Compare commits

...

15 Commits

Author SHA1 Message Date
Lance Release
9d129c7e86 Bump version: 0.25.4-beta.2 → 0.25.4-beta.3 2025-12-02 22:00:35 +00:00
Jonathan Hsieh
44878dd9a5 feat: support stable row IDs via storage_options (#2831)
Add support for enabling stable row IDs when creating tables via the
`new_table_enable_stable_row_ids` storage option.

Stable row IDs ensure that row identifiers remain constant after
compaction, update, delete, and merge operations. This is useful for
materialized views and other use cases that need to track source rows
across these operations.

The option can be set at two levels:
- Connection level: applies to all tables created with that connection
- Table level: per-table override via create_table storage_options

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude <noreply@anthropic.com>
2025-12-02 13:57:00 -08:00
LanceDB Robot
4b5bb2d76c chore: update lance dependency to v1.0.0-beta.16 (#2835)
## Summary
- bump all Lance crates to v1.0.0-beta.16 via ci/set_lance_version.py
- refresh Cargo.lock (reqwest/opendal/etc.) to satisfy the new release

## Verification
- cargo clippy --workspace --tests --all-features -- -D warnings
- cargo fmt --all

Triggered by
[refs/tags/v1.0.0-beta.16](https://github.com/lance-format/lance/releases/tag/v1.0.0-beta.16)

---------

Co-authored-by: Jack Ye <yezhaoqin@gmail.com>
2025-12-01 23:07:03 -08:00
LanceDB Robot
434f4124fc chore: update lance dependency to v1.0.0-beta.14 (#2826)
## Summary
- bump all Lance crates to 1.0.0-beta.14 via ci/set_lance_version.py
- refresh Cargo.lock to capture new transitive requirements
- verified `cargo clippy --workspace --tests --all-features -- -D
warnings` and `cargo fmt --all`

Triggered by refs/tags/v1.0.0-beta.14

---------

Co-authored-by: Jack Ye <yezhaoqin@gmail.com>
2025-12-01 14:43:03 -08:00
Rudi Floren
03a1a99270 feat: remove remote default features on lance-namespace-impls (#2828)
This tries to fix #2771. It is not a complete fix because
`lance-namespace-impls` uses `lance` which has its default features
enabled. Thus, to close #2771, the lance repo also needs an update.

The `dir-*` features are enabled by the respective remote feature
(`aws`, `gcp`, `azure`, `oss`).
The `rest` feature is enabled via `remote`.
2025-12-01 10:53:22 -08:00
Xuanwo
0110e3b6f8 chore: clippy::string_to_string has been replaced by implicit_clone (#2817)
clippy::string_to_string has been replaced by implicit_clone, so lancedb
will raise a build error in Rust 1.91. This PR suppresses it.

---

**This PR was primarily authored with Codex using GPT-5-Codex and then
hand-reviewed by me. I AM responsible for every change made in this PR.
I aimed to keep it aligned with our goals, though I may have missed
minor issues. Please flag anything that feels off, I'll fix it
quickly.**

Signed-off-by: Xuanwo <github@xuanwo.io>
2025-11-26 16:30:35 +08:00
Xuanwo
f1f85b0a84 ci: migrate macos ci runners (#2818)
This PR will migrate macos CI runners.

---

**This PR was primarily authored with Codex using GPT-5-Codex and then
hand-reviewed by me. I AM responsible for every change made in this PR.
I aimed to keep it aligned with our goals, though I may have missed
minor issues. Please flag anything that feels off, I'll fix it
quickly.**

Signed-off-by: Xuanwo <github@xuanwo.io>
2025-11-26 01:22:35 +08:00
LanceDB Robot
d6daa08b54 chore: update lance dependency to v1.0.0-beta.8 (#2813)
## Summary
- bump all Lance crates to v1.0.0-beta.8 via ci/set_lance_version.py
- verified 
- ran 

Trigger: refs/tags/v1.0.0-beta.8
2025-11-24 14:58:42 -08:00
Wyatt Alt
17b71de22e feat: update codex url key (#2812)
This previously through unknown key for htmlUrl and indicated "url" is a
valid field.
2025-11-24 13:13:18 -08:00
Prashanth Rao
a250d8e7df docs: improve docstring for RabitQ in Python (#2808)
This PR improves the docstring for `IVF_RQ` (RabitQ) in Python. The
earlier version referred to it as "residual quantization", which is
confusing to future readers of the code.

In contrast, the TypeScript and Rust codebases defined `IVF_RQ` as
RabitQ. So now the three languages use comments that are consistent with
one another.

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2025-11-24 13:35:19 +08:00
LanceDB Robot
5a2b33581e chore: update lance dependency to v1.0.0-beta.7 (#2807)
## Summary
- bump Lance dependencies to v1.0.0-beta.7 via `ci/set_lance_version.py`
- verified workspace with `cargo clippy --workspace --tests
--all-features -- -D warnings`
- formatted with `cargo fmt --all`

Trigger: refs/tags/v1.0.0-beta.7
2025-11-21 21:42:09 -08:00
Jack Ye
3d254f61b0 ci: trigger downstream verification after version bump (#2809) 2025-11-21 09:50:23 -08:00
Jack Ye
d15e380be1 ci: add support for lance-format fury index for downloading pylance (#2804)
Set the lance-format fury repo for most places that are downloading. For
uploading, it is kept unchanged since lancedb is published to lancedb
fury.
2025-11-20 23:29:36 -08:00
Jack Ye
0baf807be0 ci: use larger runner for doctest and fix failing tests (#2801)
Currently test would fail after installing to around pytorch
2025-11-20 19:44:31 -08:00
Prashanth Rao
76bcc78910 docs: nodejs failing CI is fixed (#2802)
Fixes the breaking CI for nodejs, related to the documentation of the
new Permutation API in typescript.

