tests pass; still need catalog

Closes #639 
Closes #638

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.3.11
+current_version = 0.4.0
 commit = True
 message = Bump version: {current_version} → {new_version}
 tag = True

.github/ISSUE_TEMPLATE/bug-node.yml

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

.github/ISSUE_TEMPLATE/bug-python.yml

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

.github/ISSUE_TEMPLATE/config.yml

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

.github/ISSUE_TEMPLATE/documentation.yml

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

.github/ISSUE_TEMPLATE/feature.yml

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

.github/workflows/python.yml

@@ -44,12 +44,19 @@ jobs:
       run: pytest -m "not slow" -x -v --durations=30 tests
     - name: doctest
       run: pytest --doctest-modules lancedb
-  mac:
+  platform:
+    name: "Platform: ${{ matrix.config.name }}"
     timeout-minutes: 30
     strategy:
       matrix:
-        mac-runner: [ "macos-13", "macos-13-xlarge" ]
-    runs-on: "${{ matrix.mac-runner }}"
+        config: 
+          - name: x86 Mac
+            runner: macos-13
+          - name: Arm Mac
+            runner: macos-13-xlarge
+          - name: x86 Windows
+            runner: windows-latest
+    runs-on: "${{ matrix.config.runner }}"
     defaults:
       run:
         shell: bash

.github/workflows/rust.yml

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

Cargo.toml

@@ -5,24 +5,24 @@ exclude = ["python"]
 resolver = "2"
 
 [workspace.dependencies]
-lance = { "version" = "=0.9.0", "features" = ["dynamodb"] }
-lance-index = { "version" = "=0.9.0" }
-lance-linalg = { "version" = "=0.9.0" }
-lance-testing = { "version" = "=0.9.0" }
+lance = { "version" = "=0.9.1", "features" = ["dynamodb"] }
+lance-index = { "version" = "=0.9.1" }
+lance-linalg = { "version" = "=0.9.1" }
+lance-testing = { "version" = "=0.9.1" }
 # Note that this one does not include pyarrow
-arrow = { version = "47.0.0", optional = false }
-arrow-array = "47.0"
-arrow-data = "47.0"
-arrow-ipc = "47.0"
-arrow-ord = "47.0"
-arrow-schema = "47.0"
-arrow-arith = "47.0"
-arrow-cast = "47.0"
+arrow = { version = "49.0.0", optional = false }
+arrow-array = "49.0"
+arrow-data = "49.0"
+arrow-ipc = "49.0"
+arrow-ord = "49.0"
+arrow-schema = "49.0"
+arrow-arith = "49.0"
+arrow-cast = "49.0"
 chrono = "0.4.23"
 half = { "version" = "=2.3.1", default-features = false, features = [
     "num-traits",
 ] }
 log = "0.4"
-object_store = "0.7.1"
+object_store = "0.8.0"
 snafu = "0.7.4"
 url = "2"

docs/requirements.txt

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

docs/src/basic.md

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

docs/src/guides/tables.md

@@ -201,8 +201,8 @@ A Table is a collection of Records in a LanceDB Database. You can follow along o
     ```javascript
     data
     const tb = await db.createTable("my_table",
-                  data=[{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
-                        {"vector": [5.9, 26.5], "item": "bar", "price": 20.0}])
+      [{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
+       {"vector": [5.9, 26.5], "item": "bar", "price": 20.0}])
     ```
 
     !!! info "Note"

