[python] Bump version: 0.3.4 → 0.3.5

Closes #69

Will not pass until https://github.com/lancedb/lance/pull/1585 is
released

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.3.7
+current_version = 0.3.9
 commit = True
 message = Bump version: {current_version} → {new_version}
 tag = True

.github/workflows/npm-publish.yml

@@ -37,14 +37,10 @@ jobs:
           path: |
             node/vectordb-*.tgz
 
-  node-macos:
+  node-macos-x86:
     runs-on: macos-13
     # 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
@@ -54,11 +50,8 @@ 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 x86_64-apple-darwin
       - name: Upload Darwin Artifacts
         uses: actions/upload-artifact@v3
         with:
@@ -66,6 +59,28 @@ jobs:
           path: |
             node/dist/lancedb-vectordb-darwin*.tgz
 
+  node-macos-arm64:
+    runs-on: macos-13-xlarge
+    # Only runs on tags that matches the make-release action
+    if: startsWith(github.ref, 'refs/tags/v')
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+      - name: Install system dependencies
+        run: brew install protobuf
+      - name: Install npm dependencies
+        run: |
+          cd node
+          npm ci
+      - name: Build MacOS native node modules
+        run: bash ci/build_macos_artifacts.sh aarch64-apple-darwin
+      - 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
     runs-on: ${{ matrix.config.runner }}

Cargo.toml

@@ -5,9 +5,10 @@ exclude = ["python"]
 resolver = "2"
 
 [workspace.dependencies]
-lance = { "version" = "=0.8.14", "features" = ["dynamodb"] }
-lance-linalg = { "version" = "=0.8.14" }
-lance-testing = { "version" = "=0.8.14" }
+lance = { "version" = "=0.8.20", "features" = ["dynamodb"] }
+lance-index = { "version" = "=0.8.20" }
+lance-linalg = { "version" = "=0.8.20" }
+lance-testing = { "version" = "=0.8.20" }
 # Note that this one does not include pyarrow
 arrow = { version = "47.0.0", optional = false }
 arrow-array = "47.0"

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

@@ -1,4 +1,5 @@
 site_name: LanceDB Docs
+site_url: https://lancedb.github.io/lancedb/
 repo_url: https://github.com/lancedb/lancedb
 edit_uri: https://github.com/lancedb/lancedb/tree/main/docs/src
 repo_name: lancedb/lancedb
@@ -97,6 +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:
@@ -144,7 +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/src/embeddings/default_embedding_functions.md

@@ -1,7 +1,9 @@
 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
+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.
 
 ### Sentence Transformers
 Here are the parameters that you can set when registering a `sentence-transformers` object, and their default values:
@@ -66,6 +68,56 @@ actual = table.search(query).limit(1).to_pydantic(Words)[0]
 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
+
+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.
+
+More information about the model can be found here - https://github.com/xlang-ai/instructor-embedding
+
+| Argument | Type | Default | Description |
+|---|---|---|---|
+| `name` | `str` | "hkunlp/instructor-base" | The name of the model to use |
+| `batch_size` | `int` | `32` | The batch size to use when generating embeddings |
+| `device` | `str` | `"cpu"` | The device to use when generating embeddings |
+| `show_progress_bar` | `bool` | `True` | Whether to show a progress bar when generating embeddings |
+| `normalize_embeddings` | `bool` | `True` | Whether to normalize the embeddings |
+| `quantize` | `bool` | `False` | Whether to quantize the model |
+| `source_instruction` | `str` | `"represent the docuement for retreival"` | The instruction for the source column |
+| `query_instruction` | `str` | `"represent the document for retreiving the most similar documents"` | The instruction for the query |
+
+
+
+```python
+import lancedb
+from lancedb.pydantic import LanceModel, Vector
+from lancedb.embeddings import get_registry, InstuctorEmbeddingFunction
+
+instructor = get_registry().get("instructor").create(
+                            source_instruction="represent the docuement for retreival",
+                            query_instruction="represent the document for retreiving the most similar documents"
+                            )
+
+class Schema(LanceModel):
+    vector: Vector(instructor.ndims()) = instructor.VectorField()
+    text: str = instructor.SourceField()
+
+db = lancedb.connect("~/.lancedb")
+tbl = db.create_table("test", schema=Schema, mode="overwrite")
+
+texts = [{"text": "Capitalism has been dominant in the Western world since the end of feudalism, but most feel[who?] that..."},
+        {"text": "The disparate impact theory is especially controversial under the Fair Housing Act because the Act..."},
+        {"text": "Disparate impact in United States labor law refers to practices in employment, housing, and other areas that.."}]
+
+tbl.add(texts)
+```
+
 ## Multi-modal embedding functions
 Multi-modal embedding functions allow you query your table using both images and text.
 

