7.4"
 url = "2"

README.md

@@ -5,10 +5,11 @@
 
 **Developer-friendly, serverless vector database for AI applications**
 
-<a href="https://lancedb.github.io/lancedb/">Documentation</a> •
-<a href="https://blog.lancedb.com/">Blog</a> •
-<a href="https://discord.gg/zMM32dvNtd">Discord</a> •
-<a href="https://twitter.com/lancedb">Twitter</a>
+<a href='https://github.com/lancedb/vectordb-recipes/tree/main' target="_blank"><img alt='LanceDB' src='https://img.shields.io/badge/VectorDB_Recipes-100000?style=for-the-badge&logo=LanceDB&logoColor=white&labelColor=645cfb&color=645cfb'/></a>
+<a href='https://lancedb.github.io/lancedb/' target="_blank"><img alt='lancdb' src='https://img.shields.io/badge/DOCS-100000?style=for-the-badge&logo=lancdb&logoColor=white&labelColor=645cfb&color=645cfb'/></a>
+[![Medium](https://img.shields.io/badge/Medium-12100E?style=for-the-badge&logo=medium&logoColor=white)](https://blog.lancedb.com/) 
+[![Discord](https://img.shields.io/badge/Discord-%235865F2.svg?style=for-the-badge&logo=discord&logoColor=white)](https://discord.gg/zMM32dvNtd) 
+[![Twitter](https://img.shields.io/badge/Twitter-%231DA1F2.svg?style=for-the-badge&logo=Twitter&logoColor=white)](https://twitter.com/lancedb)
 
 </p>
 

ci/build_macos_artifacts.sh

@@ -1,6 +1,7 @@
 # Builds the macOS artifacts (node binaries).
 # Usage: ./ci/build_macos_artifacts.sh [target]
 # Targets supported: x86_64-apple-darwin aarch64-apple-darwin
+set -e
 
 prebuild_rust() {
     # Building here for the sake of easier debugging.

docs/mkdocs.yml

@@ -98,6 +98,7 @@ nav:
     - YouTube Transcript Search: notebooks/youtube_transcript_search.ipynb
     - Documentation QA Bot using LangChain: notebooks/code_qa_bot.ipynb
     - Multimodal search using CLIP: notebooks/multimodal_search.ipynb
+    - Example - Calculate CLIP Embeddings with Roboflow Inference: examples/image_embeddings_roboflow.md
     - Serverless QA Bot with S3 and Lambda: examples/serverless_lancedb_with_s3_and_lambda.md
     - Serverless QA Bot with Modal: examples/serverless_qa_bot_with_modal_and_langchain.md
   - 🌐 Javascript examples:
@@ -145,7 +146,8 @@ nav:
   - Serverless Chatbot from any website: examples/serverless_website_chatbot.md
   - TransformersJS Embedding Search: examples/transformerjs_embedding_search_nodejs.md
 - API references:
-  - Python API: python/python.md
+  - OSS Python API: python/python.md
+  - SaaS Python API: python/saas-python.md
   - Javascript API: javascript/modules.md
 - LanceDB Cloud↗: https://noteforms.com/forms/lancedb-mailing-list-cloud-kty1o5?notionforms=1&utm_source=notionforms
 

docs/requirements.txt

@@ -2,3 +2,4 @@ mkdocs==1.4.2
 mkdocs-jupyter==0.24.1
 mkdocs-material==9.1.3
 mkdocstrings[python]==0.20.0
+pydantic

docs/src/basic.md

@@ -64,18 +64,26 @@ We'll cover the basics of using LanceDB on your local machine in this section.
       tbl = db.create_table("table_from_df", data=df)
       ```
 
+      !!! warning
+
+            If the table already exists, LanceDB will raise an error by default.
+            If you want to overwrite the table, you can pass in `mode="overwrite"`
+            to the `createTable` function.
+
 === "Javascript"
       ```javascript
-      const tb = await db.createTable("my_table",
-                        data=[{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
-                              {"vector": [5.9, 26.5], "item": "bar", "price": 20.0}])
+      const tb = await db.createTable(
+        "myTable",
+        [{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
+         {"vector": [5.9, 26.5], "item": "bar", "price": 20.0}])
       ```
-      
-!!! warning
 
-      If the table already exists, LanceDB will raise an error by default.
-      If you want to overwrite the table, you can pass in `mode="overwrite"`
-      to the `createTable` function.
+      !!! warning
+
+            If the table already exists, LanceDB will raise an error by default.
+            If you want to overwrite the table, you can pass in `"overwrite"`
+            to the `createTable` function like this: `await con.createTable(tableName, data, { writeMode: WriteMode.Overwrite })`
+      
 
 ??? info "Under the hood, LanceDB is converting the input data into an Apache Arrow table and persisting it to disk in [Lance format](https://www.github.com/lancedb/lance)."
 
@@ -108,7 +116,7 @@ Once created, you can open a table using the following code:
 
 === "Javascript"
       ```javascript
-      const tbl = await db.openTable("my_table");
+      const tbl = await db.openTable("myTable");
       ```
 
       If you forget the name of your table, you can always get a listing of all table names:
@@ -194,10 +202,17 @@ Use the `drop_table()` method on the database to remove a table.
       db.drop_table("my_table")
       ```
 
-This permanently removes the table and is not recoverable, unlike deleting rows.
-By default, if the table does not exist an exception is raised. To suppress this,
-you can pass in `ignore_missing=True`.
+      This permanently removes the table and is not recoverable, unlike deleting rows.
+      By default, if the table does not exist an exception is raised. To suppress this,
+      you can pass in `ignore_missing=True`.
 
+=== "JavaScript"
+      ```javascript
+      await db.dropTable('myTable')
+      ```
+
+      This permanently removes the table and is not recoverable, unlike deleting rows.
+      If the table does not exist an exception is raised. 
 
 ## What's next
 

docs/src/embeddings/default_embedding_functions.md

@@ -1,9 +1,9 @@
-There are various Embedding functions available out of the box with lancedb. We're working on supporting other popular embedding APIs.
+There are various Embedding functions available out of the box with LanceDB. We're working on supporting other popular embedding APIs.
 
 ## Text Embedding Functions
 Here are the text embedding functions registered by default.
-Embedding functions have inbuilt rate limit handler wrapper for source and query embedding function calls that retry with exponential standoff. 
-Each `EmbeddingFunction` implementation automatically takes `max_retries` as an argument which has the deafult value of 7.
+Embedding functions have an inbuilt rate limit handler wrapper for source and query embedding function calls that retry with exponential standoff. 
+Each `EmbeddingFunction` implementation automatically takes `max_retries` as an argument which has the default value of 7.
 
 ### Sentence Transformers
 Here are the parameters that you can set when registering a `sentence-transformers` object, and their default values:
@@ -69,15 +69,15 @@ print(actual.text)
 ```
 
 ### Instructor Embeddings
-Instructor is an instruction-finetuned text embedding model that can generate text embeddings tailored to any task (e.g., classification, retrieval, clustering, text evaluation, etc.) and domains (e.g., science, finance, etc.) by simply providing the task instruction, without any finetuning
+Instructor is an instruction-finetuned text embedding model that can generate text embeddings tailored to any task (e.g. classification, retrieval, clustering, text evaluation, etc.) and domains (e.g. science, finance, etc.) by simply providing the task instruction, without any finetuning.
 
 If you want to calculate customized embeddings for specific sentences, you may follow the unified template to write instructions:
 
                           Represent the `domain` `text_type` for `task_objective`:
 
-* `domain` is optional, and it specifies the domain of the text, e.g., science, finance, medicine, etc.
-* `text_type` is required, and it specifies the encoding unit, e.g., sentence, document, paragraph, etc.
-* `task_objective` is optional, and it specifies the objective of embedding, e.g., retrieve a document, classify the sentence, etc.
+* `domain` is optional, and it specifies the domain of the text, e.g. science, finance, medicine, etc.
+* `text_type` is required, and it specifies the encoding unit, e.g. sentence, document, paragraph, etc.
+* `task_objective` is optional, and it specifies the objective of embedding, e.g. retrieve a document, classify the sentence, etc.
 
 More information about the model can be found here - https://github.com/xlang-ai/instructor-embedding
 
@@ -119,10 +119,10 @@ tbl.add(texts)
 ```
 
 ## Multi-modal embedding functions
-Multi-modal embedding functions allow you query your table using both images and text.
+Multi-modal embedding functions allow you to query your table using both images and text.
 
 ### OpenClipEmbeddings
-We support CLIP model embeddings using the open souce alternbative, open-clip which support various customizations. It is registered as `open-clip` and supports following customizations.
+We support CLIP model embeddings using the open source alternative, open-clip which supports various customizations. It is registered as `open-clip` and supports the following customizations:
 
 
 | Parameter | Type | Default Value | Description |
@@ -205,4 +205,4 @@ print(actual.label)
 
 ```
 
-If you have any questions about the embeddings API, supported models, or see a relevant model missing, please raise an issue.
+If you have any questions about the embeddings API, supported models, or see a relevant model missing, please raise an issue.

docs/src/examples/image_embeddings_roboflow.md

@@ -0,0 +1,165 @@
+# How to Load Image Embeddings into LanceDB
+
+With the rise of Large Multimodal Models (LMMs) such as [GPT-4 Vision](https://blog.roboflow.com/gpt-4-vision/), the need for storing image embeddings is growing. The most effective way to store text and image embeddings is in a vector database such as LanceDB. Vector databases are a special kind of data store that enables efficient search over stored embeddings. 
+
+[CLIP](https://blog.roboflow.com/openai-clip/), a multimodal model developed by OpenAI, is commonly used to calculate image embeddings. These embeddings can then be used with a vector database to build a semantic search engine that you can query using images or text. For example, you could use LanceDB and CLIP embeddings to build a search engine for a database of folders.
+
+In this guide, we are going to show you how to use Roboflow Inference to load image embeddings into LanceDB. Without further ado, let’s get started!
+
+## Step #1: Install Roboflow Inference
+
+[Roboflow Inference](https://inference.roboflow.com) enables you to run state-of-the-art computer vision models with minimal configuration. Inference supports a range of models, from fine-tuned object detection, classification, and segmentation models to foundation models like CLIP. We will use Inference to calculate CLIP image embeddings.
+
+Inference provides a HTTP API through which you can run vision models.
+
+Inference powers the Roboflow hosted API, and is available as an open source utility. In this guide, we are going to run Inference locally, which enables you to calculate CLIP embeddings on your own hardware. We will also show you how to use the hosted Roboflow CLIP API, which is ideal if you need to scale and do not want to manage a system for calculating embeddings.
+
+To get started, first install the Inference CLI:
+
+```
+pip install inference-cli
+```
+
+Next, install Docker. Refer to the official Docker installation instructions for your operating system to get Docker set up. Once Docker is ready, you can start Inference using the following command:
+
+```
+inference server start
+```
+
+An Inference server will start running at ‘http://localhost:9001’.
+
+## Step #2: Set Up a LanceDB Vector Database
+
+Now that we have Inference running, we can set up a LanceDB vector database. You can run LanceDB in JavaScript and Python. For this guide, we will use the Python API. But, you can take the HTTP requests we make below and change them to JavaScript if required.
+
+For this guide, we are going to search the [COCO 128 dataset](https://universe.roboflow.com/team-roboflow/coco-128), which contains a wide range of objects. The variability in objects present in this dataset makes it a good dataset to demonstrate the capabilities of vector search. If you want to use this dataset, you can download [COCO 128 from Roboflow Universe](https://universe.roboflow.com/team-roboflow/coco-128). With that said, you can search whatever folder of images you want.
+
+Once you have a dataset ready, install LanceDB with the following command:
+
+```
+pip install lancedb
+```
+
+We also need to install a specific commit of `tantivy`, a dependency of the LanceDB full text search engine we will use later in this guide:
+
+```
+pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985
+```
+
+Create a new Python file and add the following code:
+
+```python
+import cv2
+import supervision as sv
+import requests
+
+import lancedb
+
+db = lancedb.connect("./embeddings")
+
+IMAGE_DIR = "images/"
+API_KEY = os.environ.get("ROBOFLOW_API_KEY")
+SERVER_URL = "http://localhost:9001"
+
+results = []
+
+for i, image in enumerate(os.listdir(IMAGE_DIR)):
+    infer_clip_payload = {
+        #Images can be provided as urls or as base64 encoded strings
+        "image": {
+            "type": "base64",
+            "value": base64.b64encode(open(IMAGE_DIR + image, "rb").read()).decode("utf-8"),
+        },
+    }
+
+    res = requests.post(
+        f"{SERVER_URL}/clip/embed_image?api_key={API_KEY}",
+        json=infer_clip_payload,
+    )
+
+    embeddings = res.json()['embeddings']
+
+    print("Calculated embedding for image: ", image)
+
+    image = {"vector": embeddings[0], "name": os.path.join(IMAGE_DIR, image)}
+
+    results.append(image)
+
+tbl = db.create_table("images", data=results)
+
+tbl.create_fts_index("name")
+```
+
+To use the code above, you will need a Roboflow API key. [Learn how to retrieve a Roboflow API key](https://docs.roboflow.com/api-reference/authentication#retrieve-an-api-key). Run the following command to set up your API key in your environment:
+
+```
+export ROBOFLOW_API_KEY=""
+```
+
+Replace the `IMAGE_DIR` value with the folder in which you are storing the images for which you want to calculate embeddings. If you want to use the Roboflow CLIP API to calculate embeddings, replace the `SERVER_URL` value with `https://infer.roboflow.com`.
+
+Run the script above to create a new LanceDB database. This database will be stored on your local machine. The database will be called `embeddings` and the table will be called `images`.
+
+The script above calculates all embeddings for a folder then creates a new table. To add additional images, use the following code:
+
+```python
+def make_batches():
+    for i in range(5):
+        yield [
+                {"vector": [3.1, 4.1], "name": "image1.png"},
+                {"vector": [5.9, 26.5], "name": "image2.png"}
+            ]
+
+tbl = db.open_table("images")
+tbl.add(make_batches())
+```
+
+Replacing the `make_batches()` function with code to load embeddings for images.
+
+## Step #3: Run a Search Query
+
+We are now ready to run a search query. To run a search query, we need a text embedding that represents a text query. We can use this embedding to search our LanceDB database for an entry.
+
+Let’s calculate a text embedding for the query “cat”, then run a search query:
+
+```python
+infer_clip_payload = {
+    "text": "cat",
+}
+
+res = requests.post(
+    f"{SERVER_URL}/clip/embed_text?api_key={API_KEY}",
+    json=infer_clip_payload,
+)
+
+embeddings = res.json()['embeddings']
+
+df = tbl.search(embeddings[0]).limit(3).to_list()
+
+print("Results:")
+
+for i in df:
+    print(i["name"])
+```
+
+This code will search for the three images most closely related to the prompt “cat”. The names of the most similar three images will be printed to the console. Here are the three top results:
+
+```
+dataset/images/train/000000000650_jpg.rf.1b74ba165c5a3513a3211d4a80b69e1c.jpg
+dataset/images/train/000000000138_jpg.rf.af439ef1c55dd8a4e4b142d186b9c957.jpg
+dataset/images/train/000000000165_jpg.rf.eae14d5509bf0c9ceccddbb53a5f0c66.jpg
+```
+
+Let’s open the top image:
+
+![Cat](https://media.roboflow.com/cat_lancedb.jpg)
+
+The top image was a cat. Our search was successful.
+
+## Conclusion
+
+LanceDB is a vector database that you can use to store and efficiently search your image embeddings. You can use Roboflow Inference, a scalable computer vision inference server, to calculate CLIP embeddings that you can store in LanceDB.
+
+You can use Inference and LanceDB together to build a range of applications with image embeddings, from a media search engine to a retrieval-augmented generation pipeline for use with LMMs.
+
+To learn more about Inference and its capabilities, refer to the Inference documentation.

docs/src/fts.md

@@ -29,8 +29,9 @@ uri = "data/sample-lancedb"
 db = lancedb.connect(uri)
 
 table = db.create_table("my_table",
-            data=[{"vector": [3.1, 4.1], "text": "Frodo was a happy puppy"},
-                  {"vector": [5.9, 26.5], "text": "There are several kittens playing"}])
+            data=[{"vector": [3.1, 4.1], "text": "Frodo was a happy puppy", "meta": "foo"},
+                  {"vector": [5.9, 26.5], "text": "Sam was a loyal puppy", "meta": "bar"},
+                  {"vector": [15.9, 6.5], "text": "There are several kittens playing"}])
 
 ```
 
@@ -64,10 +65,23 @@ table.create_fts_index(["text1", "text2"])
 
 Note that the search API call does not change - you can search over all indexed columns at once.
 
+## Filtering
+
+Currently the LanceDB full text search feature supports *post-filtering*, meaning filters are
+applied on top of the full text search results. This can be invoked via the familiar
+`where` syntax:
+
+```python
+table.search("puppy").limit(10).where("meta='foo'").to_list()
+```
+
 ## Current limitations
 
 1. Currently we do not yet support incremental writes.
-If you add data after fts index creation, it won't be reflected
-in search results until you do a full reindex.
+   If you add data after fts index creation, it won't be reflected
+   in search results until you do a full reindex.
+
+2. We currently only support local filesystem paths for the fts index. 
+   This is a tantivy limitation. We've implemented an object store plugin
+   but there's no way in tantivy-py to specify to use it.
 
-2. We currently only support local filesystem paths for the fts index.

docs/src/guides/tables.md

@@ -1,5 +1,7 @@
  <a href="https://colab.research.google.com/github/lancedb/lancedb/blob/main/docs/src/notebooks/tables_guide.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a><br/>
-A Table is a collection of Records in a LanceDB Database. You can follow along on colab!
+A Table is a collection of Records in a LanceDB Database. Tables in Lance have a schema that defines the columns and their types. These schemas can include nested columns and can evolve over time.
+
+This guide will show how to create tables, insert data into them, and update the data. You can follow along on colab!
 
 ## Creating a LanceDB Table
 
@@ -116,6 +118,84 @@ A Table is a collection of Records in a LanceDB Database. You can follow along o
     table = db.create_table(table_name, schema=Content)
     ```
 
+    #### Nested schemas
+
+    Sometimes your data model may contain nested objects.
+    For example, you may want to store the document string
+    and the document soure name as a nested Document object:
+
+    ```python
+    class Document(BaseModel):
+        content: str
+        source: str
+    ```
+
+    This can be used as the type of a LanceDB table column:
+
+    ```python
+    class NestedSchema(LanceModel):
+        id: str
+        vector: Vector(1536)
+        document: Document
+
+    tbl = db.create_table("nested_table", schema=NestedSchema, mode="overwrite")
+    ```
+
+    This creates a struct column called "document" that has two subfields 
+    called "content" and "source":
+
+    ```
+    In [28]: tbl.schema
+    Out[28]:
+    id: string not null
+    vector: fixed_size_list<item: float>[1536] not null
+        child 0, item: float
+    document: struct<content: string not null, source: string not null> not null
+        child 0, content: string not null
+        child 1, source: string not null    
+    ```
+
+    #### Validators
+
+    Note that neither pydantic nor pyarrow automatically validates that input data
+    is of the *correct* timezone, but this is easy to add as a custom field validator:
+
+    ```python
+    from datetime import datetime
+    from zoneinfo import ZoneInfo
+
+    from lancedb.pydantic import LanceModel
+    from pydantic import Field, field_validator, ValidationError, ValidationInfo
+
+    tzname = "America/New_York"
+    tz = ZoneInfo(tzname)
+
+    class TestModel(LanceModel):
+        dt_with_tz: datetime = Field(json_schema_extra={"tz": tzname})
+
+        @field_validator('dt_with_tz')
+        @classmethod
+        def tz_must_match(cls, dt: datetime) -> datetime:
+            assert dt.tzinfo == tz
+            return dt        
+
+    ok = TestModel(dt_with_tz=datetime.now(tz))
+
+    try:
+        TestModel(dt_with_tz=datetime.now(ZoneInfo("Asia/Shanghai")))
+        assert 0 == 1, "this should raise ValidationError"
+    except ValidationError:
+        print("A ValidationError was raised.")
+        pass
+    ```
+
+    When you run this code it should print "A ValidationError was raised."
+
+    #### Pydantic custom types
+
+    LanceDB does NOT yet support converting pydantic custom types. If this is something you need,
+    please file a feature request on the [LanceDB Github repo](https://github.com/lancedb/lancedb/issues/new).
+
     ### Using Iterators / Writing Large Datasets
 
     It is recommended to use itertators to add large datasets in batches when creating your table in one go. This does not create multiple versions of your dataset unlike manually adding batches using `table.add()`
@@ -151,7 +231,7 @@ A Table is a collection of Records in a LanceDB Database. You can follow along o
     You can also use iterators of other types like Pandas dataframe or Pylists directly in the above example.
 
     ## Creating Empty Table
-    You can also create empty tables in python. Initialize it with schema and later ingest data into it.
+    You can create empty tables in python. Initialize it with schema and later ingest data into it.
 
     ```python
     import lancedb
@@ -201,8 +281,8 @@ A Table is a collection of Records in a LanceDB Database. You can follow along o
     ```javascript
     data
     const tb = await db.createTable("my_table",
-                  data=[{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
-                        {"vector": [5.9, 26.5], "item": "bar", "price": 20.0}])
+      [{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
+       {"vector": [5.9, 26.5], "item": "bar", "price": 20.0}])
     ```
 
     !!! info "Note"
@@ -361,19 +441,28 @@ Use the `delete()` method on tables to delete rows from a table. To choose which
     await tbl.countRows() // Returns 1
     ```
 
-### Updating a Table [Experimental]
-EXPERIMENTAL: Update rows in the table (not threadsafe).
+## Updating a Table
 
-This can be used to update zero to all rows depending on how many rows match the where clause.
+This can be used to update zero to all rows depending on how many rows match the where clause. The update queries follow the form of a SQL UPDATE statement. The `where` parameter is a SQL filter that matches on the metadata columns. The `values` or `values_sql` parameters are used to provide the new values for the columns.
 
-| Parameter | Type | Description |
+| Parameter   | Type | Description |
 |---|---|---|
 | `where` | `str` | The SQL where clause to use when updating rows. For example, `'x = 2'` or `'x IN (1, 2, 3)'`. The filter must not be empty, or it will error. |
 | `values` | `dict` | The values to update. The keys are the column names and the values are the values to set. |
+| `values_sql` | `dict` | The values to update. The keys are the column names and the values are the SQL expressions to set. For example, `{'x': 'x + 1'}` will increment the value of the `x` column by 1. |
 
+!!! info "SQL syntax"
+
+    See [SQL filters](sql.md) for more information on the supported SQL syntax.
+
+!!! warning "Warning"
+
+    Updating nested columns is not yet supported.
 
 === "Python"
 
+    API Reference: [lancedb.table.Table.update][]
+
     ```python
     import lancedb
     import pandas as pd
@@ -403,6 +492,55 @@ This can be used to update zero to all rows depending on how many rows match the
     2  2  [10.0, 10.0]
     ```
 
+=== "Javascript/Typescript"
+
+    API Reference: [vectordb.Table.update](../../javascript/interfaces/Table/#update)
+
+    ```javascript
+    const lancedb = require("vectordb");
+
+    const db = await lancedb.connect("./.lancedb");
+
+    const data = [
+      {x: 1, vector: [1, 2]},
+      {x: 2, vector: [3, 4]},
+      {x: 3, vector: [5, 6]},
+    ];
+    const tbl = await db.createTable("my_table", data)
+
+    await tbl.update({ where: "x = 2", values: {vector: [10, 10]} })
+    ```
+
+The `values` parameter is used to provide the new values for the columns as literal values. You can also use the `values_sql` / `valuesSql` parameter to provide SQL expressions for the new values. For example, you can use `values_sql="x + 1"` to increment the value of the `x` column by 1.
+
+=== "Python"
+
+    ```python
+    # Update the table where x = 2
+    table.update(valuesSql={"x": "x + 1"})
+
+    print(table.to_pandas())
+    ```
+
+    Output
+    ```shell
+        x  vector
+    0  2  [1.0, 2.0]
+    1  4  [5.0, 6.0]
+    2  3  [10.0, 10.0]
+    ```
+
+=== "Javascript/Typescript"
+
+    ```javascript
+    await tbl.update({ valuesSql: { x: "x + 1" } })
+    ```
+
+!!! info "Note"
+
+    When rows are updated, they are moved out of the index. The row will still show up in ANN queries, but the query will not be as fast as it would be if the row was in the index. If you update a large proportion of rows, consider rebuilding the index afterwards.
+
+
 ## What's Next?
 
 Learn how to Query your tables and create indices

docs/src/javascript/README.md

@@ -11,8 +11,13 @@ npm install vectordb
 ```
 
 This will download the appropriate native library for your platform. We currently
-support x86_64 Linux, aarch64 Linux, Intel MacOS, and ARM (M1/M2) MacOS. We do not
-yet support Windows or musl-based Linux (such as Alpine Linux).
+support:
+
+* Linux (x86_64 and aarch64)
+* MacOS (Intel and ARM/M1/M2)
+* Windows (x86_64 only)
+
+We do not yet support musl-based Linux (such as Alpine Linux) or aarch64 Windows.
 
 ## Usage
 

docs/src/javascript/classes/DefaultWriteOptions.md

@@ -0,0 +1,41 @@
+[vectordb](../README.md) / [Exports](../modules.md) / DefaultWriteOptions
+
+# Class: DefaultWriteOptions
+
+Write options when creating a Table.
+
+## Implements
+
+- [`WriteOptions`](../interfaces/WriteOptions.md)
+
+## Table of contents
+
+### Constructors
+
+- [constructor](DefaultWriteOptions.md#constructor)
+
+### Properties
+
+- [writeMode](DefaultWriteOptions.md#writemode)
+
+## Constructors
+
+### constructor
+
+• **new DefaultWriteOptions**()
+
+## Properties
+
+### writeMode
+
+• **writeMode**: [`WriteMode`](../enums/WriteMode.md) = `WriteMode.Create`
+
+A [WriteMode](../enums/WriteMode.md) to use on this operation
+
+#### Implementation of
+
+[WriteOptions](../interfaces/WriteOptions.md).[writeMode](../interfaces/WriteOptions.md#writemode)
+
+#### Defined in
+
+[index.ts:778](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L778)

docs/src/javascript/classes/LocalConnection.md

@@ -26,7 +26,7 @@ A connection to a LanceDB database.
 ### Methods
 
 - [createTable](LocalConnection.md#createtable)
-- [createTableArrow](LocalConnection.md#createtablearrow)
+- [createTableImpl](LocalConnection.md#createtableimpl)
 - [dropTable](LocalConnection.md#droptable)
 - [openTable](LocalConnection.md#opentable)
 - [tableNames](LocalConnection.md#tablenames)
@@ -46,7 +46,7 @@ A connection to a LanceDB database.
 
 #### Defined in
 
-[index.ts:184](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L184)
+[index.ts:355](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L355)
 
 ## Properties
 
@@ -56,17 +56,25 @@ A connection to a LanceDB database.
 
 #### Defined in
 
-[index.ts:182](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L182)
+[index.ts:353](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L353)
 
 ___
 
 ### \_options
 
-• `Private` `Readonly` **\_options**: [`ConnectionOptions`](../interfaces/ConnectionOptions.md)
+• `Private` `Readonly` **\_options**: () => [`ConnectionOptions`](../interfaces/ConnectionOptions.md)
+
+#### Type declaration
+
+▸ (): [`ConnectionOptions`](../interfaces/ConnectionOptions.md)
+
+##### Returns
+
+[`ConnectionOptions`](../interfaces/ConnectionOptions.md)
 
 #### Defined in
 
-[index.ts:181](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L181)
+[index.ts:352](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L352)
 
 ## Accessors
 
@@ -84,27 +92,34 @@ ___
 
 #### Defined in
 
-[index.ts:189](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L189)
+[index.ts:360](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L360)
 
 ## Methods
 
 ### createTable
 
-▸ **createTable**(`name`, `data`, `mode?`): `Promise`<[`Table`](../interfaces/Table.md)<`number`[]\>\>
+▸ **createTable**\<`T`\>(`name`, `data?`, `optsOrEmbedding?`, `opt?`): `Promise`\<[`Table`](../interfaces/Table.md)\<`T`\>\>
 
-Creates a new Table and initialize it with new data.
+Creates a new Table, optionally initializing it with new data.
+
+#### Type parameters
+
+| Name |
+| :------ |
+| `T` |
 
 #### Parameters
 
-| Name | Type | Description |
-| :------ | :------ | :------ |
-| `name` | `string` | The name of the table. |
-| `data` | `Record`<`string`, `unknown`\>[] | Non-empty Array of Records to be inserted into the Table |
-| `mode?` | [`WriteMode`](../enums/WriteMode.md) | The write mode to use when creating the table. |
+| Name | Type |
+| :------ | :------ |
+| `name` | `string` \| [`CreateTableOptions`](../interfaces/CreateTableOptions.md)\<`T`\> |
+| `data?` | `Record`\<`string`, `unknown`\>[] |
+| `optsOrEmbedding?` | [`WriteOptions`](../interfaces/WriteOptions.md) \| [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)\<`T`\> |
+| `opt?` | [`WriteOptions`](../interfaces/WriteOptions.md) |
 
 #### Returns
 
-`Promise`<[`Table`](../interfaces/Table.md)<`number`[]\>\>
+`Promise`\<[`Table`](../interfaces/Table.md)\<`T`\>\>
 
 #### Implementation of
 
@@ -112,120 +127,44 @@ Creates a new Table and initialize it with new data.
 