docs/src/search.md

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

docs/src/sql.md

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

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.3.11",
+  "version": "0.4.0",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.3.11",
+      "version": "0.4.0",
       "cpu": [
         "x64",
         "arm64"
@@ -53,11 +53,11 @@
         "uuid": "^9.0.0"
       },
       "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.3.11",
-        "@lancedb/vectordb-darwin-x64": "0.3.11",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.3.11",
-        "@lancedb/vectordb-linux-x64-gnu": "0.3.11",
-        "@lancedb/vectordb-win32-x64-msvc": "0.3.11"
+        "@lancedb/vectordb-darwin-arm64": "0.4.0",
+        "@lancedb/vectordb-darwin-x64": "0.4.0",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.4.0",
+        "@lancedb/vectordb-linux-x64-gnu": "0.4.0",
+        "@lancedb/vectordb-win32-x64-msvc": "0.4.0"
       }
     },
     "node_modules/@apache-arrow/ts": {
@@ -317,9 +317,9 @@
       }
     },
     "node_modules/@lancedb/vectordb-darwin-arm64": {
-      "version": "0.3.11",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.3.11.tgz",
-      "integrity": "sha512-N0Ak0jWmSh+QIUJKgtD85+/N0UMBZxaHrd9leusWgjEdtZdQqyzd6VWYAFPR6W6p8tt1hUZiuTRQ6ugfNhyEsg==",
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.4.0.tgz",
+      "integrity": "sha512-cP6zGtBWXEcJHCI4uLNIP5ILtRvexvwmL8Uri1dnHG8dT8g12Ykug3BHO6Wt6wp/xASd2jJRIF/VAJsN9IeP1A==",
       "cpu": [
         "arm64"
       ],
@@ -329,9 +329,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-darwin-x64": {
-      "version": "0.3.11",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.3.11.tgz",
-      "integrity": "sha512-vugA+Z4XDrV1gFW5PfqJImw0w84NpGrZsaTZ9afw2oc5a37alx5zOoHEoBQimaX88j+YjWme38h3B98qoNTP5w==",
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.4.0.tgz",
+      "integrity": "sha512-ig0gV5ol1sFe2lb1HOatK0rizyj9I91WbnH79i7OdUl3nAQIcWm70CnxrPLtx0DS2NTGh2kFJbYCWcaUlu6YfA==",
       "cpu": [
         "x64"
       ],
@@ -341,9 +341,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.3.11",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.3.11.tgz",
-      "integrity": "sha512-mArXy17URht7cTdGgNc+yL6BOxvK4vAtNaPh68WBOy7e438l6++s2E4bZyaeyeoIv8sPENDmJZzBr4YuBEc7yw==",
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.4.0.tgz",
+      "integrity": "sha512-gMXIDT2kriAPDwWIRKXdaTCNdOeFGEok1S9Y30AOruHXddW1vCIo4JNJIYbBqHnwAeI4wI3ae6GRCFaf1UxO3g==",
       "cpu": [
         "arm64"
       ],
@@ -353,9 +353,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.3.11",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.3.11.tgz",
-      "integrity": "sha512-AoF0f/mUP1d2r5nirLQiajHBVnhsYCD/vDGUlTmLWH4lX4v9zVqlh9HmXjpLBcaK4klGmt5CBmcb+tj5v2/ySA==",
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.4.0.tgz",
+      "integrity": "sha512-ZQ3lDrDSz1IKdx/mS9Lz08agFO+OD5oSFrrcFNCoT1+H93eS1mCLdmCoEARu3jKbx0tMs38l5J9yXZ2QmJye3w==",
       "cpu": [
         "x64"
       ],
@@ -365,9 +365,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.3.11",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.3.11.tgz",
-      "integrity": "sha512-Zq+JHtkaGaoozHcOdXid3jRkEj6u2d1C0VD+Wg+7AIpRokzYt5zcKWPzjDnqoRuD+VTv6YFjYN58RmYwa2Ktiw==",
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.4.0.tgz",
+      "integrity": "sha512-toNcNwBRE1sdsSf5hr7W8QiqZ33csc/knVEek4CyvYkZHJGh4Z6WI+DJUIASo5wzUez4TX7qUPpRPL9HuaPMCg==",
       "cpu": [
         "x64"
       ],
@@ -4869,33 +4869,33 @@
       }
     },
     "@lancedb/vectordb-darwin-arm64": {
-      "version": "0.3.11",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.3.11.tgz",
-      "integrity": "sha512-N0Ak0jWmSh+QIUJKgtD85+/N0UMBZxaHrd9leusWgjEdtZdQqyzd6VWYAFPR6W6p8tt1hUZiuTRQ6ugfNhyEsg==",
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.4.0.tgz",
+      "integrity": "sha512-cP6zGtBWXEcJHCI4uLNIP5ILtRvexvwmL8Uri1dnHG8dT8g12Ykug3BHO6Wt6wp/xASd2jJRIF/VAJsN9IeP1A==",
       "optional": true
     },
     "@lancedb/vectordb-darwin-x64": {
-      "version": "0.3.11",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.3.11.tgz",
-      "integrity": "sha512-vugA+Z4XDrV1gFW5PfqJImw0w84NpGrZsaTZ9afw2oc5a37alx5zOoHEoBQimaX88j+YjWme38h3B98qoNTP5w==",
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.4.0.tgz",
+      "integrity": "sha512-ig0gV5ol1sFe2lb1HOatK0rizyj9I91WbnH79i7OdUl3nAQIcWm70CnxrPLtx0DS2NTGh2kFJbYCWcaUlu6YfA==",
       "optional": true
     },
     "@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.3.11",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.3.11.tgz",
-      "integrity": "sha512-mArXy17URht7cTdGgNc+yL6BOxvK4vAtNaPh68WBOy7e438l6++s2E4bZyaeyeoIv8sPENDmJZzBr4YuBEc7yw==",
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.4.0.tgz",
+      "integrity": "sha512-gMXIDT2kriAPDwWIRKXdaTCNdOeFGEok1S9Y30AOruHXddW1vCIo4JNJIYbBqHnwAeI4wI3ae6GRCFaf1UxO3g==",
       "optional": true
     },
     "@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.3.11",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.3.11.tgz",
-      "integrity": "sha512-AoF0f/mUP1d2r5nirLQiajHBVnhsYCD/vDGUlTmLWH4lX4v9zVqlh9HmXjpLBcaK4klGmt5CBmcb+tj5v2/ySA==",
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.4.0.tgz",
+      "integrity": "sha512-ZQ3lDrDSz1IKdx/mS9Lz08agFO+OD5oSFrrcFNCoT1+H93eS1mCLdmCoEARu3jKbx0tMs38l5J9yXZ2QmJye3w==",
       "optional": true
     },
     "@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.3.11",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.3.11.tgz",
-      "integrity": "sha512-Zq+JHtkaGaoozHcOdXid3jRkEj6u2d1C0VD+Wg+7AIpRokzYt5zcKWPzjDnqoRuD+VTv6YFjYN58RmYwa2Ktiw==",
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.4.0.tgz",
+      "integrity": "sha512-toNcNwBRE1sdsSf5hr7W8QiqZ33csc/knVEek4CyvYkZHJGh4Z6WI+DJUIASo5wzUez4TX7qUPpRPL9HuaPMCg==",
       "optional": true
     },
     "@neon-rs/cli": {