- Expanded the generated typings in `nodejs/lancedb/native.d.ts` to
include `SplitCalculatedOptions`, `splitNames` fields, and the
persist/options-based `splitCalculated` methods so the permutation
exports match the native API.
- The previous block comment block had an inconsistency.
`splitCalculated` takes an options object (`SplitCalculatedOptions`) in
our bindings, not a bare string. The previous example showed
`builder.splitCalculated("user_id % 3");`, which doesn’t match the
actual signature and would fail TS typecheck. I updated the comment to
`builder.splitCalculated({ calculation: "user_id % 3" });` so the
example is now correct.
- Updated the `splitCalculated` example in
`nodejs/lancedb/permutation.ts` to use the options object.
- Ran `npm docs` to ensure docs build correctly.

> [!NOTE]
> **Disclaimer**: I used GPT-5.1-Codex-Max to make these updates, but I
have read the code and run `npm run docs` to verify that they work and
are correct to the best of my knowledge.
2025-11-20 16:16:38 -08:00
29 changed files with 1156 additions and 266 deletions

View File

@@ -19,7 +19,7 @@ rustflags = [
"-Wclippy::string_add_assign",
"-Wclippy::string_add",
"-Wclippy::string_lit_as_bytes",
"-Wclippy::string_to_string",
"-Wclippy::implicit_clone",
"-Wclippy::use_self",
"-Dclippy::cargo",
"-Dclippy::dbg_macro",

View File

@@ -31,7 +31,7 @@ runs:
with:
command: build
working-directory: python
docker-options: "-e PIP_EXTRA_INDEX_URL=https://pypi.fury.io/lancedb/"
docker-options: "-e PIP_EXTRA_INDEX_URL='https://pypi.fury.io/lance-format/ https://pypi.fury.io/lancedb/'"
target: x86_64-unknown-linux-gnu
manylinux: ${{ inputs.manylinux }}
args: ${{ inputs.args }}
@@ -46,7 +46,7 @@ runs:
with:
command: build
working-directory: python
docker-options: "-e PIP_EXTRA_INDEX_URL=https://pypi.fury.io/lancedb/"
docker-options: "-e PIP_EXTRA_INDEX_URL='https://pypi.fury.io/lance-format/ https://pypi.fury.io/lancedb/'"
target: aarch64-unknown-linux-gnu
manylinux: ${{ inputs.manylinux }}
args: ${{ inputs.args }}

View File

@@ -22,5 +22,5 @@ runs:
command: build
# TODO: pass through interpreter
args: ${{ inputs.args }}
docker-options: "-e PIP_EXTRA_INDEX_URL=https://pypi.fury.io/lancedb/"
docker-options: "-e PIP_EXTRA_INDEX_URL='https://pypi.fury.io/lance-format/ https://pypi.fury.io/lancedb/'"
working-directory: python

View File

@@ -26,7 +26,7 @@ runs:
with:
command: build
args: ${{ inputs.args }}
docker-options: "-e PIP_EXTRA_INDEX_URL=https://pypi.fury.io/lancedb/"
docker-options: "-e PIP_EXTRA_INDEX_URL='https://pypi.fury.io/lance-format/ https://pypi.fury.io/lancedb/'"
working-directory: python
- uses: actions/upload-artifact@v4
with:

View File

@@ -98,3 +98,30 @@ jobs:
printenv OPENAI_API_KEY | codex login --with-api-key
codex --config shell_environment_policy.ignore_default_excludes=true exec --dangerously-bypass-approvals-and-sandbox "$(cat /tmp/codex-prompt.txt)"
- name: Trigger sophon dependency update
env:
TAG: ${{ inputs.tag }}
GH_TOKEN: ${{ secrets.ROBOT_TOKEN }}
run: |
set -euo pipefail
VERSION="${TAG#refs/tags/}"
VERSION="${VERSION#v}"
LANCEDB_BRANCH="codex/update-lance-${VERSION//[^a-zA-Z0-9]/-}"
echo "Triggering sophon workflow with:"
echo " lance_ref: ${TAG#refs/tags/}"
echo " lancedb_ref: ${LANCEDB_BRANCH}"
gh workflow run codex-bump-lancedb-lance.yml \
--repo lancedb/sophon \
-f lance_ref="${TAG#refs/tags/}" \
-f lancedb_ref="${LANCEDB_BRANCH}"
- name: Show latest sophon workflow run
env:
GH_TOKEN: ${{ secrets.ROBOT_TOKEN }}
run: |
set -euo pipefail
echo "Latest sophon workflow run:"
gh run list --repo lancedb/sophon --workflow codex-bump-lancedb-lance.yml --limit 1 --json databaseId,url,displayTitle

View File

@@ -24,7 +24,7 @@ env:
# according to: https://matklad.github.io/2021/09/04/fast-rust-builds.html
# CI builds are faster with incremental disabled.
CARGO_INCREMENTAL: "0"
PIP_EXTRA_INDEX_URL: "https://pypi.fury.io/lancedb/"
PIP_EXTRA_INDEX_URL: "https://pypi.fury.io/lance-format/ https://pypi.fury.io/lancedb/"
jobs:
# Single deploy job since we're just deploying
@@ -50,8 +50,8 @@ jobs:
- name: Build Python
working-directory: python
run: |
python -m pip install --extra-index-url https://pypi.fury.io/lancedb/ -e .
python -m pip install --extra-index-url https://pypi.fury.io/lancedb/ -r ../docs/requirements.txt
python -m pip install --extra-index-url https://pypi.fury.io/lance-format/ --extra-index-url https://pypi.fury.io/lancedb/ -e .
python -m pip install --extra-index-url https://pypi.fury.io/lance-format/ --extra-index-url https://pypi.fury.io/lancedb/ -r ../docs/requirements.txt
- name: Set up node
uses: actions/setup-node@v3
with:

View File

@@ -59,4 +59,4 @@ jobs:
GH_TOKEN: ${{ secrets.ROBOT_TOKEN }}
run: |
set -euo pipefail
gh run list --workflow codex-update-lance-dependency.yml --limit 1 --json databaseId,htmlUrl,displayTitle
gh run list --workflow codex-update-lance-dependency.yml --limit 1 --json databaseId,url,displayTitle

View File

@@ -11,7 +11,7 @@ on:
- Cargo.toml # Change in dependency frequently breaks builds
env:
PIP_EXTRA_INDEX_URL: "https://pypi.fury.io/lancedb/"
PIP_EXTRA_INDEX_URL: "https://pypi.fury.io/lance-format/ https://pypi.fury.io/lancedb/"
jobs:
linux:
@@ -65,7 +65,7 @@ jobs:
matrix:
config:
- target: x86_64-apple-darwin
runner: macos-13
runner: macos-15-large
- target: aarch64-apple-darwin
runner: warp-macos-14-arm64-6x
env:

View File

@@ -18,7 +18,7 @@ env:
# Color output for pytest is off by default.
PYTEST_ADDOPTS: "--color=yes"
FORCE_COLOR: "1"
PIP_EXTRA_INDEX_URL: "https://pypi.fury.io/lancedb/"
PIP_EXTRA_INDEX_URL: "https://pypi.fury.io/lance-format/ https://pypi.fury.io/lancedb/"
RUST_BACKTRACE: "1"
jobs:
@@ -79,7 +79,7 @@ jobs:
doctest:
name: "Doctest"
timeout-minutes: 30
runs-on: "ubuntu-24.04"
runs-on: ubuntu-2404-8x-x64
defaults:
run:
shell: bash
@@ -100,7 +100,7 @@ jobs:
sudo apt install -y protobuf-compiler
- name: Install
run: |
pip install --extra-index-url https://pypi.fury.io/lancedb/ -e .[tests,dev,embeddings]
pip install --extra-index-url https://pypi.fury.io/lance-format/ --extra-index-url https://pypi.fury.io/lancedb/ -e .[tests,dev,embeddings]
pip install tantivy
pip install mlx
- name: Doctest
@@ -149,7 +149,7 @@ jobs:
matrix:
config:
- name: x86
runner: macos-13
runner: macos-15-large
- name: Arm
runner: macos-14
runs-on: "${{ matrix.config.runner }}"
@@ -226,7 +226,7 @@ jobs:
run: |
pip install "pydantic<2"
pip install pyarrow==16
pip install --extra-index-url https://pypi.fury.io/lancedb/ -e .[tests]
pip install --extra-index-url https://pypi.fury.io/lance-format/ --extra-index-url https://pypi.fury.io/lancedb/ -e .[tests]
pip install tantivy
- name: Run tests
run: pytest -m "not slow and not s3_test" -x -v --durations=30 python/tests

View File

@@ -15,7 +15,7 @@ runs:
- name: Install lancedb
shell: bash
run: |
pip3 install --extra-index-url https://pypi.fury.io/lancedb/ $(ls target/wheels/lancedb-*.whl)[tests,dev]
pip3 install --extra-index-url https://pypi.fury.io/lance-format/ --extra-index-url https://pypi.fury.io/lancedb/ $(ls target/wheels/lancedb-*.whl)[tests,dev]
- name: Setup localstack for integration tests
if: ${{ inputs.integration == 'true' }}
shell: bash

View File

@@ -122,7 +122,7 @@ jobs:
timeout-minutes: 30
strategy:
matrix:
mac-runner: ["macos-13", "macos-14"]
mac-runner: ["macos-14", "macos-15"]
runs-on: "${{ matrix.mac-runner }}"
defaults:
run:

767
Cargo.lock generated

File diff suppressed because it is too large Load Diff

View File

@@ -15,20 +15,20 @@ categories = ["database-implementations"]
rust-version = "1.78.0"
[workspace.dependencies]
lance = { "version" = "=1.0.0-beta.5", default-features = false, "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-core = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-datagen = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-file = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-io = { "version" = "=1.0.0-beta.5", default-features = false, "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-index = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-linalg = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-namespace = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-namespace-impls = { "version" = "=1.0.0-beta.5", "features" = ["dir-aws", "dir-gcp", "dir-azure", "dir-oss", "rest"], "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-table = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-testing = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-datafusion = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-encoding = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance-arrow = { "version" = "=1.0.0-beta.5", "tag" = "v1.0.0-beta.5", "git" = "https://github.com/lance-format/lance.git" }
lance = { "version" = "=1.0.0-beta.16", default-features = false, "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-core = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-datagen = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-file = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-io = { "version" = "=1.0.0-beta.16", default-features = false, "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-index = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-linalg = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-namespace = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-namespace-impls = { "version" = "=1.0.0-beta.16", default-features = false, "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-table = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-testing = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-datafusion = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-encoding = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
lance-arrow = { "version" = "=1.0.0-beta.16", "tag" = "v1.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
ahash = "0.8"
# Note that this one does not include pyarrow
arrow = { version = "56.2", optional = false }