docs/src/embeddings/embedding_functions.md

@@ -57,6 +57,19 @@ query_image = Image.open(p)
 table.search(query_image)
 
 ```
+### Rate limit Handling
+`EmbeddingFunction` class wraps the calls for source and query embedding generation inside a rate limit handler that retries the requests with exponential backoff after successive failures. By default the maximum retires is set to 7. You can tune it by setting it to a different number or disable it by setting it to 0.
+Example
+----
+
+```python
+clip = registry.get("open-clip").create() # Defaults to 7 max retries
+clip = registry.get("open-clip").create(max_retries=10) # Increase max retries to 10
+clip = registry.get("open-clip").create(max_retries=0) # Retries disabled
+````
+
+NOTE:
+Embedding functions can also fail due to other errors that have nothing to do with rate limits. This is why the error is also logged.
 
 ### A little fun with PyDantic
 LanceDB is integrated with PyDantic. Infact we've used the integration in the above example to define the schema. It is also being used behing the scene by the embdding function API to ingest useful information as table metadata.

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/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/robots.txt

@@ -0,0 +1 @@
+User-agent: *

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.3.6",
+  "version": "0.3.9",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.3.6",
+      "version": "0.3.9",
       "cpu": [
         "x64",
         "arm64"
@@ -53,11 +53,11 @@
         "uuid": "^9.0.0"
       },
       "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.3.6",
-        "@lancedb/vectordb-darwin-x64": "0.3.6",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.3.6",
-        "@lancedb/vectordb-linux-x64-gnu": "0.3.6",
-        "@lancedb/vectordb-win32-x64-msvc": "0.3.6"
+        "@lancedb/vectordb-darwin-arm64": "0.3.9",
+        "@lancedb/vectordb-darwin-x64": "0.3.9",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.3.9",
+        "@lancedb/vectordb-linux-x64-gnu": "0.3.9",
+        "@lancedb/vectordb-win32-x64-msvc": "0.3.9"
       }
     },
     "node_modules/@apache-arrow/ts": {
@@ -317,9 +317,9 @@
       }
     },
     "node_modules/@lancedb/vectordb-darwin-arm64": {
-      "version": "0.3.6",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.3.6.tgz",
-      "integrity": "sha512-GR5v+4kHUCZ71gVxd3mLsUdlreXPUIbvBgvr+BmEXRbLfc7+JsFUjsRgxmoctQ0mXxkW67Sl7v6kQCWcBLCk/Q==",
+      "version": "0.3.9",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.3.9.tgz",
+      "integrity": "sha512-irtAdfSRQDcfnMnB8T7D0atLFfu1MMZZ1JaxMKu24DDZ8e4IMYKUplxwvWni3241yA9yDE/pliRZCNQbQCEfrg==",
       "cpu": [
         "arm64"
       ],
@@ -329,9 +329,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-darwin-x64": {
-      "version": "0.3.6",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.3.6.tgz",
-      "integrity": "sha512-4qemi4jUXG8jOk7ecECmb0+5Nm0n7YF5/1X9/5uc81I+4What+yhZE9nEsmCGRBqmtuQXkYl35ePvQgj3rCQjQ==",
+      "version": "0.3.9",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.3.9.tgz",
+      "integrity": "sha512-4xXQoPheyIl1P5kRoKmZtaAHFrYdL9pw5yq+r6ewIx0TCemN4LSvzSUTqM5nZl3QPU8FeL0CGD8Gt2gMU0HQ2A==",
       "cpu": [
         "x64"
       ],
@@ -341,9 +341,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.3.6",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.3.6.tgz",
-      "integrity": "sha512-I/lFqIUcXYxJnUG5+DILzUzcfHRGHXL3kl5bs1MGkR9a7F3oPx1IAwY9wkskVnClM7XF9H7MVcFRVTjHUqoUwA==",
+      "version": "0.3.9",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.3.9.tgz",
+      "integrity": "sha512-WIxCZKnLeSlz0PGURtKSX6hJ4CYE2o5P+IFmmuWOWB1uNapQu6zOpea6rNxcRFHUA0IJdO02lVxVfn2hDX4SMg==",
       "cpu": [
         "arm64"
       ],
@@ -353,9 +353,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.3.6",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.3.6.tgz",
-      "integrity": "sha512-UTA/4bpA3UoByhfDx//S5m4o6uQ1qfpneD0PbuftAjkt9eHg0ABIEpZdiTI3xUBdrjXSKZtpVTxOin9X39IBKQ==",
+      "version": "0.3.9",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.3.9.tgz",