node/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vectordb",
-  "version": "0.3.11",
+  "version": "0.4.0",
   "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": "0.3.11",
-    "@lancedb/vectordb-darwin-x64": "0.3.11",
-    "@lancedb/vectordb-linux-arm64-gnu": "0.3.11",
-    "@lancedb/vectordb-linux-x64-gnu": "0.3.11",
-    "@lancedb/vectordb-win32-x64-msvc": "0.3.11"
+    "@lancedb/vectordb-darwin-arm64": "0.4.0",
+    "@lancedb/vectordb-darwin-x64": "0.4.0",
+    "@lancedb/vectordb-linux-arm64-gnu": "0.4.0",
+    "@lancedb/vectordb-linux-x64-gnu": "0.4.0",
+    "@lancedb/vectordb-win32-x64-msvc": "0.4.0"
   }
 }

node/src/index.ts

@@ -744,6 +744,11 @@ export interface IvfPQIndexConfig {
    */
   replace?: boolean
 
+  /**
+   * Cache size of the index
+   */
+  index_cache_size?: number
+
   type: 'ivf_pq'
 }
 

node/src/remote/index.ts

@@ -57,8 +57,8 @@ export class RemoteConnection implements Connection {
     return 'db://' + this._client.uri
   }
 
-  async tableNames (): Promise<string[]> {
-    const response = await this._client.get('/v1/table/')
+  async tableNames (pageToken: string = '', limit: number = 10): Promise<string[]> {
+    const response = await this._client.get('/v1/table/', { limit, page_token: pageToken })
     return response.data.tables
   }
 
@@ -195,6 +195,17 @@ export class RemoteTable<T = number[]> implements Table<T> {
     return this._name
   }
 
+  get schema (): Promise<any> {
+    return this._client.post(`/v1/table/${this._name}/describe/`).then(res => {
+      if (res.status !== 200) {
+        throw new Error(`Server Error, status: ${res.status}, ` +
+        // eslint-disable-next-line @typescript-eslint/restrict-template-expressions
+          `message: ${res.statusText}: ${res.data}`)
+      }
+      return res.data?.schema
+    })
+  }
+
   search (query: T): Query<T> {
     return new RemoteQuery(query, this._client, this._name)//, this._embeddings_new)
   }
@@ -235,8 +246,41 @@ export class RemoteTable<T = number[]> implements Table<T> {
     return data.length
   }
 
-  async createIndex (indexParams: VectorIndexParams): Promise<any> {
-    throw new Error('Not implemented')
+  async createIndex (indexParams: VectorIndexParams): Promise<void> {
+    const unsupportedParams = [
+      'index_name',
+      'num_partitions',
+      'max_iters',
+      'use_opq',
+      'num_sub_vectors',
+      'num_bits',
+      'max_opq_iters',
+      'replace'
+    ]
+    for (const param of unsupportedParams) {
+      // eslint-disable-next-line @typescript-eslint/strict-boolean-expressions
+      if (indexParams[param as keyof VectorIndexParams]) {
+        throw new Error(`${param} is not supported for remote connections`)
+      }
+    }
+
+    const column = indexParams.column ?? 'vector'
+    const indexType = 'vector' // only vector index is supported for remote connections
+    const metricType = indexParams.metric_type ?? 'L2'
+    const indexCacheSize = indexParams ?? null
+
+    const data = {
+      column,
+      index_type: indexType,
+      metric_type: metricType,
+      index_cache_size: indexCacheSize
+    }
+    const res = await this._client.post(`/v1/table/${this._name}/create_index/`, data)
+    if (res.status !== 200) {
+      throw new Error(`Server Error, status: ${res.status}, ` +
+      // eslint-disable-next-line @typescript-eslint/restrict-template-expressions
+        `message: ${res.statusText}: ${res.data}`)
+    }
   }
 