View File

@@ -34,7 +34,7 @@ const results = await table.vectorSearch([0.1, 0.3]).limit(20).toArray();
console.log(results);
```
The [quickstart](https://lancedb.com/docs/quickstart/basic-usage/) contains a more complete example.
The [quickstart](https://lancedb.com/docs/quickstart/basic-usage/) contains more complete examples.
## Development

View File

@@ -147,7 +147,7 @@ A new PermutationBuilder instance
#### Example
```ts
builder.splitCalculated("user_id % 3");
builder.splitCalculated({ calculation: "user_id % 3" });
```
***

View File

@@ -118,7 +118,7 @@ export class PermutationBuilder {
* @returns A new PermutationBuilder instance
* @example
* ```ts
* builder.splitCalculated("user_id % 3");
* builder.splitCalculated({ calculation: "user_id % 3" });
* ```
*/
splitCalculated(options: SplitCalculatedOptions): PermutationBuilder {

View File

@@ -1,5 +1,5 @@
[tool.bumpversion]
current_version = "0.25.4-beta.2"
current_version = "0.25.4-beta.3"
parse = """(?x)
(?P<major>0|[1-9]\\d*)\\.
(?P<minor>0|[1-9]\\d*)\\.

View File

@@ -1,6 +1,6 @@
[package]
name = "lancedb-python"
version = "0.25.4-beta.2"
version = "0.25.4-beta.3"
edition.workspace = true
description = "Python bindings for LanceDB"
license.workspace = true

View File

@@ -1,11 +1,11 @@
PIP_EXTRA_INDEX_URL ?= https://pypi.fury.io/lancedb/
PIP_EXTRA_INDEX_URL ?= https://pypi.fury.io/lance-format/ https://pypi.fury.io/lancedb/
help: ## Show this help.
@sed -ne '/@sed/!s/## //p' $(MAKEFILE_LIST)
.PHONY: develop
develop: ## Install the package in development mode.
PIP_EXTRA_INDEX_URL=$(PIP_EXTRA_INDEX_URL) maturin develop --extras tests,dev,embeddings
PIP_EXTRA_INDEX_URL="$(PIP_EXTRA_INDEX_URL)" maturin develop --extras tests,dev,embeddings
.PHONY: format
format: ## Format the code.

View File

@@ -10,7 +10,7 @@ dependencies = [
"pyarrow>=16",
"pydantic>=1.10",
"tqdm>=4.27.0",
"lance-namespace>=0.0.21"
"lance-namespace>=0.2.1"
]
description = "lancedb"
authors = [{ name = "LanceDB Devs", email = "dev@lancedb.com" }]
@@ -45,7 +45,7 @@ repository = "https://github.com/lancedb/lancedb"
[project.optional-dependencies]
pylance = [
"pylance>=0.25",
"pylance>=1.0.0b14",
]
tests = [
"aiohttp",
@@ -59,7 +59,7 @@ tests = [
"polars>=0.19, <=1.3.0",
"tantivy",
"pyarrow-stubs",
"pylance>=1.0.0b4",
"pylance>=1.0.0b14",
"requests",
"datafusion",
]

View File

@@ -194,6 +194,10 @@ class DBConnection(EnforceOverrides):
connection will be inherited by the table, but can be overridden here.
See available options at
<https://lancedb.com/docs/storage/>
To enable stable row IDs (row IDs remain stable after compaction,
update, delete, and merges), set `new_table_enable_stable_row_ids`
to `"true"` in storage_options when connecting to the database.
data_storage_version: optional, str, default "stable"
Deprecated. Set `storage_options` when connecting to the database and set
`new_table_data_storage_version` in the options.
@@ -1079,6 +1083,10 @@ class AsyncConnection(object):
See available options at
<https://lancedb.com/docs/storage/>
To enable stable row IDs (row IDs remain stable after compaction,
update, delete, and merges), set `new_table_enable_stable_row_ids`
to `"true"` in storage_options when connecting to the database.
Returns
-------
AsyncTable

View File

@@ -609,9 +609,19 @@ class IvfPq:
class IvfRq:
"""Describes an IVF RQ Index
IVF-RQ (Residual Quantization) stores a compressed copy of each vector using
residual quantization and organizes them into IVF partitions. Parameters
largely mirror IVF-PQ for consistency.
IVF-RQ (RabitQ Quantization) compresses vectors using RabitQ quantization
and organizes them into IVF partitions.
The compression scheme is called RabitQ quantization. Each dimension is
quantized into a small number of bits. The parameters `num_bits` and
`num_partitions` control this process, providing a tradeoff between
index size (and thus search speed) and index accuracy.
The partitioning process is called IVF and the `num_partitions` parameter
controls how many groups to create.
Note that training an IVF RQ index on a large dataset is a slow operation
and currently is also a memory intensive operation.
Attributes
----------
@@ -628,7 +638,7 @@ class IvfRq:
Number of IVF partitions to create.
num_bits: int, default 1
Number of bits to encode each dimension.
Number of bits to encode each dimension in the RabitQ codebook.
max_iterations: int, default 50
Max iterations to train kmeans when computing IVF partitions.

View File

@@ -127,13 +127,17 @@ class LanceNamespaceStorageOptionsProvider(StorageOptionsProvider):
Examples
--------
>>> from lance_namespace import connect as namespace_connect
>>> namespace = namespace_connect("rest", {"url": "https://..."})
>>> provider = LanceNamespaceStorageOptionsProvider(
... namespace=namespace,
... table_id=["my_namespace", "my_table"]
... )
>>> options = provider.fetch_storage_options()
Create a provider and fetch storage options::
from lance_namespace import connect as namespace_connect
# Connect to namespace (requires a running namespace server)
namespace = namespace_connect("rest", {"uri": "https://..."})
provider = LanceNamespaceStorageOptionsProvider(
namespace=namespace,
table_id=["my_namespace", "my_table"]
)
options = provider.fetch_storage_options()
"""
def __init__(self, namespace: LanceNamespace, table_id: List[str]):

View File

@@ -2429,9 +2429,8 @@ class AsyncQueryBase(object):
