Compare commits

..

9 Commits

Author SHA1 Message Date
Wyatt Alt 3c8f9e4c9c feat: support multi-tenant azure deployments in the remote client
This forwards an x-azure-tenant-id header to the remote server when
supplied in storage_options. This makes it possible to have a server
deployment that spans multiple Azure AD tenants.
2026-02-13 14:13:31 -08:00
Weston Pace 70cbee6293 feat: improve Permutation pytorch integration (#3016)
This changes around the output format of `Permutation` in some breaking
ways but I think the API is still new enough to be considered
experimental.

1. In order to align with both huggingface's dataset and torch's
expectations the default output format is now a list of dicts
(row-major) instead of a dict of lists (column-major). I've added a
python_col option which will return the dict of lists.
2. In order to align with pytorch's expectation the `torch` format is
now a list of tensors (row-major) instead of a 2D tensor (column-major).
I've added a torch_col option which will return the 2D tensor instead.

Added tests for torch integration with Permutation

~~Leaving draft until https://github.com/lancedb/lancedb/pull/3013
merges as this is built on top of that~~
2026-02-12 13:41:14 -08:00
Weston Pace 02783bf440 feat: add a getitems implementation for the permutation (#3013) 2026-02-12 05:36:11 -08:00
Dhruv 4323ca0147 feat: show reranker info in hybrid search explain plan (#3006)
Closes #3000

The hybrid search `explain_plan` now shows the reranker as the top-level
node with
the vector and FTS sub-plans indented underneath, instead of just
listing them
separately with no reranker context.

**Before:**
```
Vector Search Plan:
ProjectionExec: ...
FTS Search Plan:
ProjectionExec: ...
```

**After:**
```
RRFReranker(K=60)
  Vector Search Plan:
  ProjectionExec: ...
  FTS Search Plan:
  ProjectionExec: ...
```

Other rerankers display similarly ; e.g.
`LinearCombinationReranker(weight=0.7, fill=1.0)`,
`MRRReranker(weight_vector=0.5, weight_fts=0.5)`,
`CohereReranker(model_name=name)`.

---------

Signed-off-by: dask-58 <googldhruv@gmail.com>
Co-authored-by: Will Jones <willjones127@gmail.com>
2026-02-10 11:45:39 -08:00
Dhruv bd3dd6a8e5 fix: improve error message for multi-field FTS index creation (#3005)
Fixes #2999

The error message previously said `"field_names must be a string when
use_tantivy=False"` implying they should use the to be deprecated
tantivy backend #2998.

Updated the error message and docstring to instead guide users to create
a separate FTS index for each field

Signed-off-by: dask-58 <googldhruv@gmail.com>
2026-02-09 16:28:50 -08:00
Abhishek 3c1162612e refactor: extract optimize logic from table.rs into submodule (#2979)
## Summary

Continues the modularization effort of table operations as outlined in
#2949.

- Extracts optimization operations (`OptimizeAction`, `OptimizeStats`,
`execute_optimize`, `compact_files_impl`, `cleanup_old_versions`,
`optimize_indices`) from
  `table.rs` into `table/optimize.rs`
  - Public API remains unchanged via re-exports
- Adds comprehensive tests including error cases with message assertions

  ## Test plan

  - [x] All new optimization tests pass
  - [x] All existing tests pass
  - [x] `cargo clippy` passes with no warnings
  - [x] `cargo fmt --check` passes