   async countRows (): Promise<number> {

python/lancedb/db.py

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

python/lancedb/fts.py

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

python/lancedb/query.py

@@ -185,14 +185,40 @@ class LanceQueryBuilder(ABC):
         """
         return self.to_pandas()
 
-    def to_pandas(self) -> "pd.DataFrame":
+    def to_pandas(self, flatten: Optional[Union[int, bool]] = None) -> "pd.DataFrame":
         """
         Execute the query and return the results as a pandas DataFrame.
         In addition to the selected columns, LanceDB also returns a vector
         and also the "_distance" column which is the distance between the query
         vector and the returned vector.
+
+        Parameters
+        ----------
+        flatten: Optional[Union[int, bool]]
+            If flatten is True, flatten all nested columns.
+            If flatten is an integer, flatten the nested columns up to the
+            specified depth.
+            If unspecified, do not flatten the nested columns.
         """
-        return self.to_arrow().to_pandas()
+        tbl = self.to_arrow()
+        if flatten is True:
+            while True:
+                tbl = tbl.flatten()
+                has_struct = False
+                # loop through all columns to check if there is any struct column
+                if any(pa.types.is_struct(col.type) for col in tbl.schema):
+                    continue
+                else:
+                    break
+        elif isinstance(flatten, int):
+            if flatten <= 0:
+                raise ValueError(
+                    "Please specify a positive integer for flatten or the boolean value `True`"
+                )
+            while flatten > 0:
+                tbl = tbl.flatten()
+                flatten -= 1
+        return tbl.to_pandas()
 
     @abstractmethod
     def to_arrow(self) -> pa.Table:

python/lancedb/table.py

@@ -23,6 +23,7 @@ import lance
 import numpy as np
 import pyarrow as pa
 import pyarrow.compute as pc
+import pyarrow.fs as pa_fs
 from lance import LanceDataset
 from lance.vector import vec_to_table
 
@@ -30,7 +31,7 @@ from .common import DATA, VEC, VECTOR_COLUMN_NAME
 from .embeddings import EmbeddingFunctionConfig, EmbeddingFunctionRegistry
 from .pydantic import LanceModel, model_to_dict
 from .query import LanceQueryBuilder, Query
-from .util import fs_from_uri, safe_import_pandas, value_to_sql
+from .util import fs_from_uri, safe_import_pandas, value_to_sql, join_uri
 from .utils.events import register_event
 
 if TYPE_CHECKING:
@@ -396,14 +397,6 @@ class LanceTable(Table):
         self.name = name
         self._version = version
 
-    def _reset_dataset(self, version=None):
-        try:
-            if "_dataset" in self.__dict__:
-                del self.__dict__["_dataset"]
-            self._version = version
-        except AttributeError:
-            pass
-
     @property
     def schema(self) -> pa.Schema:
         """Return the schema of the table.
@@ -412,16 +405,16 @@ class LanceTable(Table):
         -------
         pa.Schema
             A PyArrow schema object."""
-        return self._dataset.schema
+        return self.to_lance().schema
 
     def list_versions(self):
         """List all versions of the table"""
-        return self._dataset.versions()
+        return self.to_lance().versions()
 
     @property
     def version(self) -> int:
         """Get the current version of the table"""
-        return self._dataset.version
+        return self.to_lance().version
 
     def checkout(self, version: int):
         """Checkout a version of the table. This is an in-place operation.
@@ -454,14 +447,12 @@ class LanceTable(Table):
                vector    type
         0  [1.1, 0.9]  vector
         """
-        max_ver = max([v["version"] for v in self._dataset.versions()])
+        max_ver = max([v["version"] for v in self.to_lance().versions()])
         if version < 1 or version > max_ver:
             raise ValueError(f"Invalid version {version}")
-        self._reset_dataset(version=version)
 
         try:
-            # Accessing the property updates the cached value
-            _ = self._dataset
+            self.to_lance().checkout(version)
         except Exception as e:
             if "not found" in str(e):
                 raise ValueError(
@@ -504,7 +495,7 @@ class LanceTable(Table):
         >>> len(table.list_versions())
         4
         """
-        max_ver = max([v["version"] for v in self._dataset.versions()])
+        max_ver = max([v["version"] for v in self.to_lance().versions()])
         if version is None:
             version = self.version
         elif version < 1 or version > max_ver:
@@ -516,11 +507,10 @@ class LanceTable(Table):
             # no-op if restoring the latest version
             return
 