>>> from lancedb import connect_async
>>> async def doctest_example():
... conn = await connect_async("./.lancedb")
... table = await conn.create_table("my_table", [{"vector": [99, 99]}])
... query = [100, 100]
... plan = await table.query().nearest_to([1, 2]).explain_plan(True)
... table = await conn.create_table("my_table", [{"vector": [99.0, 99.0]}])
... plan = await table.query().nearest_to([1.0, 2.0]).explain_plan(True)
... print(plan)
>>> asyncio.run(doctest_example()) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
ProjectionExec: expr=[vector@0 as vector, _distance@2 as _distance]
@@ -2440,6 +2439,7 @@ class AsyncQueryBase(object):
SortExec: TopK(fetch=10), expr=[_distance@2 ASC NULLS LAST, _rowid@1 ASC NULLS LAST], preserve_partitioning=[false]
KNNVectorDistance: metric=l2
LanceRead: uri=..., projection=[vector], ...
<BLANKLINE>
Parameters
----------
@@ -3141,10 +3141,9 @@ class AsyncHybridQuery(AsyncStandardQuery, AsyncVectorQueryBase):
>>> from lancedb.index import FTS
>>> async def doctest_example():
... conn = await connect_async("./.lancedb")
... table = await conn.create_table("my_table", [{"vector": [99, 99], "text": "hello world"}])
... table = await conn.create_table("my_table", [{"vector": [99.0, 99.0], "text": "hello world"}])
... await table.create_index("text", config=FTS(with_position=False))
... query = [100, 100]
... plan = await table.query().nearest_to([1, 2]).nearest_to_text("hello").explain_plan(True)
... plan = await table.query().nearest_to([1.0, 2.0]).nearest_to_text("hello").explain_plan(True)
... print(plan)
>>> asyncio.run(doctest_example()) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
Vector Search Plan:
@@ -3418,9 +3417,8 @@ class BaseQueryBuilder(object):
>>> from lancedb import connect_async
>>> async def doctest_example():
... conn = await connect_async("./.lancedb")
... table = await conn.create_table("my_table", [{"vector": [99, 99]}])
... query = [100, 100]
... plan = await table.query().nearest_to([1, 2]).explain_plan(True)
... table = await conn.create_table("my_table", [{"vector": [99.0, 99.0]}])
... plan = await table.query().nearest_to([1.0, 2.0]).explain_plan(True)
... print(plan)
>>> asyncio.run(doctest_example()) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
ProjectionExec: expr=[vector@0 as vector, _distance@2 as _distance]
@@ -3429,6 +3427,7 @@ class BaseQueryBuilder(object):
SortExec: TopK(fetch=10), expr=[_distance@2 ASC NULLS LAST, _rowid@1 ASC NULLS LAST], preserve_partitioning=[false]
KNNVectorDistance: metric=l2
LanceRead: uri=..., projection=[vector], ...
<BLANKLINE>
Parameters
----------

View File

@@ -1018,7 +1018,7 @@ class Table(ABC):
... .when_not_matched_insert_all() \\
... .execute(new_data)
>>> res
MergeResult(version=2, num_updated_rows=2, num_inserted_rows=1, num_deleted_rows=0)
MergeResult(version=2, num_updated_rows=2, num_inserted_rows=1, num_deleted_rows=0, num_attempts=1)
>>> # The order of new rows is non-deterministic since we use
>>> # a hash-join as part of this operation and so we sort here
>>> table.to_arrow().sort_by("a").to_pandas()
@@ -3634,7 +3634,7 @@ class AsyncTable:
... .when_not_matched_insert_all() \\
... .execute(new_data)
>>> res
MergeResult(version=2, num_updated_rows=2, num_inserted_rows=1, num_deleted_rows=0)
MergeResult(version=2, num_updated_rows=2, num_inserted_rows=1, num_deleted_rows=0, num_attempts=1)
>>> # The order of new rows is non-deterministic since we use
>>> # a hash-join as part of this operation and so we sort here
>>> table.to_arrow().sort_by("a").to_pandas()

View File

@@ -441,6 +441,150 @@ async def test_create_table_v2_manifest_paths_async(tmp_path):
assert re.match(r"\d{20}\.manifest", manifest)
@pytest.mark.asyncio
async def test_create_table_stable_row_ids_via_storage_options(tmp_path):
"""Test stable_row_ids via storage_options at connect time."""
import lance
# Connect with stable row IDs enabled as default for new tables
db_with = await lancedb.connect_async(
tmp_path, storage_options={"new_table_enable_stable_row_ids": "true"}
)
# Connect without stable row IDs (default)
db_without = await lancedb.connect_async(
tmp_path, storage_options={"new_table_enable_stable_row_ids": "false"}
)
# Create table using connection with stable row IDs enabled
await db_with.create_table(
"with_stable_via_opts",
data=[{"id": i} for i in range(10)],
)
lance_ds_with = lance.dataset(tmp_path / "with_stable_via_opts.lance")
fragments_with = lance_ds_with.get_fragments()
assert len(fragments_with) > 0
assert fragments_with[0].metadata.row_id_meta is not None
# Create table using connection without stable row IDs
await db_without.create_table(
"without_stable_via_opts",
data=[{"id": i} for i in range(10)],
)
lance_ds_without = lance.dataset(tmp_path / "without_stable_via_opts.lance")
fragments_without = lance_ds_without.get_fragments()
assert len(fragments_without) > 0
assert fragments_without[0].metadata.row_id_meta is None
def test_create_table_stable_row_ids_via_storage_options_sync(tmp_path):
"""Test that enable_stable_row_ids can be set via storage_options (sync API)."""
# Connect with stable row IDs enabled as default for new tables
db_with = lancedb.connect(
tmp_path, storage_options={"new_table_enable_stable_row_ids": "true"}
)
# Connect without stable row IDs (default)
db_without = lancedb.connect(
tmp_path, storage_options={"new_table_enable_stable_row_ids": "false"}
)