 #### Defined in
 
-[index.ts:230](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L230)
-
-▸ **createTable**(`name`, `data`, `mode`): `Promise`<[`Table`](../interfaces/Table.md)<`number`[]\>\>
-
-#### Parameters
-
-| Name | Type |
-| :------ | :------ |
-| `name` | `string` |
-| `data` | `Record`<`string`, `unknown`\>[] |
-| `mode` | [`WriteMode`](../enums/WriteMode.md) |
-
-#### Returns
-
-`Promise`<[`Table`](../interfaces/Table.md)<`number`[]\>\>
-
-#### Implementation of
-
-Connection.createTable
-
-#### Defined in
-
-[index.ts:231](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L231)
-
-▸ **createTable**<`T`\>(`name`, `data`, `mode`, `embeddings`): `Promise`<[`Table`](../interfaces/Table.md)<`T`\>\>
-
-Creates a new Table and initialize it with new data.
-
-#### Type parameters
-
-| Name |
-| :------ |
-| `T` |
-
-#### Parameters
-
-| Name | Type | Description |
-| :------ | :------ | :------ |
-| `name` | `string` | The name of the table. |
-| `data` | `Record`<`string`, `unknown`\>[] | Non-empty Array of Records to be inserted into the Table |
-| `mode` | [`WriteMode`](../enums/WriteMode.md) | The write mode to use when creating the table. |
-| `embeddings` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)<`T`\> | An embedding function to use on this Table |
-
-#### Returns
-
-`Promise`<[`Table`](../interfaces/Table.md)<`T`\>\>
-
-#### Implementation of
-
-Connection.createTable
-
-#### Defined in
-
-[index.ts:241](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L241)
-
-▸ **createTable**<`T`\>(`name`, `data`, `mode`, `embeddings?`): `Promise`<[`Table`](../interfaces/Table.md)<`T`\>\>
-
-#### Type parameters
-
-| Name |
-| :------ |
-| `T` |
-
-#### Parameters
-
-| Name | Type |
-| :------ | :------ |
-| `name` | `string` |
-| `data` | `Record`<`string`, `unknown`\>[] |
-| `mode` | [`WriteMode`](../enums/WriteMode.md) |
-| `embeddings?` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)<`T`\> |
-
-#### Returns
-
-`Promise`<[`Table`](../interfaces/Table.md)<`T`\>\>
-
-#### Implementation of
-
-Connection.createTable
-
-#### Defined in
-
-[index.ts:242](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L242)
+[index.ts:395](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L395)
 
 ___
 
-### createTableArrow
+### createTableImpl
 
-▸ **createTableArrow**(`name`, `table`): `Promise`<[`Table`](../interfaces/Table.md)<`number`[]\>\>
+▸ `Private` **createTableImpl**\<`T`\>(`«destructured»`): `Promise`\<[`Table`](../interfaces/Table.md)\<`T`\>\>
+
+#### Type parameters
+
+| Name |
+| :------ |
+| `T` |
 
 #### Parameters
 
 | Name | Type |
 | :------ | :------ |
-| `name` | `string` |
-| `table` | `Table`<`any`\> |
+| `«destructured»` | `Object` |
+| › `data?` | `Table`\<`any`\> \| `Record`\<`string`, `unknown`\>[] |
+| › `embeddingFunction?` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)\<`T`\> |
+| › `name` | `string` |
+| › `schema?` | `Schema`\<`any`\> |
+| › `writeOptions?` | [`WriteOptions`](../interfaces/WriteOptions.md) |
 
 #### Returns
 
-`Promise`<[`Table`](../interfaces/Table.md)<`number`[]\>\>
-
-#### Implementation of
-
-[Connection](../interfaces/Connection.md).[createTableArrow](../interfaces/Connection.md#createtablearrow)
+`Promise`\<[`Table`](../interfaces/Table.md)\<`T`\>\>
 
 #### Defined in
 
-[index.ts:266](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L266)
+[index.ts:413](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L413)
 
 ___
 
 ### dropTable
 
-▸ **dropTable**(`name`): `Promise`<`void`\>
+▸ **dropTable**(`name`): `Promise`\<`void`\>
 
 Drop an existing table.
 
@@ -237,7 +176,7 @@ Drop an existing table.
 
 #### Returns
 
-`Promise`<`void`\>
+`Promise`\<`void`\>
 
 #### Implementation of
 
@@ -245,13 +184,13 @@ Drop an existing table.
 
 #### Defined in
 
-[index.ts:276](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L276)
+[index.ts:453](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L453)
 
 ___
 
 ### openTable
 
-▸ **openTable**(`name`): `Promise`<[`Table`](../interfaces/Table.md)<`number`[]\>\>
+▸ **openTable**(`name`): `Promise`\<[`Table`](../interfaces/Table.md)\<`number`[]\>\>
 
 Open a table in the database.
 
@@ -263,7 +202,7 @@ Open a table in the database.
 
 #### Returns
 
-`Promise`<[`Table`](../interfaces/Table.md)<`number`[]\>\>
+`Promise`\<[`Table`](../interfaces/Table.md)\<`number`[]\>\>
 
 #### Implementation of
 
@@ -271,9 +210,9 @@ Open a table in the database.
 
 #### Defined in
 
-[index.ts:205](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L205)
+[index.ts:376](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L376)
 
-▸ **openTable**<`T`\>(`name`, `embeddings`): `Promise`<[`Table`](../interfaces/Table.md)<`T`\>\>
+▸ **openTable**\<`T`\>(`name`, `embeddings`): `Promise`\<[`Table`](../interfaces/Table.md)\<`T`\>\>
 
 Open a table in the database.
 
@@ -288,11 +227,11 @@ Open a table in the database.
 | Name | Type | Description |
 | :------ | :------ | :------ |
 | `name` | `string` | The name of the table. |
-| `embeddings` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)<`T`\> | An embedding function to use on this Table |
+| `embeddings` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)\<`T`\> | An embedding function to use on this Table |
 
 #### Returns
 
-`Promise`<[`Table`](../interfaces/Table.md)<`T`\>\>
+`Promise`\<[`Table`](../interfaces/Table.md)\<`T`\>\>
 
 #### Implementation of
 
@@ -300,9 +239,9 @@ Connection.openTable
 
 #### Defined in
 
-[index.ts:212](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L212)
+[index.ts:384](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L384)
 
-▸ **openTable**<`T`\>(`name`, `embeddings?`): `Promise`<[`Table`](../interfaces/Table.md)<`T`\>\>
+▸ **openTable**\<`T`\>(`name`, `embeddings?`): `Promise`\<[`Table`](../interfaces/Table.md)\<`T`\>\>
 
 #### Type parameters
 
@@ -315,11 +254,11 @@ Connection.openTable
 | Name | Type |
 | :------ | :------ |
 | `name` | `string` |
-| `embeddings?` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)<`T`\> |
+| `embeddings?` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)\<`T`\> |
 
 #### Returns
 
-`Promise`<[`Table`](../interfaces/Table.md)<`T`\>\>
+`Promise`\<[`Table`](../interfaces/Table.md)\<`T`\>\>
 
 #### Implementation of
 
@@ -327,19 +266,19 @@ Connection.openTable
 
 #### Defined in
 
-[index.ts:213](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L213)
+[index.ts:385](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L385)
 
 ___
 
 ### tableNames
 
-▸ **tableNames**(): `Promise`<`string`[]\>
+▸ **tableNames**(): `Promise`\<`string`[]\>
 
 Get the names of all tables in the database.
 
 #### Returns
 
-`Promise`<`string`[]\>
+`Promise`\<`string`[]\>
 
 #### Implementation of
 
@@ -347,4 +286,4 @@ Get the names of all tables in the database.
 
 #### Defined in
 
-[index.ts:196](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L196)
+[index.ts:367](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L367)

docs/src/javascript/classes/LocalTable.md

@@ -1,6 +1,6 @@
 [vectordb](../README.md) / [Exports](../modules.md) / LocalTable
 
-# Class: LocalTable<T\>
+# Class: LocalTable\<T\>
 
 A LanceDB Table is the collection of Records. Each Record has one or more vector fields.
 
@@ -12,7 +12,7 @@ A LanceDB Table is the collection of Records. Each Record has one or more vector
 
 ## Implements
 
-- [`Table`](../interfaces/Table.md)<`T`\>
+- [`Table`](../interfaces/Table.md)\<`T`\>
 
 ## Table of contents
 
@@ -26,6 +26,7 @@ A LanceDB Table is the collection of Records. Each Record has one or more vector
 - [\_name](LocalTable.md#_name)
 - [\_options](LocalTable.md#_options)
 - [\_tbl](LocalTable.md#_tbl)
+- [where](LocalTable.md#where)
 
 ### Accessors
 
@@ -34,17 +35,23 @@ A LanceDB Table is the collection of Records. Each Record has one or more vector
 ### Methods
 
 - [add](LocalTable.md#add)
+- [cleanupOldVersions](LocalTable.md#cleanupoldversions)
+- [compactFiles](LocalTable.md#compactfiles)
 - [countRows](LocalTable.md#countrows)
 - [createIndex](LocalTable.md#createindex)
 - [delete](LocalTable.md#delete)
+- [filter](LocalTable.md#filter)
+- [indexStats](LocalTable.md#indexstats)
+- [listIndices](LocalTable.md#listindices)
 - [overwrite](LocalTable.md#overwrite)
 - [search](LocalTable.md#search)
+- [update](LocalTable.md#update)
 
 ## Constructors
 
 ### constructor
 
-• **new LocalTable**<`T`\>(`tbl`, `name`, `options`)
+• **new LocalTable**\<`T`\>(`tbl`, `name`, `options`)
 
 #### Type parameters
 
@@ -62,9 +69,9 @@ A LanceDB Table is the collection of Records. Each Record has one or more vector
 
 #### Defined in
 
-[index.ts:287](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L287)
+[index.ts:464](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L464)
 
-• **new LocalTable**<`T`\>(`tbl`, `name`, `options`, `embeddings`)
+• **new LocalTable**\<`T`\>(`tbl`, `name`, `options`, `embeddings`)
 
 #### Type parameters
 
@@ -79,21 +86,21 @@ A LanceDB Table is the collection of Records. Each Record has one or more vector
 | `tbl` | `any` |  |
 | `name` | `string` |  |
 | `options` | [`ConnectionOptions`](../interfaces/ConnectionOptions.md) |  |
-| `embeddings` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)<`T`\> | An embedding function to use when interacting with this table |
+| `embeddings` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)\<`T`\> | An embedding function to use when interacting with this table |
 
 #### Defined in
 
-[index.ts:294](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L294)
+[index.ts:471](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L471)
 
 ## Properties
 
 ### \_embeddings
 
-• `Private` `Optional` `Readonly` **\_embeddings**: [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)<`T`\>
+• `Private` `Optional` `Readonly` **\_embeddings**: [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)\<`T`\>
 
 #### Defined in
 
-[index.ts:284](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L284)
+[index.ts:461](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L461)
 
 ___
 
@@ -103,27 +110,61 @@ ___
 
 #### Defined in
 
-[index.ts:283](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L283)
+[index.ts:460](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L460)
 
 ___
 
 ### \_options
 
-• `Private` `Readonly` **\_options**: [`ConnectionOptions`](../interfaces/ConnectionOptions.md)
+• `Private` `Readonly` **\_options**: () => [`ConnectionOptions`](../interfaces/ConnectionOptions.md)
+
+#### Type declaration
+
+▸ (): [`ConnectionOptions`](../interfaces/ConnectionOptions.md)
+
+##### Returns
+
+[`ConnectionOptions`](../interfaces/ConnectionOptions.md)
 
 #### Defined in
 
-[index.ts:285](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L285)
+[index.ts:462](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L462)
 
 ___
 
 ### \_tbl
 
-• `Private` `Readonly` **\_tbl**: `any`
+• `Private` **\_tbl**: `any`
 
 #### Defined in
 
-[index.ts:282](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L282)
+[index.ts:459](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L459)
+
+___
+
+### where
+
+• **where**: (`value`: `string`) => [`Query`](Query.md)\<`T`\>
+
+#### Type declaration
+
+▸ (`value`): [`Query`](Query.md)\<`T`\>
+
+Creates a filter query to find all rows matching the specified criteria
+
+##### Parameters
+
+| Name | Type | Description |
+| :------ | :------ | :------ |
+| `value` | `string` | The filter criteria (like SQL where clause syntax) |
+
+##### Returns
+
+[`Query`](Query.md)\<`T`\>
+
+#### Defined in
+
+[index.ts:499](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L499)
 
 ## Accessors
 
@@ -141,13 +182,13 @@ ___
 
 #### Defined in
 
-[index.ts:302](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L302)
+[index.ts:479](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L479)
 
 ## Methods
 
 ### add
 
-▸ **add**(`data`): `Promise`<`number`\>
+▸ **add**(`data`): `Promise`\<`number`\>
 
 Insert records into this Table.
 
@@ -155,11 +196,11 @@ Insert records into this Table.
 
 | Name | Type | Description |
 | :------ | :------ | :------ |
-| `data` | `Record`<`string`, `unknown`\>[] | Records to be inserted into the Table |
+| `data` | `Record`\<`string`, `unknown`\>[] | Records to be inserted into the Table |
 
 #### Returns
 
-`Promise`<`number`\>
+`Promise`\<`number`\>
 
 The number of rows added to the table
 
@@ -169,19 +210,69 @@ The number of rows added to the table
 
 #### Defined in
 
-[index.ts:320](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L320)
+[index.ts:507](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L507)
+
+___
+
+### cleanupOldVersions
+
+▸ **cleanupOldVersions**(`olderThan?`, `deleteUnverified?`): `Promise`\<[`CleanupStats`](../interfaces/CleanupStats.md)\>
+
+Clean up old versions of the table, freeing disk space.
+
+#### Parameters
+
+| Name | Type | Description |
+| :------ | :------ | :------ |
+| `olderThan?` | `number` | The minimum age in minutes of the versions to delete. If not provided, defaults to two weeks. |
+| `deleteUnverified?` | `boolean` | Because they may be part of an in-progress transaction, uncommitted files newer than 7 days old are not deleted by default. This means that failed transactions can leave around data that takes up disk space for up to 7 days. You can override this safety mechanism by setting this option to `true`, only if you promise there are no in progress writes while you run this operation. Failure to uphold this promise can lead to corrupted tables. |
+
+#### Returns
+
+`Promise`\<[`CleanupStats`](../interfaces/CleanupStats.md)\>
+
+#### Defined in
+
+[index.ts:596](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L596)
+
+___
+
+### compactFiles
+
+▸ **compactFiles**(`options?`): `Promise`\<[`CompactionMetrics`](../interfaces/CompactionMetrics.md)\>
+
+Run the compaction process on the table.
+
+This can be run after making several small appends to optimize the table
+for faster reads.
+
+#### Parameters
+
+| Name | Type | Description |
+| :------ | :------ | :------ |
+| `options?` | [`CompactionOptions`](../interfaces/CompactionOptions.md) | Advanced options configuring compaction. In most cases, you can omit this arguments, as the default options are sensible for most tables. |
+
+#### Returns
+
+`Promise`\<[`CompactionMetrics`](../interfaces/CompactionMetrics.md)\>
+
+Metrics about the compaction operation.
+
+#### Defined in
+
+[index.ts:615](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L615)
 
 ___
 
 ### countRows
 
-▸ **countRows**(): `Promise`<`number`\>
+▸ **countRows**(): `Promise`\<`number`\>
 
 Returns the number of rows in this table.
 
 #### Returns
 
-`Promise`<`number`\>
+`Promise`\<`number`\>
 
 #### Implementation of
 
@@ -189,20 +280,16 @@ Returns the number of rows in this table.
 
 #### Defined in
 
-[index.ts:362](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L362)
+[index.ts:543](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L543)
 
 ___
 
 ### createIndex
 
-▸ **createIndex**(`indexParams`): `Promise`<`any`\>
+▸ **createIndex**(`indexParams`): `Promise`\<`any`\>
 
 Create an ANN index on this Table vector index.
 
-**`See`**
-
-VectorIndexParams.
-
 #### Parameters
 
 | Name | Type | Description |
@@ -211,7 +298,11 @@ VectorIndexParams.
 
 #### Returns
 
-`Promise`<`any`\>
+`Promise`\<`any`\>
+
+**`See`**
+
+VectorIndexParams.
 
 #### Implementation of
 
@@ -219,13 +310,13 @@ VectorIndexParams.
 
 #### Defined in
 
-[index.ts:355](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L355)
+[index.ts:536](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L536)
 
 ___
 
 ### delete
 
-▸ **delete**(`filter`): `Promise`<`void`\>
+▸ **delete**(`filter`): `Promise`\<`void`\>
 
 Delete rows from this table.
 
@@ -237,7 +328,7 @@ Delete rows from this table.
 
 #### Returns
 
-`Promise`<`void`\>
+`Promise`\<`void`\>
 
 #### Implementation of
 
@@ -245,13 +336,81 @@ Delete rows from this table.
 
 #### Defined in
 
-[index.ts:371](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L371)
+[index.ts:552](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L552)
+
+___
+
+### filter
+
+▸ **filter**(`value`): [`Query`](Query.md)\<`T`\>
+
+Creates a filter query to find all rows matching the specified criteria
+
+#### Parameters
+
+| Name | Type | Description |
+| :------ | :------ | :------ |
+| `value` | `string` | The filter criteria (like SQL where clause syntax) |
+
+#### Returns
+
+[`Query`](Query.md)\<`T`\>
+
+#### Defined in
+
+[index.ts:495](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L495)
+
+___
+
+### indexStats
+
+▸ **indexStats**(`indexUuid`): `Promise`\<[`IndexStats`](../interfaces/IndexStats.md)\>
+
+Get statistics about an index.
+
+#### Parameters
+
+| Name | Type |
+| :------ | :------ |
+| `indexUuid` | `string` |
+
+#### Returns
+
+`Promise`\<[`IndexStats`](../interfaces/IndexStats.md)\>
+
+#### Implementation of
+
+[Table](../interfaces/Table.md).[indexStats](../interfaces/Table.md#indexstats)
+
+#### Defined in
+
+[index.ts:628](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L628)
+
+___
+
+### listIndices
+
+▸ **listIndices**(): `Promise`\<[`VectorIndex`](../interfaces/VectorIndex.md)[]\>
+
+List the indicies on this table.
+
+#### Returns
+
+`Promise`\<[`VectorIndex`](../interfaces/VectorIndex.md)[]\>
+
+#### Implementation of
+
+[Table](../interfaces/Table.md).[listIndices](../interfaces/Table.md#listindices)
+
+#### Defined in
+
+[index.ts:624](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L624)
 
 ___
 
 ### overwrite
 
-▸ **overwrite**(`data`): `Promise`<`number`\>
+▸ **overwrite**(`data`): `Promise`\<`number`\>
 
 Insert records into this Table, replacing its contents.
 
@@ -259,11 +418,11 @@ Insert records into this Table, replacing its contents.
 
 | Name | Type | Description |
 | :------ | :------ | :------ |
-| `data` | `Record`<`string`, `unknown`\>[] | Records to be inserted into the Table |
+| `data` | `Record`\<`string`, `unknown`\>[] | Records to be inserted into the Table |
 
 #### Returns
 
-`Promise`<`number`\>
+`Promise`\<`number`\>
 
 The number of rows added to the table
 
@@ -273,13 +432,13 @@ The number of rows added to the table
 
 #### Defined in
 
-[index.ts:338](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L338)
+[index.ts:522](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L522)
 
 ___
 
 ### search
 
-▸ **search**(`query`): [`Query`](Query.md)<`T`\>
+▸ **search**(`query`): [`Query`](Query.md)\<`T`\>
 
 Creates a search query to find the nearest neighbors of the given search term
 
@@ -291,7 +450,7 @@ Creates a search query to find the nearest neighbors of the given search term
 
 #### Returns
 
-[`Query`](Query.md)<`T`\>
+[`Query`](Query.md)\<`T`\>
 
 #### Implementation of
 
@@ -299,4 +458,30 @@ Creates a search query to find the nearest neighbors of the given search term
 
 #### Defined in
 
-[index.ts:310](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L310)
+[index.ts:487](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L487)
+
+___
+
+### update
+
+▸ **update**(`args`): `Promise`\<`void`\>
+
+Update rows in this table.
+
+#### Parameters
+
+| Name | Type | Description |
+| :------ | :------ | :------ |
+| `args` | [`UpdateArgs`](../interfaces/UpdateArgs.md) \| [`UpdateSqlArgs`](../interfaces/UpdateSqlArgs.md) | see [UpdateArgs](../interfaces/UpdateArgs.md) and [UpdateSqlArgs](../interfaces/UpdateSqlArgs.md) for more details |
+
+#### Returns
+
+`Promise`\<`void`\>
+
+#### Implementation of
+
+[Table](../interfaces/Table.md).[update](../interfaces/Table.md#update)
+
+#### Defined in
+
+[index.ts:563](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L563)

docs/src/javascript/classes/OpenAIEmbeddingFunction.md

@@ -6,7 +6,7 @@ An embedding function that automatically creates vector representation for a giv
 
 ## Implements
 
-- [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)<`string`\>
+- [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)\<`string`\>
 
 ## Table of contents
 
@@ -40,7 +40,7 @@ An embedding function that automatically creates vector representation for a giv
 
 #### Defined in
 
-[embedding/openai.ts:21](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/embedding/openai.ts#L21)
+[embedding/openai.ts:21](https://github.com/lancedb/lancedb/blob/7856a94/node/src/embedding/openai.ts#L21)
 
 ## Properties
 
@@ -50,7 +50,7 @@ An embedding function that automatically creates vector representation for a giv
 
 #### Defined in
 
-[embedding/openai.ts:19](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/embedding/openai.ts#L19)
+[embedding/openai.ts:19](https://github.com/lancedb/lancedb/blob/7856a94/node/src/embedding/openai.ts#L19)
 
 ___
 
@@ -60,7 +60,7 @@ ___
 
 #### Defined in
 
-[embedding/openai.ts:18](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/embedding/openai.ts#L18)
+[embedding/openai.ts:18](https://github.com/lancedb/lancedb/blob/7856a94/node/src/embedding/openai.ts#L18)
 
 ___
 
@@ -76,13 +76,13 @@ The name of the column that will be used as input for the Embedding Function.
 
 #### Defined in
 
-[embedding/openai.ts:50](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/embedding/openai.ts#L50)
+[embedding/openai.ts:50](https://github.com/lancedb/lancedb/blob/7856a94/node/src/embedding/openai.ts#L50)
 
 ## Methods
 
 ### embed
 
-▸ **embed**(`data`): `Promise`<`number`[][]\>
+▸ **embed**(`data`): `Promise`\<`number`[][]\>
 
 Creates a vector representation for the given values.
 
@@ -94,7 +94,7 @@ Creates a vector representation for the given values.
 
 #### Returns
 
-`Promise`<`number`[][]\>
+`Promise`\<`number`[][]\>
 
 #### Implementation of
 
@@ -102,4 +102,4 @@ Creates a vector representation for the given values.
 
 #### Defined in
 
-[embedding/openai.ts:38](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/embedding/openai.ts#L38)
+[embedding/openai.ts:38](https://github.com/lancedb/lancedb/blob/7856a94/node/src/embedding/openai.ts#L38)

docs/src/javascript/classes/Query.md

@@ -1,6 +1,6 @@
 [vectordb](../README.md) / [Exports](../modules.md) / Query
 
-# Class: Query<T\>
+# Class: Query\<T\>
 
 A builder for nearest neighbor queries for LanceDB.
 
@@ -23,6 +23,7 @@ A builder for nearest neighbor queries for LanceDB.
 - [\_limit](Query.md#_limit)
 - [\_metricType](Query.md#_metrictype)
 - [\_nprobes](Query.md#_nprobes)
+- [\_prefilter](Query.md#_prefilter)
 - [\_query](Query.md#_query)
 - [\_queryVector](Query.md#_queryvector)
 - [\_refineFactor](Query.md#_refinefactor)
@@ -34,9 +35,11 @@ A builder for nearest neighbor queries for LanceDB.
 
 - [execute](Query.md#execute)
 - [filter](Query.md#filter)
+- [isElectron](Query.md#iselectron)
 - [limit](Query.md#limit)
 - [metricType](Query.md#metrictype)
 - [nprobes](Query.md#nprobes)
+- [prefilter](Query.md#prefilter)
 - [refineFactor](Query.md#refinefactor)
 - [select](Query.md#select)
 
@@ -44,7 +47,7 @@ A builder for nearest neighbor queries for LanceDB.
 
 ### constructor
 
-• **new Query**<`T`\>(`tbl`, `query`, `embeddings?`)
+• **new Query**\<`T`\>(`query?`, `tbl?`, `embeddings?`)
 
 #### Type parameters
 
@@ -56,23 +59,23 @@ A builder for nearest neighbor queries for LanceDB.
 
 | Name | Type |
 | :------ | :------ |
-| `tbl` | `any` |
-| `query` | `T` |
-| `embeddings?` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)<`T`\> |
+| `query?` | `T` |
+| `tbl?` | `any` |
+| `embeddings?` | [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)\<`T`\> |
 
 #### Defined in
 
-[index.ts:448](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L448)
+[query.ts:38](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L38)
 
 ## Properties
 
 ### \_embeddings
 
-• `Private` `Optional` `Readonly` **\_embeddings**: [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)<`T`\>
+• `Protected` `Optional` `Readonly` **\_embeddings**: [`EmbeddingFunction`](../interfaces/EmbeddingFunction.md)\<`T`\>
 
 #### Defined in
 
-[index.ts:446](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L446)
+[query.ts:36](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L36)
 
 ___
 
@@ -82,17 +85,17 @@ ___
 
 #### Defined in
 
-[index.ts:444](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L444)
+[query.ts:33](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L33)
 
 ___
 
 ### \_limit
 
-• `Private` **\_limit**: `number`
+• `Private` `Optional` **\_limit**: `number`
 
 #### Defined in
 
-[index.ts:440](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L440)
+[query.ts:29](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L29)
 
 ___
 
@@ -102,7 +105,7 @@ ___
 
 #### Defined in
 
-[index.ts:445](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L445)
+[query.ts:34](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L34)
 
 ___
 
@@ -112,17 +115,27 @@ ___
 
 #### Defined in
 
-[index.ts:442](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L442)
+[query.ts:31](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L31)
+
+___
+
+### \_prefilter
+
+• `Private` **\_prefilter**: `boolean`
+
+#### Defined in
+
+[query.ts:35](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L35)
 
 ___
 
 ### \_query
 
-• `Private` `Readonly` **\_query**: `T`
+• `Private` `Optional` `Readonly` **\_query**: `T`
 
 #### Defined in
 
-[index.ts:438](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L438)
+[query.ts:26](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L26)
 
 ___
 
@@ -132,7 +145,7 @@ ___
 
 #### Defined in
 
-[index.ts:439](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L439)
+[query.ts:28](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L28)
 
 ___
 
@@ -142,7 +155,7 @@ ___
 
 #### Defined in
 
-[index.ts:441](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L441)
+[query.ts:30](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L30)
 
 ___
 
@@ -152,27 +165,27 @@ ___
 
 #### Defined in
 
-[index.ts:443](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L443)
+[query.ts:32](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L32)
 
 ___
 
 ### \_tbl
 
-• `Private` `Readonly` **\_tbl**: `any`
+• `Private` `Optional` `Readonly` **\_tbl**: `any`
 
 #### Defined in
 
-[index.ts:437](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L437)
+[query.ts:27](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L27)
 
 ___
 
 ### where
 
-• **where**: (`value`: `string`) => [`Query`](Query.md)<`T`\>
+• **where**: (`value`: `string`) => [`Query`](Query.md)\<`T`\>
 
 #### Type declaration
 
-▸ (`value`): [`Query`](Query.md)<`T`\>
+▸ (`value`): [`Query`](Query.md)\<`T`\>
 
 A filter statement to be applied to this query.
 
@@ -184,17 +197,17 @@ A filter statement to be applied to this query.
 
 ##### Returns
 
-[`Query`](Query.md)<`T`\>
+[`Query`](Query.md)\<`T`\>
 
 #### Defined in
 
-[index.ts:496](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L496)
+[query.ts:87](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L87)
 
 ## Methods
 
 ### execute
 
-▸ **execute**<`T`\>(): `Promise`<`T`[]\>
+▸ **execute**\<`T`\>(): `Promise`\<`T`[]\>
 
 Execute the query and return the results as an Array of Objects
 
@@ -202,21 +215,21 @@ Execute the query and return the results as an Array of Objects
 
 | Name | Type |
 | :------ | :------ |
-| `T` | `Record`<`string`, `unknown`\> |
+| `T` | `Record`\<`string`, `unknown`\> |
 
 #### Returns
 
-`Promise`<`T`[]\>
+`Promise`\<`T`[]\>
 
 #### Defined in
 
-[index.ts:519](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L519)
+[query.ts:115](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L115)
 
 ___
 
 ### filter
 
-▸ **filter**(`value`): [`Query`](Query.md)<`T`\>
+▸ **filter**(`value`): [`Query`](Query.md)\<`T`\>
 
 A filter statement to be applied to this query.
 