---------

Co-authored-by: Will Jones <willjones127@gmail.com>
2026-02-09 16:22:57 -08:00
Jack Ye 53c7c560c9 feat: add third party licenses lists (#3010)
The files are generated with `make licenses`, currently expected to run
manually. In the future, some automations could be built.
2026-02-09 16:16:46 -08:00
Lance Release de4f77800d Bump version: 0.26.2-beta.0 → 0.26.2 2026-02-09 06:06:22 +00:00
Lance Release b6ab721cf7 Bump version: 0.26.1 → 0.26.2-beta.0 2026-02-09 06:06:03 +00:00
51 changed files with 52511 additions and 312 deletions
+1 -1
View File
@@ -1,5 +1,5 @@
[tool.bumpversion] [tool.bumpversion]
current_version = "0.26.1" current_version = "0.26.2"
parse = """(?x) parse = """(?x)
(?P<major>0|[1-9]\\d*)\\. (?P<major>0|[1-9]\\d*)\\.
(?P<minor>0|[1-9]\\d*)\\. (?P<minor>0|[1-9]\\d*)\\.
Generated
+3 -3
View File
@@ -4926,7 +4926,7 @@ dependencies = [
[[package]] [[package]]
name = "lancedb" name = "lancedb"
version = "0.26.1" version = "0.26.2"
dependencies = [ dependencies = [
"ahash", "ahash",
"anyhow", "anyhow",
@@ -5006,7 +5006,7 @@ dependencies = [
[[package]] [[package]]
name = "lancedb-nodejs" name = "lancedb-nodejs"
version = "0.26.1" version = "0.26.2"
dependencies = [ dependencies = [
"arrow-array", "arrow-array",
"arrow-ipc", "arrow-ipc",
@@ -5026,7 +5026,7 @@ dependencies = [
[[package]] [[package]]
name = "lancedb-python" name = "lancedb-python"
version = "0.29.1" version = "0.29.2"
dependencies = [ dependencies = [
"arrow", "arrow",
"async-trait", "async-trait",
+9
View File
@@ -0,0 +1,9 @@
.PHONY: licenses
licenses:
cargo about generate about.hbs -o RUST_THIRD_PARTY_LICENSES.html -c about.toml
cd python && cargo about generate ../about.hbs -o RUST_THIRD_PARTY_LICENSES.html -c ../about.toml
cd python && uv sync --all-extras && uv tool run pip-licenses --python .venv/bin/python --format=markdown --with-urls --output-file=PYTHON_THIRD_PARTY_LICENSES.md
cd nodejs && cargo about generate ../about.hbs -o RUST_THIRD_PARTY_LICENSES.html -c ../about.toml
cd nodejs && npx license-checker --markdown --out NODEJS_THIRD_PARTY_LICENSES.md
cd java && ./mvnw license:aggregate-add-third-party -q
File diff suppressed because it is too large Load Diff
+70
View File
@@ -0,0 +1,70 @@
<html>
<head>
<style>
@media (prefers-color-scheme: dark) {
body {
background: #333;
color: white;
}
a {
color: skyblue;
}
}
.container {
font-family: sans-serif;
max-width: 800px;
margin: 0 auto;
}
.intro {
text-align: center;
}
.licenses-list {
list-style-type: none;
margin: 0;
padding: 0;
}
.license-used-by {
margin-top: -10px;
}
.license-text {
max-height: 200px;
overflow-y: scroll;
white-space: pre-wrap;
}
</style>
</head>
<body>
<main class="container">
<div class="intro">
<h1>Third Party Licenses</h1>
<p>This page lists the licenses of the projects used in cargo-about.</p>
</div>
<h2>Overview of licenses:</h2>
<ul class="licenses-overview">
{{#each overview}}
<li><a href="#{{id}}">{{name}}</a> ({{count}})</li>
{{/each}}
</ul>
<h2>All license text:</h2>
<ul class="licenses-list">
{{#each licenses}}
<li class="license">
<h3 id="{{id}}">{{name}}</h3>
<h4>Used by:</h4>
<ul class="license-used-by">
{{#each used_by}}
<li><a href="{{#if crate.repository}} {{crate.repository}} {{else}} https://crates.io/crates/{{crate.name}} {{/if}}">{{crate.name}} {{crate.version}}</a></li>
{{/each}}
</ul>
<pre class="license-text">{{text}}</pre>
</li>
{{/each}}
</ul>
</main>
</body>
</html>
+18
View File
@@ -0,0 +1,18 @@
accepted = [
"0BSD",
"Apache-2.0",
"Apache-2.0 WITH LLVM-exception",
"BSD-2-Clause",
"BSD-3-Clause",
"BSL-1.0",
"bzip2-1.0.6",
"CC0-1.0",
"CDDL-1.0",
"CDLA-Permissive-2.0",
"ISC",
"MIT",
"MPL-2.0",
"OpenSSL",
"Unicode-3.0",
"Zlib",
]
+1 -1
View File
@@ -14,7 +14,7 @@ Add the following dependency to your `pom.xml`:
<dependency> <dependency>
<groupId>com.lancedb</groupId> <groupId>com.lancedb</groupId>
<artifactId>lancedb-core</artifactId> <artifactId>lancedb-core</artifactId>
<version>0.26.1</version> <version>0.26.2</version>
</dependency> </dependency>
``` ```
+71
View File
@@ -0,0 +1,71 @@
List of third-party dependencies grouped by their license type.
Apache 2.0:
* error-prone annotations (com.google.errorprone:error_prone_annotations:2.28.0 - https://errorprone.info/error_prone_annotations)
Apache License 2.0:
* JsonNullable Jackson module (org.openapitools:jackson-databind-nullable:0.2.6 - https://github.com/OpenAPITools/jackson-databind-nullable)
Apache License V2.0:
* FlatBuffers Java API (com.google.flatbuffers:flatbuffers-java:23.5.26 - https://github.com/google/flatbuffers)
Apache License, Version 2.0:
* Apache Commons Codec (commons-codec:commons-codec:1.15 - https://commons.apache.org/proper/commons-codec/)
* Apache HttpClient (org.apache.httpcomponents.client5:httpclient5:5.2.1 - https://hc.apache.org/httpcomponents-client-5.0.x/5.2.1/httpclient5/)
* Apache HttpComponents Core HTTP/1.1 (org.apache.httpcomponents.core5:httpcore5:5.2 - https://hc.apache.org/httpcomponents-core-5.2.x/5.2/httpcore5/)
* Apache HttpComponents Core HTTP/2 (org.apache.httpcomponents.core5:httpcore5-h2:5.2 - https://hc.apache.org/httpcomponents-core-5.2.x/5.2/httpcore5-h2/)
* Arrow Format (org.apache.arrow:arrow-format:15.0.0 - https://arrow.apache.org/arrow-format/)
* Arrow Java C Data Interface (org.apache.arrow:arrow-c-data:15.0.0 - https://arrow.apache.org/arrow-c-data/)
* Arrow Java Dataset (org.apache.arrow:arrow-dataset:15.0.0 - https://arrow.apache.org/arrow-dataset/)
* Arrow Memory - Core (org.apache.arrow:arrow-memory-core:15.0.0 - https://arrow.apache.org/arrow-memory/arrow-memory-core/)
* Arrow Memory - Netty (org.apache.arrow:arrow-memory-netty:15.0.0 - https://arrow.apache.org/arrow-memory/arrow-memory-netty/)
* Arrow Vectors (org.apache.arrow:arrow-vector:15.0.0 - https://arrow.apache.org/arrow-vector/)
* Guava: Google Core Libraries for Java (com.google.guava:guava:33.3.1-jre - https://github.com/google/guava)
* J2ObjC Annotations (com.google.j2objc:j2objc-annotations:3.0.0 - https://github.com/google/j2objc/)
* Netty/Buffer (io.netty:netty-buffer:4.1.104.Final - https://netty.io/netty-buffer/)
* Netty/Common (io.netty:netty-common:4.1.104.Final - https://netty.io/netty-common/)
Apache-2.0:
* Apache Commons Lang (org.apache.commons:commons-lang3:3.18.0 - https://commons.apache.org/proper/commons-lang/)
* lance-namespace-apache-client (org.lance:lance-namespace-apache-client:0.4.5 - https://github.com/openapitools/openapi-generator)
* lance-namespace-core (org.lance:lance-namespace-core:0.4.5 - https://lance.org/format/namespace/lance-namespace-core/)
EDL 1.0:
* Jakarta Activation API jar (jakarta.activation:jakarta.activation-api:1.2.2 - https://github.com/eclipse-ee4j/jaf/jakarta.activation-api)
Eclipse Distribution License - v 1.0:
* Eclipse Collections API (org.eclipse.collections:eclipse-collections-api:11.1.0 - https://github.com/eclipse/eclipse-collections/eclipse-collections-api)
* Eclipse Collections Main Library (org.eclipse.collections:eclipse-collections:11.1.0 - https://github.com/eclipse/eclipse-collections/eclipse-collections)
* Jakarta XML Binding API (jakarta.xml.bind:jakarta.xml.bind-api:2.3.3 - https://github.com/eclipse-ee4j/jaxb-api/jakarta.xml.bind-api)
Eclipse Public License - v 1.0:
* Eclipse Collections API (org.eclipse.collections:eclipse-collections-api:11.1.0 - https://github.com/eclipse/eclipse-collections/eclipse-collections-api)
* Eclipse Collections Main Library (org.eclipse.collections:eclipse-collections:11.1.0 - https://github.com/eclipse/eclipse-collections/eclipse-collections)
The Apache Software License, Version 2.0:
* FindBugs-jsr305 (com.google.code.findbugs:jsr305:3.0.2 - http://findbugs.sourceforge.net/)
* Guava InternalFutureFailureAccess and InternalFutures (com.google.guava:failureaccess:1.0.2 - https://github.com/google/guava/failureaccess)
* Guava ListenableFuture only (com.google.guava:listenablefuture:9999.0-empty-to-avoid-conflict-with-guava - https://github.com/google/guava/listenablefuture)
* Jackson datatype: JSR310 (com.fasterxml.jackson.datatype:jackson-datatype-jsr310:2.16.0 - https://github.com/FasterXML/jackson-modules-java8/jackson-datatype-jsr310)
* Jackson module: Old JAXB Annotations (javax.xml.bind) (com.fasterxml.jackson.module:jackson-module-jaxb-annotations:2.17.1 - https://github.com/FasterXML/jackson-modules-base)
* Jackson-annotations (com.fasterxml.jackson.core:jackson-annotations:2.16.0 - https://github.com/FasterXML/jackson)
* Jackson-core (com.fasterxml.jackson.core:jackson-core:2.16.0 - https://github.com/FasterXML/jackson-core)
* jackson-databind (com.fasterxml.jackson.core:jackson-databind:2.15.2 - https://github.com/FasterXML/jackson)
* Jackson-JAXRS: base (com.fasterxml.jackson.jaxrs:jackson-jaxrs-base:2.17.1 - https://github.com/FasterXML/jackson-jaxrs-providers/jackson-jaxrs-base)
* Jackson-JAXRS: JSON (com.fasterxml.jackson.jaxrs:jackson-jaxrs-json-provider:2.17.1 - https://github.com/FasterXML/jackson-jaxrs-providers/jackson-jaxrs-json-provider)
* JAR JNI Loader (org.questdb:jar-jni:1.1.1 - https://github.com/questdb/rust-maven-plugin)
* Lance Core (org.lance:lance-core:2.0.0 - https://lance.org/)
The MIT License:
* Checker Qual (org.checkerframework:checker-qual:3.43.0 - https://checkerframework.org/)
@@ -0,0 +1,71 @@
List of third-party dependencies grouped by their license type.
Apache 2.0:
* error-prone annotations (com.google.errorprone:error_prone_annotations:2.28.0 - https://errorprone.info/error_prone_annotations)
Apache License 2.0:
* JsonNullable Jackson module (org.openapitools:jackson-databind-nullable:0.2.6 - https://github.com/OpenAPITools/jackson-databind-nullable)
Apache License V2.0:
* FlatBuffers Java API (com.google.flatbuffers:flatbuffers-java:23.5.26 - https://github.com/google/flatbuffers)
Apache License, Version 2.0:
* Apache Commons Codec (commons-codec:commons-codec:1.15 - https://commons.apache.org/proper/commons-codec/)