+      "integrity": "sha512-bQbcV9adKzYbJLNzDjk9OYsMnT2IjmieLfb4IQ1hj5IUoWfbg80Bd0+gZUnrmrhG6fe56TIriFZYQR9i7TSE9Q==",
       "cpu": [
         "x64"
       ],
@@ -365,9 +365,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.3.6",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.3.6.tgz",
-      "integrity": "sha512-70IS0TX4BpjSX4GP1Pq835cqQ5LZpfOJuBNtGv93OxMTWTVQUxtp2MLNwOR6OJMGNQz6q84NNKrKOSf15ZGwGg==",
+      "version": "0.3.9",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.3.9.tgz",
+      "integrity": "sha512-7EXI7P1QvAfgJNPWWBMDOkoJ696gSBAClcyEJNYg0JV21jVFZRwJVI3bZXflesWduFi/mTuzPkFFA68us1u19A==",
       "cpu": [
         "x64"
       ],
@@ -4869,33 +4869,33 @@
       }
     },
     "@lancedb/vectordb-darwin-arm64": {
-      "version": "0.3.6",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.3.6.tgz",
-      "integrity": "sha512-GR5v+4kHUCZ71gVxd3mLsUdlreXPUIbvBgvr+BmEXRbLfc7+JsFUjsRgxmoctQ0mXxkW67Sl7v6kQCWcBLCk/Q==",
+      "version": "0.3.9",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.3.9.tgz",
+      "integrity": "sha512-irtAdfSRQDcfnMnB8T7D0atLFfu1MMZZ1JaxMKu24DDZ8e4IMYKUplxwvWni3241yA9yDE/pliRZCNQbQCEfrg==",
       "optional": true
     },
     "@lancedb/vectordb-darwin-x64": {
-      "version": "0.3.6",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.3.6.tgz",
-      "integrity": "sha512-4qemi4jUXG8jOk7ecECmb0+5Nm0n7YF5/1X9/5uc81I+4What+yhZE9nEsmCGRBqmtuQXkYl35ePvQgj3rCQjQ==",
+      "version": "0.3.9",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.3.9.tgz",
+      "integrity": "sha512-4xXQoPheyIl1P5kRoKmZtaAHFrYdL9pw5yq+r6ewIx0TCemN4LSvzSUTqM5nZl3QPU8FeL0CGD8Gt2gMU0HQ2A==",
       "optional": true
     },
     "@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.3.6",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.3.6.tgz",
-      "integrity": "sha512-I/lFqIUcXYxJnUG5+DILzUzcfHRGHXL3kl5bs1MGkR9a7F3oPx1IAwY9wkskVnClM7XF9H7MVcFRVTjHUqoUwA==",
+      "version": "0.3.9",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.3.9.tgz",
+      "integrity": "sha512-WIxCZKnLeSlz0PGURtKSX6hJ4CYE2o5P+IFmmuWOWB1uNapQu6zOpea6rNxcRFHUA0IJdO02lVxVfn2hDX4SMg==",
       "optional": true
     },
     "@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.3.6",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.3.6.tgz",
-      "integrity": "sha512-UTA/4bpA3UoByhfDx//S5m4o6uQ1qfpneD0PbuftAjkt9eHg0ABIEpZdiTI3xUBdrjXSKZtpVTxOin9X39IBKQ==",
+      "version": "0.3.9",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.3.9.tgz",
+      "integrity": "sha512-bQbcV9adKzYbJLNzDjk9OYsMnT2IjmieLfb4IQ1hj5IUoWfbg80Bd0+gZUnrmrhG6fe56TIriFZYQR9i7TSE9Q==",
       "optional": true
     },
     "@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.3.6",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.3.6.tgz",
-      "integrity": "sha512-70IS0TX4BpjSX4GP1Pq835cqQ5LZpfOJuBNtGv93OxMTWTVQUxtp2MLNwOR6OJMGNQz6q84NNKrKOSf15ZGwGg==",
+      "version": "0.3.9",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.3.9.tgz",
+      "integrity": "sha512-7EXI7P1QvAfgJNPWWBMDOkoJ696gSBAClcyEJNYg0JV21jVFZRwJVI3bZXflesWduFi/mTuzPkFFA68us1u19A==",
       "optional": true
     },
     "@neon-rs/cli": {

node/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vectordb",
-  "version": "0.3.7",
+  "version": "0.3.9",
   "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": "0.3.7",
-    "@lancedb/vectordb-darwin-x64": "0.3.7",
-    "@lancedb/vectordb-linux-arm64-gnu": "0.3.7",
-    "@lancedb/vectordb-linux-x64-gnu": "0.3.7",
-    "@lancedb/vectordb-win32-x64-msvc": "0.3.7"
+    "@lancedb/vectordb-darwin-arm64": "0.3.9",
+    "@lancedb/vectordb-darwin-x64": "0.3.9",
+    "@lancedb/vectordb-linux-arm64-gnu": "0.3.9",
+    "@lancedb/vectordb-linux-x64-gnu": "0.3.9",
+    "@lancedb/vectordb-win32-x64-msvc": "0.3.9"
   }
 }

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, tableCreateVectorIndex, tableCountRows, tableDelete, tableUpdate, tableCleanupOldVersions, tableCompactFiles, tableListIndices, tableIndexStats } = require('../native.js')
 
 export { Query }
 export type { EmbeddingFunction }
@@ -261,6 +262,39 @@ 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({