@@ -228,17 +241,31 @@ A filter statement to be applied to this query.
 
 #### Returns
 
-[`Query`](Query.md)<`T`\>
+[`Query`](Query.md)\<`T`\>
 
 #### Defined in
 
-[index.ts:491](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L491)
+[query.ts:82](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L82)
+
+___
+
+### isElectron
+
+▸ `Private` **isElectron**(): `boolean`
+
+#### Returns
+
+`boolean`
+
+#### Defined in
+
+[query.ts:142](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L142)
 
 ___
 
 ### limit
 
-▸ **limit**(`value`): [`Query`](Query.md)<`T`\>
+▸ **limit**(`value`): [`Query`](Query.md)\<`T`\>
 
 Sets the number of results that will be returned
 
@@ -250,24 +277,20 @@ Sets the number of results that will be returned
 
 #### Returns
 
-[`Query`](Query.md)<`T`\>
+[`Query`](Query.md)\<`T`\>
 
 #### Defined in
 
-[index.ts:464](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L464)
+[query.ts:55](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L55)
 
 ___
 
 ### metricType
 
-▸ **metricType**(`value`): [`Query`](Query.md)<`T`\>
+▸ **metricType**(`value`): [`Query`](Query.md)\<`T`\>
 
 The MetricType used for this Query.
 
-**`See`**
-
-MetricType for the different options
-
 #### Parameters
 
 | Name | Type | Description |
@@ -276,17 +299,21 @@ MetricType for the different options
 
 #### Returns
 
-[`Query`](Query.md)<`T`\>
+[`Query`](Query.md)\<`T`\>
+
+**`See`**
+
+MetricType for the different options
 
 #### Defined in
 
-[index.ts:511](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L511)
+[query.ts:102](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L102)
 
 ___
 
 ### nprobes
 
-▸ **nprobes**(`value`): [`Query`](Query.md)<`T`\>
+▸ **nprobes**(`value`): [`Query`](Query.md)\<`T`\>
 
 The number of probes used. A higher number makes search more accurate but also slower.
 
@@ -298,17 +325,37 @@ The number of probes used. A higher number makes search more accurate but also s
 
 #### Returns
 
-[`Query`](Query.md)<`T`\>
+[`Query`](Query.md)\<`T`\>
 
 #### Defined in
 
-[index.ts:482](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L482)
+[query.ts:73](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L73)
+
+___
+
+### prefilter
+
+▸ **prefilter**(`value`): [`Query`](Query.md)\<`T`\>
+
+#### Parameters
+
+| Name | Type |
+| :------ | :------ |
+| `value` | `boolean` |
+
+#### Returns
+
+[`Query`](Query.md)\<`T`\>
+
+#### Defined in
+
+[query.ts:107](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L107)
 
 ___
 
 ### refineFactor
 
-▸ **refineFactor**(`value`): [`Query`](Query.md)<`T`\>
+▸ **refineFactor**(`value`): [`Query`](Query.md)\<`T`\>
 
 Refine the results by reading extra elements and re-ranking them in memory.
 
@@ -320,17 +367,17 @@ Refine the results by reading extra elements and re-ranking them in memory.
 
 #### Returns
 
-[`Query`](Query.md)<`T`\>
+[`Query`](Query.md)\<`T`\>
 
 #### Defined in
 
-[index.ts:473](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L473)
+[query.ts:64](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L64)
 
 ___
 
 ### select
 
-▸ **select**(`value`): [`Query`](Query.md)<`T`\>
+▸ **select**(`value`): [`Query`](Query.md)\<`T`\>
 
 Return only the specified columns.
 
@@ -342,8 +389,8 @@ Return only the specified columns.
 
 #### Returns
 
-[`Query`](Query.md)<`T`\>
+[`Query`](Query.md)\<`T`\>
 
 #### Defined in
 
-[index.ts:502](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L502)
+[query.ts:93](https://github.com/lancedb/lancedb/blob/7856a94/node/src/query.ts#L93)

docs/src/javascript/enums/MetricType.md

@@ -22,7 +22,7 @@ Cosine distance
 
 #### Defined in
 
-[index.ts:567](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L567)
+[index.ts:798](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L798)
 
 ___
 
@@ -34,7 +34,7 @@ Dot product
 
 #### Defined in
 
-[index.ts:572](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L572)
+[index.ts:803](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L803)
 
 ___
 
@@ -46,4 +46,4 @@ Euclidean distance
 
 #### Defined in
 
-[index.ts:562](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L562)
+[index.ts:793](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L793)

docs/src/javascript/enums/WriteMode.md

@@ -22,7 +22,7 @@ Append new data to the table.
 
 #### Defined in
 
-[index.ts:552](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L552)
+[index.ts:766](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L766)
 
 ___
 
@@ -34,7 +34,7 @@ Create a new [Table](../interfaces/Table.md).
 
 #### Defined in
 
-[index.ts:548](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L548)
+[index.ts:762](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L762)
 
 ___
 
@@ -46,4 +46,4 @@ Overwrite the existing [Table](../interfaces/Table.md) if presented.
 
 #### Defined in
 
-[index.ts:550](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L550)
+[index.ts:764](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L764)

docs/src/javascript/interfaces/AwsCredentials.md

@@ -18,7 +18,7 @@
 
 #### Defined in
 
-[index.ts:31](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L31)
+[index.ts:34](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L34)
 
 ___
 
@@ -28,7 +28,7 @@ ___
 
 #### Defined in
 
-[index.ts:33](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L33)
+[index.ts:36](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L36)
 
 ___
 
@@ -38,4 +38,4 @@ ___
 
 #### Defined in
 
-[index.ts:35](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L35)
+[index.ts:38](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L38)

docs/src/javascript/interfaces/CleanupStats.md

@@ -0,0 +1,34 @@
+[vectordb](../README.md) / [Exports](../modules.md) / CleanupStats
+
+# Interface: CleanupStats
+
+## Table of contents
+
+### Properties
+
+- [bytesRemoved](CleanupStats.md#bytesremoved)
+- [oldVersions](CleanupStats.md#oldversions)
+
+## Properties
+
+### bytesRemoved
+
+• **bytesRemoved**: `number`
+
+The number of bytes removed from disk.
+
+#### Defined in
+
+[index.ts:637](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L637)
+
+___
+
+### oldVersions
+
+• **oldVersions**: `number`
+
+The number of old table versions removed.
+
+#### Defined in
+
+[index.ts:641](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L641)

docs/src/javascript/interfaces/CompactionMetrics.md

@@ -0,0 +1,62 @@
+[vectordb](../README.md) / [Exports](../modules.md) / CompactionMetrics
+
+# Interface: CompactionMetrics
+
+## Table of contents
+
+### Properties
+
+- [filesAdded](CompactionMetrics.md#filesadded)
+- [filesRemoved](CompactionMetrics.md#filesremoved)
+- [fragmentsAdded](CompactionMetrics.md#fragmentsadded)
+- [fragmentsRemoved](CompactionMetrics.md#fragmentsremoved)
+
+## Properties
+
+### filesAdded
+
+• **filesAdded**: `number`
+
+The number of files added. This is typically equal to the number of
+fragments added.
+
+#### Defined in
+
+[index.ts:692](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L692)
+
+___
+
+### filesRemoved
+
+• **filesRemoved**: `number`
+
+The number of files that were removed. Each fragment may have more than one
+file.
+
+#### Defined in
+
+[index.ts:687](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L687)
+
+___
+
+### fragmentsAdded
+
+• **fragmentsAdded**: `number`
+
+The number of new fragments that were created.
+
+#### Defined in
+
+[index.ts:682](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L682)
+
+___
+
+### fragmentsRemoved
+
+• **fragmentsRemoved**: `number`
+
+The number of fragments that were removed.
+
+#### Defined in
+
+[index.ts:678](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L678)

docs/src/javascript/interfaces/CompactionOptions.md

@@ -0,0 +1,80 @@
+[vectordb](../README.md) / [Exports](../modules.md) / CompactionOptions
+
+# Interface: CompactionOptions
+
+## Table of contents
+
+### Properties
+
+- [materializeDeletions](CompactionOptions.md#materializedeletions)
+- [materializeDeletionsThreshold](CompactionOptions.md#materializedeletionsthreshold)
+- [maxRowsPerGroup](CompactionOptions.md#maxrowspergroup)
+- [numThreads](CompactionOptions.md#numthreads)
+- [targetRowsPerFragment](CompactionOptions.md#targetrowsperfragment)
+
+## Properties
+
+### materializeDeletions
+
+• `Optional` **materializeDeletions**: `boolean`
+
+If true, fragments that have rows that are deleted may be compacted to
+remove the deleted rows. This can improve the performance of queries.
+Default is true.
+
+#### Defined in
+
+[index.ts:660](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L660)
+
+___
+
+### materializeDeletionsThreshold
+
+• `Optional` **materializeDeletionsThreshold**: `number`
+
+A number between 0 and 1, representing the proportion of rows that must be
+marked deleted before a fragment is a candidate for compaction to remove
+the deleted rows. Default is 10%.
+
+#### Defined in
+
+[index.ts:666](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L666)
+
+___
+
+### maxRowsPerGroup
+
+• `Optional` **maxRowsPerGroup**: `number`
+
+The maximum number of rows per group. Defaults to 1024.
+
+#### Defined in
+
+[index.ts:654](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L654)
+
+___
+
+### numThreads
+
+• `Optional` **numThreads**: `number`
+
+The number of threads to use for compaction. If not provided, defaults to
+the number of cores on the machine.
+
+#### Defined in
+
+[index.ts:671](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L671)
+
+___
+
+### targetRowsPerFragment
+
+• `Optional` **targetRowsPerFragment**: `number`
+
+The number of rows per fragment to target. Fragments that have fewer rows
+will be compacted into adjacent fragments to produce larger fragments.
+Defaults to 1024 * 1024.
+
+#### Defined in
+
+[index.ts:650](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L650)

docs/src/javascript/interfaces/Connection.md

@@ -19,7 +19,6 @@ Connection could be local against filesystem or remote against a server.
 ### Methods
 
 - [createTable](Connection.md#createtable)
-- [createTableArrow](Connection.md#createtablearrow)
 - [dropTable](Connection.md#droptable)
 - [openTable](Connection.md#opentable)
 - [tableNames](Connection.md#tablenames)
@@ -32,13 +31,76 @@ Connection could be local against filesystem or remote against a server.
 
 #### Defined in
 
-[index.ts:70](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L70)
+[index.ts:125](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L125)
 
 ## Methods
 
 ### createTable
 
-▸ **createTable**<`T`\>(`name`, `data`, `mode?`, `embeddings?`): `Promise`<[`Table`](Table.md)<`T`\>\>
+▸ **createTable**\<`T`\>(`«destructured»`): `Promise`\<[`Table`](Table.md)\<`T`\>\>
+
+Creates a new Table, optionally initializing it with new data.
+
+#### Type parameters
+
+| Name |
+| :------ |
+| `T` |
+
+#### Parameters
+
+| Name | Type |
+| :------ | :------ |
+| `«destructured»` | [`CreateTableOptions`](CreateTableOptions.md)\<`T`\> |
+
+#### Returns
+
+`Promise`\<[`Table`](Table.md)\<`T`\>\>
+
+#### Defined in
+
+[index.ts:146](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L146)
+
+▸ **createTable**(`name`, `data`): `Promise`\<[`Table`](Table.md)\<`number`[]\>\>
+
+Creates a new Table and initialize it with new data.
+
+#### Parameters
+
+| Name | Type | Description |
+| :------ | :------ | :------ |
+| `name` | `string` | The name of the table. |
+| `data` | `Record`\<`string`, `unknown`\>[] | Non-empty Array of Records to be inserted into the table |
+
+#### Returns
+
+`Promise`\<[`Table`](Table.md)\<`number`[]\>\>
+
+#### Defined in
+
+[index.ts:154](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L154)
+
+▸ **createTable**(`name`, `data`, `options`): `Promise`\<[`Table`](Table.md)\<`number`[]\>\>
+
+Creates a new Table and initialize it with new data.
+
+#### Parameters
+
+| Name | Type | Description |
+| :------ | :------ | :------ |
+| `name` | `string` | The name of the table. |
+| `data` | `Record`\<`string`, `unknown`\>[] | Non-empty Array of Records to be inserted into the table |
+| `options` | [`WriteOptions`](WriteOptions.md) | The write options to use when creating the table. |
+
+#### Returns
+
+`Promise`\<[`Table`](Table.md)\<`number`[]\>\>
+
+#### Defined in
+
+[index.ts:163](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L163)
+
+▸ **createTable**\<`T`\>(`name`, `data`, `embeddings`): `Promise`\<[`Table`](Table.md)\<`T`\>\>
 
 Creates a new Table and initialize it with new data.
 
@@ -53,44 +115,49 @@ Creates a new Table and initialize it with new data.
 | Name | Type | Description |
 | :------ | :------ | :------ |
 | `name` | `string` | The name of the table. |
-| `data` | `Record`<`string`, `unknown`\>[] | Non-empty Array of Records to be inserted into the table |
-| `mode?` | [`WriteMode`](../enums/WriteMode.md) | The write mode to use when creating the table. |
-| `embeddings?` | [`EmbeddingFunction`](EmbeddingFunction.md)<`T`\> | An embedding function to use on this table |
+| `data` | `Record`\<`string`, `unknown`\>[] | Non-empty Array of Records to be inserted into the table |
+| `embeddings` | [`EmbeddingFunction`](EmbeddingFunction.md)\<`T`\> | An embedding function to use on this table |
 
 #### Returns
 
-`Promise`<[`Table`](Table.md)<`T`\>\>
+`Promise`\<[`Table`](Table.md)\<`T`\>\>
 
 #### Defined in
 
-[index.ts:90](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L90)
+[index.ts:172](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L172)
 
-___
+▸ **createTable**\<`T`\>(`name`, `data`, `embeddings`, `options`): `Promise`\<[`Table`](Table.md)\<`T`\>\>
 
-### createTableArrow
+Creates a new Table and initialize it with new data.
 
-▸ **createTableArrow**(`name`, `table`): `Promise`<[`Table`](Table.md)<`number`[]\>\>
+#### Type parameters
+
+| Name |
+| :------ |
+| `T` |
 
 #### Parameters
 
-| Name | Type |
-| :------ | :------ |
-| `name` | `string` |
-| `table` | `Table`<`any`\> |
+| Name | Type | Description |
+| :------ | :------ | :------ |
+| `name` | `string` | The name of the table. |
+| `data` | `Record`\<`string`, `unknown`\>[] | Non-empty Array of Records to be inserted into the table |
+| `embeddings` | [`EmbeddingFunction`](EmbeddingFunction.md)\<`T`\> | An embedding function to use on this table |
+| `options` | [`WriteOptions`](WriteOptions.md) | The write options to use when creating the table. |
 
 #### Returns
 
-`Promise`<[`Table`](Table.md)<`number`[]\>\>
+`Promise`\<[`Table`](Table.md)\<`T`\>\>
 
 #### Defined in
 
-[index.ts:92](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L92)
+[index.ts:181](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L181)
 
 ___
 
 ### dropTable
 
-▸ **dropTable**(`name`): `Promise`<`void`\>
+▸ **dropTable**(`name`): `Promise`\<`void`\>
 
 Drop an existing table.
 
@@ -102,17 +169,17 @@ Drop an existing table.
 
 #### Returns
 
-`Promise`<`void`\>
+`Promise`\<`void`\>
 
 #### Defined in
 
-[index.ts:98](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L98)
+[index.ts:187](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L187)
 
 ___
 
 ### openTable
 
-▸ **openTable**<`T`\>(`name`, `embeddings?`): `Promise`<[`Table`](Table.md)<`T`\>\>
+▸ **openTable**\<`T`\>(`name`, `embeddings?`): `Promise`\<[`Table`](Table.md)\<`T`\>\>
 
 Open a table in the database.
 
@@ -127,26 +194,26 @@ Open a table in the database.
 | Name | Type | Description |
 | :------ | :------ | :------ |
 | `name` | `string` | The name of the table. |
-| `embeddings?` | [`EmbeddingFunction`](EmbeddingFunction.md)<`T`\> | An embedding function to use on this table |
+| `embeddings?` | [`EmbeddingFunction`](EmbeddingFunction.md)\<`T`\> | An embedding function to use on this table |
 
 #### Returns
 
-`Promise`<[`Table`](Table.md)<`T`\>\>
+`Promise`\<[`Table`](Table.md)\<`T`\>\>
 
 #### Defined in
 
-[index.ts:80](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L80)
+[index.ts:135](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L135)
 
 ___
 
 ### tableNames
 
-▸ **tableNames**(): `Promise`<`string`[]\>
+▸ **tableNames**(): `Promise`\<`string`[]\>
 
 #### Returns
 
-`Promise`<`string`[]\>
+`Promise`\<`string`[]\>
 
 #### Defined in
 
-[index.ts:72](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L72)
+[index.ts:127](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L127)

docs/src/javascript/interfaces/ConnectionOptions.md

@@ -6,18 +6,62 @@
 
 ### Properties
 
+- [apiKey](ConnectionOptions.md#apikey)
 - [awsCredentials](ConnectionOptions.md#awscredentials)
+- [awsRegion](ConnectionOptions.md#awsregion)
+- [hostOverride](ConnectionOptions.md#hostoverride)
+- [region](ConnectionOptions.md#region)
 - [uri](ConnectionOptions.md#uri)
 
 ## Properties
 
+### apiKey
+
+• `Optional` **apiKey**: `string`
+
+#### Defined in
+
+[index.ts:49](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L49)
+
+___
+
 ### awsCredentials
 
 • `Optional` **awsCredentials**: [`AwsCredentials`](AwsCredentials.md)
 
 #### Defined in
 
-[index.ts:40](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L40)
+[index.ts:44](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L44)
+
+___
+
+### awsRegion
+
+• `Optional` **awsRegion**: `string`
+
+#### Defined in
+
+[index.ts:46](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L46)
+
+___
+
+### hostOverride
+
+• `Optional` **hostOverride**: `string`
+
+#### Defined in
+
+[index.ts:54](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L54)
+
+___
+
+### region
+
+• `Optional` **region**: `string`
+
+#### Defined in
+
+[index.ts:51](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L51)
 
 ___
 
@@ -27,4 +71,4 @@ ___
 
 #### Defined in
 
-[index.ts:39](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L39)
+[index.ts:42](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L42)

docs/src/javascript/interfaces/CreateTableOptions.md

@@ -0,0 +1,69 @@
+[vectordb](../README.md) / [Exports](../modules.md) / CreateTableOptions
+
+# Interface: CreateTableOptions\<T\>
+
+## Type parameters
+
+| Name |
+| :------ |
+| `T` |
+
+## Table of contents
+
+### Properties
+
+- [data](CreateTableOptions.md#data)
+- [embeddingFunction](CreateTableOptions.md#embeddingfunction)
+- [name](CreateTableOptions.md#name)
+- [schema](CreateTableOptions.md#schema)
+- [writeOptions](CreateTableOptions.md#writeoptions)
+
+## Properties
+
+### data
+
+• `Optional` **data**: `Table`\<`any`\> \| `Record`\<`string`, `unknown`\>[]
+
+#### Defined in
+
+[index.ts:79](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L79)
+
+___
+
+### embeddingFunction
+
+• `Optional` **embeddingFunction**: [`EmbeddingFunction`](EmbeddingFunction.md)\<`T`\>
+
+#### Defined in
+
+[index.ts:85](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L85)
+
+___
+
+### name
+
+• **name**: `string`
+
+#### Defined in
+
+[index.ts:76](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L76)
+
+___
+
+### schema
+
+• `Optional` **schema**: `Schema`\<`any`\>
+
+#### Defined in
+
+[index.ts:82](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L82)
+
+___
+
+### writeOptions
+
+• `Optional` **writeOptions**: [`WriteOptions`](WriteOptions.md)
+
+#### Defined in
+
+[index.ts:88](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L88)

docs/src/javascript/interfaces/EmbeddingFunction.md

@@ -1,6 +1,6 @@
 [vectordb](../README.md) / [Exports](../modules.md) / EmbeddingFunction
 
-# Interface: EmbeddingFunction<T\>
+# Interface: EmbeddingFunction\<T\>
 
 An embedding function that automatically creates vector representation for a given column.
 
@@ -25,11 +25,11 @@ An embedding function that automatically creates vector representation for a giv
 
 ### embed
 
-• **embed**: (`data`: `T`[]) => `Promise`<`number`[][]\>
+• **embed**: (`data`: `T`[]) => `Promise`\<`number`[][]\>
 
 #### Type declaration
 
-▸ (`data`): `Promise`<`number`[][]\>
+▸ (`data`): `Promise`\<`number`[][]\>
 
 Creates a vector representation for the given values.
 
@@ -41,11 +41,11 @@ Creates a vector representation for the given values.
 
 ##### Returns
 
-`Promise`<`number`[][]\>
+`Promise`\<`number`[][]\>
 
 #### Defined in
 
-[embedding/embedding_function.ts:27](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/embedding/embedding_function.ts#L27)
+[embedding/embedding_function.ts:27](https://github.com/lancedb/lancedb/blob/7856a94/node/src/embedding/embedding_function.ts#L27)
 
 ___
 
@@ -57,4 +57,4 @@ The name of the column that will be used as input for the Embedding Function.
 
 #### Defined in
 
-[embedding/embedding_function.ts:22](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/embedding/embedding_function.ts#L22)
+[embedding/embedding_function.ts:22](https://github.com/lancedb/lancedb/blob/7856a94/node/src/embedding/embedding_function.ts#L22)

docs/src/javascript/interfaces/IndexStats.md

@@ -0,0 +1,30 @@
+[vectordb](../README.md) / [Exports](../modules.md) / IndexStats
+
+# Interface: IndexStats
+
+## Table of contents
+
+### Properties
+
+- [numIndexedRows](IndexStats.md#numindexedrows)
+- [numUnindexedRows](IndexStats.md#numunindexedrows)
+
+## Properties
+
+### numIndexedRows
+
+• **numIndexedRows**: ``null`` \| `number`
+
+#### Defined in
+
+[index.ts:344](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L344)
+
+___
+
+### numUnindexedRows
+
+• **numUnindexedRows**: ``null`` \| `number`
+
+#### Defined in
+
+[index.ts:345](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L345)

docs/src/javascript/interfaces/IvfPQIndexConfig.md

@@ -7,6 +7,7 @@
 ### Properties
 
 - [column](IvfPQIndexConfig.md#column)
+- [index\_cache\_size](IvfPQIndexConfig.md#index_cache_size)
 - [index\_name](IvfPQIndexConfig.md#index_name)
 - [max\_iters](IvfPQIndexConfig.md#max_iters)
 - [max\_opq\_iters](IvfPQIndexConfig.md#max_opq_iters)
@@ -28,7 +29,19 @@ The column to be indexed
 
 #### Defined in
 
-[index.ts:382](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L382)
+[index.ts:701](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L701)
+
+___
+
+### index\_cache\_size
+
+• `Optional` **index\_cache\_size**: `number`
+
+Cache size of the index
+
+#### Defined in
+
+[index.ts:750](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L750)
 
 ___
 
@@ -40,7 +53,7 @@ A unique name for the index
 
 #### Defined in
 
-[index.ts:387](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L387)
+[index.ts:706](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L706)
 
 ___
 
@@ -52,7 +65,7 @@ The max number of iterations for kmeans training.
 
 #### Defined in
 
-[index.ts:402](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L402)
+[index.ts:721](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L721)
 
 ___
 
@@ -64,7 +77,7 @@ Max number of iterations to train OPQ, if `use_opq` is true.
 
 #### Defined in
 
-[index.ts:421](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L421)
+[index.ts:740](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L740)
 
 ___
 
@@ -76,7 +89,7 @@ Metric type, L2 or Cosine
 
 #### Defined in
 
-[index.ts:392](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L392)
+[index.ts:711](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L711)
 
 ___
 
@@ -88,7 +101,7 @@ The number of bits to present one PQ centroid.
 
 #### Defined in
 
-[index.ts:416](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L416)
+[index.ts:735](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L735)
 
 ___
 
@@ -100,7 +113,7 @@ The number of partitions this index
 
 #### Defined in
 
-[index.ts:397](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L397)
+[index.ts:716](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L716)
 
 ___
 
@@ -112,7 +125,7 @@ Number of subvectors to build PQ code
 
 #### Defined in
 
-[index.ts:412](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L412)
+[index.ts:731](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L731)
 
 ___
 
@@ -124,7 +137,7 @@ Replace an existing index with the same name if it exists.
 
 #### Defined in
 
-[index.ts:426](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L426)
+[index.ts:745](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L745)
 
 ___
 
@@ -134,7 +147,7 @@ ___
 
 #### Defined in
 
-[index.ts:428](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L428)
+[index.ts:752](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L752)
 
 ___
 
@@ -146,4 +159,4 @@ Train as optimized product quantization.
 
 #### Defined in
 
-[index.ts:407](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L407)
+[index.ts:726](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L726)

docs/src/javascript/interfaces/Table.md

@@ -1,6 +1,6 @@
 [vectordb](../README.md) / [Exports](../modules.md) / Table
 
-# Interface: Table<T\>
+# Interface: Table\<T\>
 
 A LanceDB Table is the collection of Records. Each Record has one or more vector fields.
 
@@ -22,19 +22,22 @@ A LanceDB Table is the collection of Records. Each Record has one or more vector
 - [countRows](Table.md#countrows)
 - [createIndex](Table.md#createindex)
 - [delete](Table.md#delete)
+- [indexStats](Table.md#indexstats)
+- [listIndices](Table.md#listindices)
 - [name](Table.md#name)
 - [overwrite](Table.md#overwrite)
 - [search](Table.md#search)
+- [update](Table.md#update)
 
 ## Properties
 
 ### add
 
-• **add**: (`data`: `Record`<`string`, `unknown`\>[]) => `Promise`<`number`\>
+• **add**: (`data`: `Record`\<`string`, `unknown`\>[]) => `Promise`\<`number`\>
 
 #### Type declaration
 
-▸ (`data`): `Promise`<`number`\>
+▸ (`data`): `Promise`\<`number`\>
 
 Insert records into this Table.
 
@@ -42,54 +45,50 @@ Insert records into this Table.
 
 | Name | Type | Description |
 | :------ | :------ | :------ |
-| `data` | `Record`<`string`, `unknown`\>[] | Records to be inserted into the Table |
+| `data` | `Record`\<`string`, `unknown`\>[] | Records to be inserted into the Table |
 
 ##### Returns
 
-`Promise`<`number`\>
+`Promise`\<`number`\>
 
 The number of rows added to the table
 
 #### Defined in
 
-[index.ts:120](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L120)
+[index.ts:209](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L209)
 
 ___
 
 ### countRows
 
-• **countRows**: () => `Promise`<`number`\>
+• **countRows**: () => `Promise`\<`number`\>
 
 #### Type declaration
 
-▸ (): `Promise`<`number`\>
+▸ (): `Promise`\<`number`\>
 
 Returns the number of rows in this table.
 
 ##### Returns
 
-`Promise`<`number`\>
+`Promise`\<`number`\>
 
 #### Defined in
 
-[index.ts:140](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L140)
+[index.ts:229](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L229)
 
 ___
 
 ### createIndex
 
-• **createIndex**: (`indexParams`: [`IvfPQIndexConfig`](IvfPQIndexConfig.md)) => `Promise`<`any`\>
+• **createIndex**: (`indexParams`: [`IvfPQIndexConfig`](IvfPQIndexConfig.md)) => `Promise`\<`any`\>
 
 #### Type declaration
 
-▸ (`indexParams`): `Promise`<`any`\>
+▸ (`indexParams`): `Promise`\<`any`\>
 
 Create an ANN index on this Table vector index.
 
-**`See`**
-
-VectorIndexParams.
-
 ##### Parameters
 
 | Name | Type | Description |
@@ -98,27 +97,41 @@ VectorIndexParams.
 
 ##### Returns
 
-`Promise`<`any`\>
+`Promise`\<`any`\>
+
+**`See`**
+
+VectorIndexParams.
 
 #### Defined in
 
-[index.ts:135](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L135)
+[index.ts:224](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L224)
 
 ___
 
 ### delete
 
-• **delete**: (`filter`: `string`) => `Promise`<`void`\>
+• **delete**: (`filter`: `string`) => `Promise`\<`void`\>
 
 #### Type declaration
 
-▸ (`filter`): `Promise`<`void`\>
+▸ (`filter`): `Promise`\<`void`\>
 
 Delete rows from this table.
 
 This can be used to delete a single row, many rows, all rows, or
 sometimes no rows (if your predicate matches nothing).
 