-        self._dataset.restore()
-        self._reset_dataset()
+        self.to_lance().restore()
 
     def __len__(self):
-        return self._dataset.count_rows()
+        return self.to_lance().count_rows()
 
     def __repr__(self) -> str:
         return f"LanceTable({self.name})"
@@ -530,7 +520,7 @@ class LanceTable(Table):
 
     def head(self, n=5) -> pa.Table:
         """Return the first n rows of the table."""
-        return self._dataset.head(n)
+        return self.to_lance().head(n)
 
     def to_pandas(self) -> "pd.DataFrame":
         """Return the table as a pandas DataFrame.
@@ -547,11 +537,11 @@ class LanceTable(Table):
         Returns
         -------
         pa.Table"""
-        return self._dataset.to_table()
+        return self.to_lance().to_table()
 
     @property
     def _dataset_uri(self) -> str:
-        return os.path.join(self._conn.uri, f"{self.name}.lance")
+        return join_uri(self._conn.uri, f"{self.name}.lance")
 
     def create_index(
         self,
@@ -574,10 +564,11 @@ class LanceTable(Table):
             accelerator=accelerator,
             index_cache_size=index_cache_size,
         )
-        self._reset_dataset()
         register_event("create_index")
 
-    def create_fts_index(self, field_names: Union[str, List[str]]):
+    def create_fts_index(
+        self, field_names: Union[str, List[str]], *, replace: bool = False
+    ):
         """Create a full-text search index on the table.
 
         Warning - this API is highly experimental and is highly likely to change
@@ -587,17 +578,35 @@ class LanceTable(Table):
         ----------
         field_names: str or list of str
             The name(s) of the field to index.
+        replace: bool, default False
+            If True, replace the existing index if it exists. Note that this is
+            not yet an atomic operation; the index will be temporarily
+            unavailable while the new index is being created.
         """
         from .fts import create_index, populate_index
 
         if isinstance(field_names, str):
             field_names = [field_names]
+
+        fs, path = fs_from_uri(self._get_fts_index_path())
+        index_exists = fs.get_file_info(path).type != pa_fs.FileType.NotFound
+        if index_exists:
+            if not replace:
+                raise ValueError(
+                    f"Index already exists. Use replace=True to overwrite."
+                )
+            try:
+                fs.delete_dir(path)
+            except FileNotFoundError as e:
+                if "Cannot get information for path" in str(e):
+                    pass
+
         index = create_index(self._get_fts_index_path(), field_names)
         populate_index(index, self, field_names)
         register_event("create_fts_index")
 
     def _get_fts_index_path(self):
-        return os.path.join(self._dataset_uri, "_indices", "tantivy")
+        return join_uri(self._dataset_uri, "_indices", "tantivy")
 
     @cached_property
     def _dataset(self) -> LanceDataset:
@@ -645,8 +654,7 @@ class LanceTable(Table):
             on_bad_vectors=on_bad_vectors,
             fill_value=fill_value,
         )
-        lance.write_dataset(data, self._dataset_uri, schema=self.schema, mode=mode)
-        self._reset_dataset()
+        self.to_lance().write(data, mode=mode)
         register_event("add")
 
     def merge(
@@ -707,10 +715,9 @@ class LanceTable(Table):
             other_table = other_table.to_lance()
         if isinstance(other_table, LanceDataset):
             other_table = other_table.to_table()
-        self._dataset.merge(
+        self.to_lance().merge(
             other_table, left_on=left_on, right_on=right_on, schema=schema
         )
-        self._reset_dataset()
         register_event("merge")
 
     @cached_property
@@ -913,7 +920,7 @@ class LanceTable(Table):
         return tbl
 
     def delete(self, where: str):
-        self._dataset.delete(where)
+        self.to_lance().delete(where)
 
     def update(
         self,
@@ -968,7 +975,6 @@ class LanceTable(Table):
             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")
 
     def _execute_query(self, query: Query) -> pa.Table:

python/lancedb/util.py

@@ -14,7 +14,8 @@
 import os
 from datetime import date, datetime
 from functools import singledispatch
-from typing import Tuple
+import pathlib
+from typing import Tuple, Union
 from urllib.parse import urlparse
 
 import numpy as np
@@ -62,6 +63,12 @@ def get_uri_location(uri: str) -> str:
     str: Location part of the URL, without scheme
     """
     parsed = urlparse(uri)
+    if len(parsed.scheme) == 1:
+        # Windows drive names are parsed as the scheme
+        # e.g. "c:\path" -> ParseResult(scheme="c", netloc="", path="/path", ...)
+        # So we add special handling here for schemes that are a single character
+        return uri
+
     if not parsed.netloc:
         return parsed.path
     else:
@@ -84,6 +91,29 @@ def fs_from_uri(uri: str) -> Tuple[pa_fs.FileSystem, str]:
     return pa_fs.FileSystem.from_uri(uri)
 