+   *   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'}
+   * // ]
+   * ```
+   *
+   */
+  update: (args: UpdateArgs | UpdateSqlArgs) => Promise<void>
+
   /**
    * List the indicies on this table.
    */
@@ -272,6 +306,34 @@ 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 +488,16 @@ 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.
    *
@@ -481,6 +553,31 @@ 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.
    *

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'
 
@@ -154,6 +155,7 @@ 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
@@ -245,6 +247,10 @@ 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> {
+    throw new Error('Not implemented')
+  }
+
   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 +121,20 @@ 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('select only a subset of columns', async function () {
       const uri = await createTestDB()
       const con = await lancedb.connect(uri)
@@ -246,6 +279,46 @@ 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)
@@ -282,7 +355,8 @@ describe('LanceDB client', function () {
       )
       const table = await con.createTable({ name: 'vectors', schema })
       await table.add([{ vector: Array(128).fill(0.1) }])
-      await table.delete('vector IS NOT NULL')
+      // https://github.com/lancedb/lance/issues/1635
+      await table.delete('true')
       const result = await table.search(Array(128).fill(0.1)).execute()
       assert.isEmpty(result)
     })
@@ -527,7 +601,7 @@ 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 = 0.3.3
+current_version = 0.3.5
 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/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

@@ -37,7 +37,10 @@ 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/")
         )
@@ -53,26 +56,62 @@ 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_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,
     ):
-        raise NotImplementedError
+        """Create an index on the table.
+        Currently, the only parameters that matter are
+        the metric and the vector column name.
+
+        Parameters
+        ----------
+        metric : str
+            The metric to use for the index. Default is "L2".
+        vector_column_name : str
+            The name of the vector column. Default is "vector".
+
+        Examples
+        --------
+        >>> 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"
+
+        data = {
+            "column": vector_column_name,
+            "index_type": index_type,
+            "metric_type": metric,
+            "index_cache_size": index_cache_size,
+        }
+        resp = self._conn._loop.run_until_complete(
+            self._conn._client.post(f"/v1/table/{self._name}/create_index/", data=data)
+        )
+        return resp
 
     def add(
         self,
@@ -81,6 +120,28 @@ 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,
@@ -104,6 +165,58 @@ 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:
@@ -111,7 +224,51 @@ class RemoteTable(Table):
         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)

python/lancedb/table.py

@@ -17,7 +17,7 @@ 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
@@ -30,7 +30,7 @@ from .common import DATA, VEC, VECTOR_COLUMN_NAME
 from .embeddings import EmbeddingFunctionConfig, EmbeddingFunctionRegistry
 from .pydantic import LanceModel
 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
 from .utils.events import register_event
 
 if TYPE_CHECKING:
@@ -188,6 +188,7 @@ class Table(ABC):
         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.
 
@@ -212,6 +213,8 @@ class Table(ABC):
         accelerator: str, default None
             If set, use the given accelerator to create the index.
             Only support "cuda" for now.
+        index_cache_size : int, optional
+            The size of the index cache in number of entries. Default value is 256.
         """
         raise NotImplementedError
 
@@ -556,6 +559,7 @@ class LanceTable(Table):
         vector_column_name=VECTOR_COLUMN_NAME,
         replace: bool = True,
         accelerator: Optional[str] = None,