* Apache HttpClient (org.apache.httpcomponents.client5:httpclient5:5.2.1 - https://hc.apache.org/httpcomponents-client-5.0.x/5.2.1/httpclient5/)
* Apache HttpComponents Core HTTP/1.1 (org.apache.httpcomponents.core5:httpcore5:5.2 - https://hc.apache.org/httpcomponents-core-5.2.x/5.2/httpcore5/)
* Apache HttpComponents Core HTTP/2 (org.apache.httpcomponents.core5:httpcore5-h2:5.2 - https://hc.apache.org/httpcomponents-core-5.2.x/5.2/httpcore5-h2/)
* Arrow Format (org.apache.arrow:arrow-format:15.0.0 - https://arrow.apache.org/arrow-format/)
* Arrow Java C Data Interface (org.apache.arrow:arrow-c-data:15.0.0 - https://arrow.apache.org/arrow-c-data/)
* Arrow Java Dataset (org.apache.arrow:arrow-dataset:15.0.0 - https://arrow.apache.org/arrow-dataset/)
* Arrow Memory - Core (org.apache.arrow:arrow-memory-core:15.0.0 - https://arrow.apache.org/arrow-memory/arrow-memory-core/)
* Arrow Memory - Netty (org.apache.arrow:arrow-memory-netty:15.0.0 - https://arrow.apache.org/arrow-memory/arrow-memory-netty/)
* Arrow Vectors (org.apache.arrow:arrow-vector:15.0.0 - https://arrow.apache.org/arrow-vector/)
* Guava: Google Core Libraries for Java (com.google.guava:guava:33.3.1-jre - https://github.com/google/guava)
* J2ObjC Annotations (com.google.j2objc:j2objc-annotations:3.0.0 - https://github.com/google/j2objc/)
* Netty/Buffer (io.netty:netty-buffer:4.1.104.Final - https://netty.io/netty-buffer/)
* Netty/Common (io.netty:netty-common:4.1.104.Final - https://netty.io/netty-common/)
Apache-2.0:
* Apache Commons Lang (org.apache.commons:commons-lang3:3.18.0 - https://commons.apache.org/proper/commons-lang/)
* lance-namespace-apache-client (org.lance:lance-namespace-apache-client:0.4.5 - https://github.com/openapitools/openapi-generator)
* lance-namespace-core (org.lance:lance-namespace-core:0.4.5 - https://lance.org/format/namespace/lance-namespace-core/)
EDL 1.0:
* Jakarta Activation API jar (jakarta.activation:jakarta.activation-api:1.2.2 - https://github.com/eclipse-ee4j/jaf/jakarta.activation-api)
Eclipse Distribution License - v 1.0:
* Eclipse Collections API (org.eclipse.collections:eclipse-collections-api:11.1.0 - https://github.com/eclipse/eclipse-collections/eclipse-collections-api)
* Eclipse Collections Main Library (org.eclipse.collections:eclipse-collections:11.1.0 - https://github.com/eclipse/eclipse-collections/eclipse-collections)
* Jakarta XML Binding API (jakarta.xml.bind:jakarta.xml.bind-api:2.3.3 - https://github.com/eclipse-ee4j/jaxb-api/jakarta.xml.bind-api)
Eclipse Public License - v 1.0:
* Eclipse Collections API (org.eclipse.collections:eclipse-collections-api:11.1.0 - https://github.com/eclipse/eclipse-collections/eclipse-collections-api)
* Eclipse Collections Main Library (org.eclipse.collections:eclipse-collections:11.1.0 - https://github.com/eclipse/eclipse-collections/eclipse-collections)
The Apache Software License, Version 2.0:
* FindBugs-jsr305 (com.google.code.findbugs:jsr305:3.0.2 - http://findbugs.sourceforge.net/)
* Guava InternalFutureFailureAccess and InternalFutures (com.google.guava:failureaccess:1.0.2 - https://github.com/google/guava/failureaccess)
* Guava ListenableFuture only (com.google.guava:listenablefuture:9999.0-empty-to-avoid-conflict-with-guava - https://github.com/google/guava/listenablefuture)
* Jackson datatype: JSR310 (com.fasterxml.jackson.datatype:jackson-datatype-jsr310:2.16.0 - https://github.com/FasterXML/jackson-modules-java8/jackson-datatype-jsr310)
* Jackson module: Old JAXB Annotations (javax.xml.bind) (com.fasterxml.jackson.module:jackson-module-jaxb-annotations:2.17.1 - https://github.com/FasterXML/jackson-modules-base)
* Jackson-annotations (com.fasterxml.jackson.core:jackson-annotations:2.16.0 - https://github.com/FasterXML/jackson)
* Jackson-core (com.fasterxml.jackson.core:jackson-core:2.16.0 - https://github.com/FasterXML/jackson-core)
* jackson-databind (com.fasterxml.jackson.core:jackson-databind:2.15.2 - https://github.com/FasterXML/jackson)
* Jackson-JAXRS: base (com.fasterxml.jackson.jaxrs:jackson-jaxrs-base:2.17.1 - https://github.com/FasterXML/jackson-jaxrs-providers/jackson-jaxrs-base)
* Jackson-JAXRS: JSON (com.fasterxml.jackson.jaxrs:jackson-jaxrs-json-provider:2.17.1 - https://github.com/FasterXML/jackson-jaxrs-providers/jackson-jaxrs-json-provider)
* JAR JNI Loader (org.questdb:jar-jni:1.1.1 - https://github.com/questdb/rust-maven-plugin)
* Lance Core (org.lance:lance-core:2.0.0 - https://lance.org/)
The MIT License:
* Checker Qual (org.checkerframework:checker-qual:3.43.0 - https://checkerframework.org/)
+1 -1
View File
@@ -8,7 +8,7 @@
<parent> <parent>
<groupId>com.lancedb</groupId> <groupId>com.lancedb</groupId>
<artifactId>lancedb-parent</artifactId> <artifactId>lancedb-parent</artifactId>
<version>0.26.1-final.0</version> <version>0.26.2-final.0</version>
<relativePath>../pom.xml</relativePath> <relativePath>../pom.xml</relativePath>
</parent> </parent>
+14 -1
View File
@@ -6,7 +6,7 @@
<groupId>com.lancedb</groupId> <groupId>com.lancedb</groupId>
<artifactId>lancedb-parent</artifactId> <artifactId>lancedb-parent</artifactId>
<version>0.26.1-final.0</version> <version>0.26.2-final.0</version>
<packaging>pom</packaging> <packaging>pom</packaging>
<name>${project.artifactId}</name> <name>${project.artifactId}</name>
<description>LanceDB Java SDK Parent POM</description> <description>LanceDB Java SDK Parent POM</description>
@@ -160,6 +160,19 @@
<groupId>com.diffplug.spotless</groupId> <groupId>com.diffplug.spotless</groupId>
<artifactId>spotless-maven-plugin</artifactId> <artifactId>spotless-maven-plugin</artifactId>
</plugin> </plugin>
<plugin>
<groupId>org.codehaus.mojo</groupId>
<artifactId>license-maven-plugin</artifactId>
<version>2.4.0</version>
<configuration>
<outputDirectory>${project.basedir}</outputDirectory>
<thirdPartyFilename>JAVA_THIRD_PARTY_LICENSES.md</thirdPartyFilename>
<fileTemplate>/org/codehaus/mojo/license/third-party-file-groupByLicense.ftl</fileTemplate>
<includedScopes>compile,runtime</includedScopes>
<excludedScopes>test,provided</excludedScopes>
<sortArtifactByName>true</sortArtifactByName>
</configuration>
</plugin>
</plugins> </plugins>
<pluginManagement> <pluginManagement>
<plugins> <plugins>
+1 -1
View File
@@ -1,7 +1,7 @@
[package] [package]
name = "lancedb-nodejs" name = "lancedb-nodejs"
edition.workspace = true edition.workspace = true
version = "0.26.1" version = "0.26.2"
license.workspace = true license.workspace = true
description.workspace = true description.workspace = true
repository.workspace = true repository.workspace = true
+668
View File
@@ -0,0 +1,668 @@
[@75lb/deep-merge@1.1.2](https://github.com/75lb/deep-merge) - MIT
[@aashutoshrathi/word-wrap@1.2.6](https://github.com/aashutoshrathi/word-wrap) - MIT
[@ampproject/remapping@2.2.1](https://github.com/ampproject/remapping) - Apache-2.0
[@aws-crypto/crc32@3.0.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/crc32c@3.0.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/ie11-detection@3.0.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/sha1-browser@3.0.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/sha256-browser@3.0.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/sha256-browser@5.2.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/sha256-js@3.0.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/sha256-js@5.2.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/supports-web-crypto@3.0.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/supports-web-crypto@5.2.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/util@3.0.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-crypto/util@5.2.0](https://github.com/aws/aws-sdk-js-crypto-helpers) - Apache-2.0
[@aws-sdk/client-dynamodb@3.602.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/client-kms@3.549.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/client-s3@3.550.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/client-sso-oidc@3.549.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/client-sso-oidc@3.600.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/client-sso@3.549.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/client-sso@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/client-sts@3.549.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/client-sts@3.600.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/core@3.549.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/core@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-env@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-env@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-http@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-http@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-ini@3.549.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-ini@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-node@3.549.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-node@3.600.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-process@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-process@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-sso@3.549.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-sso@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-web-identity@3.549.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/credential-provider-web-identity@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/endpoint-cache@3.572.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-bucket-endpoint@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-endpoint-discovery@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-expect-continue@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-flexible-checksums@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-host-header@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-host-header@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-location-constraint@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-logger@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-logger@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-recursion-detection@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-recursion-detection@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-sdk-s3@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-signing@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-ssec@3.537.