 
+def join_uri(base: Union[str, pathlib.Path], *parts: str) -> str:
+    """
+    Join a URI with multiple parts, handles both local and remote paths
+
+    Parameters
+    ----------
+    base : str
+        The base URI
+    parts : str
+        The parts to join to the base URI, each separated by the
+        appropriate path separator for the URI scheme and OS
+    """
+    if isinstance(base, pathlib.Path):
+        return base.joinpath(*parts)
+    base = str(base)
+    if get_uri_scheme(base) == "file":
+        # using pathlib for local paths make this windows compatible
+        # `get_uri_scheme` returns `file` for windows drive names (e.g. `c:\path`)
+        return str(pathlib.Path(base, *parts))
+    # for remote paths, just use os.path.join
+    return "/".join([p.rstrip("/") for p in [base, *parts]])
+
+
 def safe_import_pandas():
     try:
         import pandas as pd

python/pyproject.toml

@@ -3,7 +3,7 @@ name = "lancedb"
 version = "0.4.0"
 dependencies = [
     "deprecation",
-    "pylance==0.9.0",
+    "pylance==0.9.1",
     "ratelimiter~=1.0",
     "retry>=0.9.2",
     "tqdm>=4.27.0",

python/tests/test_fts.py

@@ -43,7 +43,15 @@ def table(tmp_path) -> ldb.table.LanceTable:
         for _ in range(100)
     ]
     table = db.create_table(
-        "test", data=pd.DataFrame({"vector": vectors, "text": text, "text2": text})
+        "test",
+        data=pd.DataFrame(
+            {
+                "vector": vectors,
+                "text": text,
+                "text2": text,
+                "nested": [{"text": t} for t in text],
+            }
+        ),
     )
     return table
 
@@ -75,6 +83,24 @@ def test_create_index_from_table(tmp_path, table):
     assert len(df) == 10
     assert "text" in df.columns
 
+    # Check whether it can be updated
+    table.add(
+        [
+            {
+                "vector": np.random.randn(128),
+                "text": "gorilla",
+                "text2": "gorilla",
+                "nested": {"text": "gorilla"},
+            }
+        ]
+    )
+
+    table.create_fts_index("text", replace=True)
+    assert len(table.search("gorilla").limit(1).to_pandas()) == 1
+
+    with pytest.raises(ValueError, match="already exists"):
+        table.create_fts_index("text")
+
 
 def test_create_index_multiple_columns(tmp_path, table):
     table.create_fts_index(["text", "text2"])
@@ -89,3 +115,9 @@ def test_empty_rs(tmp_path, table, mocker):
     mocker.patch("lancedb.fts.search_index", return_value=([], []))
     df = table.search("puppy").limit(10).to_pandas()
     assert len(df) == 0
+
+
+def test_nested_schema(tmp_path, table):
+    table.create_fts_index("nested.text")
+    rs = table.search("puppy").limit(10).to_list()
+    assert len(rs) == 10

python/tests/test_table.py

@@ -21,8 +21,8 @@ import lance
 import numpy as np
 import pandas as pd
 import pyarrow as pa
-from pydantic import BaseModel
 import pytest
+from pydantic import BaseModel
 
 from lancedb.conftest import MockTextEmbeddingFunction
 from lancedb.db import LanceDBConnection
@@ -144,9 +144,13 @@ def test_add(db):
 def test_add_pydantic_model(db):
     # https://github.com/lancedb/lancedb/issues/562
 
+    class Metadata(BaseModel):
+        source: str
+        timestamp: datetime
+
     class Document(BaseModel):
         content: str
-        source: str
+        meta: Metadata
 
     class LanceSchema(LanceModel):
         id: str
@@ -162,13 +166,21 @@ def test_add_pydantic_model(db):
         id="id",
         vector=[0.0, 0.0],
         li=[1, 2, 3],
-        payload=Document(content="foo", source="bar"),
+        payload=Document(
+            content="foo", meta=Metadata(source="bar", timestamp=datetime.now())
+        ),
     )
     tbl.add([expected])
 
     result = tbl.search([0.0, 0.0]).limit(1).to_pydantic(LanceSchema)[0]
     assert result == expected
 