+        index_cache_size: Optional[int] = None,
     ):
         """Create an index on the table."""
         self._dataset.create_index(
@@ -566,6 +570,7 @@ class LanceTable(Table):
             num_sub_vectors=num_sub_vectors,
             replace=replace,
             accelerator=accelerator,
+            index_cache_size=index_cache_size,
         )
         self._reset_dataset()
         register_event("create_index")
@@ -780,7 +785,7 @@ 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
         )
@@ -901,35 +906,42 @@ 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()
@@ -945,18 +957,15 @@ 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 +12,12 @@
 #  limitations under the License.
 
 import os
+from datetime import date, datetime
+from functools import singledispatch
 from typing import Tuple
 from urllib.parse import urlparse
 
+import numpy as np
 import pyarrow.fs as pa_fs
 
 
@@ -88,3 +91,53 @@ 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 = "0.3.3"
+version = "0.3.5"
 dependencies = [
     "deprecation",
-    "pylance==0.8.10",
+    "pylance==0.8.21",
     "ratelimiter~=1.0",
     "retry>=0.9.2",
-    "tqdm>=4.1.0",
+    "tqdm>=4.27.0",
     "aiohttp",
     "pydantic>=1.10",
     "attrs>=21.3.0",

python/tests/test_db.py

@@ -301,6 +301,7 @@ def test_replace_index(tmp_path):
         num_partitions=2,
         num_sub_vectors=4,
         replace=True,
+        index_cache_size=10,
     )
 
 

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
@@ -213,6 +213,7 @@ def test_create_index_method():
                 num_sub_vectors=96,
                 vector_column_name="vector",
                 replace=True,
+                index_cache_size=256,
             )
 
             # Check that the _dataset.create_index method was called
@@ -225,6 +226,7 @@ def test_create_index_method():
                 num_sub_vectors=96,
                 replace=True,
                 accelerator=None,
+                index_cache_size=256,
             )
 
 
@@ -346,14 +348,79 @@ 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

rust/ffi/node/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "vectordb-node"
-version = "0.3.7"
+version = "0.3.9"
 description = "Serverless, low-latency vector database for AI applications"
 license = "Apache-2.0"
 edition = "2018"
@@ -19,6 +19,7 @@ once_cell = "1"
 futures = "0.3"
 half = { workspace = true }
 lance = { workspace = true }
+lance-index = { workspace = true }
 lance-linalg = { workspace = true }
 vectordb = { path = "../../vectordb" }
 tokio = { version = "1.23", features = ["rt-multi-thread"] }

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

@@ -12,7 +12,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use lance::index::vector::{ivf::IvfBuildParams, pq::PQBuildParams};
+use lance_index::vector::{ivf::IvfBuildParams, pq::PQBuildParams};
 use lance_linalg::distance::MetricType;
 use neon::context::FunctionContext;
 use neon::prelude::*;

rust/ffi/node/src/lib.rs

@@ -237,6 +237,7 @@ 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)?;

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) as u64;
+                if limit <= 0 {
+                    panic!("Limit must be a positive integer");
+                }
+                limit
+            });
         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(|q| Float32Array::from(q)))
                 .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

@@ -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.clone())?
+                .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_ref().map(|s| s.as_str());
+
+            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 = "0.3.7"
+version = "0.3.9"
 edition = "2021"
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license = "Apache-2.0"
@@ -21,6 +21,7 @@ object_store = { workspace = true }
 snafu = { workspace = true }
 half = { workspace = true }
 lance = { workspace = true }
+lance-index = { workspace = true }
 lance-linalg = { workspace = true }
 lance-testing = { workspace = true }
 tokio = { version = "1.23", features = ["rt-multi-thread"] }

rust/vectordb/src/index/vector.rs