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-user-agent@3.540.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/middleware-user-agent@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/region-config-resolver@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/region-config-resolver@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/signature-v4-multi-region@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/token-providers@3.549.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/token-providers@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/types@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/types@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/util-arn-parser@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/util-endpoints@3.540.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/util-endpoints@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/util-locate-window@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/util-user-agent-browser@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/util-user-agent-browser@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/util-user-agent-node@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/util-user-agent-node@3.598.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/util-utf8-browser@3.259.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@aws-sdk/xml-builder@3.535.0](https://github.com/aws/aws-sdk-js-v3) - Apache-2.0
[@babel/code-frame@7.26.2](https://github.com/babel/babel) - MIT
[@babel/compat-data@7.23.5](https://github.com/babel/babel) - MIT
[@babel/core@7.23.7](https://github.com/babel/babel) - MIT
[@babel/generator@7.23.6](https://github.com/babel/babel) - MIT
[@babel/helper-compilation-targets@7.23.6](https://github.com/babel/babel) - MIT
[@babel/helper-environment-visitor@7.22.20](https://github.com/babel/babel) - MIT
[@babel/helper-function-name@7.23.0](https://github.com/babel/babel) - MIT
[@babel/helper-hoist-variables@7.22.5](https://github.com/babel/babel) - MIT
[@babel/helper-module-imports@7.22.15](https://github.com/babel/babel) - MIT
[@babel/helper-module-transforms@7.23.3](https://github.com/babel/babel) - MIT
[@babel/helper-plugin-utils@7.22.5](https://github.com/babel/babel) - MIT
[@babel/helper-simple-access@7.22.5](https://github.com/babel/babel) - MIT
[@babel/helper-split-export-declaration@7.22.6](https://github.com/babel/babel) - MIT
[@babel/helper-string-parser@7.25.9](https://github.com/babel/babel) - MIT
[@babel/helper-validator-identifier@7.25.9](https://github.com/babel/babel) - MIT
[@babel/helper-validator-option@7.23.5](https://github.com/babel/babel) - MIT
[@babel/helpers@7.27.0](https://github.com/babel/babel) - MIT
[@babel/parser@7.27.0](https://github.com/babel/babel) - MIT
[@babel/plugin-syntax-async-generators@7.8.4](https://github.com/babel/babel/tree/master/packages/babel-plugin-syntax-async-generators) - MIT
[@babel/plugin-syntax-bigint@7.8.3](https://github.com/babel/babel/tree/master/packages/babel-plugin-syntax-bigint) - MIT
[@babel/plugin-syntax-class-properties@7.12.13](https://github.com/babel/babel) - MIT
[@babel/plugin-syntax-import-meta@7.10.4](https://github.com/babel/babel) - MIT
[@babel/plugin-syntax-json-strings@7.8.3](https://github.com/babel/babel/tree/master/packages/babel-plugin-syntax-json-strings) - MIT
[@babel/plugin-syntax-jsx@7.23.3](https://github.com/babel/babel) - MIT
[@babel/plugin-syntax-logical-assignment-operators@7.10.4](https://github.com/babel/babel) - MIT
[@babel/plugin-syntax-nullish-coalescing-operator@7.8.3](https://github.com/babel/babel/tree/master/packages/babel-plugin-syntax-nullish-coalescing-operator) - MIT
[@babel/plugin-syntax-numeric-separator@7.10.4](https://github.com/babel/babel) - MIT
[@babel/plugin-syntax-object-rest-spread@7.8.3](https://github.com/babel/babel/tree/master/packages/babel-plugin-syntax-object-rest-spread) - MIT
[@babel/plugin-syntax-optional-catch-binding@7.8.3](https://github.com/babel/babel/tree/master/packages/babel-plugin-syntax-optional-catch-binding) - MIT
[@babel/plugin-syntax-optional-chaining@7.8.3](https://github.com/babel/babel/tree/master/packages/babel-plugin-syntax-optional-chaining) - MIT
[@babel/plugin-syntax-top-level-await@7.14.5](https://github.com/babel/babel) - MIT
[@babel/plugin-syntax-typescript@7.23.3](https://github.com/babel/babel) - MIT
[@babel/template@7.27.0](https://github.com/babel/babel) - MIT
[@babel/traverse@7.23.7](https://github.com/babel/babel) - MIT
[@babel/types@7.27.0](https://github.com/babel/babel) - MIT
[@bcoe/v8-coverage@0.2.3](https://github.com/demurgos/v8-coverage) - MIT
[@biomejs/biome@1.8.3](https://github.com/biomejs/biome) - MIT OR Apache-2.0
[@biomejs/cli-darwin-arm64@1.8.3](https://github.com/biomejs/biome) - MIT OR Apache-2.0
[@eslint-community/eslint-utils@4.4.0](https://github.com/eslint-community/eslint-utils) - MIT
[@eslint-community/regexpp@4.10.0](https://github.com/eslint-community/regexpp) - MIT
[@eslint/eslintrc@2.1.4](https://github.com/eslint/eslintrc) - MIT
[@eslint/js@8.57.0](https://github.com/eslint/eslint) - MIT
[@huggingface/jinja@0.3.2](https://github.com/huggingface/huggingface.js) - MIT
[@huggingface/transformers@3.0.2](https://github.com/huggingface/transformers.js) - Apache-2.0
[@humanwhocodes/config-array@0.11.14](https://github.com/humanwhocodes/config-array) - Apache-2.0
[@humanwhocodes/module-importer@1.0.1](https://github.com/humanwhocodes/module-importer) - Apache-2.0
[@humanwhocodes/object-schema@2.0.2](https://github.com/humanwhocodes/object-schema) - BSD-3-Clause
[@img/sharp-darwin-arm64@0.33.5](https://github.com/lovell/sharp) - Apache-2.0
[@img/sharp-libvips-darwin-arm64@1.0.4](https://github.com/lovell/sharp-libvips) - LGPL-3.0-or-later
[@isaacs/cliui@8.0.2](https://github.com/yargs/cliui) - ISC
[@isaacs/fs-minipass@4.0.1](https://github.com/npm/fs-minipass) - ISC
[@istanbuljs/load-nyc-config@1.1.0](https://github.com/istanbuljs/load-nyc-config) - ISC
[@istanbuljs/schema@0.1.3](https://github.com/istanbuljs/schema) - MIT
[@jest/console@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/core@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/environment@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/expect-utils@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/expect@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/fake-timers@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/globals@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/reporters@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/schemas@29.6.3](https://github.com/jestjs/jest) - MIT
[@jest/source-map@29.6.3](https://github.com/jestjs/jest) - MIT
[@jest/test-result@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/test-sequencer@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/transform@29.7.0](https://github.com/jestjs/jest) - MIT
[@jest/types@29.6.3](https://github.com/jestjs/jest) - MIT
[@jridgewell/gen-mapping@0.3.3](https://github.com/jridgewell/gen-mapping) - MIT
[@jridgewell/resolve-uri@3.1.1](https://github.com/jridgewell/resolve-uri) - MIT
[@jridgewell/set-array@1.1.2](https://github.com/jridgewell/set-array) - MIT
[@jridgewell/sourcemap-codec@1.4.15](https://github.com/jridgewell/sourcemap-codec) - MIT
[@jridgewell/trace-mapping@0.3.22](https://github.com/jridgewell/trace-mapping) - MIT
[@lancedb/lancedb@0.26.2](https://github.com/lancedb/lancedb) - Apache-2.0
[@napi-rs/cli@2.18.3](https://github.com/napi-rs/napi-rs) - MIT
[@nodelib/fs.scandir@2.1.5](https://github.com/nodelib/nodelib/tree/master/packages/fs/fs.scandir) - MIT
[@nodelib/fs.stat@2.0.5](https://github.com/nodelib/nodelib/tree/master/packages/fs/fs.stat) - MIT
[@nodelib/fs.walk@1.2.8](https://github.com/nodelib/nodelib/tree/master/packages/fs/fs.walk) - MIT
[@pkgjs/parseargs@0.11.0](https://github.com/pkgjs/parseargs) - MIT
[@protobufjs/aspromise@1.1.2](https://github.com/dcodeIO/protobuf.js) - BSD-3-Clause
[@protobufjs/base64@1.1.2](https://github.com/dcodeIO/protobuf.js) - BSD-3-Clause
[@protobufjs/codegen@2.0.4](https://github.com/dcodeIO/protobuf.js) - BSD-3-Clause
[@protobufjs/eventemitter@1.1.0](https://github.com/dcodeIO/protobuf.js) - BSD-3-Clause
[@protobufjs/fetch@1.1.0](https://github.com/dcodeIO/protobuf.js) - BSD-3-Clause
[@protobufjs/float@1.0.2](https://github.com/dcodeIO/protobuf.js) - BSD-3-Clause
[@protobufjs/inquire@1.1.0](https://github.com/dcodeIO/protobuf.js) - BSD-3-Clause
[@protobufjs/path@1.1.2](https://github.com/dcodeIO/protobuf.js) - BSD-3-Clause
[@protobufjs/pool@1.1.0](https://github.com/dcodeIO/protobuf.js) - BSD-3-Clause
[@protobufjs/utf8@1.1.0](https://github.com/dcodeIO/protobuf.js) - BSD-3-Clause
[@shikijs/core@1.10.3](https://github.com/shikijs/shiki) - MIT
[@sinclair/typebox@0.27.8](https://github.com/sinclairzx81/typebox) - MIT
[@sinonjs/commons@3.0.1](https://github.com/sinonjs/commons) - BSD-3-Clause
[@sinonjs/fake-timers@10.3.0](https://github.com/sinonjs/fake-timers) - BSD-3-Clause