+    flattened = tbl.search([0.0, 0.0]).limit(1).to_pandas(flatten=1)
+    assert len(flattened.columns) == 6  # _distance is automatically added
+
+    really_flattened = tbl.search([0.0, 0.0]).limit(1).to_pandas(flatten=True)
+    assert len(really_flattened.columns) == 7
+
 
 def _add(table, schema):
     # table = LanceTable(db, "test")
@@ -214,39 +226,38 @@ def test_versioning(db):
 
 
 def test_create_index_method():
-    with patch.object(LanceTable, "_reset_dataset", return_value=None):
-        with patch.object(
-            LanceTable, "_dataset", new_callable=PropertyMock
-        ) as mock_dataset:
-            # Setup mock responses
-            mock_dataset.return_value.create_index.return_value = None
+    with patch.object(
+        LanceTable, "_dataset", new_callable=PropertyMock
+    ) as mock_dataset:
+        # Setup mock responses
+        mock_dataset.return_value.create_index.return_value = None
 
-            # Create a LanceTable object
-            connection = LanceDBConnection(uri="mock.uri")
-            table = LanceTable(connection, "test_table")
+        # Create a LanceTable object
+        connection = LanceDBConnection(uri="mock.uri")
+        table = LanceTable(connection, "test_table")
 
-            # Call the create_index method
-            table.create_index(
-                metric="L2",
-                num_partitions=256,
-                num_sub_vectors=96,
-                vector_column_name="vector",
-                replace=True,
-                index_cache_size=256,
-            )
+        # Call the create_index method
+        table.create_index(
+            metric="L2",
+            num_partitions=256,
+            num_sub_vectors=96,
+            vector_column_name="vector",
+            replace=True,
+            index_cache_size=256,
+        )
 
-            # Check that the _dataset.create_index method was called
-            # with the right parameters
-            mock_dataset.return_value.create_index.assert_called_once_with(
-                column="vector",
-                index_type="IVF_PQ",
-                metric="L2",
-                num_partitions=256,
-                num_sub_vectors=96,
-                replace=True,
-                accelerator=None,
-                index_cache_size=256,
-            )
+        # Check that the _dataset.create_index method was called
+        # with the right parameters
+        mock_dataset.return_value.create_index.assert_called_once_with(
+            column="vector",
+            index_type="IVF_PQ",
+            metric="L2",
+            num_partitions=256,
+            num_sub_vectors=96,
+            replace=True,
+            accelerator=None,
+            index_cache_size=256,
+        )
 
 
 def test_add_with_nans(db):

python/tests/test_util.py

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

rust/ffi/node/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "vectordb-node"
-version = "0.3.11"
+version = "0.4.0"
 description = "Serverless, low-latency vector database for AI applications"
 license = "Apache-2.0"
 edition = "2018"

rust/ffi/node/src/error.rs

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

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

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

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

@@ -47,15 +47,15 @@ fn f64_to_u32_safe(n: f64, key: &str) -> Result<u32> {
     use conv::*;
 
     n.approx_as::<u32>().map_err(|e| match e {
-        FloatError::NegOverflow(_) => Error::RangeError {
+        FloatError::NegOverflow(_) => Error::OutOfRange {
             name: key.into(),
             message: "must be > 0".to_string(),
         },
-        FloatError::PosOverflow(_) => Error::RangeError {
+        FloatError::PosOverflow(_) => Error::OutOfRange {
             name: key.into(),
             message: format!("must be < {}", u32::MAX),
         },
-        FloatError::NotANumber(_) => Error::RangeError {
+        FloatError::NotANumber(_) => Error::OutOfRange {
             name: key.into(),
             message: "not a valid number".to_string(),
         },
@@ -66,15 +66,15 @@ fn f64_to_usize_safe(n: f64, key: &str) -> Result<usize> {
     use conv::*;
 
     n.approx_as::<usize>().map_err(|e| match e {
-        FloatError::NegOverflow(_) => Error::RangeError {
+        FloatError::NegOverflow(_) => Error::OutOfRange {
             name: key.into(),
             message: "must be > 0".to_string(),
         },
-        FloatError::PosOverflow(_) => Error::RangeError {
+        FloatError::PosOverflow(_) => Error::OutOfRange {
             name: key.into(),
             message: format!("must be < {}", usize::MAX),
         },
-        FloatError::NotANumber(_) => Error::RangeError {
+        FloatError::NotANumber(_) => Error::OutOfRange {
             name: key.into(),
             message: "not a valid number".to_string(),
         },

rust/ffi/node/src/query.rs

@@ -25,11 +25,11 @@ impl JsQuery {
         let limit = query_obj
             .get_opt::<JsNumber, _, _>(&mut cx, "_limit")?
             .map(|value| {
-                let limit = value.value(&mut cx) as u64;
-                if limit <= 0 {
+                let limit = value.value(&mut cx);
+                if limit <= 0.0 {
                     panic!("Limit must be a positive integer");
                 }
-                limit
+                limit as u64
             });
         let select = query_obj
             .get_opt::<JsArray, _, _>(&mut cx, "_select")?
@@ -73,7 +73,7 @@ impl JsQuery {
 