+##### Parameters
+
+| Name | Type | Description |
+| :------ | :------ | :------ |
+| `filter` | `string` | A filter in the same format used by a sql WHERE clause. The filter must not be empty. |
+
+##### Returns
+
+`Promise`\<`void`\>
+
 **`Examples`**
 
 ```ts
@@ -142,19 +155,55 @@ await tbl.delete(`id IN (${to_remove.join(",")})`)
 await tbl.countRows() // Returns 1
 ```
 
+#### Defined in
+
+[index.ts:263](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L263)
+
+___
+
+### indexStats
+
+• **indexStats**: (`indexUuid`: `string`) => `Promise`\<[`IndexStats`](IndexStats.md)\>
+
+#### Type declaration
+
+▸ (`indexUuid`): `Promise`\<[`IndexStats`](IndexStats.md)\>
+
+Get statistics about an index.
+
 ##### Parameters
 
-| Name | Type | Description |
-| :------ | :------ | :------ |
-| `filter` | `string` | A filter in the same format used by a sql WHERE clause. The filter must not be empty. |
+| Name | Type |
+| :------ | :------ |
+| `indexUuid` | `string` |
 
 ##### Returns
 
-`Promise`<`void`\>
+`Promise`\<[`IndexStats`](IndexStats.md)\>
 
 #### Defined in
 
-[index.ts:174](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L174)
+[index.ts:306](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L306)
+
+___
+
+### listIndices
+
+• **listIndices**: () => `Promise`\<[`VectorIndex`](VectorIndex.md)[]\>
+
+#### Type declaration
+
+▸ (): `Promise`\<[`VectorIndex`](VectorIndex.md)[]\>
+
+List the indicies on this table.
+
+##### Returns
+
+`Promise`\<[`VectorIndex`](VectorIndex.md)[]\>
+
+#### Defined in
+
+[index.ts:301](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L301)
 
 ___
 
@@ -164,17 +213,17 @@ ___
 
 #### Defined in
 
-[index.ts:106](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L106)
+[index.ts:195](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L195)
 
 ___
 
 ### overwrite
 
-• **overwrite**: (`data`: `Record`<`string`, `unknown`\>[]) => `Promise`<`number`\>
+• **overwrite**: (`data`: `Record`\<`string`, `unknown`\>[]) => `Promise`\<`number`\>
 
 #### Type declaration
 
-▸ (`data`): `Promise`<`number`\>
+▸ (`data`): `Promise`\<`number`\>
 
 Insert records into this Table, replacing its contents.
 
@@ -182,27 +231,27 @@ Insert records into this Table, replacing its contents.
 
 | Name | Type | Description |
 | :------ | :------ | :------ |
-| `data` | `Record`<`string`, `unknown`\>[] | Records to be inserted into the Table |
+| `data` | `Record`\<`string`, `unknown`\>[] | Records to be inserted into the Table |
 
 ##### Returns
 
-`Promise`<`number`\>
+`Promise`\<`number`\>
 
 The number of rows added to the table
 
 #### Defined in
 
-[index.ts:128](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L128)
+[index.ts:217](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L217)
 
 ___
 
 ### search
 
-• **search**: (`query`: `T`) => [`Query`](../classes/Query.md)<`T`\>
+• **search**: (`query`: `T`) => [`Query`](../classes/Query.md)\<`T`\>
 
 #### Type declaration
 
-▸ (`query`): [`Query`](../classes/Query.md)<`T`\>
+▸ (`query`): [`Query`](../classes/Query.md)\<`T`\>
 
 Creates a search query to find the nearest neighbors of the given search term
 
@@ -214,8 +263,59 @@ Creates a search query to find the nearest neighbors of the given search term
 
 ##### Returns
 
-[`Query`](../classes/Query.md)<`T`\>
+[`Query`](../classes/Query.md)\<`T`\>
 
 #### Defined in
 
-[index.ts:112](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L112)
+[index.ts:201](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L201)
+
+___
+
+### update
+
+• **update**: (`args`: [`UpdateArgs`](UpdateArgs.md) \| [`UpdateSqlArgs`](UpdateSqlArgs.md)) => `Promise`\<`void`\>
+
+#### Type declaration
+
+▸ (`args`): `Promise`\<`void`\>
+
+Update rows in this table.
+
+This can be used to update a single row, many rows, all rows, or
+sometimes no rows (if your predicate matches nothing).
+
+##### Parameters
+
+| Name | Type | Description |
+| :------ | :------ | :------ |
+| `args` | [`UpdateArgs`](UpdateArgs.md) \| [`UpdateSqlArgs`](UpdateSqlArgs.md) | see [UpdateArgs](UpdateArgs.md) and [UpdateSqlArgs](UpdateSqlArgs.md) for more details |
+
+##### Returns
+
+`Promise`\<`void`\>
+
+**`Examples`**
+
+```ts
+const con = await lancedb.connect("./.lancedb")
+const data = [
+   {id: 1, vector: [3, 3], name: 'Ye'},
+   {id: 2, vector: [4, 4], name: 'Mike'},
+];
+const tbl = await con.createTable("my_table", data)
+
+await tbl.update({
+  filter: "id = 2",
+  updates: { vector: [2, 2], name: "Michael" },
+})
+
+let results = await tbl.search([1, 1]).execute();
+// Returns [
+//   {id: 2, vector: [2, 2], name: 'Michael'}
+//   {id: 1, vector: [3, 3], name: 'Ye'}
+// ]
+```
+
+#### Defined in
+
+[index.ts:296](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L296)

docs/src/javascript/interfaces/UpdateArgs.md

@@ -0,0 +1,36 @@
+[vectordb](../README.md) / [Exports](../modules.md) / UpdateArgs
+
+# Interface: UpdateArgs
+
+## Table of contents
+
+### Properties
+
+- [values](UpdateArgs.md#values)
+- [where](UpdateArgs.md#where)
+
+## Properties
+
+### values
+
+• **values**: `Record`\<`string`, `Literal`\>
+
+A key-value map of updates. The keys are the column names, and the values are the
+new values to set
+
+#### Defined in
+
+[index.ts:320](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L320)
+
+___
+
+### where
+
+• `Optional` **where**: `string`
+
+A filter in the same format used by a sql WHERE clause. The filter may be empty,
+in which case all rows will be updated.
+
+#### Defined in
+
+[index.ts:314](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L314)

docs/src/javascript/interfaces/UpdateSqlArgs.md

@@ -0,0 +1,36 @@
+[vectordb](../README.md) / [Exports](../modules.md) / UpdateSqlArgs
+
+# Interface: UpdateSqlArgs
+
+## Table of contents
+
+### Properties
+
+- [valuesSql](UpdateSqlArgs.md#valuessql)
+- [where](UpdateSqlArgs.md#where)
+
+## Properties
+
+### valuesSql
+
+• **valuesSql**: `Record`\<`string`, `string`\>
+
+A key-value map of updates. The keys are the column names, and the values are the
+new values to set as SQL expressions.
+
+#### Defined in
+
+[index.ts:334](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L334)
+
+___
+
+### where
+
+• `Optional` **where**: `string`
+
+A filter in the same format used by a sql WHERE clause. The filter may be empty,
+in which case all rows will be updated.
+
+#### Defined in
+
+[index.ts:328](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L328)

docs/src/javascript/interfaces/VectorIndex.md

@@ -0,0 +1,41 @@
+[vectordb](../README.md) / [Exports](../modules.md) / VectorIndex
+
+# Interface: VectorIndex
+
+## Table of contents
+
+### Properties
+
+- [columns](VectorIndex.md#columns)
+- [name](VectorIndex.md#name)
+- [uuid](VectorIndex.md#uuid)
+
+## Properties
+
+### columns
+
+• **columns**: `string`[]
+
+#### Defined in
+
+[index.ts:338](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L338)
+
+___
+
+### name
+
+• **name**: `string`
+
+#### Defined in
+
+[index.ts:339](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L339)
+
+___
+
+### uuid
+
+• **uuid**: `string`
+
+#### Defined in
+
+[index.ts:340](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L340)

docs/src/javascript/interfaces/WriteOptions.md

@@ -0,0 +1,27 @@
+[vectordb](../README.md) / [Exports](../modules.md) / WriteOptions
+
+# Interface: WriteOptions
+
+Write options when creating a Table.
+
+## Implemented by
+
+- [`DefaultWriteOptions`](../classes/DefaultWriteOptions.md)
+
+## Table of contents
+
+### Properties
+
+- [writeMode](WriteOptions.md#writemode)
+
+## Properties
+
+### writeMode
+
+• `Optional` **writeMode**: [`WriteMode`](../enums/WriteMode.md)
+
+A [WriteMode](../enums/WriteMode.md) to use on this operation
+
+#### Defined in
+
+[index.ts:774](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L774)

docs/src/javascript/modules.md

@@ -11,6 +11,7 @@
 
 ### Classes
 
+- [DefaultWriteOptions](classes/DefaultWriteOptions.md)
 - [LocalConnection](classes/LocalConnection.md)
 - [LocalTable](classes/LocalTable.md)
 - [OpenAIEmbeddingFunction](classes/OpenAIEmbeddingFunction.md)
@@ -19,11 +20,20 @@
 ### Interfaces
 
 - [AwsCredentials](interfaces/AwsCredentials.md)
+- [CleanupStats](interfaces/CleanupStats.md)
+- [CompactionMetrics](interfaces/CompactionMetrics.md)
+- [CompactionOptions](interfaces/CompactionOptions.md)
 - [Connection](interfaces/Connection.md)
 - [ConnectionOptions](interfaces/ConnectionOptions.md)
+- [CreateTableOptions](interfaces/CreateTableOptions.md)
 - [EmbeddingFunction](interfaces/EmbeddingFunction.md)
+- [IndexStats](interfaces/IndexStats.md)
 - [IvfPQIndexConfig](interfaces/IvfPQIndexConfig.md)
 - [Table](interfaces/Table.md)
+- [UpdateArgs](interfaces/UpdateArgs.md)
+- [UpdateSqlArgs](interfaces/UpdateSqlArgs.md)
+- [VectorIndex](interfaces/VectorIndex.md)
+- [WriteOptions](interfaces/WriteOptions.md)
 
 ### Type Aliases
 
@@ -32,6 +42,7 @@
 ### Functions
 
 - [connect](modules.md#connect)
+- [isWriteOptions](modules.md#iswriteoptions)
 
 ## Type Aliases
 
@@ -41,13 +52,13 @@
 
 #### Defined in
 
-[index.ts:431](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L431)
+[index.ts:755](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L755)
 
 ## Functions
 
 ### connect
 
-▸ **connect**(`uri`): `Promise`<[`Connection`](interfaces/Connection.md)\>
+▸ **connect**(`uri`): `Promise`\<[`Connection`](interfaces/Connection.md)\>
 
 Connect to a LanceDB instance at the given URI
 
@@ -59,24 +70,44 @@ Connect to a LanceDB instance at the given URI
 
 #### Returns
 
-`Promise`<[`Connection`](interfaces/Connection.md)\>
+`Promise`\<[`Connection`](interfaces/Connection.md)\>
 
 #### Defined in
 
-[index.ts:47](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L47)
+[index.ts:95](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L95)
 
-▸ **connect**(`opts`): `Promise`<[`Connection`](interfaces/Connection.md)\>
+▸ **connect**(`opts`): `Promise`\<[`Connection`](interfaces/Connection.md)\>
 
 #### Parameters
 
 | Name | Type |
 | :------ | :------ |
-| `opts` | `Partial`<[`ConnectionOptions`](interfaces/ConnectionOptions.md)\> |
+| `opts` | `Partial`\<[`ConnectionOptions`](interfaces/ConnectionOptions.md)\> |
 
 #### Returns
 
-`Promise`<[`Connection`](interfaces/Connection.md)\>
+`Promise`\<[`Connection`](interfaces/Connection.md)\>
 
 #### Defined in
 
-[index.ts:48](https://github.com/lancedb/lancedb/blob/b1eeb90/node/src/index.ts#L48)
+[index.ts:96](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L96)
+
+___
+
+### isWriteOptions
+
+▸ **isWriteOptions**(`value`): value is WriteOptions
+
+#### Parameters
+
+| Name | Type |
+| :------ | :------ |
+| `value` | `any` |
+
+#### Returns
+
+value is WriteOptions
+
+#### Defined in
+
+[index.ts:781](https://github.com/lancedb/lancedb/blob/7856a94/node/src/index.ts#L781)

docs/src/python/pydantic.md

@@ -7,7 +7,7 @@ LanceDB integrates with Pydantic for schema inference, data ingestion, and query
 
 LanceDB supports to create Apache Arrow Schema from a
 [Pydantic BaseModel](https://docs.pydantic.dev/latest/api/main/#pydantic.main.BaseModel)
-via [pydantic_to_schema()](python.md##lancedb.pydantic.pydantic_to_schema) method.
+via [pydantic_to_schema()](python.md#lancedb.pydantic.pydantic_to_schema) method.
 
 ::: lancedb.pydantic.pydantic_to_schema
 

docs/src/python/saas-python.md

@@ -0,0 +1,18 @@
+# LanceDB Python API Reference
+
+## Installation
+
+```shell
+pip install lancedb
+```
+
+## Connection
+
+::: lancedb.connect
+
+::: lancedb.remote.db.RemoteDBConnection
+
+## Table
+
+::: lancedb.remote.table.RemoteTable
+

docs/src/search.md

@@ -118,4 +118,101 @@ However, fast vector search using indices often entails making a trade-off with
 This is why it is often called **Approximate Nearest Neighbors (ANN)** search, while the Flat Search (KNN)
 always returns 100% recall.
 
-See [ANN Index](ann_indexes.md) for more details.
+See [ANN Index](ann_indexes.md) for more details.
+
+
+### Output formats
+
+LanceDB returns results in many different formats commonly used in python.
+Let's create a LanceDB table with a nested schema:
+
+```python
+from datetime import datetime
+import lancedb
+from lancedb.pydantic import LanceModel, Vector
+import numpy as np
+from pydantic import BaseModel
+uri = "data/sample-lancedb-nested"
+
+class Metadata(BaseModel):
+    source: str
+    timestamp: datetime
+
+class Document(BaseModel):
+    content: str
+    meta: Metadata
+
+class LanceSchema(LanceModel):
+    id: str
+    vector: Vector(1536)
+    payload: Document
+
+# Let's add 100 sample rows to our dataset
+data = [LanceSchema(
+    id=f"id{i}",
+    vector=np.random.randn(1536),
+    payload=Document(
+        content=f"document{i}", meta=Metadata(source=f"source{i%10}", timestamp=datetime.now())
+    ),
+) for i in range(100)]
+
+tbl = db.create_table("documents", data=data)
+```
+
+#### As a pyarrow table
+
+Using `to_arrow()` we can get the results back as a pyarrow Table.
+This result table has the same columns as the LanceDB table, with 
+the addition of an `_distance` column for vector search or a `score`
+column for full text search.
+
+```python
+tbl.search(np.random.randn(1536)).to_arrow()
+```
+
+#### As a pandas dataframe
+
+You can also get the results as a pandas dataframe.
+
+```python
+tbl.search(np.random.randn(1536)).to_pandas()
+```
+
+While other formats like Arrow/Pydantic/Python dicts have a natural 
+way to handle nested schemas, pandas can only store nested data as a 
+python dict column, which makes it difficult to support nested references.
+So for convenience, you can also tell LanceDB to flatten a nested schema 
+when creating the pandas dataframe. 
+
+```python
+tbl.search(np.random.randn(1536)).to_pandas(flatten=True)
+```
+
+If your table has a deeply nested struct, you can control how many levels
+of nesting to flatten by passing in a positive integer.
+
+```python
+tbl.search(np.random.randn(1536)).to_pandas(flatten=1)
+```
+
+
+#### As a list of python dicts
+
+You can of course return results as a list of python dicts.
+
+```python
+tbl.search(np.random.randn(1536)).to_list()
+```
+
+#### As a list of pydantic models
+
+We can add data using pydantic models, and we can certainly
+retrieve results as pydantic models
+
+```python
+tbl.search(np.random.randn(1536)).to_pydantic(LanceSchema)
+```
+
+Note that in this case the extra `_distance` field is discarded since
+it's not part of the LanceSchema.
+

docs/src/sql.md

@@ -1,7 +1,7 @@
 # SQL filters
 
 LanceDB embraces the utilization of standard SQL expressions as predicates for hybrid
-filters. It can be used during hybrid vector search and deletion operations.
+filters. It can be used during hybrid vector search, update, and deletion operations.
 
 Currently, Lance supports a growing list of expressions.
 
@@ -22,7 +22,7 @@ import numpy as np
 uri = "data/sample-lancedb"
 db = lancedb.connect(uri)
 
-data = [{"vector": row, "item": f"item {i}"}
+data = [{"vector": row, "item": f"item {i}", "id": i}
      for i, row in enumerate(np.random.random((10_000, 2)).astype('int'))]
 
 tbl = db.create_table("my_vectors", data=data)
@@ -35,33 +35,25 @@ const db = await vectordb.connect('data/sample-lancedb')
 
 let data = []
 for (let i = 0; i < 10_000; i++) {
-     data.push({vector: Array(1536).fill(i), id: `${i}`, content: "", longId: `${i}`},)
+     data.push({vector: Array(1536).fill(i), id: i, item: `item ${i}`, strId: `${i}`})
 }
-const tbl = await db.createTable('my_vectors', data)
+const tbl = await db.createTable('myVectors', data)
 ```
 -->
 === "Python"
 
     ```python
     tbl.search([100, 102]) \
-       .where("""(
-        (label IN [10, 20])
-        AND
-        (note.email IS NOT NULL)
-        ) OR NOT note.created
-        """)
-
+       .where("(item IN ('item 0', 'item 2')) AND (id > 10)") \
+       .to_arrow()
     ```
+
 === "Javascript"
 
     ```javascript
-    tbl.search([100, 102])
-       .where(`(
-            (label IN [10, 20])
-            AND
-            (note.email IS NOT NULL)
-        ) OR NOT note.created
-       `)
+    await tbl.search(Array(1536).fill(0))
+       .where("(item IN ('item 0', 'item 2')) AND (id > 10)")
+       .execute()
     ```
 
 
@@ -118,3 +110,22 @@ The mapping from SQL types to Arrow types is:
 
 [^1]: See precision mapping in previous table.
 
+
+## Filtering without Vector Search
+
+You can also filter your data without search.
+
+=== "Python"
+      ```python
+      tbl.search().where("id=10").limit(10).to_arrow()
+      ```
+
+=== "JavaScript"
+      ```javascript
+      await tbl.where('id=10').limit(10).execute()
+      ```
+
+!!! warning
+    If your table is large, this could potentially return a very large
+    amount of data. Please be sure to use a `limit` clause unless
+    you're sure you want to return the whole result set.

node/README.md

@@ -9,8 +9,13 @@ npm install vectordb
 ```
 
 This will download the appropriate native library for your platform. We currently
-support x86_64 Linux, aarch64 Linux, Intel MacOS, and ARM (M1/M2) MacOS. We do not
-yet support musl-based Linux (such as Alpine Linux).
+support:
+
+* Linux (x86_64 and aarch64)
+* MacOS (Intel and ARM/M1/M2)
+* Windows (x86_64 only)
+
+We do not yet support musl-based Linux (such as Alpine Linux) or aarch64 Windows.
 
 ## Usage
 

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.3.8",
+  "version": "0.4.1",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.3.8",
+      "version": "0.4.1",
       "cpu": [
         "x64",
         "arm64"
@@ -53,11 +53,11 @@
         "uuid": "^9.0.0"
       },
       "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.3.8",
-        "@lancedb/vectordb-darwin-x64": "0.3.8",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.3.8",
-        "@lancedb/vectordb-linux-x64-gnu": "0.3.8",
-        "@lancedb/vectordb-win32-x64-msvc": "0.3.8"
+        "@lancedb/vectordb-darwin-arm64": "0.4.1",
+        "@lancedb/vectordb-darwin-x64": "0.4.1",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.4.1",
+        "@lancedb/vectordb-linux-x64-gnu": "0.4.1",
+        "@lancedb/vectordb-win32-x64-msvc": "0.4.1"
       }
     },
     "node_modules/@apache-arrow/ts": {
@@ -316,6 +316,66 @@
         "@jridgewell/sourcemap-codec": "^1.4.10"
       }
     },
+    "node_modules/@lancedb/vectordb-darwin-arm64": {
+      "version": "0.4.1",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.4.1.tgz",
+      "integrity": "sha512-ul/Hvv5RX2RThpKSuiUjJRVrmXuBPvpU+HrLjcBmu4dzpuWN4+IeHIUM6xe79gLxOKlwkscVweTOuZnmMfsZeg==",
+      "cpu": [
+        "arm64"
+      ],
+      "optional": true,
+      "os": [
+        "darwin"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-darwin-x64": {
+      "version": "0.4.1",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.4.1.tgz",
+      "integrity": "sha512-sJtF2Cv6T9RhUpdeHNkryiJwPuW9QPQ3aMs5fID1hMCJA2U3BX27t/WlkiPT2+kTLeUcwF1JvAOgsfvZkfvI8w==",
+      "cpu": [
+        "x64"
+      ],
+      "optional": true,
+      "os": [
+        "darwin"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-linux-arm64-gnu": {
+      "version": "0.4.1",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.4.1.tgz",
+      "integrity": "sha512-tNnziT0BRjPsznKI4GgWROFdCOsCGx0inFu0z+WV1UomwXKcMWGslpWBqKE8IUiCq14duPVx/ie7Wwcf51IeJQ==",
+      "cpu": [
+        "arm64"
+      ],
+      "optional": true,
+      "os": [
+        "linux"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-linux-x64-gnu": {
+      "version": "0.4.1",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.4.1.tgz",
+      "integrity": "sha512-PAcF2p1FUsC0AD+qkLfgE5+ZlQwlHe9eTP9dSsX43V/NGPDQ9+gBzaBTEDbvyHj1wl2Wft2NwOqB1HAFhilSDg==",
+      "cpu": [
+        "x64"
+      ],
+      "optional": true,
+      "os": [
+        "linux"
+      ]
+    },
+    "node_modules/@lancedb/vectordb-win32-x64-msvc": {
+      "version": "0.4.1",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.4.1.tgz",
+      "integrity": "sha512-8mvThCppI/AfSPby6Y3t6xpCfbo8IY6JH5exO8fDGTwBFHOqgwR4Izb2K7FgXxkwUYcN4EfGSsk/6B1GpwMudg==",
+      "cpu": [
+        "x64"
+      ],
+      "optional": true,
+      "os": [
+        "win32"
+      ]
+    },
     "node_modules/@neon-rs/cli": {
       "version": "0.0.160",
       "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.160.tgz",
@@ -4808,6 +4868,36 @@
         "@jridgewell/sourcemap-codec": "^1.4.10"
       }
     },
+    "@lancedb/vectordb-darwin-arm64": {
+      "version": "0.4.1",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.4.1.tgz",
+      "integrity": "sha512-ul/Hvv5RX2RThpKSuiUjJRVrmXuBPvpU+HrLjcBmu4dzpuWN4+IeHIUM6xe79gLxOKlwkscVweTOuZnmMfsZeg==",
+      "optional": true
+    },
+    "@lancedb/vectordb-darwin-x64": {
+      "version": "0.4.1",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.4.1.tgz",
+      "integrity": "sha512-sJtF2Cv6T9RhUpdeHNkryiJwPuW9QPQ3aMs5fID1hMCJA2U3BX27t/WlkiPT2+kTLeUcwF1JvAOgsfvZkfvI8w==",
+      "optional": true
+    },
+    "@lancedb/vectordb-linux-arm64-gnu": {
+      "version": "0.4.1",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.4.1.tgz",
+      "integrity": "sha512-tNnziT0BRjPsznKI4GgWROFdCOsCGx0inFu0z+WV1UomwXKcMWGslpWBqKE8IUiCq14duPVx/ie7Wwcf51IeJQ==",
+      "optional": true
+    },
+    "@lancedb/vectordb-linux-x64-gnu": {
+      "version": "0.4.1",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.4.1.tgz",
+      "integrity": "sha512-PAcF2p1FUsC0AD+qkLfgE5+ZlQwlHe9eTP9dSsX43V/NGPDQ9+gBzaBTEDbvyHj1wl2Wft2NwOqB1HAFhilSDg==",
+      "optional": true
+    },
+    "@lancedb/vectordb-win32-x64-msvc": {
+      "version": "0.4.1",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.4.1.tgz",
+      "integrity": "sha512-8mvThCppI/AfSPby6Y3t6xpCfbo8IY6JH5exO8fDGTwBFHOqgwR4Izb2K7FgXxkwUYcN4EfGSsk/6B1GpwMudg==",
+      "optional": true
+    },
     "@neon-rs/cli": {
       "version": "0.0.160",
       "resolved": "https://registry.npmjs.org/@neon-rs/cli/-/cli-0.0.160.tgz",

node/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vectordb",
-  "version": "0.3.8",
+  "version": "0.4.1",
   "description": " Serverless, low-latency vector database for AI applications",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",
@@ -81,10 +81,10 @@
     }
   },
   "optionalDependencies": {
-    "@lancedb/vectordb-darwin-arm64": "0.3.8",
-    "@lancedb/vectordb-darwin-x64": "0.3.8",
-    "@lancedb/vectordb-linux-arm64-gnu": "0.3.8",
-    "@lancedb/vectordb-linux-x64-gnu": "0.3.8",
-    "@lancedb/vectordb-win32-x64-msvc": "0.3.8"
+    "@lancedb/vectordb-darwin-arm64": "0.4.1",
+    "@lancedb/vectordb-darwin-x64": "0.4.1",
+    "@lancedb/vectordb-linux-arm64-gnu": "0.4.1",
+    "@lancedb/vectordb-linux-x64-gnu": "0.4.1",
+    "@lancedb/vectordb-win32-x64-msvc": "0.4.1"
   }
 }

node/src/index.ts

@@ -21,9 +21,10 @@ import type { EmbeddingFunction } from './embedding/embedding_function'
 import { RemoteConnection } from './remote'
 import { Query } from './query'
 import { isEmbeddingFunction } from './embedding/embedding_function'
+import { type Literal, toSQL } from './util'
 
 // eslint-disable-next-line @typescript-eslint/no-var-requires
-const { databaseNew, databaseTableNames, databaseOpenTable, databaseDropTable, tableCreate, tableAdd, tableCreateVectorIndex, tableCountRows, tableDelete, tableCleanupOldVersions, tableCompactFiles, tableListIndices, tableIndexStats } = require('../native.js')
+const { databaseNew, databaseTableNames, databaseOpenTable, databaseDropTable, tableCreate, tableAdd, tableCreateScalarIndex, tableCreateVectorIndex, tableCountRows, tableDelete, tableUpdate, tableCleanupOldVersions, tableCompactFiles, tableListIndices, tableIndexStats } = require('../native.js')
 
 export { Query }
 export type { EmbeddingFunction }
@@ -222,6 +223,56 @@ export interface Table<T = number[]> {
    */
   createIndex: (indexParams: VectorIndexParams) => Promise<any>
 
+  /**
+   * Create a scalar index on this Table for the given column
+   *
+   * @param column The column to index
+   * @param replace If false, fail if an index already exists on the column
+   *
+   * Scalar indices, like vector indices, can be used to speed up scans.  A scalar
+   * index can speed up scans that contain filter expressions on the indexed column.
+   * For example, the following scan will be faster if the column `my_col` has
+   * a scalar index:
+   *
+   * ```ts
+   * const con = await lancedb.connect('./.lancedb');
+   * const table = await con.openTable('images');
+   * const results = await table.where('my_col = 7').execute();
+   * ```
+   *
+   * Scalar indices can also speed up scans containing a vector search and a
+   * prefilter:
+   *
+   * ```ts
+   * const con = await lancedb.connect('././lancedb');
+   * const table = await con.openTable('images');
+   * const results = await table.search([1.0, 2.0]).where('my_col != 7').prefilter(true);
+   * ```
+   *
+   * Scalar indices can only speed up scans for basic filters using
+   * equality, comparison, range (e.g. `my_col BETWEEN 0 AND 100`), and set
+   * membership (e.g. `my_col IN (0, 1, 2)`)
+   *
+   * Scalar indices can be used if the filter contains multiple indexed columns and
+   * the filter criteria are AND'd or OR'd together
+   * (e.g. `my_col < 0 AND other_col> 100`)
+   *
+   * Scalar indices may be used if the filter contains non-indexed columns but,
+   * depending on the structure of the filter, they may not be usable.  For example,
+   * if the column `not_indexed` does not have a scalar index then the filter
+   * `my_col = 0 OR not_indexed = 1` will not be able to use any scalar index on
+   * `my_col`.
+   *
+   * @examples
+   *
+   * ```ts
+   * const con = await lancedb.connect('././lancedb')
+   * const table = await con.openTable('images')
+   * await table.createScalarIndex('my_col')
+   * ```
+   */
+  createScalarIndex: (column: string, replace: boolean) => Promise<void>
+
   /**
    * Returns the number of rows in this table.
    */
@@ -261,6 +312,39 @@ export interface Table<T = number[]> {
    */
   delete: (filter: string) => Promise<void>
 
+  /**
+   * Update rows in this table.
+   *
+   * This can be used to update a single row, many rows, all rows, or
+   * sometimes no rows (if your predicate matches nothing).
+   *
+   * @param args see {@link UpdateArgs} and {@link UpdateSqlArgs} for more details
+   *
+   * @examples
+   *
+   * ```ts
+   * const con = await lancedb.connect("./.lancedb")
+   * const data = [
+   *    {id: 1, vector: [3, 3], name: 'Ye'},
+   *    {id: 2, vector: [4, 4], name: 'Mike'},
+   * ];
+   * const tbl = await con.createTable("my_table", data)
+   *
+   * await tbl.update({
+   *   where: "id = 2",
+   *   values: { vector: [2, 2], name: "Michael" },
+   * })
+   *
+   * let results = await tbl.search([1, 1]).execute();
+   * // Returns [
+   * //   {id: 2, vector: [2, 2], name: 'Michael'}
+   * //   {id: 1, vector: [3, 3], name: 'Ye'}
+   * // ]
+   * ```
+   *
+   */
+  update: (args: UpdateArgs | UpdateSqlArgs) => Promise<void>
+
   /**
    * List the indicies on this table.
    */
@@ -272,6 +356,34 @@ export interface Table<T = number[]> {
   indexStats: (indexUuid: string) => Promise<IndexStats>
 }
 