[@smithy/abort-controller@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/abort-controller@3.1.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/chunked-blob-reader-native@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/chunked-blob-reader@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/config-resolver@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/config-resolver@3.0.3](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/core@1.4.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/core@2.2.3](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/credential-provider-imds@2.3.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/credential-provider-imds@3.1.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/eventstream-codec@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/eventstream-serde-browser@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/eventstream-serde-config-resolver@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/eventstream-serde-node@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/eventstream-serde-universal@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/fetch-http-handler@2.5.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/fetch-http-handler@3.1.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/hash-blob-browser@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/hash-node@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/hash-node@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/hash-stream-node@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/invalid-dependency@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/invalid-dependency@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/is-array-buffer@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/is-array-buffer@3.0.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/md5-js@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/middleware-content-length@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/middleware-content-length@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/middleware-endpoint@2.5.1](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/middleware-endpoint@3.0.3](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/middleware-retry@2.3.1](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/middleware-retry@3.0.6](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/middleware-serde@2.3.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/middleware-serde@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/middleware-stack@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/middleware-stack@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/node-config-provider@2.3.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/node-config-provider@3.1.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/node-http-handler@2.5.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/node-http-handler@3.1.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/property-provider@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/property-provider@3.1.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/protocol-http@3.3.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/protocol-http@4.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/querystring-builder@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/querystring-builder@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/querystring-parser@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/querystring-parser@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/service-error-classification@2.1.5](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/service-error-classification@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/shared-ini-file-loader@2.4.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/shared-ini-file-loader@3.1.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/signature-v4@2.2.1](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/signature-v4@3.1.1](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/smithy-client@2.5.1](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/smithy-client@3.1.4](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/types@2.12.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/types@3.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/url-parser@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/url-parser@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-base64@2.3.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-base64@3.0.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-body-length-browser@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-body-length-browser@3.0.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-body-length-node@2.3.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-body-length-node@3.0.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-buffer-from@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-buffer-from@3.0.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-config-provider@2.3.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-config-provider@3.0.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-defaults-mode-browser@2.2.1](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-defaults-mode-browser@3.0.6](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-defaults-mode-node@2.3.1](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-defaults-mode-node@3.0.6](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-endpoints@1.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-endpoints@2.0.3](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-hex-encoding@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-hex-encoding@3.0.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-middleware@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-middleware@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-retry@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-retry@3.0.2](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-stream@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-stream@3.0.4](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-uri-escape@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-uri-escape@3.0.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-utf8@2.3.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-utf8@3.0.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-waiter@2.2.0](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@smithy/util-waiter@3.1.1](https://github.com/awslabs/smithy-typescript) - Apache-2.0
[@swc/helpers@0.5.12](https://github.com/swc-project/swc) - Apache-2.0
[@types/axios@0.14.0](https://github.com/mzabriskie/axios) - MIT
[@types/babel__core@7.20.5](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/babel__generator@7.6.8](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/babel__template@7.4.4](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/babel__traverse@7.20.5](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/command-line-args@5.2.3](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/command-line-usage@5.0.2](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/command-line-usage@5.0.4](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/graceful-fs@4.1.9](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/hast@3.0.4](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/istanbul-lib-coverage@2.0.6](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/istanbul-lib-report@3.0.3](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/istanbul-reports@3.0.4](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/jest@29.5.12](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/json-schema@7.0.15](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/node-fetch@2.6.11](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/node@18.19.26](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/node@20.16.10](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/node@20.17.9](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/node@22.7.4](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/semver@7.5.6](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/stack-utils@2.0.3](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/tmp@0.2.6](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/unist@3.0.2](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/yargs-parser@21.0.3](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@types/yargs@17.0.32](https://github.com/DefinitelyTyped/DefinitelyTyped) - MIT
[@typescript-eslint/eslint-plugin@7.1.0](https://github.com/typescript-eslint/typescript-eslint) - MIT
[@typescript-eslint/parser@7.1.0](https://github.com/typescript-eslint/typescript-eslint) - BSD-2-Clause
[@typescript-eslint/scope-manager@7.1.0](https://github.com/typescript-eslint/typescript-eslint) - MIT
[@typescript-eslint/type-utils@7.1.0](https://github.com/typescript-eslint/typescript-eslint) - MIT
[@typescript-eslint/types@7.1.0](https://github.com/typescript-eslint/typescript-eslint) - MIT
[@typescript-eslint/typescript-estree@7.1.0](https://github.com/typescript-eslint/typescript-eslint) - BSD-2-Clause
[@typescript-eslint/utils@7.1.0](https://github.com/typescript-eslint/typescript-eslint) - MIT
[@typescript-eslint/visitor-keys@7.1.0](https://github.com/typescript-eslint/typescript-eslint) - MIT
[@ungap/structured-clone@1.2.0](https://github.com/ungap/structured-clone) - ISC
[abort-controller@3.0.0](https://github.com/mysticatea/abort-controller) - MIT
[acorn-jsx@5.3.2](https://github.com/acornjs/acorn-jsx) - MIT
[acorn@8.11.3](https://github.com/acornjs/acorn) - MIT
[agentkeepalive@4.5.0](https://github.com/node-modules/agentkeepalive) - MIT
[ajv@6.12.6](https://github.com/ajv-validator/ajv) - MIT
[ansi-escapes@4.3.2](https://github.com/sindresorhus/ansi-escapes) - MIT
[ansi-regex@5.0.1](https://github.com/chalk/ansi-regex) - MIT
[ansi-regex@6.1.0](https://github.com/chalk/ansi-regex) - MIT
[ansi-styles@4.3.0](https://github.com/chalk/ansi-styles) - MIT
[ansi-styles@5.2.0](https://github.com/chalk/ansi-styles) - MIT
[ansi-styles@6.2.1](https://github.com/chalk/ansi-styles) - MIT
[anymatch@3.1.3](https://github.com/micromatch/anymatch) - ISC
[apache-arrow@15.0.0](https://github.com/apache/arrow) - Apache-2.0
[apache-arrow@16.0.0](https://github.com/apache/arrow) - Apache-2.0
[apache-arrow@17.0.0](https://github.com/apache/arrow) - Apache-2.0
[apache-arrow@18.0.0](https://github.com/apache/arrow) - Apache-2.0
[argparse@1.0.10](https://github.com/nodeca/argparse) - MIT