# Create table using connection with stable row IDs enabled
tbl_with = db_with.create_table(
"with_stable_sync",
data=[{"id": i} for i in range(10)],
)
lance_ds_with = tbl_with.to_lance()
fragments_with = lance_ds_with.get_fragments()
assert len(fragments_with) > 0
assert fragments_with[0].metadata.row_id_meta is not None
# Create table using connection without stable row IDs
tbl_without = db_without.create_table(
"without_stable_sync",
data=[{"id": i} for i in range(10)],
)
lance_ds_without = tbl_without.to_lance()
fragments_without = lance_ds_without.get_fragments()
assert len(fragments_without) > 0
assert fragments_without[0].metadata.row_id_meta is None
@pytest.mark.asyncio
async def test_create_table_stable_row_ids_table_level_override(tmp_path):
"""Test that stable_row_ids can be enabled/disabled at create_table level."""
import lance
# Connect without any stable row ID setting
db_default = await lancedb.connect_async(tmp_path)
# Connect with stable row IDs enabled at connection level
db_with_stable = await lancedb.connect_async(
tmp_path, storage_options={"new_table_enable_stable_row_ids": "true"}
)
# Case 1: No connection setting, enable at table level
await db_default.create_table(
"table_level_enabled",
data=[{"id": i} for i in range(10)],
storage_options={"new_table_enable_stable_row_ids": "true"},
)
lance_ds = lance.dataset(tmp_path / "table_level_enabled.lance")
fragments = lance_ds.get_fragments()
assert len(fragments) > 0
assert fragments[0].metadata.row_id_meta is not None, (
"Table should have stable row IDs when enabled at table level"
)
# Case 2: Connection has stable row IDs, override with false at table level
await db_with_stable.create_table(
"table_level_disabled",
data=[{"id": i} for i in range(10)],
storage_options={"new_table_enable_stable_row_ids": "false"},
)
lance_ds = lance.dataset(tmp_path / "table_level_disabled.lance")
fragments = lance_ds.get_fragments()
assert len(fragments) > 0
assert fragments[0].metadata.row_id_meta is None, (
"Table should NOT have stable row IDs when disabled at table level"
)
def test_create_table_stable_row_ids_table_level_override_sync(tmp_path):
"""Test that stable_row_ids can be enabled/disabled at create_table level (sync)."""
# Connect without any stable row ID setting
db_default = lancedb.connect(tmp_path)
# Connect with stable row IDs enabled at connection level
db_with_stable = lancedb.connect(
tmp_path, storage_options={"new_table_enable_stable_row_ids": "true"}
)
# Case 1: No connection setting, enable at table level
tbl = db_default.create_table(
"table_level_enabled_sync",
data=[{"id": i} for i in range(10)],
storage_options={"new_table_enable_stable_row_ids": "true"},
)
lance_ds = tbl.to_lance()
fragments = lance_ds.get_fragments()
assert len(fragments) > 0
assert fragments[0].metadata.row_id_meta is not None, (
"Table should have stable row IDs when enabled at table level"
)
# Case 2: Connection has stable row IDs, override with false at table level
tbl = db_with_stable.create_table(
"table_level_disabled_sync",
data=[{"id": i} for i in range(10)],
storage_options={"new_table_enable_stable_row_ids": "false"},
)
lance_ds = tbl.to_lance()
fragments = lance_ds.get_fragments()
assert len(fragments) > 0
assert fragments[0].metadata.row_id_meta is None, (
"Table should NOT have stable row IDs when disabled at table level"
)
def test_open_table_sync(tmp_db: lancedb.DBConnection):
tmp_db.create_table("test", data=[{"id": 0}])
assert tmp_db.open_table("test").count_rows() == 1

View File

@@ -690,7 +690,7 @@ impl FTSQuery {
}
pub fn get_query(&self) -> String {
self.fts_query.query.query().to_owned()
self.fts_query.query.query().clone()
}
pub fn to_query_request(&self) -> PyQueryRequest {

View File

@@ -105,12 +105,12 @@ test-log = "0.2"
[features]
default = ["aws", "gcs", "azure", "dynamodb", "oss"]
aws = ["lance/aws", "lance-io/aws"]
oss = ["lance/oss", "lance-io/oss"]
gcs = ["lance/gcp", "lance-io/gcp"]
azure = ["lance/azure", "lance-io/azure"]
aws = ["lance/aws", "lance-io/aws", "lance-namespace-impls/dir-aws"]
oss = ["lance/oss", "lance-io/oss", "lance-namespace-impls/dir-oss"]
gcs = ["lance/gcp", "lance-io/gcp", "lance-namespace-impls/dir-gcp"]
azure = ["lance/azure", "lance-io/azure", "lance-namespace-impls/dir-azure"]
dynamodb = ["lance/dynamodb", "aws"]
remote = ["dep:reqwest", "dep:http"]
remote = ["dep:reqwest", "dep:http", "lance-namespace-impls/rest"]
fp16kernels = ["lance-linalg/fp16kernels"]
s3-test = []
bedrock = ["dep:aws-sdk-bedrockruntime"]

View File

@@ -35,6 +35,7 @@ pub const LANCE_FILE_EXTENSION: &str = "lance";
pub const OPT_NEW_TABLE_STORAGE_VERSION: &str = "new_table_data_storage_version";
pub const OPT_NEW_TABLE_V2_MANIFEST_PATHS: &str = "new_table_enable_v2_manifest_paths";
pub const OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS: &str = "new_table_enable_stable_row_ids";
/// Controls how new tables should be created
#[derive(Clone, Debug, Default)]
@@ -48,6 +49,12 @@ pub struct NewTableConfig {
/// V2 manifest paths are more efficient than V2 manifest paths but are not
/// supported by old clients.
pub enable_v2_manifest_paths: Option<bool>,
/// Whether to enable stable row IDs for new tables
///
/// When enabled, row IDs remain stable after compaction, update, delete,
/// and merges. This is useful for materialized views and other use cases
/// that need to track source rows across these operations.