+export interface UpdateArgs {
+  /**
+   * A filter in the same format used by a sql WHERE clause. The filter may be empty,
+   * in which case all rows will be updated.
+   */
+  where?: string
+
+  /**
+   * A key-value map of updates. The keys are the column names, and the values are the
+   * new values to set
+   */
+  values: Record<string, Literal>
+}
+
+export interface UpdateSqlArgs {
+  /**
+   * A filter in the same format used by a sql WHERE clause. The filter may be empty,
+   * in which case all rows will be updated.
+   */
+  where?: string
+
+  /**
+   * A key-value map of updates. The keys are the column names, and the values are the
+   * new values to set as SQL expressions.
+   */
+  valuesSql: Record<string, string>
+}
+
 export interface VectorIndex {
   columns: string[]
   name: string
@@ -426,6 +538,16 @@ export class LocalTable<T = number[]> implements Table<T> {
     return new Query(query, this._tbl, this._embeddings)
   }
 
+  /**
+   * Creates a filter query to find all rows matching the specified criteria
+   * @param value The filter criteria (like SQL where clause syntax)
+   */
+  filter (value: string): Query<T> {
+    return new Query(undefined, this._tbl, this._embeddings).filter(value)
+  }
+
+  where = this.filter
+
   /**
    * Insert records into this Table.
    *
@@ -465,6 +587,10 @@ export class LocalTable<T = number[]> implements Table<T> {
     return tableCreateVectorIndex.call(this._tbl, indexParams).then((newTable: any) => { this._tbl = newTable })
   }
 
+  async createScalarIndex (column: string, replace: boolean): Promise<void> {
+    return tableCreateScalarIndex.call(this._tbl, column, replace)
+  }
+
   /**
    * Returns the number of rows in this table.
    */
@@ -481,6 +607,31 @@ export class LocalTable<T = number[]> implements Table<T> {
     return tableDelete.call(this._tbl, filter).then((newTable: any) => { this._tbl = newTable })
   }
 
+  /**
+   * Update rows in this table.
+   *
+   * @param args see {@link UpdateArgs} and {@link UpdateSqlArgs} for more details
+   *
+   * @returns
+   */
+  async update (args: UpdateArgs | UpdateSqlArgs): Promise<void> {
+    let filter: string | null
+    let updates: Record<string, string>
+
+    if ('valuesSql' in args) {
+      filter = args.where ?? null
+      updates = args.valuesSql
+    } else {
+      filter = args.where ?? null
+      updates = {}
+      for (const [key, value] of Object.entries(args.values)) {
+        updates[key] = toSQL(value)
+      }
+    }
+
+    return tableUpdate.call(this._tbl, filter, updates).then((newTable: any) => { this._tbl = newTable })
+  }
+
   /**
    * Clean up old versions of the table, freeing disk space.
    *
@@ -647,6 +798,11 @@ export interface IvfPQIndexConfig {
    */
   replace?: boolean
 
+  /**
+   * Cache size of the index
+   */
+  index_cache_size?: number
+
   type: 'ivf_pq'
 }
 

node/src/query.ts

@@ -23,27 +23,29 @@ const { tableSearch } = require('../native.js')
  * A builder for nearest neighbor queries for LanceDB.
  */
 export class Query<T = number[]> {
-  private readonly _query: T
+  private readonly _query?: T
   private readonly _tbl?: any
   private _queryVector?: number[]
-  private _limit: number
+  private _limit?: number
   private _refineFactor?: number
   private _nprobes: number
   private _select?: string[]
   private _filter?: string
   private _metricType?: MetricType
+  private _prefilter: boolean
   protected readonly _embeddings?: EmbeddingFunction<T>
 
-  constructor (query: T, tbl?: any, embeddings?: EmbeddingFunction<T>) {
+  constructor (query?: T, tbl?: any, embeddings?: EmbeddingFunction<T>) {
     this._tbl = tbl
     this._query = query
-    this._limit = 10
+    this._limit = undefined
     this._nprobes = 20
     this._refineFactor = undefined
     this._select = undefined
     this._filter = undefined
     this._metricType = undefined
     this._embeddings = embeddings
+    this._prefilter = false
   }
 
   /***
@@ -102,14 +104,21 @@ export class Query<T = number[]> {
     return this
   }
 
+  prefilter (value: boolean): Query<T> {
+    this._prefilter = value
+    return this
+  }
+
   /**
      * Execute the query and return the results as an Array of Objects
      */
   async execute<T = Record<string, unknown>> (): Promise<T[]> {
-    if (this._embeddings !== undefined) {
-      this._queryVector = (await this._embeddings.embed([this._query]))[0]
-    } else {
-      this._queryVector = this._query as number[]
+    if (this._query !== undefined) {
+      if (this._embeddings !== undefined) {
+        this._queryVector = (await this._embeddings.embed([this._query]))[0]
+      } else {
+        this._queryVector = this._query as number[]
+      }
     }
 
     const isElectron = this.isElectron()

node/src/remote/client.ts

@@ -38,6 +38,7 @@ export class HttpLancedbClient {
     vector: number[],
     k: number,
     nprobes: number,
+    prefilter: boolean,
     refineFactor?: number,
     columns?: string[],
     filter?: string
@@ -50,7 +51,8 @@ export class HttpLancedbClient {
                 nprobes,
                 refineFactor,
                 columns,
-                filter
+                filter,
+                prefilter
               },
               {
                 headers: {

node/src/remote/index.ts

@@ -16,7 +16,8 @@ import {
   type EmbeddingFunction, type Table, type VectorIndexParams, type Connection,
   type ConnectionOptions, type CreateTableOptions, type VectorIndex,
   type WriteOptions,
-  type IndexStats
+  type IndexStats,
+  type UpdateArgs, type UpdateSqlArgs
 } from '../index'
 import { Query } from '../query'
 
@@ -24,6 +25,7 @@ import { Vector, Table as ArrowTable } from 'apache-arrow'
 import { HttpLancedbClient } from './client'
 import { isEmbeddingFunction } from '../embedding/embedding_function'
 import { createEmptyTable, fromRecordsToStreamBuffer, fromTableToStreamBuffer } from '../arrow'
+import { toSQL } from '../util'
 
 /**
  * Remote connection.
@@ -55,8 +57,8 @@ export class RemoteConnection implements Connection {
     return 'db://' + this._client.uri
   }
 
-  async tableNames (): Promise<string[]> {
-    const response = await this._client.get('/v1/table/')
+  async tableNames (pageToken: string = '', limit: number = 10): Promise<string[]> {
+    const response = await this._client.get('/v1/table/', { limit, page_token: pageToken })
     return response.data.tables
   }
 
@@ -154,6 +156,7 @@ export class RemoteQuery<T = number[]> extends Query<T> {
       queryVector,
       (this as any)._limit,
       (this as any)._nprobes,
+      (this as any)._prefilter,
       (this as any)._refineFactor,
       (this as any)._select,
       (this as any)._filter
@@ -192,6 +195,17 @@ export class RemoteTable<T = number[]> implements Table<T> {
     return this._name
   }
 
+  get schema (): Promise<any> {
+    return this._client.post(`/v1/table/${this._name}/describe/`).then(res => {
+      if (res.status !== 200) {
+        throw new Error(`Server Error, status: ${res.status}, ` +
+        // eslint-disable-next-line @typescript-eslint/restrict-template-expressions
+          `message: ${res.statusText}: ${res.data}`)
+      }
+      return res.data?.schema
+    })
+  }
+
   search (query: T): Query<T> {
     return new RemoteQuery(query, this._client, this._name)//, this._embeddings_new)
   }
@@ -232,7 +246,44 @@ export class RemoteTable<T = number[]> implements Table<T> {
     return data.length
   }
 
-  async createIndex (indexParams: VectorIndexParams): Promise<any> {
+  async createIndex (indexParams: VectorIndexParams): Promise<void> {
+    const unsupportedParams = [
+      'index_name',
+      'num_partitions',
+      'max_iters',
+      'use_opq',
+      'num_sub_vectors',
+      'num_bits',
+      'max_opq_iters',
+      'replace'
+    ]
+    for (const param of unsupportedParams) {
+      // eslint-disable-next-line @typescript-eslint/strict-boolean-expressions
+      if (indexParams[param as keyof VectorIndexParams]) {
+        throw new Error(`${param} is not supported for remote connections`)
+      }
+    }
+
+    const column = indexParams.column ?? 'vector'
+    const indexType = 'vector' // only vector index is supported for remote connections
+    const metricType = indexParams.metric_type ?? 'L2'
+    const indexCacheSize = indexParams.index_cache_size ?? null
+
+    const data = {
+      column,
+      index_type: indexType,
+      metric_type: metricType,
+      index_cache_size: indexCacheSize
+    }
+    const res = await this._client.post(`/v1/table/${this._name}/create_index/`, data)
+    if (res.status !== 200) {
+      throw new Error(`Server Error, status: ${res.status}, ` +
+      // eslint-disable-next-line @typescript-eslint/restrict-template-expressions
+        `message: ${res.statusText}: ${res.data}`)
+    }
+  }
+
+  async createScalarIndex (column: string, replace: boolean): Promise<void> {
     throw new Error('Not implemented')
   }
 
@@ -245,6 +296,26 @@ export class RemoteTable<T = number[]> implements Table<T> {
     await this._client.post(`/v1/table/${this._name}/delete/`, { predicate: filter })
   }
 
+  async update (args: UpdateArgs | UpdateSqlArgs): Promise<void> {
+    let filter: string | null
+    let updates: Record<string, string>
+
+    if ('valuesSql' in args) {
+      filter = args.where ?? null
+      updates = args.valuesSql
+    } else {
+      filter = args.where ?? null
+      updates = {}
+      for (const [key, value] of Object.entries(args.values)) {
+        updates[key] = toSQL(value)
+      }
+    }
+    await this._client.post(`/v1/table/${this._name}/update/`, {
+      predicate: filter,
+      updates: Object.entries(updates).map(([key, value]) => [key, value])
+    })
+  }
+
   async listIndices (): Promise<VectorIndex[]> {
     const results = await this._client.post(`/v1/table/${this._name}/index/list/`)
     return results.data.indexes?.map((index: any) => ({

node/src/test/test.ts

@@ -78,12 +78,31 @@ describe('LanceDB client', function () {
     })
 
     it('limits # of results', async function () {
-      const uri = await createTestDB()
+      const uri = await createTestDB(2, 100)
       const con = await lancedb.connect(uri)
       const table = await con.openTable('vectors')
-      const results = await table.search([0.1, 0.3]).limit(1).execute()
+      let results = await table.search([0.1, 0.3]).limit(1).execute()
       assert.equal(results.length, 1)
       assert.equal(results[0].id, 1)
+
+      // there is a default limit if unspecified
+      results = await table.search([0.1, 0.3]).execute()
+      assert.equal(results.length, 10)
+    })
+
+    it('uses a filter / where clause without vector search', async function () {
+      // eslint-disable-next-line @typescript-eslint/explicit-function-return-type
+      const assertResults = (results: Array<Record<string, unknown>>) => {
+        assert.equal(results.length, 50)
+      }
+
+      const uri = await createTestDB(2, 100)
+      const con = await lancedb.connect(uri)
+      const table = (await con.openTable('vectors')) as LocalTable
+      let results = await table.filter('id % 2 = 0').execute()
+      assertResults(results)
+      results = await table.where('id % 2 = 0').execute()
+      assertResults(results)
     })
 
     it('uses a filter / where clause', async function () {
@@ -102,6 +121,31 @@ describe('LanceDB client', function () {
       assertResults(results)
     })
 
+    it('should correctly process prefilter/postfilter', async function () {
+      const uri = await createTestDB(16, 300)
+      const con = await lancedb.connect(uri)
+      const table = await con.openTable('vectors')
+      await table.createIndex({ type: 'ivf_pq', column: 'vector', num_partitions: 2, max_iters: 2, num_sub_vectors: 2 })
+      // post filter should return less than the limit
+      let results = await table.search(new Array(16).fill(0.1)).limit(10).filter('id >= 10').prefilter(false).execute()
+      assert.isTrue(results.length < 10)
+
+      // pre filter should return exactly the limit
+      results = await table.search(new Array(16).fill(0.1)).limit(10).filter('id >= 10').prefilter(true).execute()
+      assert.isTrue(results.length === 10)
+    })
+
+    it('should allow creation and use of scalar indices', async function () {
+      const uri = await createTestDB(16, 300)
+      const con = await lancedb.connect(uri)
+      const table = await con.openTable('vectors')
+      await table.createScalarIndex('id', true)
+
+      // Prefiltering should still work the same
+      const results = await table.search(new Array(16).fill(0.1)).limit(10).filter('id >= 10').prefilter(true).execute()
+      assert.isTrue(results.length === 10)
+    })
+
     it('select only a subset of columns', async function () {
       const uri = await createTestDB()
       const con = await lancedb.connect(uri)
@@ -246,6 +290,46 @@ describe('LanceDB client', function () {
       assert.equal(await table.countRows(), 2)
     })
 
+    it('can update records in the table', async function () {
+      const uri = await createTestDB()
+      const con = await lancedb.connect(uri)
+
+      const table = await con.openTable('vectors')
+      assert.equal(await table.countRows(), 2)
+
+      await table.update({ where: 'price = 10', valuesSql: { price: '100' } })
+      const results = await table.search([0.1, 0.2]).execute()
+      assert.equal(results[0].price, 100)
+      assert.equal(results[1].price, 11)
+    })
+
+    it('can update the records using a literal value', async function () {
+      const uri = await createTestDB()
+      const con = await lancedb.connect(uri)
+
+      const table = await con.openTable('vectors')
+      assert.equal(await table.countRows(), 2)
+
+      await table.update({ where: 'price = 10', values: { price: 100 } })
+      const results = await table.search([0.1, 0.2]).execute()
+      assert.equal(results[0].price, 100)
+      assert.equal(results[1].price, 11)
+    })
+
+    it('can update every record in the table', async function () {
+      const uri = await createTestDB()
+      const con = await lancedb.connect(uri)
+
+      const table = await con.openTable('vectors')
+      assert.equal(await table.countRows(), 2)
+
+      await table.update({ valuesSql: { price: '100' } })
+      const results = await table.search([0.1, 0.2]).execute()
+
+      assert.equal(results[0].price, 100)
+      assert.equal(results[1].price, 100)
+    })
+
     it('can delete records from a table', async function () {
       const uri = await createTestDB()
       const con = await lancedb.connect(uri)
@@ -528,7 +612,7 @@ describe('Compact and cleanup', function () {
 
     // should have no effect, but this validates the arguments are parsed.
     await table.compactFiles({
-      targetRowsPerFragment: 1024 * 10,
+      targetRowsPerFragment: 102410,
       maxRowsPerGroup: 1024,
       materializeDeletions: true,
       materializeDeletionsThreshold: 0.5,

node/src/test/util.ts

@@ -0,0 +1,45 @@
+// Copyright 2023 LanceDB 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.
+
+import { toSQL } from '../util'
+import * as chai from 'chai'
+
+const expect = chai.expect
+
+describe('toSQL', function () {
+  it('should turn string to SQL expression', function () {
+    expect(toSQL('foo')).to.equal("'foo'")
+  })
+
+  it('should turn number to SQL expression', function () {
+    expect(toSQL(123)).to.equal('123')
+  })
+
+  it('should turn boolean to SQL expression', function () {
+    expect(toSQL(true)).to.equal('TRUE')
+  })
+
+  it('should turn null to SQL expression', function () {
+    expect(toSQL(null)).to.equal('NULL')
+  })
+
+  it('should turn Date to SQL expression', function () {
+    const date = new Date('05 October 2011 14:48 UTC')
+    expect(toSQL(date)).to.equal("'2011-10-05T14:48:00.000Z'")
+  })
+
+  it('should turn array to SQL expression', function () {
+    expect(toSQL(['foo', 'bar', true, 1])).to.equal("['foo', 'bar', TRUE, 1]")
+  })
+})

node/src/util.ts

@@ -0,0 +1,44 @@
+// Copyright 2023 LanceDB 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.
+
+export type Literal = string | number | boolean | null | Date | Literal[]
+
+export function toSQL (value: Literal): string {
+  if (typeof value === 'string') {
+    return `'${value}'`
+  }
+
+  if (typeof value === 'number') {
+    return value.toString()
+  }
+
+  if (typeof value === 'boolean') {
+    return value ? 'TRUE' : 'FALSE'
+  }
+
+  if (value === null) {
+    return 'NULL'
+  }
+
+  if (value instanceof Date) {
+    return `'${value.toISOString()}'`
+  }
+
+  if (Array.isArray(value)) {
+    return `[${value.map(toSQL).join(', ')}]`
+  }
+
+  // eslint-disable-next-line @typescript-eslint/restrict-template-expressions
+  throw new Error(`Unsupported value type: ${typeof value} value: (${value})`)
+}

python/.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.3.4
+current_version = 0.4.3
 commit = True
 message = [python] Bump version: {current_version} → {new_version}
 tag = True

python/lancedb/__init__.py

@@ -27,7 +27,7 @@ def connect(
     uri: URI,
     *,
     api_key: Optional[str] = None,
-    region: str = "us-west-2",
+    region: str = "us-east-1",
     host_override: Optional[str] = None,
 ) -> DBConnection:
     """Connect to a LanceDB database.
@@ -39,7 +39,7 @@ def connect(
     api_key: str, optional
         If presented, connect to LanceDB cloud.
         Otherwise, connect to a database on file system or cloud storage.
-    region: str, default "us-west-2"
+    region: str, default "us-east-1"
         The region to use for LanceDB Cloud.
     host_override: str, optional
         The override url for LanceDB Cloud.

python/lancedb/db.py

@@ -23,7 +23,7 @@ from overrides import EnforceOverrides, override
 from pyarrow import fs
 
 from .table import LanceTable, Table
-from .util import fs_from_uri, get_uri_location, get_uri_scheme
+from .util import fs_from_uri, get_uri_location, get_uri_scheme, join_uri
 
 if TYPE_CHECKING:
     from .common import DATA, URI
@@ -288,14 +288,13 @@ class LanceDBConnection(DBConnection):
             A list of table names.
         """
         try:
-            filesystem, path = fs_from_uri(self.uri)
+            filesystem = fs_from_uri(self.uri)[0]
         except pa.ArrowInvalid:
             raise NotImplementedError("Unsupported scheme: " + self.uri)
 
         try:
-            paths = filesystem.get_file_info(
-                fs.FileSelector(get_uri_location(self.uri))
-            )
+            loc = get_uri_location(self.uri)
+            paths = filesystem.get_file_info(fs.FileSelector(loc))
         except FileNotFoundError:
             # It is ok if the file does not exist since it will be created
             paths = []
@@ -373,7 +372,7 @@ class LanceDBConnection(DBConnection):
         """
         try:
             filesystem, path = fs_from_uri(self.uri)
-            table_path = os.path.join(path, name + ".lance")
+            table_path = join_uri(path, name + ".lance")
             filesystem.delete_dir(table_path)
         except FileNotFoundError:
             if not ignore_missing:

python/lancedb/embeddings/openai.py

@@ -10,6 +10,7 @@
 #  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.
+from functools import cached_property
 from typing import List, Union
 
 import numpy as np
@@ -44,6 +45,10 @@ class OpenAIEmbeddings(TextEmbeddingFunction):
             The texts to embed
         """
         # TODO retry, rate limit, token limit
+        rs = self._openai_client.embeddings.create(input=texts, model=self.name)
+        return [v.embedding for v in rs.data]
+
+    @cached_property
+    def _openai_client(self):
         openai = self.safe_import("openai")
-        rs = openai.Embedding.create(input=texts, model=self.name)["data"]
-        return [v["embedding"] for v in rs]
+        return openai.OpenAI()

python/lancedb/embeddings/utils.py

@@ -249,7 +249,7 @@ def retry_with_exponential_backoff(
 
                 if num_retries > max_retries:
                     raise Exception(
-                        f"Maximum number of retries ({max_retries}) exceeded."
+                        f"Maximum number of retries ({max_retries}) exceeded.", e
                     )
 
                 delay *= exponential_base * (1 + jitter * random.random())

python/lancedb/fts.py

@@ -75,8 +75,14 @@ def populate_index(index: tantivy.Index, table: LanceTable, fields: List[str]) -
         The number of rows indexed
     """
     # first check the fields exist and are string or large string type
+    nested = []
     for name in fields:
-        f = table.schema.field(name)  # raises KeyError if not found
+        try:
+            f = table.schema.field(name)  # raises KeyError if not found
+        except KeyError:
+            f = resolve_path(table.schema, name)
+            nested.append(name)
+
         if not pa.types.is_string(f.type) and not pa.types.is_large_string(f.type):
             raise TypeError(f"Field {name} is not a string type")
 
@@ -85,7 +91,16 @@ def populate_index(index: tantivy.Index, table: LanceTable, fields: List[str]) -
     # write data into index
     dataset = table.to_lance()
     row_id = 0
+
+    max_nested_level = 0
+    if len(nested) > 0:
+        max_nested_level = max([len(name.split(".")) for name in nested])
+
     for b in dataset.to_batches(columns=fields):
+        if max_nested_level > 0:
+            b = pa.Table.from_batches([b])
+            for _ in range(max_nested_level - 1):
+                b = b.flatten()
         for i in range(b.num_rows):
             doc = tantivy.Document()
             doc.add_integer("doc_id", row_id)
@@ -98,6 +113,30 @@ def populate_index(index: tantivy.Index, table: LanceTable, fields: List[str]) -
     return row_id
 
 
+def resolve_path(schema, field_name: str) -> pa.Field:
+    """
+    Resolve a nested field path to a list of field names
+
+    Parameters
+    ----------
+    field_name : str
+        The field name to resolve
+
+    Returns
+    -------
+    List[str]
+        The resolved path
+    """
+    path = field_name.split(".")
+    field = schema.field(path.pop(0))
+    for segment in path:
+        if pa.types.is_struct(field.type):
+            field = field.type.field(segment)
+        else:
+            raise KeyError(f"field {field_name} not found in schema {schema}")
+    return field
+
+
 def search_index(
     index: tantivy.Index, query: str, limit: int = 10
 ) -> Tuple[Tuple[int], Tuple[float]]:

python/lancedb/pydantic.py

@@ -26,6 +26,7 @@ import numpy as np
 import pyarrow as pa
 import pydantic
 import semver
+from pydantic.fields import FieldInfo
 
 from .embeddings import EmbeddingFunctionRegistry
 
@@ -142,8 +143,8 @@ def Vector(
     return FixedSizeList
 
 
-def _py_type_to_arrow_type(py_type: Type[Any]) -> pa.DataType:
-    """Convert Python Type to Arrow DataType.
+def _py_type_to_arrow_type(py_type: Type[Any], field: FieldInfo) -> pa.DataType:
+    """Convert a field with native Python type to Arrow data type.
 
     Raises
     ------
@@ -163,9 +164,13 @@ def _py_type_to_arrow_type(py_type: Type[Any]) -> pa.DataType:
     elif py_type == date:
         return pa.date32()
     elif py_type == datetime:
-        return pa.timestamp("us")
+        tz = get_extras(field, "tz")
+        return pa.timestamp("us", tz=tz)
+    elif getattr(py_type, "__origin__", None) in (list, tuple):
+        child = py_type.__args__[0]
+        return pa.list_(_py_type_to_arrow_type(child, field))
     raise TypeError(
-        f"Converting Pydantic type to Arrow Type: unsupported type {py_type}"
+        f"Converting Pydantic type to Arrow Type: unsupported type {py_type}."
     )
 
 
@@ -194,10 +199,10 @@ def _pydantic_to_arrow_type(field: pydantic.fields.FieldInfo) -> pa.DataType:
         args = field.annotation.__args__
         if origin == list:
             child = args[0]
-            return pa.list_(_py_type_to_arrow_type(child))
+            return pa.list_(_py_type_to_arrow_type(child, field))
         elif origin == Union:
             if len(args) == 2 and args[1] == type(None):
-                return _py_type_to_arrow_type(args[0])
+                return _py_type_to_arrow_type(args[0], field)
     elif inspect.isclass(field.annotation):
         if issubclass(field.annotation, pydantic.BaseModel):
             # Struct
@@ -205,7 +210,7 @@ def _pydantic_to_arrow_type(field: pydantic.fields.FieldInfo) -> pa.DataType:
             return pa.struct(fields)
         elif issubclass(field.annotation, FixedSizeListMixin):
             return pa.list_(field.annotation.value_arrow_type(), field.annotation.dim())
-    return _py_type_to_arrow_type(field.annotation)
+    return _py_type_to_arrow_type(field.annotation, field)
 
 
 def is_nullable(field: pydantic.fields.FieldInfo) -> bool:
@@ -348,3 +353,20 @@ def get_extras(field_info: pydantic.fields.FieldInfo, key: str) -> Any:
     if PYDANTIC_VERSION.major >= 2:
         return (field_info.json_schema_extra or {}).get(key)
     return (field_info.field_info.extra or {}).get("json_schema_extra", {}).get(key)
+
+
+if PYDANTIC_VERSION.major < 2:
+
+    def model_to_dict(model: pydantic.BaseModel) -> Dict[str, Any]:
+        """
+        Convert a Pydantic model to a dictionary.
+        """
+        return model.dict()
+
+else:
+
+    def model_to_dict(model: pydantic.BaseModel) -> Dict[str, Any]:
+        """
+        Convert a Pydantic model to a dictionary.
+        """
+        return model.model_dump()

python/lancedb/query.py

@@ -70,7 +70,7 @@ class Query(pydantic.BaseModel):
     vector_column: str = VECTOR_COLUMN_NAME
 
     # vector to search for
-    vector: List[float]
+    vector: Union[List[float], List[List[float]]]
 
     # sql filter to refine the query with
     filter: Optional[str] = None
@@ -185,14 +185,40 @@ class LanceQueryBuilder(ABC):
         """
         return self.to_pandas()
 
-    def to_pandas(self) -> "pd.DataFrame":
+    def to_pandas(self, flatten: Optional[Union[int, bool]] = None) -> "pd.DataFrame":
         """
         Execute the query and return the results as a pandas DataFrame.
         In addition to the selected columns, LanceDB also returns a vector
         and also the "_distance" column which is the distance between the query
         vector and the returned vector.
+
+        Parameters
+        ----------
+        flatten: Optional[Union[int, bool]]
+            If flatten is True, flatten all nested columns.
+            If flatten is an integer, flatten the nested columns up to the
+            specified depth.
+            If unspecified, do not flatten the nested columns.
         """
-        return self.to_arrow().to_pandas()
+        tbl = self.to_arrow()
+        if flatten is True:
+            while True:
+                tbl = tbl.flatten()
+                has_struct = False
+                # loop through all columns to check if there is any struct column
+                if any(pa.types.is_struct(col.type) for col in tbl.schema):
+                    continue
+                else:
+                    break
+        elif isinstance(flatten, int):
+            if flatten <= 0:
+                raise ValueError(
+                    "Please specify a positive integer for flatten or the boolean value `True`"
+                )
+            while flatten > 0:
+                tbl = tbl.flatten()
+                flatten -= 1
+        return tbl.to_pandas()
 
     @abstractmethod
     def to_arrow(self) -> pa.Table:
@@ -395,6 +421,8 @@ class LanceVectorQueryBuilder(LanceQueryBuilder):
         vector and the returned vectors.
         """
         vector = self._query if isinstance(self._query, list) else self._query.tolist()
+        if isinstance(vector[0], np.ndarray):
+            vector = [v.tolist() for v in vector]
         query = Query(
             vector=vector,
             filter=self._where,
@@ -462,6 +490,27 @@ class LanceFtsQueryBuilder(LanceQueryBuilder):
         scores = pa.array(scores)
         output_tbl = self._table.to_lance().take(row_ids, columns=self._columns)
         output_tbl = output_tbl.append_column("score", scores)
+
+        if self._where is not None:
+            try:
+                # TODO would be great to have Substrait generate pyarrow compute expressions
+                # or conversely have pyarrow support SQL expressions using Substrait
+                import duckdb
+
+                output_tbl = (
+                    duckdb.sql(f"SELECT * FROM output_tbl")
+                    .filter(self._where)
+                    .to_arrow_table()
+                )
+            except ImportError:
+                import lance
+                import tempfile
+
+                # TODO Use "memory://" instead once that's supported
+                with tempfile.TemporaryDirectory() as tmp:
+                    ds = lance.write_dataset(output_tbl, tmp)
+                    output_tbl = ds.to_table(filter=self._where)
+
         return output_tbl
 
 

python/lancedb/remote/__init__.py

@@ -18,6 +18,8 @@ import attrs
 import pyarrow as pa
 from pydantic import BaseModel
 
+from lancedb.common import VECTOR_COLUMN_NAME
+
 __all__ = ["LanceDBClient", "VectorQuery", "VectorQueryResult"]
 
 
@@ -43,6 +45,8 @@ class VectorQuery(BaseModel):
 
     refine_factor: Optional[int] = None
 
+    vector_column: str = VECTOR_COLUMN_NAME
+
 
 @attrs.define
 class VectorQueryResult:

python/lancedb/remote/db.py

@@ -28,6 +28,7 @@ from ..pydantic import LanceModel
 from ..table import Table, _sanitize_data
 from .arrow import to_ipc_binary
 from .client import ARROW_STREAM_CONTENT_TYPE, RestfulLanceDBClient
+from .errors import LanceDBClientError
 
 
 class RemoteDBConnection(DBConnection):
@@ -55,16 +56,20 @@ class RemoteDBConnection(DBConnection):
             self._loop = asyncio.get_event_loop()
 
     def __repr__(self) -> str:
-        return f"RemoveConnect(name={self.db_name})"
+        return f"RemoteConnect(name={self.db_name})"
 
     @override
-    def table_names(self, page_token: Optional[str] = None, limit=10) -> Iterable[str]:
+    def table_names(
+        self, page_token: Optional[str] = None, limit: int = 10
+    ) -> Iterable[str]:
         """List the names of all tables in the database.
 