[argparse@2.0.1](https://github.com/nodeca/argparse) - Python-2.0
[array-back@3.1.0](https://github.com/75lb/array-back) - MIT
[array-back@6.2.2](https://github.com/75lb/array-back) - MIT
[array-union@2.1.0](https://github.com/sindresorhus/array-union) - MIT
[asynckit@0.4.0](https://github.com/alexindigo/asynckit) - MIT
[axios@1.8.4](https://github.com/axios/axios) - MIT
[babel-jest@29.7.0](https://github.com/jestjs/jest) - MIT
[babel-plugin-istanbul@6.1.1](https://github.com/istanbuljs/babel-plugin-istanbul) - BSD-3-Clause
[babel-plugin-jest-hoist@29.6.3](https://github.com/jestjs/jest) - MIT
[babel-preset-current-node-syntax@1.0.1](https://github.com/nicolo-ribaudo/babel-preset-current-node-syntax) - MIT
[babel-preset-jest@29.6.3](https://github.com/jestjs/jest) - MIT
[balanced-match@1.0.2](https://github.com/juliangruber/balanced-match) - MIT
[base-64@0.1.0](https://github.com/mathiasbynens/base64) - MIT
[bowser@2.11.0](https://github.com/lancedikson/bowser) - MIT
[brace-expansion@1.1.11](https://github.com/juliangruber/brace-expansion) - MIT
[brace-expansion@2.0.1](https://github.com/juliangruber/brace-expansion) - MIT
[braces@3.0.3](https://github.com/micromatch/braces) - MIT
[browserslist@4.22.2](https://github.com/browserslist/browserslist) - MIT
[bs-logger@0.2.6](https://github.com/huafu/bs-logger) - MIT
[bser@2.1.1](https://github.com/facebook/watchman) - Apache-2.0
[buffer-from@1.1.2](https://github.com/LinusU/buffer-from) - MIT
[callsites@3.1.0](https://github.com/sindresorhus/callsites) - MIT
[camelcase@5.3.1](https://github.com/sindresorhus/camelcase) - MIT
[camelcase@6.3.0](https://github.com/sindresorhus/camelcase) - MIT
[caniuse-lite@1.0.30001579](https://github.com/browserslist/caniuse-lite) - CC-BY-4.0
[chalk-template@0.4.0](https://github.com/chalk/chalk-template) - MIT
[chalk@4.1.2](https://github.com/chalk/chalk) - MIT
[char-regex@1.0.2](https://github.com/Richienb/char-regex) - MIT
[charenc@0.0.2](https://github.com/pvorb/node-charenc) - BSD-3-Clause
[chownr@3.0.0](https://github.com/isaacs/chownr) - BlueOak-1.0.0
[ci-info@3.9.0](https://github.com/watson/ci-info) - MIT
[cjs-module-lexer@1.2.3](https://github.com/nodejs/cjs-module-lexer) - MIT
[cliui@8.0.1](https://github.com/yargs/cliui) - ISC
[co@4.6.0](https://github.com/tj/co) - MIT
[collect-v8-coverage@1.0.2](https://github.com/SimenB/collect-v8-coverage) - MIT
[color-convert@2.0.1](https://github.com/Qix-/color-convert) - MIT
[color-name@1.1.4](https://github.com/colorjs/color-name) - MIT
[color-string@1.9.1](https://github.com/Qix-/color-string) - MIT
[color@4.2.3](https://github.com/Qix-/color) - MIT
[combined-stream@1.0.8](https://github.com/felixge/node-combined-stream) - MIT
[command-line-args@5.2.1](https://github.com/75lb/command-line-args) - MIT
[command-line-usage@7.0.1](https://github.com/75lb/command-line-usage) - MIT
[concat-map@0.0.1](https://github.com/substack/node-concat-map) - MIT
[convert-source-map@2.0.0](https://github.com/thlorenz/convert-source-map) - MIT
[create-jest@29.7.0](https://github.com/jestjs/jest) - MIT
[cross-spawn@7.0.6](https://github.com/moxystudio/node-cross-spawn) - MIT
[crypt@0.0.2](https://github.com/pvorb/node-crypt) - BSD-3-Clause
[debug@4.3.4](https://github.com/debug-js/debug) - MIT
[dedent@1.5.1](https://github.com/dmnd/dedent) - MIT
[deep-is@0.1.4](https://github.com/thlorenz/deep-is) - MIT
[deepmerge@4.3.1](https://github.com/TehShrike/deepmerge) - MIT
[delayed-stream@1.0.0](https://github.com/felixge/node-delayed-stream) - MIT
[detect-libc@2.0.3](https://github.com/lovell/detect-libc) - Apache-2.0
[detect-newline@3.1.0](https://github.com/sindresorhus/detect-newline) - MIT
[diff-sequences@29.6.3](https://github.com/jestjs/jest) - MIT
[digest-fetch@1.3.0](https://github.com/devfans/digest-fetch) - ISC
[dir-glob@3.0.1](https://github.com/kevva/dir-glob) - MIT
[doctrine@3.0.0](https://github.com/eslint/doctrine) - Apache-2.0
[eastasianwidth@0.2.0](https://github.com/komagata/eastasianwidth) - MIT
[electron-to-chromium@1.4.642](https://github.com/kilian/electron-to-chromium) - ISC
[emittery@0.13.1](https://github.com/sindresorhus/emittery) - MIT
[emoji-regex@8.0.0](https://github.com/mathiasbynens/emoji-regex) - MIT
[emoji-regex@9.2.2](https://github.com/mathiasbynens/emoji-regex) - MIT
[entities@4.5.0](https://github.com/fb55/entities) - BSD-2-Clause
[error-ex@1.3.2](https://github.com/qix-/node-error-ex) - MIT
[escalade@3.1.1](https://github.com/lukeed/escalade) - MIT
[escape-string-regexp@2.0.0](https://github.com/sindresorhus/escape-string-regexp) - MIT
[escape-string-regexp@4.0.0](https://github.com/sindresorhus/escape-string-regexp) - MIT
[eslint-scope@7.2.2](https://github.com/eslint/eslint-scope) - BSD-2-Clause
[eslint-visitor-keys@3.4.3](https://github.com/eslint/eslint-visitor-keys) - Apache-2.0
[eslint@8.57.0](https://github.com/eslint/eslint) - MIT
[espree@9.6.1](https://github.com/eslint/espree) - BSD-2-Clause
[esprima@4.0.1](https://github.com/jquery/esprima) - BSD-2-Clause
[esquery@1.5.0](https://github.com/estools/esquery) - BSD-3-Clause
[esrecurse@4.3.0](https://github.com/estools/esrecurse) - BSD-2-Clause
[estraverse@5.3.0](https://github.com/estools/estraverse) - BSD-2-Clause
[esutils@2.0.3](https://github.com/estools/esutils) - BSD-2-Clause
[event-target-shim@5.0.1](https://github.com/mysticatea/event-target-shim) - MIT
[execa@5.1.1](https://github.com/sindresorhus/execa) - MIT
[exit@0.1.2](https://github.com/cowboy/node-exit) - MIT
[expect@29.7.0](https://github.com/jestjs/jest) - MIT
[fast-deep-equal@3.1.3](https://github.com/epoberezkin/fast-deep-equal) - MIT
[fast-glob@3.3.2](https://github.com/mrmlnc/fast-glob) - MIT
[fast-json-stable-stringify@2.1.0](https://github.com/epoberezkin/fast-json-stable-stringify) - MIT
[fast-levenshtein@2.0.6](https://github.com/hiddentao/fast-levenshtein) - MIT
[fast-xml-parser@4.2.5](https://github.com/NaturalIntelligence/fast-xml-parser) - MIT
[fastq@1.16.0](https://github.com/mcollina/fastq) - ISC
[fb-watchman@2.0.2](https://github.com/facebook/watchman) - Apache-2.0
[file-entry-cache@6.0.1](https://github.com/royriojas/file-entry-cache) - MIT
[fill-range@7.1.1](https://github.com/jonschlinkert/fill-range) - MIT
[find-replace@3.0.0](https://github.com/75lb/find-replace) - MIT
[find-up@4.1.0](https://github.com/sindresorhus/find-up) - MIT
[find-up@5.0.0](https://github.com/sindresorhus/find-up) - MIT
[flat-cache@3.2.0](https://github.com/jaredwray/flat-cache) - MIT
[flatbuffers@1.12.0](https://github.com/google/flatbuffers) - Apache*
[flatbuffers@23.5.26](https://github.com/google/flatbuffers) - Apache*
[flatbuffers@24.3.25](https://github.com/google/flatbuffers) - Apache-2.0
[flatted@3.2.9](https://github.com/WebReflection/flatted) - ISC
[follow-redirects@1.15.6](https://github.com/follow-redirects/follow-redirects) - MIT
[foreground-child@3.3.0](https://github.com/tapjs/foreground-child) - ISC
[form-data-encoder@1.7.2](https://github.com/octet-stream/form-data-encoder) - MIT
[form-data@4.0.0](https://github.com/form-data/form-data) - MIT
[formdata-node@4.4.1](https://github.com/octet-stream/form-data) - MIT
[fs.realpath@1.0.0](https://github.com/isaacs/fs.realpath) - ISC
[fsevents@2.3.3](https://github.com/fsevents/fsevents) - MIT
[function-bind@1.1.2](https://github.com/Raynos/function-bind) - MIT
[gensync@1.0.0-beta.2](https://github.com/loganfsmyth/gensync) - MIT
[get-caller-file@2.0.5](https://github.com/stefanpenner/get-caller-file) - ISC
[get-package-type@0.1.0](https://github.com/cfware/get-package-type) - MIT
[get-stream@6.0.1](https://github.com/sindresorhus/get-stream) - MIT
[glob-parent@5.1.2](https://github.com/gulpjs/glob-parent) - ISC
[glob-parent@6.0.2](https://github.com/gulpjs/glob-parent) - ISC
[glob@10.4.5](https://github.com/isaacs/node-glob) - ISC
[glob@7.2.3](https://github.com/isaacs/node-glob) - ISC
[globals@11.12.0](https://github.com/sindresorhus/globals) - MIT
[globals@13.24.0](https://github.com/sindresorhus/globals) - MIT
[globby@11.1.0](https://github.com/sindresorhus/globby) - MIT
[graceful-fs@4.2.11](https://github.com/isaacs/node-graceful-fs) - ISC
[graphemer@1.4.0](https://github.com/flmnt/graphemer) - MIT
[guid-typescript@1.0.9](https://github.com/NicolasDeveloper/guid-typescript) - ISC
[has-flag@4.0.0](https://github.com/sindresorhus/has-flag) - MIT
[hasown@2.0.0](https://github.com/inspect-js/hasOwn) - MIT
[html-escaper@2.0.2](https://github.com/WebReflection/html-escaper) - MIT
[human-signals@2.1.0](https://github.com/ehmicky/human-signals) - Apache-2.0
[humanize-ms@1.2.1](https://github.com/node-modules/humanize-ms) - MIT
[ignore@5.3.0](https://github.com/kaelzhang/node-ignore) - MIT
[import-fresh@3.3.0](https://github.com/sindresorhus/import-fresh) - MIT
[import-local@3.1.0](https://github.com/sindresorhus/import-local) - MIT
[imurmurhash@0.1.4](https://github.com/jensyt/imurmurhash-js) - MIT
[inflight@1.0.6](https://github.com/npm/inflight) - ISC
[inherits@2.0.4](https://github.com/isaacs/inherits) - ISC
[interpret@1.4.0](https://github.com/gulpjs/interpret) - MIT
[is-arrayish@0.2.1](https://github.com/qix-/node-is-arrayish) - MIT
[is-arrayish@0.3.2](https://github.com/qix-/node-is-arrayish) - MIT
[is-buffer@1.1.6](https://github.com/feross/is-buffer) - MIT
[is-core-module@2.13.1](https://github.com/inspect-js/is-core-module) - MIT
[is-extglob@2.1.1](https://github.com/jonschlinkert/is-extglob) - MIT
[is-fullwidth-code-point@3.0.0](https://github.com/sindresorhus/is-fullwidth-code-point) - MIT
[is-generator-fn@2.1.0](https://github.com/sindresorhus/is-generator-fn) - MIT
[is-glob@4.0.3](https://github.com/micromatch/is-glob) - MIT
[is-number@7.0.0](https://github.com/jonschlinkert/is-number) - MIT
[is-path-inside@3.0.3](https://github.com/sindresorhus/is-path-inside) - MIT
[is-stream@2.0.1](https://github.com/sindresorhus/is-stream) - MIT
[isexe@2.0.0](https://github.com/isaacs/isexe) - ISC
[istanbul-lib-coverage@3.2.2](https://github.com/istanbuljs/istanbuljs) - BSD-3-Clause