pub enable_stable_row_ids: Option<bool>,
}
/// Options specific to the listing database
@@ -87,6 +94,14 @@ impl ListingDatabaseOptions {
})
})
.transpose()?,
enable_stable_row_ids: map
.get(OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS)
.map(|s| {
s.parse::<bool>().map_err(|_| Error::InvalidInput {
message: format!("enable_stable_row_ids must be a boolean, received {}", s),
})
})
.transpose()?,
};
// We just assume that any options that are not new table config options are storage options
let storage_options = map
@@ -94,6 +109,7 @@ impl ListingDatabaseOptions {
.filter(|(key, _)| {
key.as_str() != OPT_NEW_TABLE_STORAGE_VERSION
&& key.as_str() != OPT_NEW_TABLE_V2_MANIFEST_PATHS
&& key.as_str() != OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS
})
.map(|(key, value)| (key.clone(), value.clone()))
.collect();
@@ -118,6 +134,12 @@ impl DatabaseOptions for ListingDatabaseOptions {
enable_v2_manifest_paths.to_string(),
);
}
if let Some(enable_stable_row_ids) = self.new_table_config.enable_stable_row_ids {
map.insert(
OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS.to_string(),
enable_stable_row_ids.to_string(),
);
}
}
}
@@ -475,7 +497,7 @@ impl ListingDatabase {
// this error is not lance::Error::DatasetNotFound, as the method
// `remove_dir_all` may be used to remove something not be a dataset
lance::Error::NotFound { .. } => Error::TableNotFound {
name: name.to_owned(),
name: name.clone(),
source: Box::new(err),
},
_ => Error::from(err),
@@ -497,7 +519,7 @@ impl ListingDatabase {
fn extract_storage_overrides(
&self,
request: &CreateTableRequest,
) -> Result<(Option<LanceFileVersion>, Option<bool>)> {
) -> Result<(Option<LanceFileVersion>, Option<bool>, Option<bool>)> {
let storage_options = request
.write_options
.lance_write_params
@@ -518,7 +540,19 @@ impl ListingDatabase {
message: "enable_v2_manifest_paths must be a boolean".to_string(),
})?;
Ok((storage_version_override, v2_manifest_override))
let stable_row_ids_override = storage_options
.and_then(|opts| opts.get(OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS))
.map(|s| s.parse::<bool>())
.transpose()
.map_err(|_| Error::InvalidInput {
message: "enable_stable_row_ids must be a boolean".to_string(),
})?;
Ok((
storage_version_override,
v2_manifest_override,
stable_row_ids_override,
))
}
/// Prepare write parameters for table creation
@@ -527,6 +561,7 @@ impl ListingDatabase {
request: &CreateTableRequest,
storage_version_override: Option<LanceFileVersion>,
v2_manifest_override: Option<bool>,
stable_row_ids_override: Option<bool>,
) -> lance::dataset::WriteParams {
let mut write_params = request
.write_options
@@ -571,6 +606,13 @@ impl ListingDatabase {
write_params.enable_v2_manifest_paths = enable_v2_manifest_paths;
}
// Apply enable_stable_row_ids: table-level override takes precedence over connection config
if let Some(enable_stable_row_ids) =
stable_row_ids_override.or(self.new_table_config.enable_stable_row_ids)
{
write_params.enable_stable_row_ids = enable_stable_row_ids;
}
if matches!(&request.mode, CreateTableMode::Overwrite) {
write_params.mode = WriteMode::Overwrite;
}
@@ -706,11 +748,15 @@ impl Database for ListingDatabase {
.clone()
.unwrap_or_else(|| self.table_uri(&request.name).unwrap());
let (storage_version_override, v2_manifest_override) =
let (storage_version_override, v2_manifest_override, stable_row_ids_override) =
self.extract_storage_overrides(&request)?;
let write_params =
self.prepare_write_params(&request, storage_version_override, v2_manifest_override);
let write_params = self.prepare_write_params(
&request,
storage_version_override,
v2_manifest_override,
stable_row_ids_override,
);
let data_schema = request.data.arrow_schema();
@@ -921,7 +967,7 @@ impl Database for ListingDatabase {
mod tests {
use super::*;
use crate::connection::ConnectRequest;
use crate::database::{CreateTableData, CreateTableMode, CreateTableRequest};
use crate::database::{CreateTableData, CreateTableMode, CreateTableRequest, WriteOptions};
use crate::table::{Table, TableDefinition};
use arrow_array::{Int32Array, RecordBatch, StringArray};
use arrow_schema::{DataType, Field, Schema};
@@ -1621,4 +1667,267 @@ mod tests {
// Cloned table should have all 8 rows from the latest version
assert_eq!(cloned_table.count_rows(None).await.unwrap(), 8);
}
#[tokio::test]
async fn test_create_table_with_stable_row_ids_connection_level() {
let tempdir = tempdir().unwrap();
let uri = tempdir.path().to_str().unwrap();
// Create database with stable row IDs enabled at connection level
let mut options = HashMap::new();
options.insert(
OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS.to_string(),
"true".to_string(),
);
let request = ConnectRequest {
uri: uri.to_string(),
#[cfg(feature = "remote")]
client_config: Default::default(),
options,
read_consistency_interval: None,
session: None,
};
let db = ListingDatabase::connect_with_options(&request)
.await
.unwrap();
// Verify the config was parsed correctly
assert_eq!(db.new_table_config.enable_stable_row_ids, Some(true));
// Create a table - it should inherit the stable row IDs setting
let schema = Arc::new(Schema::new(vec![Field::new("id", DataType::Int32, false)]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from(vec![1, 2, 3]))],
)
.unwrap();
let reader = Box::new(arrow_array::RecordBatchIterator::new(
vec![Ok(batch)],
schema.clone(),
));
let table = db