         Parameters
         ----------
         page_token: str
             The last token to start the new page.
+        limit: int, default 10
+            The maximum number of tables to return for each page.
 
         Returns
         -------
@@ -101,11 +106,12 @@ class RemoteDBConnection(DBConnection):
             self._loop.run_until_complete(
                 self._client.post(f"/v1/table/{name}/describe/")
             )
-        except Exception:
-            logging.error(
-                "Table {name} does not exist."
-                "Please first call db.create_table({name}, data)"
-            )
+        except LanceDBClientError as err:
+            if str(err).startswith("Not found"):
+                logging.error(
+                    f"Table {name} does not exist. "
+                    f"Please first call db.create_table({name}, data)"
+                )
         return RemoteTable(self, name)
 
     @override
@@ -118,6 +124,97 @@ class RemoteDBConnection(DBConnection):
         fill_value: float = 0.0,
         embedding_functions: Optional[List[EmbeddingFunctionConfig]] = None,
     ) -> Table:
+        """Create a [Table][lancedb.table.Table] in the database.
+
+        Parameters
+        ----------
+        name: str
+            The name of the table.
+        data: The data to initialize the table, *optional*
+            User must provide at least one of `data` or `schema`.
+            Acceptable types are:
+
+            - dict or list-of-dict
+
+            - pandas.DataFrame
+
+            - pyarrow.Table or pyarrow.RecordBatch
+        schema: The schema of the table, *optional*
+            Acceptable types are:
+
+            - pyarrow.Schema
+
+            - [LanceModel][lancedb.pydantic.LanceModel]
+        on_bad_vectors: str, default "error"
+            What to do if any of the vectors are not the same size or contains NaNs.
+            One of "error", "drop", "fill".
+        fill_value: float
+            The value to use when filling vectors. Only used if on_bad_vectors="fill".
+
+        Returns
+        -------
+        LanceTable
+            A reference to the newly created table.
+
+        !!! note
+
+            The vector index won't be created by default.
+            To create the index, call the `create_index` method on the table.
+
+        Examples
+        --------
+
+        Can create with list of tuples or dictionaries:
+
+        >>> import lancedb
+        >>> db = lancedb.connect("db://...", api_key="...", region="...") # doctest: +SKIP
+        >>> data = [{"vector": [1.1, 1.2], "lat": 45.5, "long": -122.7},
+        ...         {"vector": [0.2, 1.8], "lat": 40.1, "long":  -74.1}]
+        >>> db.create_table("my_table", data) # doctest: +SKIP
+        LanceTable(my_table)
+
+        You can also pass a pandas DataFrame:
+
+        >>> import pandas as pd
+        >>> data = pd.DataFrame({
+        ...    "vector": [[1.1, 1.2], [0.2, 1.8]],
+        ...    "lat": [45.5, 40.1],
+        ...    "long": [-122.7, -74.1]
+        ... })
+        >>> db.create_table("table2", data) # doctest: +SKIP
+        LanceTable(table2)
+
+        >>> custom_schema = pa.schema([
+        ...   pa.field("vector", pa.list_(pa.float32(), 2)),
+        ...   pa.field("lat", pa.float32()),
+        ...   pa.field("long", pa.float32())
+        ... ])
+        >>> db.create_table("table3", data, schema = custom_schema) # doctest: +SKIP
+        LanceTable(table3)
+
+        It is also possible to create an table from `[Iterable[pa.RecordBatch]]`:
+
+        >>> import pyarrow as pa
+        >>> def make_batches():
+        ...     for i in range(5):
+        ...         yield pa.RecordBatch.from_arrays(
+        ...             [
+        ...                 pa.array([[3.1, 4.1], [5.9, 26.5]],
+        ...                     pa.list_(pa.float32(), 2)),
+        ...                 pa.array(["foo", "bar"]),
+        ...                 pa.array([10.0, 20.0]),
+        ...             ],
+        ...             ["vector", "item", "price"],
+        ...         )
+        >>> schema=pa.schema([
+        ...     pa.field("vector", pa.list_(pa.float32(), 2)),
+        ...     pa.field("item", pa.utf8()),
+        ...     pa.field("price", pa.float32()),
+        ... ])
+        >>> db.create_table("table4", make_batches(), schema=schema) # doctest: +SKIP
+        LanceTable(table4)
+
+        """
         if data is None and schema is None:
             raise ValueError("Either data or schema must be provided.")
         if embedding_functions is not None:

python/lancedb/remote/table.py

@@ -11,9 +11,10 @@
 #  See the License for the specific language governing permissions and
 #  limitations under the License.
 
+import asyncio
 import uuid
 from functools import cached_property
-from typing import Optional, Union
+from typing import Dict, Optional, Union
 
 import pyarrow as pa
 from lance import json_to_schema
@@ -22,6 +23,7 @@ from lancedb.common import DATA, VEC, VECTOR_COLUMN_NAME
 
 from ..query import LanceVectorQueryBuilder
 from ..table import Query, Table, _sanitize_data
+from ..util import value_to_sql
 from .arrow import to_ipc_binary
 from .client import ARROW_STREAM_CONTENT_TYPE
 from .db import RemoteDBConnection
@@ -37,7 +39,10 @@ class RemoteTable(Table):
 
     @cached_property
     def schema(self) -> pa.Schema:
-        """Return the schema of the table."""
+        """The [Arrow Schema](https://arrow.apache.org/docs/python/api/datatypes.html#)
+        of this Table
+
+        """
         resp = self._conn._loop.run_until_complete(
             self._conn._client.post(f"/v1/table/{self._name}/describe/")
         )
@@ -53,24 +58,23 @@ class RemoteTable(Table):
         return resp["version"]
 
     def to_arrow(self) -> pa.Table:
-        """Return the table as an Arrow table."""
+        """to_arrow() is not supported on the LanceDB cloud"""
         raise NotImplementedError("to_arrow() is not supported on the LanceDB cloud")
 
     def to_pandas(self):
-        """Return the table as a Pandas DataFrame.
-
-        Intercept `to_arrow()` for better error message.
-        """
+        """to_pandas() is not supported on the LanceDB cloud"""
         return NotImplementedError("to_pandas() is not supported on the LanceDB cloud")
 
+    def create_scalar_index(self, *args, **kwargs):
+        """Creates a scalar index"""
+        return NotImplementedError(
+            "create_scalar_index() is not supported on the LanceDB cloud"
+        )
+
     def create_index(
         self,
         metric="L2",
-        num_partitions=256,
-        num_sub_vectors=96,
         vector_column_name: str = VECTOR_COLUMN_NAME,
-        replace: bool = True,
-        accelerator: Optional[str] = None,
         index_cache_size: Optional[int] = None,
     ):
         """Create an index on the table.
@@ -81,39 +85,28 @@ class RemoteTable(Table):
         ----------
         metric : str
             The metric to use for the index. Default is "L2".
-        num_partitions : int
-            The number of partitions to use for the index. Default is 256.
-        num_sub_vectors : int
-            The number of sub-vectors to use for the index. Default is 96.
         vector_column_name : str
             The name of the vector column. Default is "vector".
-        replace : bool
-            Whether to replace the existing index. Default is True.
-        accelerator : str, optional
-            If set, use the given accelerator to create the index.
-            Default is None. Currently not supported.
-        index_cache_size : int, optional
-            The size of the index cache in number of entries. Default value is 256.
 
         Examples
         --------
-        import lancedb
-        import uuid
-        from lancedb.schema import vector
-        conn = lancedb.connect("db://...", api_key="...", region="...")
-        table_name = uuid.uuid4().hex
-        schema = pa.schema(
-            [
-                    pa.field("id", pa.uint32(), False),
-                    pa.field("vector", vector(128), False),
-                    pa.field("s", pa.string(), False),
-            ]
-        )
-        table = conn.create_table(
-            table_name,
-            schema=schema,
-        )
-        table.create_index()
+        >>> import lancedb
+        >>> import uuid
+        >>> from lancedb.schema import vector
+        >>> db = lancedb.connect("db://...", api_key="...", region="...") # doctest: +SKIP
+        >>> table_name = uuid.uuid4().hex
+        >>> schema = pa.schema(
+        ...     [
+        ...             pa.field("id", pa.uint32(), False),
+        ...            pa.field("vector", vector(128), False),
+        ...             pa.field("s", pa.string(), False),
+        ...     ]
+        ... )
+        >>> table = db.create_table( # doctest: +SKIP
+        ...     table_name, # doctest: +SKIP
+        ...     schema=schema, # doctest: +SKIP
+        ... )
+        >>> table.create_index("L2", "vector") # doctest: +SKIP
         """
         index_type = "vector"
 
@@ -135,6 +128,28 @@ class RemoteTable(Table):
         on_bad_vectors: str = "error",
         fill_value: float = 0.0,
     ) -> int:
+        """Add more data to the [Table](Table). It has the same API signature as the OSS version.
+
+        Parameters
+        ----------
+        data: DATA
+            The data to insert into the table. Acceptable types are:
+
+            - dict or list-of-dict
+
+            - pandas.DataFrame
+
+            - pyarrow.Table or pyarrow.RecordBatch
+        mode: str
+            The mode to use when writing the data. Valid values are
+            "append" and "overwrite".
+        on_bad_vectors: str, default "error"
+            What to do if any of the vectors are not the same size or contains NaNs.
+            One of "error", "drop", "fill".
+        fill_value: float, default 0.
+            The value to use when filling vectors. Only used if on_bad_vectors="fill".
+
+        """
         data = _sanitize_data(
             data,
             self.schema,
@@ -158,15 +173,197 @@ class RemoteTable(Table):
     def search(
         self, query: Union[VEC, str], vector_column_name: str = VECTOR_COLUMN_NAME
     ) -> LanceVectorQueryBuilder:
+        """Create a search query to find the nearest neighbors
+        of the given query vector. We currently support [vector search][search]
+
+        All query options are defined in [Query][lancedb.query.Query].
+
+        Examples
+        --------
+        >>> import lancedb
+        >>> db = lancedb.connect("db://...", api_key="...", region="...") # doctest: +SKIP
+        >>> data = [
+        ...    {"original_width": 100, "caption": "bar", "vector": [0.1, 2.3, 4.5]},
+        ...    {"original_width": 2000, "caption": "foo",  "vector": [0.5, 3.4, 1.3]},
+        ...    {"original_width": 3000, "caption": "test", "vector": [0.3, 6.2, 2.6]}
+        ... ]
+        >>> table = db.create_table("my_table", data) # doctest: +SKIP
+        >>> query = [0.4, 1.4, 2.4]
+        >>> (table.search(query, vector_column_name="vector") # doctest: +SKIP
+        ...     .where("original_width > 1000", prefilter=True) # doctest: +SKIP
+        ...     .select(["caption", "original_width"]) # doctest: +SKIP
+        ...     .limit(2) # doctest: +SKIP
+        ...     .to_pandas()) # doctest: +SKIP
+          caption  original_width           vector  _distance # doctest: +SKIP
+        0     foo            2000  [0.5, 3.4, 1.3]   5.220000 # doctest: +SKIP
+        1    test            3000  [0.3, 6.2, 2.6]  23.089996 # doctest: +SKIP
+
+        Parameters
+        ----------
+        query: list/np.ndarray/str/PIL.Image.Image, default None
+            The targetted vector to search for.
+
+            - *default None*.
+            Acceptable types are: list, np.ndarray, PIL.Image.Image
+
+            - If None then the select/where/limit clauses are applied to filter
+            the table
+        vector_column_name: str
+            The name of the vector column to search.
+            *default "vector"*
+
+        Returns
+        -------
+        LanceQueryBuilder
+            A query builder object representing the query.
+            Once executed, the query returns
+
+            - selected columns
+
+            - the vector
+
+            - and also the "_distance" column which is the distance between the query
+            vector and the returned vector.
+        """
         return LanceVectorQueryBuilder(self, query, vector_column_name)
 
     def _execute_query(self, query: Query) -> pa.Table:
-        result = self._conn._client.query(self._name, query)
-        return self._conn._loop.run_until_complete(result).to_arrow()
+        if (
+            query.vector is not None
+            and len(query.vector) > 0
+            and not isinstance(query.vector[0], float)
+        ):
+            futures = []
+            for v in query.vector:
+                v = list(v)
+                q = query.copy()
+                q.vector = v
+                futures.append(self._conn._client.query(self._name, q))
+            result = self._conn._loop.run_until_complete(asyncio.gather(*futures))
+            return pa.concat_tables(
+                [add_index(r.to_arrow(), i) for i, r in enumerate(result)]
+            )
+        else:
+            result = self._conn._client.query(self._name, query)
+            return self._conn._loop.run_until_complete(result).to_arrow()
 
     def delete(self, predicate: str):
-        """Delete rows from the table."""
+        """Delete rows from the table.
+
+        This can be used to delete a single row, many rows, all rows, or
+        sometimes no rows (if your predicate matches nothing).
+
+        Parameters
+        ----------
+        predicate: str
+            The SQL where clause to use when deleting rows.
+
+            - For example, 'x = 2' or 'x IN (1, 2, 3)'.
+
+            The filter must not be empty, or it will error.
+
+        Examples
+        --------
+        >>> import lancedb
+        >>> data = [
+        ...    {"x": 1, "vector": [1, 2]},
+        ...    {"x": 2, "vector": [3, 4]},
+        ...    {"x": 3, "vector": [5, 6]}
+        ... ]
+        >>> db = lancedb.connect("db://...", api_key="...", region="...") # doctest: +SKIP
+        >>> table = db.create_table("my_table", data) # doctest: +SKIP
+        >>> table.search([10,10]).to_pandas() # doctest: +SKIP
+           x      vector  _distance # doctest: +SKIP
+        0  3  [5.0, 6.0]       41.0 # doctest: +SKIP
+        1  2  [3.0, 4.0]       85.0 # doctest: +SKIP
+        2  1  [1.0, 2.0]      145.0 # doctest: +SKIP
+        >>> table.delete("x = 2") # doctest: +SKIP
+        >>> table.search([10,10]).to_pandas() # doctest: +SKIP
+           x      vector  _distance # doctest: +SKIP
+        0  3  [5.0, 6.0]       41.0 # doctest: +SKIP
+        1  1  [1.0, 2.0]      145.0 # doctest: +SKIP
+
+        If you have a list of values to delete, you can combine them into a
+        stringified list and use the `IN` operator:
+
+        >>> to_remove = [1, 3] # doctest: +SKIP
+        >>> to_remove = ", ".join([str(v) for v in to_remove]) # doctest: +SKIP
+        >>> table.delete(f"x IN ({to_remove})") # doctest: +SKIP
+        >>> table.search([10,10]).to_pandas() # doctest: +SKIP
+           x      vector  _distance # doctest: +SKIP
+        0  2  [3.0, 4.0]       85.0 # doctest: +SKIP
+        """
         payload = {"predicate": predicate}
         self._conn._loop.run_until_complete(
             self._conn._client.post(f"/v1/table/{self._name}/delete/", data=payload)
         )
+
+    def update(
+        self,
+        where: Optional[str] = None,
+        values: Optional[dict] = None,
+        *,
+        values_sql: Optional[Dict[str, str]] = None,
+    ):
+        """
+        This can be used to update zero to all rows depending on how many
+        rows match the where clause.
+
+        Parameters
+        ----------
+        where: str, optional
+            The SQL where clause to use when updating rows. For example, 'x = 2'
+            or 'x IN (1, 2, 3)'. The filter must not be empty, or it will error.
+        values: dict, optional
+            The values to update. The keys are the column names and the values
+            are the values to set.
+        values_sql: dict, optional
+            The values to update, expressed as SQL expression strings. These can
+            reference existing columns. For example, {"x": "x + 1"} will increment
+            the x column by 1.
+
+        Examples
+        --------
+        >>> import lancedb
+        >>> data = [
+        ...    {"x": 1, "vector": [1, 2]},
+        ...    {"x": 2, "vector": [3, 4]},
+        ...    {"x": 3, "vector": [5, 6]}
+        ... ]
+        >>> db = lancedb.connect("db://...", api_key="...", region="...") # doctest: +SKIP
+        >>> table = db.create_table("my_table", data) # doctest: +SKIP
+        >>> table.to_pandas() # doctest: +SKIP
+           x      vector # doctest: +SKIP
+        0  1  [1.0, 2.0] # doctest: +SKIP
+        1  2  [3.0, 4.0] # doctest: +SKIP
+        2  3  [5.0, 6.0] # doctest: +SKIP
+        >>> table.update(where="x = 2", values={"vector": [10, 10]}) # doctest: +SKIP
+        >>> table.to_pandas() # doctest: +SKIP
+           x        vector # doctest: +SKIP
+        0  1    [1.0, 2.0] # doctest: +SKIP
+        1  3    [5.0, 6.0] # doctest: +SKIP
+        2  2  [10.0, 10.0] # doctest: +SKIP
+
+        """
+        if values is not None and values_sql is not None:
+            raise ValueError("Only one of values or values_sql can be provided")
+        if values is None and values_sql is None:
+            raise ValueError("Either values or values_sql must be provided")
+
+        if values is not None:
+            updates = [[k, value_to_sql(v)] for k, v in values.items()]
+        else:
+            updates = [[k, v] for k, v in values_sql.items()]
+
+        payload = {"predicate": where, "updates": updates}
+        self._conn._loop.run_until_complete(
+            self._conn._client.post(f"/v1/table/{self._name}/update/", data=payload)
+        )
+
+
+def add_index(tbl: pa.Table, i: int) -> pa.Table:
+    return tbl.add_column(
+        0,
+        pa.field("query_index", pa.uint32()),
+        pa.array([i] * len(tbl), pa.uint32()),
+    )

python/lancedb/table.py

@@ -17,20 +17,21 @@ import inspect
 import os
 from abc import ABC, abstractmethod
 from functools import cached_property
-from typing import TYPE_CHECKING, Any, Iterable, List, Optional, Union
+from typing import TYPE_CHECKING, Any, Dict, Iterable, List, Optional, Union
 
 import lance
 import numpy as np
 import pyarrow as pa
 import pyarrow.compute as pc
+import pyarrow.fs as pa_fs
 from lance import LanceDataset
 from lance.vector import vec_to_table
 
 from .common import DATA, VEC, VECTOR_COLUMN_NAME
 from .embeddings import EmbeddingFunctionConfig, EmbeddingFunctionRegistry
-from .pydantic import LanceModel
+from .pydantic import LanceModel, model_to_dict
 from .query import LanceQueryBuilder, Query
-from .util import fs_from_uri, safe_import_pandas
+from .util import fs_from_uri, safe_import_pandas, value_to_sql, join_uri
 from .utils.events import register_event
 
 if TYPE_CHECKING:
@@ -53,8 +54,10 @@ def _sanitize_data(
         # 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 = [model_to_dict(d) for d in data]
+            data = pa.Table.from_pylist(data, schema=schema)
+        else:
+            data = pa.Table.from_pylist(data)
     elif isinstance(data, dict):
         data = vec_to_table(data)
     elif pd is not None and isinstance(data, pd.DataFrame):
@@ -218,6 +221,77 @@ class Table(ABC):
         """
         raise NotImplementedError
 
+    @abstractmethod
+    def create_scalar_index(
+        self,
+        column: str,
+        *,
+        replace: bool = True,
+    ):
+        """Create a scalar index on a column.
+
+        Scalar indices, like vector indices, can be used to speed up scans.  A scalar
+        index can speed up scans that contain filter expressions on the indexed column.
+        For example, the following scan will be faster if the column ``my_col`` has
+        a scalar index:
+
+        .. code-block:: python
+
+            import lancedb
+
+            db = lancedb.connect("/data/lance")
+            img_table = db.open_table("images")
+            my_df = img_table.search().where("my_col = 7", prefilter=True).to_pandas()
+
+        Scalar indices can also speed up scans containing a vector search and a
+        prefilter:
+
+        .. code-block::python
+
+            import lancedb
+
+            db = lancedb.connect("/data/lance")
+            img_table = db.open_table("images")
+            img_table.search([1, 2, 3, 4], vector_column_name="vector")
+                .where("my_col != 7", prefilter=True)
+                .to_pandas()
+
+        Scalar indices can only speed up scans for basic filters using
+        equality, comparison, range (e.g. ``my_col BETWEEN 0 AND 100``), and set
+        membership (e.g. `my_col IN (0, 1, 2)`)
+
+        Scalar indices can be used if the filter contains multiple indexed columns and
+        the filter criteria are AND'd or OR'd together
+        (e.g. ``my_col < 0 AND other_col> 100``)
+
+        Scalar indices may be used if the filter contains non-indexed columns but,
+        depending on the structure of the filter, they may not be usable.  For example,
+        if the column ``not_indexed`` does not have a scalar index then the filter
+        ``my_col = 0 OR not_indexed = 1`` will not be able to use any scalar index on
+        ``my_col``.
+
+        **Experimental API**
+
+        Parameters
+        ----------
+        column : str
+            The column to be indexed.  Must be a boolean, integer, float,
+            or string column.
+        replace : bool, default True
+            Replace the existing index if it exists.
+
+        Examples
+        --------
+
+        .. code-block:: python
+
+            import lance
+
+            dataset = lance.dataset("/tmp/images.lance")
+            dataset.create_scalar_index("category")
+        """
+        raise NotImplementedError
+
     @abstractmethod
     def add(
         self,
@@ -381,6 +455,62 @@ class Table(ABC):
         """
         raise NotImplementedError
 
+    @abstractmethod
+    def update(
+        self,
+        where: Optional[str] = None,
+        values: Optional[dict] = None,
+        *,
+        values_sql: Optional[Dict[str, str]] = None,
+    ):
+        """
+        This can be used to update zero to all rows depending on how many
+        rows match the where clause. If no where clause is provided, then
+        all rows will be updated.
+
+        Either `values` or `values_sql` must be provided. You cannot provide
+        both.
+
+        Parameters
+        ----------
+        where: str, optional
+            The SQL where clause to use when updating rows. For example, 'x = 2'
+            or 'x IN (1, 2, 3)'. The filter must not be empty, or it will error.
+        values: dict, optional
+            The values to update. The keys are the column names and the values
+            are the values to set.
+        values_sql: dict, optional
+            The values to update, expressed as SQL expression strings. These can
+            reference existing columns. For example, {"x": "x + 1"} will increment
+            the x column by 1.
+
+        Examples
+        --------
+        >>> import lancedb
+        >>> import pandas as pd
+        >>> data = pd.DataFrame({"x": [1, 2, 3], "vector": [[1, 2], [3, 4], [5, 6]]})
+        >>> db = lancedb.connect("./.lancedb")
+        >>> table = db.create_table("my_table", data)
+        >>> table.to_pandas()
+           x      vector
+        0  1  [1.0, 2.0]
+        1  2  [3.0, 4.0]
+        2  3  [5.0, 6.0]
+        >>> table.update(where="x = 2", values={"vector": [10, 10]})
+        >>> table.to_pandas()
+           x        vector
+        0  1    [1.0, 2.0]
+        1  3    [5.0, 6.0]
+        2  2  [10.0, 10.0]
+        >>> table.update(values_sql={"x": "x + 1"})
+        >>> table.to_pandas()
+           x        vector
+        0  2    [1.0, 2.0]
+        1  4    [5.0, 6.0]
+        2  3  [10.0, 10.0]
+        """
+        raise NotImplementedError
+
 
 class LanceTable(Table):
     """
@@ -549,7 +679,7 @@ class LanceTable(Table):
 
     @property
     def _dataset_uri(self) -> str:
-        return os.path.join(self._conn.uri, f"{self.name}.lance")
+        return join_uri(self._conn.uri, f"{self.name}.lance")
 
     def create_index(
         self,
@@ -575,7 +705,12 @@ class LanceTable(Table):
         self._reset_dataset()
         register_event("create_index")
 
-    def create_fts_index(self, field_names: Union[str, List[str]]):
+    def create_scalar_index(self, column: str, *, replace: bool = True):
+        self._dataset.create_scalar_index(column, index_type="BTREE", replace=replace)
+
+    def create_fts_index(
+        self, field_names: Union[str, List[str]], *, replace: bool = False
+    ):
         """Create a full-text search index on the table.
 
         Warning - this API is highly experimental and is highly likely to change
@@ -585,17 +720,31 @@ class LanceTable(Table):
         ----------
         field_names: str or list of str
             The name(s) of the field to index.
+        replace: bool, default False
+            If True, replace the existing index if it exists. Note that this is
+            not yet an atomic operation; the index will be temporarily
+            unavailable while the new index is being created.
         """
         from .fts import create_index, populate_index
 
         if isinstance(field_names, str):
             field_names = [field_names]
+
+        fs, path = fs_from_uri(self._get_fts_index_path())
+        index_exists = fs.get_file_info(path).type != pa_fs.FileType.NotFound
+        if index_exists:
+            if not replace:
+                raise ValueError(
+                    f"Index already exists. Use replace=True to overwrite."
+                )
+            fs.delete_dir(path)
+
         index = create_index(self._get_fts_index_path(), field_names)
         populate_index(index, self, field_names)
         register_event("create_fts_index")
 
     def _get_fts_index_path(self):
-        return os.path.join(self._dataset_uri, "_indices", "tantivy")
+        return join_uri(self._dataset_uri, "_indices", "tantivy")
 
     @cached_property
     def _dataset(self) -> LanceDataset:
@@ -785,7 +934,7 @@ class LanceTable(Table):
             and also the "_distance" column which is the distance between the query
             vector and the returned vector.
         """
-        register_event("search")
+        register_event("search_table")
         return LanceQueryBuilder.create(
             self, query, query_type, vector_column_name=vector_column_name
         )
@@ -906,35 +1055,42 @@ class LanceTable(Table):
                 f"Table {name} does not exist."
                 f"Please first call db.create_table({name}, data)"
             )
+        register_event("open_table")
+
         return tbl
 
     def delete(self, where: str):
         self._dataset.delete(where)
 
-    def update(self, where: str, values: dict):
+    def update(
+        self,
+        where: Optional[str] = None,
+        values: Optional[dict] = None,
+        *,
+        values_sql: Optional[Dict[str, str]] = None,
+    ):
         """
-        EXPERIMENTAL: Update rows in the table (not threadsafe).
-
         This can be used to update zero to all rows depending on how many
         rows match the where clause.
 
         Parameters
         ----------
-        where: str
+        where: str, optional
             The SQL where clause to use when updating rows. For example, 'x = 2'
             or 'x IN (1, 2, 3)'. The filter must not be empty, or it will error.
-        values: dict
+        values: dict, optional
             The values to update. The keys are the column names and the values
             are the values to set.
+        values_sql: dict, optional
+            The values to update, expressed as SQL expression strings. These can
+            reference existing columns. For example, {"x": "x + 1"} will increment
+            the x column by 1.
 