[istanbul-lib-instrument@5.2.1](https://github.com/istanbuljs/istanbuljs) - BSD-3-Clause
[istanbul-lib-instrument@6.0.1](https://github.com/istanbuljs/istanbuljs) - BSD-3-Clause
[istanbul-lib-report@3.0.1](https://github.com/istanbuljs/istanbuljs) - BSD-3-Clause
[istanbul-lib-source-maps@4.0.1](https://github.com/istanbuljs/istanbuljs) - BSD-3-Clause
[istanbul-reports@3.1.6](https://github.com/istanbuljs/istanbuljs) - BSD-3-Clause
[jackspeak@3.4.3](https://github.com/isaacs/jackspeak) - BlueOak-1.0.0
[jest-changed-files@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-circus@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-cli@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-config@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-diff@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-docblock@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-each@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-environment-node@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-get-type@29.6.3](https://github.com/jestjs/jest) - MIT
[jest-haste-map@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-leak-detector@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-matcher-utils@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-message-util@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-mock@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-pnp-resolver@1.2.3](https://github.com/arcanis/jest-pnp-resolver) - MIT
[jest-regex-util@29.6.3](https://github.com/jestjs/jest) - MIT
[jest-resolve-dependencies@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-resolve@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-runner@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-runtime@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-snapshot@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-util@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-validate@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-watcher@29.7.0](https://github.com/jestjs/jest) - MIT
[jest-worker@29.7.0](https://github.com/jestjs/jest) - MIT
[jest@29.7.0](https://github.com/jestjs/jest) - MIT
[js-tokens@4.0.0](https://github.com/lydell/js-tokens) - MIT
[js-yaml@3.14.1](https://github.com/nodeca/js-yaml) - MIT
[js-yaml@4.1.0](https://github.com/nodeca/js-yaml) - MIT
[jsesc@2.5.2](https://github.com/mathiasbynens/jsesc) - MIT
[json-bignum@0.0.3](https://github.com/datalanche/json-bignum) - MIT
[json-buffer@3.0.1](https://github.com/dominictarr/json-buffer) - MIT
[json-parse-even-better-errors@2.3.1](https://github.com/npm/json-parse-even-better-errors) - MIT
[json-schema-traverse@0.4.1](https://github.com/epoberezkin/json-schema-traverse) - MIT
[json-stable-stringify-without-jsonify@1.0.1](https://github.com/samn/json-stable-stringify) - MIT
[json5@2.2.3](https://github.com/json5/json5) - MIT
[keyv@4.5.4](https://github.com/jaredwray/keyv) - MIT
[kleur@3.0.3](https://github.com/lukeed/kleur) - MIT
[leven@3.1.0](https://github.com/sindresorhus/leven) - MIT
[levn@0.4.1](https://github.com/gkz/levn) - MIT
[lines-and-columns@1.2.4](https://github.com/eventualbuddha/lines-and-columns) - MIT
[linkify-it@5.0.0](https://github.com/markdown-it/linkify-it) - MIT
[locate-path@5.0.0](https://github.com/sindresorhus/locate-path) - MIT
[locate-path@6.0.0](https://github.com/sindresorhus/locate-path) - MIT
[lodash.camelcase@4.3.0](https://github.com/lodash/lodash) - MIT
[lodash.memoize@4.1.2](https://github.com/lodash/lodash) - MIT
[lodash.merge@4.6.2](https://github.com/lodash/lodash) - MIT
[lodash@4.17.21](https://github.com/lodash/lodash) - MIT
[long@5.2.3](https://github.com/dcodeIO/long.js) - Apache-2.0
[lru-cache@10.4.3](https://github.com/isaacs/node-lru-cache) - ISC
[lru-cache@5.1.1](https://github.com/isaacs/node-lru-cache) - ISC
[lunr@2.3.9](https://github.com/olivernn/lunr.js) - MIT
[make-dir@4.0.0](https://github.com/sindresorhus/make-dir) - MIT
[make-error@1.3.6](https://github.com/JsCommunity/make-error) - ISC
[makeerror@1.0.12](https://github.com/daaku/nodejs-makeerror) - BSD-3-Clause
[markdown-it@14.1.0](https://github.com/markdown-it/markdown-it) - MIT
[md5@2.3.0](https://github.com/pvorb/node-md5) - BSD-3-Clause
[mdurl@2.0.0](https://github.com/markdown-it/mdurl) - MIT
[merge-stream@2.0.0](https://github.com/grncdr/merge-stream) - MIT
[merge2@1.4.1](https://github.com/teambition/merge2) - MIT
[micromatch@4.0.8](https://github.com/micromatch/micromatch) - MIT
[mime-db@1.52.0](https://github.com/jshttp/mime-db) - MIT
[mime-types@2.1.35](https://github.com/jshttp/mime-types) - MIT
[mimic-fn@2.1.0](https://github.com/sindresorhus/mimic-fn) - MIT
[minimatch@3.1.2](https://github.com/isaacs/minimatch) - ISC
[minimatch@9.0.3](https://github.com/isaacs/minimatch) - ISC
[minimatch@9.0.5](https://github.com/isaacs/minimatch) - ISC
[minimist@1.2.8](https://github.com/minimistjs/minimist) - MIT
[minipass@7.1.2](https://github.com/isaacs/minipass) - ISC
[minizlib@3.0.1](https://github.com/isaacs/minizlib) - MIT
[mkdirp@3.0.1](https://github.com/isaacs/node-mkdirp) - MIT
[mnemonist@0.38.3](https://github.com/yomguithereal/mnemonist) - MIT
[ms@2.1.2](https://github.com/zeit/ms) - MIT
[ms@2.1.3](https://github.com/vercel/ms) - MIT
[natural-compare@1.4.0](https://github.com/litejs/natural-compare-lite) - MIT
[node-domexception@1.0.0](https://github.com/jimmywarting/node-domexception) - MIT
[node-fetch@2.7.0](https://github.com/bitinn/node-fetch) - MIT
[node-int64@0.4.0](https://github.com/broofa/node-int64) - MIT
[node-releases@2.0.14](https://github.com/chicoxyzzy/node-releases) - MIT
[normalize-path@3.0.0](https://github.com/jonschlinkert/normalize-path) - MIT
[npm-run-path@4.0.1](https://github.com/sindresorhus/npm-run-path) - MIT
[obliterator@1.6.1](https://github.com/yomguithereal/obliterator) - MIT
[once@1.4.0](https://github.com/isaacs/once) - ISC
[onetime@5.1.2](https://github.com/sindresorhus/onetime) - MIT
[onnxruntime-common@1.19.2](https://github.com/Microsoft/onnxruntime) - MIT
[onnxruntime-common@1.20.0-dev.20241016-2b8fc5529b](https://github.com/Microsoft/onnxruntime) - MIT
[onnxruntime-node@1.19.2](https://github.com/Microsoft/onnxruntime) - MIT
[onnxruntime-web@1.21.0-dev.20241024-d9ca84ef96](https://github.com/Microsoft/onnxruntime) - MIT
[openai@4.29.2](https://github.com/openai/openai-node) - Apache-2.0
[optionator@0.9.3](https://github.com/gkz/optionator) - MIT
[p-limit@2.3.0](https://github.com/sindresorhus/p-limit) - MIT
[p-limit@3.1.0](https://github.com/sindresorhus/p-limit) - MIT
[p-locate@4.1.0](https://github.com/sindresorhus/p-locate) - MIT
[p-locate@5.0.0](https://github.com/sindresorhus/p-locate) - MIT
[p-try@2.2.0](https://github.com/sindresorhus/p-try) - MIT
[package-json-from-dist@1.0.1](https://github.com/isaacs/package-json-from-dist) - BlueOak-1.0.0
[parent-module@1.0.1](https://github.com/sindresorhus/parent-module) - MIT
[parse-json@5.2.0](https://github.com/sindresorhus/parse-json) - MIT
[path-exists@4.0.0](https://github.com/sindresorhus/path-exists) - MIT
[path-is-absolute@1.0.1](https://github.com/sindresorhus/path-is-absolute) - MIT
[path-key@3.1.1](https://github.com/sindresorhus/path-key) - MIT
[path-parse@1.0.7](https://github.com/jbgutierrez/path-parse) - MIT
[path-scurry@1.11.1](https://github.com/isaacs/path-scurry) - BlueOak-1.0.0
[path-type@4.0.0](https://github.com/sindresorhus/path-type) - MIT
[picocolors@1.0.0](https://github.com/alexeyraspopov/picocolors) - ISC
[picomatch@2.3.1](https://github.com/micromatch/picomatch) - MIT
[pirates@4.0.6](https://github.com/danez/pirates) - MIT
[pkg-dir@4.2.0](https://github.com/sindresorhus/pkg-dir) - MIT
[platform@1.3.6](https://github.com/bestiejs/platform.js) - MIT
[prelude-ls@1.2.1](https://github.com/gkz/prelude-ls) - MIT
[pretty-format@29.7.0](https://github.com/jestjs/jest) - MIT
[prompts@2.4.2](https://github.com/terkelg/prompts) - MIT
[protobufjs@7.4.0](https://github.com/protobufjs/protobuf.js) - BSD-3-Clause
[proxy-from-env@1.1.0](https://github.com/Rob--W/proxy-from-env) - MIT
[punycode.js@2.3.1](https://github.com/mathiasbynens/punycode.js) - MIT
[punycode@2.3.1](https://github.com/mathiasbynens/punycode.js) - MIT
[pure-rand@6.0.4](https://github.com/dubzzz/pure-rand) - MIT
[queue-microtask@1.2.3](https://github.com/feross/queue-microtask) - MIT
[react-is@18.2.0](https://github.com/facebook/react) - MIT
[rechoir@0.6.2](https://github.com/tkellen/node-rechoir) - MIT
[reflect-metadata@0.2.2](https://github.com/rbuckton/reflect-metadata) - Apache-2.0
[require-directory@2.1.1](https://github.com/troygoode/node-require-directory) - MIT
[resolve-cwd@3.0.0](https://github.com/sindresorhus/resolve-cwd) - MIT
[resolve-from@4.0.0](https://github.com/sindresorhus/resolve-from) - MIT
[resolve-from@5.0.0](https://github.com/sindresorhus/resolve-from) - MIT
[resolve.exports@2.0.2](https://github.com/lukeed/resolve.exports) - MIT
[resolve@1.22.8](https://github.com/browserify/resolve) - MIT
[reusify@1.0.4](https://github.com/mcollina/reusify) - MIT
[rimraf@3.0.2](https://github.com/isaacs/rimraf) - ISC
[rimraf@5.0.10](https://github.com/isaacs/rimraf) - ISC
[run-parallel@1.2.0](https://github.com/feross/run-parallel) - MIT
[semver@6.3.1](https://github.com/npm/node-semver) - ISC
[semver@7.6.3](https://github.com/npm/node-semver) - ISC
[sharp@0.33.5](https://github.com/lovell/sharp) - Apache-2.0
[shebang-command@2.0.0](https://github.com/kevva/shebang-command) - MIT
[shebang-regex@3.0.0](https://github.com/sindresorhus/shebang-regex) - MIT
[shelljs@0.8.5](https://github.com/shelljs/shelljs) - BSD-3-Clause
[shiki@1.10.3](https://github.com/shikijs/shiki) - MIT
[shx@0.3.4](https://github.com/shelljs/shx) - MIT
[signal-exit@3.0.7](https://github.com/tapjs/signal-exit) - ISC
[signal-exit@4.1.0](https://github.com/tapjs/signal-exit) - ISC
[simple-swizzle@0.2.2](https://github.com/qix-/node-simple-swizzle) - MIT