.create_table(CreateTableRequest {
name: "test_stable".to_string(),
namespace: vec![],
data: CreateTableData::Data(reader),
mode: CreateTableMode::Create,
write_options: Default::default(),
location: None,
})
.await
.unwrap();
// Verify table was created successfully
assert_eq!(table.count_rows(None).await.unwrap(), 3);
}
#[tokio::test]
async fn test_create_table_with_stable_row_ids_table_level() {
let (_tempdir, db) = setup_database().await;
// Verify connection has no stable row IDs config
assert_eq!(db.new_table_config.enable_stable_row_ids, None);
// Create a table with stable row IDs enabled at table level via storage_options
let schema = Arc::new(Schema::new(vec![Field::new("id", DataType::Int32, false)]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from(vec![1, 2, 3]))],
)
.unwrap();
let reader = Box::new(arrow_array::RecordBatchIterator::new(
vec![Ok(batch)],
schema.clone(),
));
let mut storage_options = HashMap::new();
storage_options.insert(
OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS.to_string(),
"true".to_string(),
);
let write_options = WriteOptions {
lance_write_params: Some(lance::dataset::WriteParams {
store_params: Some(lance::io::ObjectStoreParams {
storage_options: Some(storage_options),
..Default::default()
}),
..Default::default()
}),
};
let table = db
.create_table(CreateTableRequest {
name: "test_stable_table_level".to_string(),
namespace: vec![],
data: CreateTableData::Data(reader),
mode: CreateTableMode::Create,
write_options,
location: None,
})
.await
.unwrap();
// Verify table was created successfully
assert_eq!(table.count_rows(None).await.unwrap(), 3);
}
#[tokio::test]
async fn test_create_table_stable_row_ids_table_overrides_connection() {
let tempdir = tempdir().unwrap();
let uri = tempdir.path().to_str().unwrap();
// Create database with stable row IDs enabled at connection level
let mut options = HashMap::new();
options.insert(
OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS.to_string(),
"true".to_string(),
);
let request = ConnectRequest {
uri: uri.to_string(),
#[cfg(feature = "remote")]
client_config: Default::default(),
options,
read_consistency_interval: None,
session: None,
};
let db = ListingDatabase::connect_with_options(&request)
.await
.unwrap();
assert_eq!(db.new_table_config.enable_stable_row_ids, Some(true));
// Create table with stable row IDs disabled at table level (overrides connection)
let schema = Arc::new(Schema::new(vec![Field::new("id", DataType::Int32, false)]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from(vec![1, 2, 3]))],
)
.unwrap();
let reader = Box::new(arrow_array::RecordBatchIterator::new(
vec![Ok(batch)],
schema.clone(),
));
let mut storage_options = HashMap::new();
storage_options.insert(
OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS.to_string(),
"false".to_string(),
);
let write_options = WriteOptions {
lance_write_params: Some(lance::dataset::WriteParams {
store_params: Some(lance::io::ObjectStoreParams {
storage_options: Some(storage_options),
..Default::default()
}),
..Default::default()
}),
};
let table = db
.create_table(CreateTableRequest {
name: "test_override".to_string(),
namespace: vec![],
data: CreateTableData::Data(reader),
mode: CreateTableMode::Create,
write_options,
location: None,
})
.await
.unwrap();
// Verify table was created successfully
assert_eq!(table.count_rows(None).await.unwrap(), 3);
}
#[tokio::test]
async fn test_stable_row_ids_invalid_value() {
let tempdir = tempdir().unwrap();
let uri = tempdir.path().to_str().unwrap();
// Try to create database with invalid stable row IDs value
let mut options = HashMap::new();
options.insert(
OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS.to_string(),
"not_a_boolean".to_string(),
);
let request = ConnectRequest {
uri: uri.to_string(),
#[cfg(feature = "remote")]
client_config: Default::default(),
options,
read_consistency_interval: None,
session: None,
};
let result = ListingDatabase::connect_with_options(&request).await;
assert!(result.is_err());
assert!(matches!(
result.unwrap_err(),
Error::InvalidInput { message } if message.contains("enable_stable_row_ids must be a boolean")
));
}
#[test]
fn test_stable_row_ids_config_serialization() {
// Test that ListingDatabaseOptions correctly serializes stable_row_ids
let mut options = HashMap::new();
options.insert(
OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS.to_string(),
"true".to_string(),
);
// Parse the options
let db_options = ListingDatabaseOptions::parse_from_map(&options).unwrap();
assert_eq!(
db_options.new_table_config.enable_stable_row_ids,
Some(true)
);
// Serialize back to map
let mut serialized = HashMap::new();
db_options.serialize_into_map(&mut serialized);
assert_eq!(
serialized.get(OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS),
Some(&"true".to_string())
);
}
#[test]
fn test_stable_row_ids_config_parse_false() {
let mut options = HashMap::new();
options.insert(
OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS.to_string(),
"false".to_string(),
);
let db_options = ListingDatabaseOptions::parse_from_map(&options).unwrap();
assert_eq!(
db_options.new_table_config.enable_stable_row_ids,
Some(false)
);
}
#[test]
fn test_stable_row_ids_config_not_set() {
let options = HashMap::new();
let db_options = ListingDatabaseOptions::parse_from_map(&options).unwrap();
assert_eq!(db_options.new_table_config.enable_stable_row_ids, None);
}
}