         Examples
         --------
         >>> import lancedb
-        >>> data = [
-        ...    {"x": 1, "vector": [1, 2]},
-        ...    {"x": 2, "vector": [3, 4]},
-        ...    {"x": 3, "vector": [5, 6]}
-        ... ]
+        >>> import pandas as pd
+        >>> data = pd.DataFrame({"x": [1, 2, 3], "vector": [[1, 2], [3, 4], [5, 6]]})
         >>> db = lancedb.connect("./.lancedb")
         >>> table = db.create_table("my_table", data)
         >>> table.to_pandas()
@@ -950,18 +1106,15 @@ class LanceTable(Table):
         2  2  [10.0, 10.0]
 
         """
-        orig_data = self._dataset.to_table(filter=where).combine_chunks()
-        if len(orig_data) == 0:
-            return
-        for col, val in values.items():
-            i = orig_data.column_names.index(col)
-            if i < 0:
-                raise ValueError(f"Column {col} does not exist")
-            orig_data = orig_data.set_column(
-                i, col, pa.array([val] * len(orig_data), type=orig_data[col].type)
-            )
-        self.delete(where)
-        self.add(orig_data, mode="append")
+        if values is not None and values_sql is not None:
+            raise ValueError("Only one of values or values_sql can be provided")
+        if values is None and values_sql is None:
+            raise ValueError("Either values or values_sql must be provided")
+
+        if values is not None:
+            values_sql = {k: value_to_sql(v) for k, v in values.items()}
+
+        self.to_lance().update(values_sql, where)
         self._reset_dataset()
         register_event("update")
 

python/lancedb/util.py

@@ -12,9 +12,13 @@
 #  limitations under the License.
 
 import os
-from typing import Tuple
+from datetime import date, datetime
+from functools import singledispatch
+import pathlib
+from typing import Tuple, Union
 from urllib.parse import urlparse
 
+import numpy as np
 import pyarrow.fs as pa_fs
 
 
@@ -59,6 +63,12 @@ def get_uri_location(uri: str) -> str:
     str: Location part of the URL, without scheme
     """
     parsed = urlparse(uri)
+    if len(parsed.scheme) == 1:
+        # Windows drive names are parsed as the scheme
+        # e.g. "c:\path" -> ParseResult(scheme="c", netloc="", path="/path", ...)
+        # So we add special handling here for schemes that are a single character
+        return uri
+
     if not parsed.netloc:
         return parsed.path
     else:
@@ -81,6 +91,29 @@ def fs_from_uri(uri: str) -> Tuple[pa_fs.FileSystem, str]:
     return pa_fs.FileSystem.from_uri(uri)
 
 
+def join_uri(base: Union[str, pathlib.Path], *parts: str) -> str:
+    """
+    Join a URI with multiple parts, handles both local and remote paths
+
+    Parameters
+    ----------
+    base : str
+        The base URI
+    parts : str
+        The parts to join to the base URI, each separated by the
+        appropriate path separator for the URI scheme and OS
+    """
+    if isinstance(base, pathlib.Path):
+        return base.joinpath(*parts)
+    base = str(base)
+    if get_uri_scheme(base) == "file":
+        # using pathlib for local paths make this windows compatible
+        # `get_uri_scheme` returns `file` for windows drive names (e.g. `c:\path`)
+        return str(pathlib.Path(base, *parts))
+    # for remote paths, just use os.path.join
+    return "/".join([p.rstrip("/") for p in [base, *parts]])
+
+
 def safe_import_pandas():
     try:
         import pandas as pd
@@ -88,3 +121,53 @@ def safe_import_pandas():
         return pd
     except ImportError:
         return None
+
+
+@singledispatch
+def value_to_sql(value):
+    raise NotImplementedError("SQL conversion is not implemented for this type")
+
+
+@value_to_sql.register(str)
+def _(value: str):
+    return f"'{value}'"
+
+
+@value_to_sql.register(int)
+def _(value: int):
+    return str(value)
+
+
+@value_to_sql.register(float)
+def _(value: float):
+    return str(value)
+
+
+@value_to_sql.register(bool)
+def _(value: bool):
+    return str(value).upper()
+
+
+@value_to_sql.register(type(None))
+def _(value: type(None)):
+    return "NULL"
+
+
+@value_to_sql.register(datetime)
+def _(value: datetime):
+    return f"'{value.isoformat()}'"
+
+
+@value_to_sql.register(date)
+def _(value: date):
+    return f"'{value.isoformat()}'"
+
+
+@value_to_sql.register(list)
+def _(value: list):
+    return "[" + ", ".join(map(value_to_sql, value)) + "]"
+
+
+@value_to_sql.register(np.ndarray)
+def _(value: np.ndarray):
+    return value_to_sql(value.tolist())

python/lancedb/utils/events.py

@@ -64,8 +64,10 @@ class _Events:
         Initializes the Events object with default values for events, rate_limit, and metadata.
         """
         self.events = []  # events list
-        self.max_events = 25  # max events to store in memory
-        self.rate_limit = 60.0  # rate limit (seconds)
+        self.throttled_event_names = ["search_table"]
+        self.throttled_events = set()
+        self.max_events = 5  # max events to store in memory
+        self.rate_limit = 60.0 * 5  # rate limit (seconds)
         self.time = 0.0
 
         if is_git_dir():
@@ -112,18 +114,21 @@ class _Events:
             return
         if (
             len(self.events) < self.max_events
-        ):  # Events list limited to 25 events (drop any events past this)
+        ):  # Events list limited to self.max_events (drop any events past this)
             params.update(self.metadata)
-            self.events.append(
-                {
-                    "event": event_name,
-                    "properties": params,
-                    "timestamp": datetime.datetime.now(
-                        tz=datetime.timezone.utc
-                    ).isoformat(),
-                    "distinct_id": CONFIG["uuid"],
-                }
-            )
+            event = {
+                "event": event_name,
+                "properties": params,
+                "timestamp": datetime.datetime.now(
+                    tz=datetime.timezone.utc
+                ).isoformat(),
+                "distinct_id": CONFIG["uuid"],
+            }
+            if event_name not in self.throttled_event_names:
+                self.events.append(event)
+            elif event_name not in self.throttled_events:
+                self.throttled_events.add(event_name)
+                self.events.append(event)
 
         # Check rate limit
         t = time.time()
@@ -135,7 +140,6 @@ class _Events:
             "distinct_id": CONFIG["uuid"],  # posthog needs this to accepts the event
             "batch": self.events,
         }
-
         # POST equivalent to requests.post(self.url, json=data).
         # threaded request is used to avoid blocking, retries are disabled, and verbose is disabled
         # to avoid any possible disruption in the console.
@@ -150,6 +154,7 @@ class _Events:
 
         # Flush & Reset
         self.events = []
+        self.throttled_events = set()
         self.time = t
 
 

python/pyproject.toml

@@ -1,12 +1,12 @@
 [project]
 name = "lancedb"
-version = "0.3.4"
+version = "0.4.3"
 dependencies = [
     "deprecation",
-    "pylance==0.8.17",
+    "pylance==0.9.2",
     "ratelimiter~=1.0",
     "retry>=0.9.2",
-    "tqdm>=4.1.0",
+    "tqdm>=4.27.0",
     "aiohttp",
     "pydantic>=1.10",
     "attrs>=21.3.0",
@@ -49,11 +49,11 @@ classifiers = [
 repository = "https://github.com/lancedb/lancedb"
 
 [project.optional-dependencies]
-tests = ["pandas>=1.4", "pytest", "pytest-mock", "pytest-asyncio", "requests"]
+tests = ["pandas>=1.4", "pytest", "pytest-mock", "pytest-asyncio", "requests", "duckdb", "pytz"]
 dev = ["ruff", "pre-commit", "black"]
 docs = ["mkdocs", "mkdocs-jupyter", "mkdocs-material", "mkdocstrings[python]"]
 clip = ["torch", "pillow", "open-clip"]
-embeddings = ["openai", "sentence-transformers", "torch", "pillow", "open-clip-torch", "cohere", "InstructorEmbedding"]
+embeddings = ["openai>=1.6.1", "sentence-transformers", "torch", "pillow", "open-clip-torch", "cohere", "InstructorEmbedding"]
 
 [project.scripts]
 lancedb = "lancedb.cli.cli:cli"

python/tests/test_embeddings_slow.py

@@ -29,7 +29,7 @@ from lancedb.pydantic import LanceModel, Vector
 
 @pytest.mark.slow
 @pytest.mark.parametrize("alias", ["sentence-transformers", "openai"])
-def test_sentence_transformer(alias, tmp_path):
+def test_basic_text_embeddings(alias, tmp_path):
     db = lancedb.connect(tmp_path)
     registry = get_registry()
     func = registry.get(alias).create(max_retries=0)

python/tests/test_fts.py

@@ -12,6 +12,7 @@
 #  limitations under the License.
 import os
 import random
+from unittest import mock
 
 import numpy as np
 import pandas as pd
@@ -43,7 +44,16 @@ def table(tmp_path) -> ldb.table.LanceTable:
         for _ in range(100)
     ]
     table = db.create_table(
-        "test", data=pd.DataFrame({"vector": vectors, "text": text, "text2": text})
+        "test",
+        data=pd.DataFrame(
+            {
+                "vector": vectors,
+                "id": [i % 2 for i in range(100)],
+                "text": text,
+                "text2": text,
+                "nested": [{"text": t} for t in text],
+            }
+        ),
     )
     return table
 
@@ -75,6 +85,25 @@ def test_create_index_from_table(tmp_path, table):
     assert len(df) == 10
     assert "text" in df.columns
 
+    # Check whether it can be updated
+    table.add(
+        [
+            {
+                "vector": np.random.randn(128),
+                "id": 101,
+                "text": "gorilla",
+                "text2": "gorilla",
+                "nested": {"text": "gorilla"},
+            }
+        ]
+    )
+
+    with pytest.raises(ValueError, match="already exists"):
+        table.create_fts_index("text")
+
+    table.create_fts_index("text", replace=True)
+    assert len(table.search("gorilla").limit(1).to_pandas()) == 1
+
 
 def test_create_index_multiple_columns(tmp_path, table):
     table.create_fts_index(["text", "text2"])
@@ -89,3 +118,32 @@ def test_empty_rs(tmp_path, table, mocker):
     mocker.patch("lancedb.fts.search_index", return_value=([], []))
     df = table.search("puppy").limit(10).to_pandas()
     assert len(df) == 0
+
+
+def test_nested_schema(tmp_path, table):
+    table.create_fts_index("nested.text")
+    rs = table.search("puppy").limit(10).to_list()
+    assert len(rs) == 10
+
+
+def test_search_index_with_filter(table):
+    table.create_fts_index("text")
+    orig_import = __import__
+
+    def import_mock(name, *args):
+        if name == "duckdb":
+            raise ImportError
+        return orig_import(name, *args)
+
+    # no duckdb
+    with mock.patch("builtins.__import__", side_effect=import_mock):
+        rs = table.search("puppy").where("id=1").limit(10).to_list()
+        for r in rs:
+            assert r["id"] == 1
+
+    # yes duckdb
+    rs2 = table.search("puppy").where("id=1").limit(10).to_list()
+    for r in rs2:
+        assert r["id"] == 1
+
+    assert rs == rs2

python/tests/test_pydantic.py

@@ -13,9 +13,10 @@
 
 
 import json
+import pytz
 import sys
 from datetime import date, datetime
-from typing import List, Optional
+from typing import List, Optional, Tuple
 
 import pyarrow as pa
 import pydantic
@@ -38,11 +39,14 @@ def test_pydantic_to_arrow():
         id: int
         s: str
         vec: list[float]
-        li: List[int]
+        li: list[int]
+        lili: list[list[float]]
+        litu: list[tuple[float, float]]
         opt: Optional[str] = None
         st: StructModel
         dt: date
         dtt: datetime
+        dt_with_tz: datetime = Field(json_schema_extra={"tz": "Asia/Shanghai"})
         # d: dict
 
     m = TestModel(
@@ -50,9 +54,12 @@ def test_pydantic_to_arrow():
         s="hello",
         vec=[1.0, 2.0, 3.0],
         li=[2, 3, 4],
+        lili=[[2.5, 1.5], [3.5, 4.5], [5.5, 6.5]],
+        litu=[(2.5, 1.5), (3.5, 4.5), (5.5, 6.5)],
         st=StructModel(a="a", b=1.0),
         dt=date.today(),
         dtt=datetime.now(),
+        dt_with_tz=datetime.now(pytz.timezone("Asia/Shanghai")),
     )
 
     schema = pydantic_to_schema(TestModel)
@@ -63,6 +70,8 @@ def test_pydantic_to_arrow():
             pa.field("s", pa.utf8(), False),
             pa.field("vec", pa.list_(pa.float64()), False),
             pa.field("li", pa.list_(pa.int64()), False),
+            pa.field("lili", pa.list_(pa.list_(pa.float64())), False),
+            pa.field("litu", pa.list_(pa.list_(pa.float64())), False),
             pa.field("opt", pa.utf8(), True),
             pa.field(
                 "st",
@@ -73,11 +82,16 @@ def test_pydantic_to_arrow():
             ),
             pa.field("dt", pa.date32(), False),
             pa.field("dtt", pa.timestamp("us"), False),
+            pa.field("dt_with_tz", pa.timestamp("us", tz="Asia/Shanghai"), False),
         ]
     )
     assert schema == expect_schema
 
 
+@pytest.mark.skipif(
+    sys.version_info > (3, 8),
+    reason="using native type alias requires python3.9 or higher",
+)
 def test_pydantic_to_arrow_py38():
     class StructModel(pydantic.BaseModel):
         a: str
@@ -88,10 +102,13 @@ def test_pydantic_to_arrow_py38():
         s: str
         vec: List[float]
         li: List[int]
+        lili: List[List[float]]
+        litu: List[Tuple[float, float]]
         opt: Optional[str] = None
         st: StructModel
         dt: date
         dtt: datetime
+        dt_with_tz: datetime = Field(json_schema_extra={"tz": "Asia/Shanghai"})
         # d: dict
 
     m = TestModel(
@@ -99,9 +116,12 @@ def test_pydantic_to_arrow_py38():
         s="hello",
         vec=[1.0, 2.0, 3.0],
         li=[2, 3, 4],
+        lili=[[2.5, 1.5], [3.5, 4.5], [5.5, 6.5]],
+        litu=[(2.5, 1.5), (3.5, 4.5), (5.5, 6.5)],
         st=StructModel(a="a", b=1.0),
         dt=date.today(),
         dtt=datetime.now(),
+        dt_with_tz=datetime.now(pytz.timezone("Asia/Shanghai")),
     )
 
     schema = pydantic_to_schema(TestModel)
@@ -112,6 +132,8 @@ def test_pydantic_to_arrow_py38():
             pa.field("s", pa.utf8(), False),
             pa.field("vec", pa.list_(pa.float64()), False),
             pa.field("li", pa.list_(pa.int64()), False),
+            pa.field("lili", pa.list_(pa.list_(pa.float64())), False),
+            pa.field("litu", pa.list_(pa.list_(pa.float64())), False),
             pa.field("opt", pa.utf8(), True),
             pa.field(
                 "st",
@@ -122,6 +144,7 @@ def test_pydantic_to_arrow_py38():
             ),
             pa.field("dt", pa.date32(), False),
             pa.field("dtt", pa.timestamp("us"), False),
+            pa.field("dt_with_tz", pa.timestamp("us", tz="Asia/Shanghai"), False),
         ]
     )
     assert schema == expect_schema

python/tests/test_remote_db.py

@@ -26,6 +26,9 @@ class FakeLanceDBClient:
         t = pa.schema([]).empty_table()
         return VectorQueryResult(t)
 
+    async def post(self, path: str):
+        pass
+
 
 def test_remote_db():
     conn = lancedb.connect("db://client-will-be-injected", api_key="fake")

python/tests/test_table.py

@@ -12,7 +12,7 @@
 #  limitations under the License.
 
 import functools
-from datetime import timedelta
+from datetime import date, datetime, timedelta
 from pathlib import Path
 from typing import List
 from unittest.mock import PropertyMock, patch
@@ -22,6 +22,7 @@ import numpy as np
 import pandas as pd
 import pyarrow as pa
 import pytest
+from pydantic import BaseModel
 
 from lancedb.conftest import MockTextEmbeddingFunction
 from lancedb.db import LanceDBConnection
@@ -141,14 +142,44 @@ def test_add(db):
 
 
 def test_add_pydantic_model(db):
-    class TestModel(LanceModel):
-        vector: Vector(16)
-        li: List[int]
+    # https://github.com/lancedb/lancedb/issues/562
 
-    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()
+    class Metadata(BaseModel):
+        source: str
+        timestamp: datetime
+
+    class Document(BaseModel):
+        content: str
+        meta: Metadata
+
+    class LanceSchema(LanceModel):
+        id: str
+        vector: Vector(2)
+        li: List[int]
+        payload: Document
+
+    tbl = LanceTable.create(db, "mytable", schema=LanceSchema, mode="overwrite")
+    assert tbl.schema == LanceSchema.to_arrow_schema()
+
+    # add works
+    expected = LanceSchema(
+        id="id",
+        vector=[0.0, 0.0],
+        li=[1, 2, 3],
+        payload=Document(
+            content="foo", meta=Metadata(source="bar", timestamp=datetime.now())
+        ),
+    )
+    tbl.add([expected])
+
+    result = tbl.search([0.0, 0.0]).limit(1).to_pydantic(LanceSchema)[0]
+    assert result == expected
+
+    flattened = tbl.search([0.0, 0.0]).limit(1).to_pandas(flatten=1)
+    assert len(flattened.columns) == 6  # _distance is automatically added
+
+    really_flattened = tbl.search([0.0, 0.0]).limit(1).to_pandas(flatten=True)
+    assert len(really_flattened.columns) == 7
 
 
 def _add(table, schema):
@@ -348,14 +379,79 @@ def test_update(db):
     assert len(table) == 2
     assert len(table.list_versions()) == 2
     table.update(where="id=0", values={"vector": [1.1, 1.1]})
-    assert len(table.list_versions()) == 4
-    assert table.version == 4
+    assert len(table.list_versions()) == 3
+    assert table.version == 3
     assert len(table) == 2
     v = table.to_arrow()["vector"].combine_chunks()
     v = v.values.to_numpy().reshape(2, 2)
     assert np.allclose(v, np.array([[1.2, 1.9], [1.1, 1.1]]))
 
 
+def test_update_types(db):
+    table = LanceTable.create(
+        db,
+        "my_table",
+        data=[
+            {
+                "id": 0,
+                "str": "foo",
+                "float": 1.1,
+                "timestamp": datetime(2021, 1, 1),
+                "date": date(2021, 1, 1),
+                "vector1": [1.0, 0.0],
+                "vector2": [1.0, 1.0],
+            }
+        ],
+    )
+    # Update with SQL
+    table.update(
+        values_sql=dict(
+            id="1",
+            str="'bar'",
+            float="2.2",
+            timestamp="TIMESTAMP '2021-01-02 00:00:00'",
+            date="DATE '2021-01-02'",
+            vector1="[2.0, 2.0]",
+            vector2="[3.0, 3.0]",
+        )
+    )
+    actual = table.to_arrow().to_pylist()[0]
+    expected = dict(
+        id=1,
+        str="bar",
+        float=2.2,
+        timestamp=datetime(2021, 1, 2),
+        date=date(2021, 1, 2),
+        vector1=[2.0, 2.0],
+        vector2=[3.0, 3.0],
+    )
+    assert actual == expected
+
+    # Update with values
+    table.update(
+        values=dict(
+            id=2,
+            str="baz",
+            float=3.3,
+            timestamp=datetime(2021, 1, 3),
+            date=date(2021, 1, 3),
+            vector1=[3.0, 3.0],
+            vector2=np.array([4.0, 4.0]),
+        )
+    )
+    actual = table.to_arrow().to_pylist()[0]
+    expected = dict(
+        id=2,
+        str="baz",
+        float=3.3,
+        timestamp=datetime(2021, 1, 3),
+        date=date(2021, 1, 3),
+        vector1=[3.0, 3.0],
+        vector2=[4.0, 4.0],
+    )
+    assert actual == expected
+
+
 def test_create_with_embedding_function(db):
     class MyTable(LanceModel):
         text: str
@@ -436,6 +532,33 @@ def test_multiple_vector_columns(db):
     assert result1["text"].iloc[0] != result2["text"].iloc[0]
 
 
+def test_create_scalar_index(db):
+    vec_array = pa.array(
+        [[1, 1], [2, 2], [3, 3], [4, 4], [5, 5]], pa.list_(pa.float32(), 2)
+    )
+    test_data = pa.Table.from_pydict(
+        {"x": ["c", "b", "a", "e", "b"], "y": [1, 2, 3, 4, 5], "vector": vec_array}
+    )
+    table = LanceTable.create(
+        db,
+        "my_table",
+        data=test_data,
+    )
+    table.create_scalar_index("x")
+    indices = table.to_lance().list_indices()
+    assert len(indices) == 1
+    scalar_index = indices[0]
+    assert scalar_index["type"] == "Scalar"
+
+    # Confirm that prefiltering still works with the scalar index column
+    results = table.search().where("x = 'c'").to_arrow()
+    assert results == test_data.slice(0, 1)
+    results = table.search([5, 5]).to_arrow()
+    assert results["_distance"][0].as_py() == 0
+    results = table.search([5, 5]).where("x != 'b'").to_arrow()
+    assert results["_distance"][0].as_py() > 0
+
+
 def test_empty_query(db):
     table = LanceTable.create(
         db,

python/tests/test_util.py

@@ -11,7 +11,12 @@
 #  See the License for the specific language governing permissions and
 #  limitations under the License.
 
-from lancedb.util import get_uri_scheme
+import os
+import pathlib
+
+import pytest
+
+from lancedb.util import get_uri_scheme, join_uri
 
 
 def test_normalize_uri():
@@ -28,3 +33,55 @@ def test_normalize_uri():
     for uri, expected_scheme in zip(uris, schemes):
         parsed_scheme = get_uri_scheme(uri)
         assert parsed_scheme == expected_scheme
+
+
+def test_join_uri_remote():
+    schemes = ["s3", "az", "gs"]
+    for scheme in schemes:
+        expected = f"{scheme}://bucket/path/to/table.lance"
+        base_uri = f"{scheme}://bucket/path/to/"
+        parts = ["table.lance"]
+        assert join_uri(base_uri, *parts) == expected
+
+        base_uri = f"{scheme}://bucket"
+        parts = ["path", "to", "table.lance"]
+        assert join_uri(base_uri, *parts) == expected
+
+
+# skip this test if on windows
+@pytest.mark.skipif(os.name == "nt", reason="Windows paths are not POSIX")
+def test_join_uri_posix():
+    for base in [
+        # relative path
+        "relative/path",
+        "relative/path/",
+        # an absolute path
+        "/absolute/path",
+        "/absolute/path/",
+        # a file URI
+        "file:///absolute/path",
+        "file:///absolute/path/",
+    ]:
+        joined = join_uri(base, "table.lance")
+        assert joined == str(pathlib.Path(base) / "table.lance")
+        joined = join_uri(pathlib.Path(base), "table.lance")
+        assert joined == pathlib.Path(base) / "table.lance"
+
+
+# skip this test if not on windows
+@pytest.mark.skipif(os.name != "nt", reason="Windows paths are not POSIX")
+def test_local_join_uri_windows():
+    # https://learn.microsoft.com/en-us/dotnet/standard/io/file-path-formats
+    for base in [
+        # windows relative path
+        "relative\\path",
+        "relative\\path\\",
+        # windows absolute path from current drive
+        "c:\\absolute\\path",
+        # relative path from root of current drive
+        "\\relative\\path",
+    ]:
+        joined = join_uri(base, "table.lance")
+        assert joined == str(pathlib.Path(base) / "table.lance")
+        joined = join_uri(pathlib.Path(base), "table.lance")
+        assert joined == pathlib.Path(base) / "table.lance"

rust/ffi/node/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "vectordb-node"
-version = "0.3.8"
+version = "0.4.1"
 description = "Serverless, low-latency vector database for AI applications"
 license = "Apache-2.0"
 edition = "2018"

rust/ffi/node/src/error.rs

@@ -23,7 +23,7 @@ pub enum Error {
     #[snafu(display("column '{name}' is missing"))]
     MissingColumn { name: String },
     #[snafu(display("{name}: {message}"))]
-    RangeError { name: String, message: String },
+    OutOfRange { name: String, message: String },
     #[snafu(display("{index_type} is not a valid index type"))]
     InvalidIndexType { index_type: String },
 

rust/ffi/node/src/index.rs

@@ -12,4 +12,5 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+pub mod scalar;
 pub mod vector;

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

@@ -0,0 +1,43 @@
+// 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 neon::{
+    context::{Context, FunctionContext},
+    result::JsResult,
+    types::{JsBoolean, JsBox, JsPromise, JsString},
+};
+
+use crate::{error::ResultExt, runtime, table::JsTable};
+
+pub(crate) fn table_create_scalar_index(mut cx: FunctionContext) -> JsResult<JsPromise> {
+    let js_table = cx.this().downcast_or_throw::<JsBox<JsTable>, _>(&mut cx)?;
+    let column = cx.argument::<JsString>(0)?.value(&mut cx);
+    let replace = cx.argument::<JsBoolean>(1)?.value(&mut cx);
+
+    let rt = runtime(&mut cx)?;
+
+    let (deferred, promise) = cx.promise();
+    let channel = cx.channel();
+    let mut table = js_table.table.clone();
+
+    rt.spawn(async move {
+        let idx_result = table.create_scalar_index(&column, replace).await;
+
+        deferred.settle_with(&channel, move |mut cx| {
+            idx_result.or_throw(&mut cx)?;
+            Ok(cx.undefined())
+        });
+    });
+    Ok(promise)
+}

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

@@ -65,12 +65,10 @@ fn get_index_params_builder(
             obj.get_opt::<JsString, _, _>(cx, "index_name")?
                 .map(|s| index_builder.index_name(s.value(cx)));
 
-            obj.get_opt::<JsString, _, _>(cx, "metric_type")?
-                .map(|s| MetricType::try_from(s.value(cx).as_str()))
-                .map(|mt| {
-                    let metric_type = mt.unwrap();
-                    index_builder.metric_type(metric_type);
-                });
+            if let Some(metric_type) = obj.get_opt::<JsString, _, _>(cx, "metric_type")? {
+                let metric_type = MetricType::try_from(metric_type.value(cx).as_str()).unwrap();
+                index_builder.metric_type(metric_type);
+            }
 
             let num_partitions = obj.get_opt_usize(cx, "num_partitions")?;
             let max_iters = obj.get_opt_usize(cx, "max_iters")?;
@@ -85,23 +83,29 @@ fn get_index_params_builder(
                 index_builder.ivf_params(ivf_params)
             });
 
-            obj.get_opt::<JsBoolean, _, _>(cx, "use_opq")?
-                .map(|s| pq_params.use_opq = s.value(cx));
+            if let Some(use_opq) = obj.get_opt::<JsBoolean, _, _>(cx, "use_opq")? {
+                pq_params.use_opq = use_opq.value(cx);
+            }
 
-            obj.get_opt_usize(cx, "num_sub_vectors")?
-                .map(|s| pq_params.num_sub_vectors = s);
+            if let Some(num_sub_vectors) = obj.get_opt_usize(cx, "num_sub_vectors")? {
+                pq_params.num_sub_vectors = num_sub_vectors;
+            }
 
-            obj.get_opt_usize(cx, "num_bits")?
-                .map(|s| pq_params.num_bits = s);
+            if let Some(num_bits) = obj.get_opt_usize(cx, "num_bits")? {
+                pq_params.num_bits = num_bits;
+            }
 
-            obj.get_opt_usize(cx, "max_iters")?
-                .map(|s| pq_params.max_iters = s);
+            if let Some(max_iters) = obj.get_opt_usize(cx, "max_iters")? {
+                pq_params.max_iters = max_iters;
+            }
 
-            obj.get_opt_usize(cx, "max_opq_iters")?
-                .map(|s| pq_params.max_opq_iters = s);
+            if let Some(max_opq_iters) = obj.get_opt_usize(cx, "max_opq_iters")? {
+                pq_params.max_opq_iters = max_opq_iters;
+            }
 
-            obj.get_opt::<JsBoolean, _, _>(cx, "replace")?
-                .map(|s| index_builder.replace(s.value(cx)));
+            if let Some(replace) = obj.get_opt::<JsBoolean, _, _>(cx, "replace")? {
+                index_builder.replace(replace.value(cx));
+            }
 
             Ok(index_builder)
         }

rust/ffi/node/src/lib.rs

@@ -237,10 +237,15 @@ fn main(mut cx: ModuleContext) -> NeonResult<()> {
     cx.export_function("tableAdd", JsTable::js_add)?;
     cx.export_function("tableCountRows", JsTable::js_count_rows)?;
     cx.export_function("tableDelete", JsTable::js_delete)?;
+    cx.export_function("tableUpdate", JsTable::js_update)?;
     cx.export_function("tableCleanupOldVersions", JsTable::js_cleanup)?;
     cx.export_function("tableCompactFiles", JsTable::js_compact)?;
     cx.export_function("tableListIndices", JsTable::js_list_indices)?;
     cx.export_function("tableIndexStats", JsTable::js_index_stats)?;
+    cx.export_function(
+        "tableCreateScalarIndex",
+        index::scalar::table_create_scalar_index,
+    )?;
     cx.export_function(
         "tableCreateVectorIndex",
         index::vector::table_create_vector_index,

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

@@ -47,15 +47,15 @@ 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 {
+        FloatError::NegOverflow(_) => Error::OutOfRange {
             name: key.into(),
             message: "must be > 0".to_string(),
         },
-        FloatError::PosOverflow(_) => Error::RangeError {
+        FloatError::PosOverflow(_) => Error::OutOfRange {
             name: key.into(),
             message: format!("must be < {}", u32::MAX),
         },
-        FloatError::NotANumber(_) => Error::RangeError {
+        FloatError::NotANumber(_) => Error::OutOfRange {
             name: key.into(),
             message: "not a valid number".to_string(),
         },
@@ -66,15 +66,15 @@ 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 {
+        FloatError::NegOverflow(_) => Error::OutOfRange {
             name: key.into(),
             message: "must be > 0".to_string(),
         },
-        FloatError::PosOverflow(_) => Error::RangeError {
+        FloatError::PosOverflow(_) => Error::OutOfRange {
             name: key.into(),
             message: format!("must be < {}", usize::MAX),
         },
-        FloatError::NotANumber(_) => Error::RangeError {
+        FloatError::NotANumber(_) => Error::OutOfRange {
             name: key.into(),
             message: "not a valid number".to_string(),
         },

rust/ffi/node/src/query.rs

@@ -23,8 +23,14 @@ impl JsQuery {
         let query_obj = cx.argument::<JsObject>(0)?;
 
         let limit = query_obj
-            .get::<JsNumber, _, _>(&mut cx, "_limit")?
-            .value(&mut cx);
+            .get_opt::<JsNumber, _, _>(&mut cx, "_limit")?
+            .map(|value| {
+                let limit = value.value(&mut cx);
+                if limit <= 0.0 {
+                    panic!("Limit must be a positive integer");
+                }
+                limit as u64
+            });
         let select = query_obj
             .get_opt::<JsArray, _, _>(&mut cx, "_select")?
             .map(|arr| {
@@ -48,6 +54,10 @@ impl JsQuery {
             .map(|s| s.value(&mut cx))
             .map(|s| MetricType::try_from(s.as_str()).unwrap());
 