         rt.spawn(async move {
             let mut builder = table
-                .search(query.map(|q| Float32Array::from(q)))
+                .search(query.map(Float32Array::from))
                 .refine_factor(refine_factor)
                 .nprobes(nprobes)
                 .filter(filter)

rust/ffi/node/src/table.rs

@@ -45,7 +45,7 @@ impl JsTable {
         let table_name = cx.argument::<JsString>(0)?.value(&mut cx);
         let buffer = cx.argument::<JsBuffer>(1)?;
         let (batches, schema) =
-            arrow_buffer_to_record_batch(buffer.as_slice(&mut cx)).or_throw(&mut cx)?;
+            arrow_buffer_to_record_batch(buffer.as_slice(&cx)).or_throw(&mut cx)?;
 
         // Write mode
         let mode = match cx.argument::<JsString>(2)?.value(&mut cx).as_str() {
@@ -93,7 +93,7 @@ impl JsTable {
         let buffer = cx.argument::<JsBuffer>(0)?;
         let write_mode = cx.argument::<JsString>(1)?.value(&mut cx);
         let (batches, schema) =
-            arrow_buffer_to_record_batch(buffer.as_slice(&mut cx)).or_throw(&mut cx)?;
+            arrow_buffer_to_record_batch(buffer.as_slice(&cx)).or_throw(&mut cx)?;
         let rt = runtime(&mut cx)?;
         let channel = cx.channel();
         let mut table = js_table.table.clone();
@@ -186,7 +186,7 @@ impl JsTable {
                 .downcast_or_throw::<JsString, _>(&mut cx)?;
 
             let value = updates_arg
-                .get_value(&mut cx, property.clone())?
+                .get_value(&mut cx, property)?
                 .downcast_or_throw::<JsString, _>(&mut cx)?;
 
             let property = property.value(&mut cx);
@@ -216,7 +216,7 @@ impl JsTable {
                 .map(|(k, v)| (k.as_str(), v.as_str()))
                 .collect::<Vec<_>>();
 
-            let predicate = predicate.as_ref().map(|s| s.as_str());
+            let predicate = predicate.as_deref();
 
             let update_result = table.update(predicate, updates_arg).await;
             deferred.settle_with(&channel, move |mut cx| {

rust/vectordb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "vectordb"
-version = "0.3.11"
+version = "0.4.0"
 edition = "2021"
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license = "Apache-2.0"

rust/vectordb/src/io/object_store.rs

@@ -26,7 +26,7 @@ use futures::{stream::BoxStream, FutureExt, StreamExt};
 use lance::io::object_store::WrappingObjectStore;
 use object_store::{
     path::Path, Error, GetOptions, GetResult, ListResult, MultipartId, ObjectMeta, ObjectStore,
-    Result,
+    PutOptions, PutResult, Result,
 };
 
 use async_trait::async_trait;
@@ -72,13 +72,28 @@ impl PrimaryOnly for Path {
 /// Note: this object store does not mirror writes to *.manifest files
 #[async_trait]
 impl ObjectStore for MirroringObjectStore {
-    async fn put(&self, location: &Path, bytes: Bytes) -> Result<()> {
+    async fn put(&self, location: &Path, bytes: Bytes) -> Result<PutResult> {
         if location.primary_only() {
             self.primary.put(location, bytes).await
         } else {
             self.secondary.put(location, bytes.clone()).await?;
-            self.primary.put(location, bytes).await?;
-            Ok(())
+            self.primary.put(location, bytes).await
+        }
+    }
+
+    async fn put_opts(
+        &self,
+        location: &Path,
+        bytes: Bytes,
+        options: PutOptions,
+    ) -> Result<PutResult> {
+        if location.primary_only() {
+            self.primary.put_opts(location, bytes, options).await
+        } else {
+            self.secondary
+                .put_opts(location, bytes.clone(), options.clone())
+                .await?;
+            self.primary.put_opts(location, bytes, options).await
         }
     }
 
@@ -129,8 +144,8 @@ impl ObjectStore for MirroringObjectStore {
         self.primary.delete(location).await
     }
 
-    async fn list(&self, prefix: Option<&Path>) -> Result<BoxStream<'_, Result<ObjectMeta>>> {
-        self.primary.list(prefix).await
+    fn list(&self, prefix: Option<&Path>) -> BoxStream<'_, Result<ObjectMeta>> {
+        self.primary.list(prefix)
     }
 
     async fn list_with_delimiter(&self, prefix: Option<&Path>) -> Result<ListResult> {