@@ -13,9 +13,9 @@
 // limitations under the License.
 
 use lance::format::{Index, Manifest};
-use lance::index::vector::ivf::IvfBuildParams;
 use lance::index::vector::pq::PQBuildParams;
 use lance::index::vector::VectorIndexParams;
+use lance_index::vector::ivf::IvfBuildParams;
 use lance_linalg::distance::MetricType;
 
 pub trait VectorIndexBuilder {
@@ -136,9 +136,9 @@ impl VectorIndex {
 mod tests {
     use super::*;
 
-    use lance::index::vector::ivf::IvfBuildParams;
-    use lance::index::vector::pq::PQBuildParams;
     use lance::index::vector::StageParams;
+    use lance_index::vector::ivf::IvfBuildParams;
+    use lance_index::vector::pq::PQBuildParams;
 
     use crate::index::vector::{IvfPQIndexBuilder, VectorIndexBuilder};
 

rust/vectordb/src/io/object_store.rs

@@ -359,7 +359,7 @@ 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 +24,15 @@ 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 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 +46,18 @@ 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,
+            limit: None,
             nprobes: 20,
             refine_factor: None,
             metric_type: None,
             use_index: true,
             filter: None,
             select: None,
+            prefilter: false,
         }
     }
 
@@ -67,13 +69,17 @@ 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(crate::table::VECTOR_COLUMN_NAME, 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));
@@ -87,7 +93,7 @@ impl Query {
     ///
     /// * `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 +103,7 @@ 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 +164,11 @@ impl Query {
         self.select = columns;
         self
     }
+
+    pub fn prefilter(mut self, prefilter: bool) -> Query {
+        self.prefilter = prefilter;
+        self
+    }
 }
 
 #[cfg(test)]
@@ -165,9 +176,11 @@ mod tests {
     use std::sync::Arc;
 
     use super::*;
-    use arrow_array::{Float32Array, RecordBatch, RecordBatchIterator, RecordBatchReader};
+    use arrow_array::{Float32Array, RecordBatch, RecordBatchIterator, RecordBatchReader, cast::AsArray, Int32Array};
     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 +189,7 @@ 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 +203,8 @@ 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 +213,64 @@ 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

@@ -13,6 +13,8 @@
 // limitations under the License.
 
 use chrono::Duration;
+use lance::dataset::builder::DatasetBuilder;
+use lance_index::IndexType;
 use std::sync::Arc;
 
 use arrow_array::{Float32Array, RecordBatchReader};
@@ -21,8 +23,8 @@ use lance::dataset::cleanup::RemovalStats;
 use lance::dataset::optimize::{
     compact_files, CompactionMetrics, CompactionOptions, IndexRemapperOptions,
 };
-use lance::dataset::{Dataset, WriteParams};
-use lance::index::{DatasetIndexExt, IndexType};
+use lance::dataset::{Dataset, UpdateBuilder, WriteParams};
+use lance::index::DatasetIndexExt;
 use lance::io::object_store::WrappingObjectStore;
 use std::path::Path;
 
@@ -96,7 +98,10 @@ impl Table {
             Some(wrapper) => params.patch_with_store_wrapper(wrapper)?,
             None => params,
         };
-        let dataset = Dataset::open_with_params(uri, &params)
+
+        let dataset = DatasetBuilder::from_uri(uri)
+            .with_read_params(params)
+            .load()
             .await
             .map_err(|e| match e {
                 lance::Error::DatasetNotFound { .. } => Error::TableNotFound {
@@ -303,10 +308,14 @@ 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?)
@@ -333,6 +342,27 @@ 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
@@ -408,15 +438,18 @@ 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::ivf::IvfBuildParams;
     use lance::index::vector::pq::PQBuildParams;
     use lance::io::object_store::{ObjectStoreParams, WrappingObjectStore};
+    use lance_index::vector::ivf::IvfBuildParams;
     use rand::Rng;
     use tempfile::tempdir;
 
@@ -535,6 +568,272 @@ 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();
@@ -549,8 +848,8 @@ 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)]