+        let prefilter = query_obj
+            .get::<JsBoolean, _, _>(&mut cx, "_prefilter")?
+            .value(&mut cx);
+
         let is_electron = cx
             .argument::<JsBoolean>(1)
             .or_throw(&mut cx)?
@@ -57,19 +67,23 @@ impl JsQuery {
 
         let (deferred, promise) = cx.promise();
         let channel = cx.channel();
-        let query_vector = query_obj.get::<JsArray, _, _>(&mut cx, "_queryVector")?;
-        let query = convert::js_array_to_vec(query_vector.deref(), &mut cx);
+        let query_vector = query_obj.get_opt::<JsArray, _, _>(&mut cx, "_queryVector")?;
         let table = js_table.table.clone();
+        let query = query_vector.map(|q| convert::js_array_to_vec(q.deref(), &mut cx));
 
         rt.spawn(async move {
-            let builder = table
-                .search(Float32Array::from(query))
-                .limit(limit as usize)
+            let mut builder = table
+                .search(query.map(Float32Array::from))
                 .refine_factor(refine_factor)
                 .nprobes(nprobes)
                 .filter(filter)
                 .metric_type(metric_type)
-                .select(select);
+                .select(select)
+                .prefilter(prefilter);
+            if let Some(limit) = limit {
+                builder = builder.limit(limit as usize);
+            };
+
             let record_batch_stream = builder.execute();
             let results = record_batch_stream
                 .and_then(|stream| {

rust/ffi/node/src/table.rs

@@ -45,7 +45,7 @@ impl JsTable {
         let table_name = cx.argument::<JsString>(0)?.value(&mut cx);
         let buffer = cx.argument::<JsBuffer>(1)?;
         let (batches, schema) =
-            arrow_buffer_to_record_batch(buffer.as_slice(&mut cx)).or_throw(&mut cx)?;
+            arrow_buffer_to_record_batch(buffer.as_slice(&cx)).or_throw(&mut cx)?;
 
         // Write mode
         let mode = match cx.argument::<JsString>(2)?.value(&mut cx).as_str() {
@@ -93,7 +93,7 @@ impl JsTable {
         let buffer = cx.argument::<JsBuffer>(0)?;
         let write_mode = cx.argument::<JsString>(1)?.value(&mut cx);
         let (batches, schema) =
-            arrow_buffer_to_record_batch(buffer.as_slice(&mut cx)).or_throw(&mut cx)?;
+            arrow_buffer_to_record_batch(buffer.as_slice(&cx)).or_throw(&mut cx)?;
         let rt = runtime(&mut cx)?;
         let channel = cx.channel();
         let mut table = js_table.table.clone();
@@ -165,6 +165,69 @@ impl JsTable {
         Ok(promise)
     }
 
+    pub(crate) fn js_update(mut cx: FunctionContext) -> JsResult<JsPromise> {
+        let js_table = cx.this().downcast_or_throw::<JsBox<JsTable>, _>(&mut cx)?;
+        let mut table = js_table.table.clone();
+
+        let rt = runtime(&mut cx)?;
+        let (deferred, promise) = cx.promise();
+        let channel = cx.channel();
+
+        // create a vector of updates from the passed map
+        let updates_arg = cx.argument::<JsObject>(1)?;
+        let properties = updates_arg.get_own_property_names(&mut cx)?;
+        let mut updates: Vec<(String, String)> =
+            Vec::with_capacity(properties.len(&mut cx) as usize);
+
+        let len_properties = properties.len(&mut cx);
+        for i in 0..len_properties {
+            let property = properties
+                .get_value(&mut cx, i)?
+                .downcast_or_throw::<JsString, _>(&mut cx)?;
+
+            let value = updates_arg
+                .get_value(&mut cx, property)?
+                .downcast_or_throw::<JsString, _>(&mut cx)?;
+
+            let property = property.value(&mut cx);
+            let value = value.value(&mut cx);
+            updates.push((property, value));
+        }
+
+        // get the filter/predicate if the user passed one
+        let predicate = cx.argument_opt(0);
+        let predicate = predicate.unwrap().downcast::<JsString, _>(&mut cx);
+        let predicate = match predicate {
+            Ok(_) => {
+                let val = predicate.map(|s| s.value(&mut cx)).unwrap();
+                Some(val)
+            }
+            Err(_) => {
+                // if the predicate is not string, check it's null otherwise an invalid
+                // type was passed
+                cx.argument::<JsNull>(0)?;
+                None
+            }
+        };
+
+        rt.spawn(async move {
+            let updates_arg = updates
+                .iter()
+                .map(|(k, v)| (k.as_str(), v.as_str()))
+                .collect::<Vec<_>>();
+
+            let predicate = predicate.as_deref();
+
+            let update_result = table.update(predicate, updates_arg).await;
+            deferred.settle_with(&channel, move |mut cx| {
+                update_result.or_throw(&mut cx)?;
+                Ok(cx.boxed(JsTable::from(table)))
+            })
+        });
+
+        Ok(promise)
+    }
+
     pub(crate) fn js_cleanup(mut cx: FunctionContext) -> JsResult<JsPromise> {
         let js_table = cx.this().downcast_or_throw::<JsBox<JsTable>, _>(&mut cx)?;
         let rt = runtime(&mut cx)?;

rust/vectordb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "vectordb"
-version = "0.3.8"
+version = "0.4.1"
 edition = "2021"
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license = "Apache-2.0"

rust/vectordb/src/io/object_store.rs

@@ -26,7 +26,7 @@ use futures::{stream::BoxStream, FutureExt, StreamExt};
 use lance::io::object_store::WrappingObjectStore;
 use object_store::{
     path::Path, Error, GetOptions, GetResult, ListResult, MultipartId, ObjectMeta, ObjectStore,
-    Result,
+    PutOptions, PutResult, Result,
 };
 
 use async_trait::async_trait;
@@ -72,13 +72,28 @@ impl PrimaryOnly for Path {
 /// Note: this object store does not mirror writes to *.manifest files
 #[async_trait]
 impl ObjectStore for MirroringObjectStore {
-    async fn put(&self, location: &Path, bytes: Bytes) -> Result<()> {
+    async fn put(&self, location: &Path, bytes: Bytes) -> Result<PutResult> {
         if location.primary_only() {
             self.primary.put(location, bytes).await
         } else {
             self.secondary.put(location, bytes.clone()).await?;
-            self.primary.put(location, bytes).await?;
-            Ok(())
+            self.primary.put(location, bytes).await
+        }
+    }
+
+    async fn put_opts(
+        &self,
+        location: &Path,
+        bytes: Bytes,
+        options: PutOptions,
+    ) -> Result<PutResult> {
+        if location.primary_only() {
+            self.primary.put_opts(location, bytes, options).await
+        } else {
+            self.secondary
+                .put_opts(location, bytes.clone(), options.clone())
+                .await?;
+            self.primary.put_opts(location, bytes, options).await
         }
     }
 
@@ -129,8 +144,8 @@ impl ObjectStore for MirroringObjectStore {
         self.primary.delete(location).await
     }
 
-    async fn list(&self, prefix: Option<&Path>) -> Result<BoxStream<'_, Result<ObjectMeta>>> {
-        self.primary.list(prefix).await
+    fn list(&self, prefix: Option<&Path>) -> BoxStream<'_, Result<ObjectMeta>> {
+        self.primary.list(prefix)
     }
 
     async fn list_with_delimiter(&self, prefix: Option<&Path>) -> Result<ListResult> {
@@ -359,7 +374,9 @@ mod test {
         assert_eq!(t.count_rows().await.unwrap(), 100);
 
         let q = t
-            .search(PrimitiveArray::from_iter_values(vec![0.1, 0.1, 0.1, 0.1]))
+            .search(Some(PrimitiveArray::from_iter_values(vec![
+                0.1, 0.1, 0.1, 0.1,
+            ])))
             .limit(10)
             .execute()
             .await

rust/vectordb/src/query.rs

@@ -24,14 +24,16 @@ use crate::error::Result;
 /// A builder for nearest neighbor queries for LanceDB.
 pub struct Query {
     pub dataset: Arc<Dataset>,
-    pub query_vector: Float32Array,
-    pub limit: usize,
+    pub query_vector: Option<Float32Array>,
+    pub column: String,
+    pub limit: Option<usize>,
     pub filter: Option<String>,
     pub select: Option<Vec<String>>,
     pub nprobes: usize,
     pub refine_factor: Option<u32>,
     pub metric_type: Option<MetricType>,
     pub use_index: bool,
+    pub prefilter: bool,
 }
 
 impl Query {
@@ -45,17 +47,19 @@ impl Query {
     /// # Returns
     ///
     /// * A [Query] object.
-    pub(crate) fn new(dataset: Arc<Dataset>, vector: Float32Array) -> Self {
+    pub(crate) fn new(dataset: Arc<Dataset>, vector: Option<Float32Array>) -> Self {
         Query {
             dataset,
             query_vector: vector,
-            limit: 10,
+            column: crate::table::VECTOR_COLUMN_NAME.to_string(),
+            limit: None,
             nprobes: 20,
             refine_factor: None,
             metric_type: None,
             use_index: true,
             filter: None,
             select: None,
+            prefilter: false,
         }
     }
 
@@ -67,13 +71,17 @@ impl Query {
     pub async fn execute(&self) -> Result<DatasetRecordBatchStream> {
         let mut scanner: Scanner = self.dataset.scan();
 
-        scanner.nearest(
-            crate::table::VECTOR_COLUMN_NAME,
-            &self.query_vector,
-            self.limit,
-        )?;
+        if let Some(query) = self.query_vector.as_ref() {
+            // If there is a vector query, default to limit=10 if unspecified
+            scanner.nearest(&self.column, query, self.limit.unwrap_or(10))?;
+        } else {
+            // If there is no vector query, it's ok to not have a limit
+            scanner.limit(self.limit.map(|limit| limit as i64), None)?;
+        }
         scanner.nprobs(self.nprobes);
         scanner.use_index(self.use_index);
+        scanner.prefilter(self.prefilter);
+
         self.select.as_ref().map(|p| scanner.project(p.as_slice()));
         self.filter.as_ref().map(|f| scanner.filter(f));
         self.refine_factor.map(|rf| scanner.refine(rf));
@@ -81,13 +89,23 @@ impl Query {
         Ok(scanner.try_into_stream().await?)
     }
 
+    /// Set the column to query
+    ///
+    /// # Arguments
+    ///
+    /// * `column` - The column name
+    pub fn column(mut self, column: &str) -> Query {
+        self.column = column.into();
+        self
+    }
+
     /// Set the maximum number of results to return.
     ///
     /// # Arguments
     ///
     /// * `limit` - The maximum number of results to return.
     pub fn limit(mut self, limit: usize) -> Query {
-        self.limit = limit;
+        self.limit = Some(limit);
         self
     }
 
@@ -97,7 +115,7 @@ impl Query {
     ///
     /// * `vector` - The vector that will be used for search.
     pub fn query_vector(mut self, query_vector: Float32Array) -> Query {
-        self.query_vector = query_vector;
+        self.query_vector = Some(query_vector);
         self
     }
 
@@ -158,6 +176,11 @@ impl Query {
         self.select = columns;
         self
     }
+
+    pub fn prefilter(mut self, prefilter: bool) -> Query {
+        self.prefilter = prefilter;
+        self
+    }
 }
 
 #[cfg(test)]
@@ -165,9 +188,14 @@ mod tests {
     use std::sync::Arc;
 
     use super::*;
-    use arrow_array::{Float32Array, RecordBatch, RecordBatchIterator, RecordBatchReader};
+    use arrow_array::{
+        cast::AsArray, Float32Array, Int32Array, RecordBatch, RecordBatchIterator,
+        RecordBatchReader,
+    };
     use arrow_schema::{DataType, Field as ArrowField, Schema as ArrowSchema};
+    use futures::StreamExt;
     use lance::dataset::Dataset;
+    use lance_testing::datagen::{BatchGenerator, IncrementingInt32, RandomVector};
 
     use crate::query::Query;
 
@@ -176,7 +204,7 @@ mod tests {
         let batches = make_test_batches();
         let ds = Dataset::write(batches, "memory://foo", None).await.unwrap();
 
-        let vector = Float32Array::from_iter_values([0.1, 0.2]);
+        let vector = Some(Float32Array::from_iter_values([0.1, 0.2]));
         let query = Query::new(Arc::new(ds), vector.clone());
         assert_eq!(query.query_vector, vector);
 
@@ -190,8 +218,8 @@ mod tests {
             .metric_type(Some(MetricType::Cosine))
             .refine_factor(Some(999));
 
-        assert_eq!(query.query_vector, new_vector);
-        assert_eq!(query.limit, 100);
+        assert_eq!(query.query_vector.unwrap(), new_vector);
+        assert_eq!(query.limit.unwrap(), 100);
         assert_eq!(query.nprobes, 1000);
         assert_eq!(query.use_index, true);
         assert_eq!(query.metric_type, Some(MetricType::Cosine));
@@ -200,13 +228,64 @@ mod tests {
 
     #[tokio::test]
     async fn test_execute() {
-        let batches = make_test_batches();
-        let ds = Dataset::write(batches, "memory://foo", None).await.unwrap();
+        let batches = make_non_empty_batches();
+        let ds = Arc::new(Dataset::write(batches, "memory://foo", None).await.unwrap());
 
-        let vector = Float32Array::from_iter_values([0.1; 128]);
-        let query = Query::new(Arc::new(ds), vector.clone());
-        let result = query.execute().await;
-        assert_eq!(result.is_ok(), true);
+        let vector = Some(Float32Array::from_iter_values([0.1; 4]));
+
+        let query = Query::new(ds.clone(), vector.clone());
+        let result = query
+            .limit(10)
+            .filter(Some("id % 2 == 0".to_string()))
+            .execute()
+            .await;
+        let mut stream = result.expect("should have result");
+        // should only have one batch
+        while let Some(batch) = stream.next().await {
+            // post filter should have removed some rows
+            assert!(batch.expect("should be Ok").num_rows() < 10);
+        }
+
+        let query = Query::new(ds, vector.clone());
+        let result = query
+            .limit(10)
+            .filter(Some("id % 2 == 0".to_string()))
+            .prefilter(true)
+            .execute()
+            .await;
+        let mut stream = result.expect("should have result");
+        // should only have one batch
+        while let Some(batch) = stream.next().await {
+            // pre filter should return 10 rows
+            assert!(batch.expect("should be Ok").num_rows() == 10);
+        }
+    }
+
+    #[tokio::test]
+    async fn test_execute_no_vector() {
+        // test that it's ok to not specify a query vector (just filter / limit)
+        let batches = make_non_empty_batches();
+        let ds = Arc::new(Dataset::write(batches, "memory://foo", None).await.unwrap());
+
+        let query = Query::new(ds.clone(), None);
+        let result = query
+            .filter(Some("id % 2 == 0".to_string()))
+            .execute()
+            .await;
+        let mut stream = result.expect("should have result");
+        // should only have one batch
+        while let Some(batch) = stream.next().await {
+            let b = batch.expect("should be Ok");
+            // cast arr into Int32Array
+            let arr: &Int32Array = b["id"].as_primitive();
+            assert!(arr.iter().all(|x| x.unwrap() % 2 == 0));
+        }
+    }
+
+    fn make_non_empty_batches() -> impl RecordBatchReader + Send + 'static {
+        let vec = Box::new(RandomVector::new().named("vector".to_string()));
+        let id = Box::new(IncrementingInt32::new().named("id".to_string()));
+        BatchGenerator::new().col(vec).col(id).batch(512)
     }
 
     fn make_test_batches() -> impl RecordBatchReader + Send + 'static {

rust/vectordb/src/table.rs

@@ -14,6 +14,7 @@
 
 use chrono::Duration;
 use lance::dataset::builder::DatasetBuilder;
+use lance::index::scalar::ScalarIndexParams;
 use lance_index::IndexType;
 use std::sync::Arc;
 
@@ -23,7 +24,7 @@ use lance::dataset::cleanup::RemovalStats;
 use lance::dataset::optimize::{
     compact_files, CompactionMetrics, CompactionOptions, IndexRemapperOptions,
 };
-use lance::dataset::{Dataset, WriteParams};
+use lance::dataset::{Dataset, UpdateBuilder, WriteParams};
 use lance::index::DatasetIndexExt;
 use lance::io::object_store::WrappingObjectStore;
 use std::path::Path;
@@ -262,6 +263,16 @@ impl Table {
         Ok(())
     }
 
+    /// Create a scalar index on the table
+    pub async fn create_scalar_index(&mut self, column: &str, replace: bool) -> Result<()> {
+        let mut dataset = self.dataset.as_ref().clone();
+        let params = ScalarIndexParams::default();
+        dataset
+            .create_index(&[column], IndexType::Scalar, None, &params, replace)
+            .await?;
+        Ok(())
+    }
+
     pub async fn optimize_indices(&mut self) -> Result<()> {
         let mut dataset = self.dataset.as_ref().clone();
 
@@ -308,10 +319,14 @@ impl Table {
     /// # Returns
     ///
     /// * A [Query] object.
-    pub fn search(&self, query_vector: Float32Array) -> Query {
+    pub fn search(&self, query_vector: Option<Float32Array>) -> Query {
         Query::new(self.dataset.clone(), query_vector)
     }
 
+    pub fn filter(&self, expr: String) -> Query {
+        Query::new(self.dataset.clone(), None).filter(Some(expr))
+    }
+
     /// Returns the number of rows in this Table
     pub async fn count_rows(&self) -> Result<usize> {
         Ok(self.dataset.count_rows().await?)
@@ -338,6 +353,27 @@ impl Table {
         Ok(())
     }
 
+    pub async fn update(
+        &mut self,
+        predicate: Option<&str>,
+        updates: Vec<(&str, &str)>,
+    ) -> Result<()> {
+        let mut builder = UpdateBuilder::new(self.dataset.clone());
+        if let Some(predicate) = predicate {
+            builder = builder.update_where(predicate)?;
+        }
+
+        for (column, value) in updates {
+            builder = builder.set(column, value)?;
+        }
+
+        let operation = builder.build()?;
+        let new_ds = operation.execute().await?;
+        self.dataset = new_ds;
+
+        Ok(())
+    }
+
     /// Remove old versions of the dataset from disk.
     ///
     /// # Arguments
@@ -413,11 +449,14 @@ mod tests {
     use std::sync::Arc;
 
     use arrow_array::{
-        Array, FixedSizeListArray, Float32Array, Int32Array, RecordBatch, RecordBatchIterator,
-        RecordBatchReader,
+        Array, BooleanArray, Date32Array, FixedSizeListArray, Float32Array, Float64Array,
+        Int32Array, Int64Array, LargeStringArray, RecordBatch, RecordBatchIterator,
+        RecordBatchReader, StringArray, TimestampMillisecondArray, TimestampNanosecondArray,
+        UInt32Array,
     };
     use arrow_data::ArrayDataBuilder;
-    use arrow_schema::{DataType, Field, Schema};
+    use arrow_schema::{DataType, Field, Schema, TimeUnit};
+    use futures::TryStreamExt;
     use lance::dataset::{Dataset, WriteMode};
     use lance::index::vector::pq::PQBuildParams;
     use lance::io::object_store::{ObjectStoreParams, WrappingObjectStore};
@@ -540,6 +579,272 @@ mod tests {
         assert_eq!(table.name, "test");
     }
 
+    #[tokio::test]
+    async fn test_update_with_predicate() {
+        let tmp_dir = tempdir().unwrap();
+        let dataset_path = tmp_dir.path().join("test.lance");
+        let uri = dataset_path.to_str().unwrap();
+
+        let schema = Arc::new(Schema::new(vec![
+            Field::new("id", DataType::Int32, false),
+            Field::new("name", DataType::Utf8, false),
+        ]));
+
+        let record_batch_iter = RecordBatchIterator::new(
+            vec![RecordBatch::try_new(
+                schema.clone(),
+                vec![
+                    Arc::new(Int32Array::from_iter_values(0..10)),
+                    Arc::new(StringArray::from_iter_values(vec![
+                        "a", "b", "c", "d", "e", "f", "g", "h", "i", "j",
+                    ])),
+                ],
+            )
+            .unwrap()]
+            .into_iter()
+            .map(Ok),
+            schema.clone(),
+        );
+
+        Dataset::write(record_batch_iter, uri, None).await.unwrap();
+        let mut table = Table::open(uri).await.unwrap();
+
+        table
+            .update(Some("id > 5"), vec![("name", "'foo'")])
+            .await
+            .unwrap();
+
+        let ds_after = Dataset::open(uri).await.unwrap();
+        let mut batches = ds_after
+            .scan()
+            .project(&["id", "name"])
+            .unwrap()
+            .try_into_stream()
+            .await
+            .unwrap()
+            .try_collect::<Vec<_>>()
+            .await
+            .unwrap();
+
+        while let Some(batch) = batches.pop() {
+            let ids = batch
+                .column(0)
+                .as_any()
+                .downcast_ref::<Int32Array>()
+                .unwrap()
+                .iter()
+                .collect::<Vec<_>>();
+            let names = batch
+                .column(1)
+                .as_any()
+                .downcast_ref::<StringArray>()
+                .unwrap()
+                .iter()
+                .collect::<Vec<_>>();
+            for (i, name) in names.iter().enumerate() {
+                let id = ids[i].unwrap();
+                let name = name.unwrap();
+                if id > 5 {
+                    assert_eq!(name, "foo");
+                } else {
+                    assert_eq!(name, &format!("{}", (b'a' + id as u8) as char));
+                }
+            }
+        }
+    }
+
+    #[tokio::test]
+    async fn test_update_all_types() {
+        let tmp_dir = tempdir().unwrap();
+        let dataset_path = tmp_dir.path().join("test.lance");
+        let uri = dataset_path.to_str().unwrap();
+
+        let schema = Arc::new(Schema::new(vec![
+            Field::new("int32", DataType::Int32, false),
+            Field::new("int64", DataType::Int64, false),
+            Field::new("uint32", DataType::UInt32, false),
+            Field::new("string", DataType::Utf8, false),
+            Field::new("large_string", DataType::LargeUtf8, false),
+            Field::new("float32", DataType::Float32, false),
+            Field::new("float64", DataType::Float64, false),
+            Field::new("bool", DataType::Boolean, false),
+            Field::new("date32", DataType::Date32, false),
+            Field::new(
+                "timestamp_ns",
+                DataType::Timestamp(TimeUnit::Nanosecond, None),
+                false,
+            ),
+            Field::new(
+                "timestamp_ms",
+                DataType::Timestamp(TimeUnit::Millisecond, None),
+                false,
+            ),
+            Field::new(
+                "vec_f32",
+                DataType::FixedSizeList(Arc::new(Field::new("item", DataType::Float32, true)), 2),
+                false,
+            ),
+            Field::new(
+                "vec_f64",
+                DataType::FixedSizeList(Arc::new(Field::new("item", DataType::Float64, true)), 2),
+                false,
+            ),
+        ]));
+
+        let record_batch_iter = RecordBatchIterator::new(
+            vec![RecordBatch::try_new(
+                schema.clone(),
+                vec![
+                    Arc::new(Int32Array::from_iter_values(0..10)),
+                    Arc::new(Int64Array::from_iter_values(0..10)),
+                    Arc::new(UInt32Array::from_iter_values(0..10)),
+                    Arc::new(StringArray::from_iter_values(vec![
+                        "a", "b", "c", "d", "e", "f", "g", "h", "i", "j",
+                    ])),
+                    Arc::new(LargeStringArray::from_iter_values(vec![
+                        "a", "b", "c", "d", "e", "f", "g", "h", "i", "j",
+                    ])),
+                    Arc::new(Float32Array::from_iter_values(
+                        (0..10).into_iter().map(|i| i as f32),
+                    )),
+                    Arc::new(Float64Array::from_iter_values(
+                        (0..10).into_iter().map(|i| i as f64),
+                    )),
+                    Arc::new(Into::<BooleanArray>::into(vec![
+                        true, false, true, false, true, false, true, false, true, false,
+                    ])),
+                    Arc::new(Date32Array::from_iter_values(0..10)),
+                    Arc::new(TimestampNanosecondArray::from_iter_values(0..10)),
+                    Arc::new(TimestampMillisecondArray::from_iter_values(0..10)),
+                    Arc::new(
+                        create_fixed_size_list(
+                            Float32Array::from_iter_values((0..20).into_iter().map(|i| i as f32)),
+                            2,
+                        )
+                        .unwrap(),
+                    ),
+                    Arc::new(
+                        create_fixed_size_list(
+                            Float64Array::from_iter_values((0..20).into_iter().map(|i| i as f64)),
+                            2,
+                        )
+                        .unwrap(),
+                    ),
+                ],
+            )
+            .unwrap()]
+            .into_iter()
+            .map(Ok),
+            schema.clone(),
+        );
+
+        Dataset::write(record_batch_iter, uri, None).await.unwrap();
+        let mut table = Table::open(uri).await.unwrap();
+
+        // check it can do update for each type
+        let updates: Vec<(&str, &str)> = vec![
+            ("string", "'foo'"),
+            ("large_string", "'large_foo'"),
+            ("int32", "1"),
+            ("int64", "1"),
+            ("uint32", "1"),
+            ("float32", "1.0"),
+            ("float64", "1.0"),
+            ("bool", "true"),
+            ("date32", "1"),
+            ("timestamp_ns", "1"),
+            ("timestamp_ms", "1"),
+            ("vec_f32", "[1.0, 1.0]"),
+            ("vec_f64", "[1.0, 1.0]"),
+        ];
+
+        // for (column, value) in test_cases {
+        table.update(None, updates).await.unwrap();
+
+        let ds_after = Dataset::open(uri).await.unwrap();
+        let mut batches = ds_after
+            .scan()
+            .project(&[
+                "string",
+                "large_string",
+                "int32",
+                "int64",
+                "uint32",
+                "float32",
+                "float64",
+                "bool",
+                "date32",
+                "timestamp_ns",
+                "timestamp_ms",
+                "vec_f32",
+                "vec_f64",
+            ])
+            .unwrap()
+            .try_into_stream()
+            .await
+            .unwrap()
+            .try_collect::<Vec<_>>()
+            .await
+            .unwrap();
+        let batch = batches.pop().unwrap();
+
+        macro_rules! assert_column {
+            ($column:expr, $array_type:ty, $expected:expr) => {
+                let array = $column
+                    .as_any()
+                    .downcast_ref::<$array_type>()
+                    .unwrap()
+                    .iter()
+                    .collect::<Vec<_>>();
+                for v in array {
+                    assert_eq!(v, Some($expected));
+                }
+            };
+        }
+
+        assert_column!(batch.column(0), StringArray, "foo");
+        assert_column!(batch.column(1), LargeStringArray, "large_foo");
+        assert_column!(batch.column(2), Int32Array, 1);
+        assert_column!(batch.column(3), Int64Array, 1);
+        assert_column!(batch.column(4), UInt32Array, 1);
+        assert_column!(batch.column(5), Float32Array, 1.0);
+        assert_column!(batch.column(6), Float64Array, 1.0);
+        assert_column!(batch.column(7), BooleanArray, true);
+        assert_column!(batch.column(8), Date32Array, 1);
+        assert_column!(batch.column(9), TimestampNanosecondArray, 1);
+        assert_column!(batch.column(10), TimestampMillisecondArray, 1);
+
+        let array = batch
+            .column(11)
+            .as_any()
+            .downcast_ref::<FixedSizeListArray>()
+            .unwrap()
+            .iter()
+            .collect::<Vec<_>>();
+        for v in array {
+            let v = v.unwrap();
+            let f32array = v.as_any().downcast_ref::<Float32Array>().unwrap();
+            for v in f32array {
+                assert_eq!(v, Some(1.0));
+            }
+        }
+
+        let array = batch
+            .column(12)
+            .as_any()
+            .downcast_ref::<FixedSizeListArray>()
+            .unwrap()
+            .iter()
+            .collect::<Vec<_>>();
+        for v in array {
+            let v = v.unwrap();
+            let f64array = v.as_any().downcast_ref::<Float64Array>().unwrap();
+            for v in f64array {
+                assert_eq!(v, Some(1.0));
+            }
+        }
+    }
+
     #[tokio::test]
     async fn test_search() {
         let tmp_dir = tempdir().unwrap();
@@ -554,8 +859,8 @@ mod tests {
         let table = Table::open(uri).await.unwrap();
 
         let vector = Float32Array::from_iter_values([0.1, 0.2]);
-        let query = table.search(vector.clone());
-        assert_eq!(vector, query.query_vector);
+        let query = table.search(Some(vector.clone()));
+        assert_eq!(vector, query.query_vector.unwrap());
     }
 
     #[derive(Default, Debug)]