[sisteransi@1.0.5](https://github.com/terkelg/sisteransi) - MIT
[slash@3.0.0](https://github.com/sindresorhus/slash) - MIT
[source-map-support@0.5.13](https://github.com/evanw/node-source-map-support) - MIT
[source-map@0.6.1](https://github.com/mozilla/source-map) - BSD-3-Clause
[sprintf-js@1.0.3](https://github.com/alexei/sprintf.js) - BSD-3-Clause
[stack-utils@2.0.6](https://github.com/tapjs/stack-utils) - MIT
[stream-read-all@3.0.1](https://github.com/75lb/stream-read-all) - MIT
[string-length@4.0.2](https://github.com/sindresorhus/string-length) - MIT
[string-width@4.2.3](https://github.com/sindresorhus/string-width) - MIT
[string-width@5.1.2](https://github.com/sindresorhus/string-width) - MIT
[strip-ansi@6.0.1](https://github.com/chalk/strip-ansi) - MIT
[strip-ansi@7.1.0](https://github.com/chalk/strip-ansi) - MIT
[strip-bom@4.0.0](https://github.com/sindresorhus/strip-bom) - MIT
[strip-final-newline@2.0.0](https://github.com/sindresorhus/strip-final-newline) - MIT
[strip-json-comments@3.1.1](https://github.com/sindresorhus/strip-json-comments) - MIT
[strnum@1.0.5](https://github.com/NaturalIntelligence/strnum) - MIT
[supports-color@7.2.0](https://github.com/chalk/supports-color) - MIT
[supports-color@8.1.1](https://github.com/chalk/supports-color) - MIT
[supports-preserve-symlinks-flag@1.0.0](https://github.com/inspect-js/node-supports-preserve-symlinks-flag) - MIT
[table-layout@3.0.2](https://github.com/75lb/table-layout) - MIT
[tar@7.4.3](https://github.com/isaacs/node-tar) - ISC
[test-exclude@6.0.0](https://github.com/istanbuljs/test-exclude) - ISC
[text-table@0.2.0](https://github.com/substack/text-table) - MIT
[tmp@0.2.3](https://github.com/raszi/node-tmp) - MIT
[tmpl@1.0.5](https://github.com/daaku/nodejs-tmpl) - BSD-3-Clause
[to-regex-range@5.0.1](https://github.com/micromatch/to-regex-range) - MIT
[tr46@0.0.3](https://github.com/Sebmaster/tr46.js) - MIT
[ts-api-utils@1.0.3](https://github.com/JoshuaKGoldberg/ts-api-utils) - MIT
[ts-jest@29.1.2](https://github.com/kulshekhar/ts-jest) - MIT
[tslib@1.14.1](https://github.com/Microsoft/tslib) - 0BSD
[tslib@2.6.2](https://github.com/Microsoft/tslib) - 0BSD
[type-check@0.4.0](https://github.com/gkz/type-check) - MIT
[type-detect@4.0.8](https://github.com/chaijs/type-detect) - MIT
[type-fest@0.20.2](https://github.com/sindresorhus/type-fest) - (MIT OR CC0-1.0)
[type-fest@0.21.3](https://github.com/sindresorhus/type-fest) - (MIT OR CC0-1.0)
[typedoc-plugin-markdown@4.2.1](https://github.com/typedoc2md/typedoc-plugin-markdown) - MIT
[typedoc@0.26.4](https://github.com/TypeStrong/TypeDoc) - Apache-2.0
[typescript-eslint@7.1.0](https://github.com/typescript-eslint/typescript-eslint) - MIT
[typescript@5.5.4](https://github.com/Microsoft/TypeScript) - Apache-2.0
[typical@4.0.0](https://github.com/75lb/typical) - MIT
[typical@7.1.1](https://github.com/75lb/typical) - MIT
[uc.micro@2.1.0](https://github.com/markdown-it/uc.micro) - MIT
[undici-types@5.26.5](https://github.com/nodejs/undici) - MIT
[undici-types@6.19.8](https://github.com/nodejs/undici) - MIT
[update-browserslist-db@1.0.13](https://github.com/browserslist/update-db) - MIT
[uri-js@4.4.1](https://github.com/garycourt/uri-js) - BSD-2-Clause
[uuid@9.0.1](https://github.com/uuidjs/uuid) - MIT
[v8-to-istanbul@9.2.0](https://github.com/istanbuljs/v8-to-istanbul) - ISC
[walker@1.0.8](https://github.com/daaku/nodejs-walker) - Apache-2.0
[web-streams-polyfill@3.3.3](https://github.com/MattiasBuelens/web-streams-polyfill) - MIT
[web-streams-polyfill@4.0.0-beta.3](https://github.com/MattiasBuelens/web-streams-polyfill) - MIT
[webidl-conversions@3.0.1](https://github.com/jsdom/webidl-conversions) - BSD-2-Clause
[whatwg-url@5.0.0](https://github.com/jsdom/whatwg-url) - MIT
[which@2.0.2](https://github.com/isaacs/node-which) - ISC
[wordwrapjs@5.1.0](https://github.com/75lb/wordwrapjs) - MIT
[wrap-ansi@7.0.0](https://github.com/chalk/wrap-ansi) - MIT
[wrap-ansi@8.1.0](https://github.com/chalk/wrap-ansi) - MIT
[wrappy@1.0.2](https://github.com/npm/wrappy) - ISC
[write-file-atomic@4.0.2](https://github.com/npm/write-file-atomic) - ISC
[y18n@5.0.8](https://github.com/yargs/y18n) - ISC
[yallist@3.1.1](https://github.com/isaacs/yallist) - ISC
[yallist@5.0.0](https://github.com/isaacs/yallist) - BlueOak-1.0.0
[yaml@2.4.5](https://github.com/eemeli/yaml) - ISC
[yargs-parser@21.1.1](https://github.com/yargs/yargs-parser) - ISC
[yargs@17.7.2](https://github.com/yargs/yargs) - MIT
[yocto-queue@0.1.0](https://github.com/sindresorhus/yocto-queue) - MIT
File diff suppressed because it is too large Load Diff
+1 -1
View File
@@ -1,6 +1,6 @@
{ {
"name": "@lancedb/lancedb-darwin-arm64", "name": "@lancedb/lancedb-darwin-arm64",
"version": "0.26.1", "version": "0.26.2",
"os": ["darwin"], "os": ["darwin"],
"cpu": ["arm64"], "cpu": ["arm64"],
"main": "lancedb.darwin-arm64.node", "main": "lancedb.darwin-arm64.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{ {
"name": "@lancedb/lancedb-linux-arm64-gnu", "name": "@lancedb/lancedb-linux-arm64-gnu",
"version": "0.26.1", "version": "0.26.2",
"os": ["linux"], "os": ["linux"],
"cpu": ["arm64"], "cpu": ["arm64"],
"main": "lancedb.linux-arm64-gnu.node", "main": "lancedb.linux-arm64-gnu.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{ {
"name": "@lancedb/lancedb-linux-arm64-musl", "name": "@lancedb/lancedb-linux-arm64-musl",
"version": "0.26.1", "version": "0.26.2",
"os": ["linux"], "os": ["linux"],
"cpu": ["arm64"], "cpu": ["arm64"],
"main": "lancedb.linux-arm64-musl.node", "main": "lancedb.linux-arm64-musl.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{ {
"name": "@lancedb/lancedb-linux-x64-gnu", "name": "@lancedb/lancedb-linux-x64-gnu",
"version": "0.26.1", "version": "0.26.2",
"os": ["linux"], "os": ["linux"],
"cpu": ["x64"], "cpu": ["x64"],
"main": "lancedb.linux-x64-gnu.node", "main": "lancedb.linux-x64-gnu.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{ {
"name": "@lancedb/lancedb-linux-x64-musl", "name": "@lancedb/lancedb-linux-x64-musl",
"version": "0.26.1", "version": "0.26.2",
"os": ["linux"], "os": ["linux"],
"cpu": ["x64"], "cpu": ["x64"],
"main": "lancedb.linux-x64-musl.node", "main": "lancedb.linux-x64-musl.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{ {
"name": "@lancedb/lancedb-win32-arm64-msvc", "name": "@lancedb/lancedb-win32-arm64-msvc",
"version": "0.26.1", "version": "0.26.2",
"os": [ "os": [
"win32" "win32"
], ],
+1 -1
View File
@@ -1,6 +1,6 @@
{ {
"name": "@lancedb/lancedb-win32-x64-msvc", "name": "@lancedb/lancedb-win32-x64-msvc",
"version": "0.26.1", "version": "0.26.2",
"os": ["win32"], "os": ["win32"],
"cpu": ["x64"], "cpu": ["x64"],
"main": "lancedb.win32-x64-msvc.node", "main": "lancedb.win32-x64-msvc.node",
+2 -2
View File
@@ -1,12 +1,12 @@
{ {
"name": "@lancedb/lancedb", "name": "@lancedb/lancedb",
"version": "0.26.1", "version": "0.26.2",
"lockfileVersion": 3, "lockfileVersion": 3,
"requires": true, "requires": true,
"packages": { "packages": {
"": { "": {
"name": "@lancedb/lancedb", "name": "@lancedb/lancedb",
"version": "0.26.1", "version": "0.26.2",
"cpu": [ "cpu": [
"x64", "x64",
"arm64" "arm64"
+1 -1
View File
@@ -11,7 +11,7 @@
"ann" "ann"
], ],
"private": false, "private": false,
"version": "0.26.1", "version": "0.26.2",
"main": "dist/index.js", "main": "dist/index.js",
"exports": { "exports": {
".": "./dist/index.js", ".": "./dist/index.js",
+206
View File
@@ -0,0 +1,206 @@
| Name | Version | License | URL |
|--------------------------------|-----------------|--------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------|
| InstructorEmbedding | 1.0.1 | Apache License 2.0 | https://github.com/HKUNLP/instructor-embedding |
| Jinja2 | 3.1.6 | BSD License | https://github.com/pallets/jinja/ |
| Markdown | 3.10.2 | BSD-3-Clause | https://Python-Markdown.github.io/ |
| MarkupSafe | 3.0.3 | BSD-3-Clause | https://github.com/pallets/markupsafe/ |
| PyJWT | 2.11.0 | MIT | https://github.com/jpadilla/pyjwt |
| PyYAML | 6.0.3 | MIT License | https://pyyaml.org/ |
| Pygments | 2.19.2 | BSD License | https://pygments.org |
| accelerate | 1.12.0 | Apache Software License | https://github.com/huggingface/accelerate |
| adlfs | 2026.2.0 | BSD License | UNKNOWN |
| aiohappyeyeballs | 2.6.1 | Python Software Foundation License | https://github.com/aio-libs/aiohappyeyeballs |
| aiohttp | 3.13.3 | Apache-2.0 AND MIT | https://github.com/aio-libs/aiohttp |
| aiosignal | 1.4.0 | Apache Software License | https://github.com/aio-libs/aiosignal |
| annotated-types | 0.7.0 | MIT License | https://github.com/annotated-types/annotated-types |
| anyio | 4.12.1 | MIT | https://anyio.readthedocs.io/en/stable/versionhistory.html |
| appnope | 0.1.4 | BSD License | http://github.com/minrk/appnope |
| asttokens | 3.0.1 | Apache 2.0 | https://github.com/gristlabs/asttokens |
| attrs | 25.4.0 | MIT | https://www.attrs.org/en/stable/changelog.html |
| awscli | 1.44.35 | Apache Software License | http://aws.amazon.com/cli/ |
| azure-core | 1.38.0 | MIT License | https://github.com/Azure/azure-sdk-for-python/tree/main/sdk/core/azure-core |
| azure-datalake-store | 0.0.53 | MIT License | https://github.com/Azure/azure-data-lake-store-python |
| azure-identity | 1.25.1 | MIT | https://github.com/Azure/azure-sdk-for-python |
| azure-storage-blob | 12.28.0 | MIT License | https://github.com/Azure/azure-sdk-for-python/tree/main/sdk/storage/azure-storage-blob |
| babel | 2.18.0 | BSD License | https://babel.pocoo.org/ |
| backrefs | 6.1 | MIT | https://github.com/facelessuser/backrefs |
| beautifulsoup4 | 4.14.3 | MIT License | https://www.crummy.com/software/BeautifulSoup/bs4/ |
| bleach | 6.3.0 | Apache Software License | https://github.com/mozilla/bleach |
| boto3 | 1.42.45 | Apache-2.0 | https://github.com/boto/boto3 |
| botocore | 1.42.45 | Apache-2.0 | https://github.com/boto/botocore |
| cachetools | 7.0.0 | MIT | https://github.com/tkem/cachetools/ |
| certifi | 2026.1.4 | Mozilla Public License 2.0 (MPL 2.0) | https://github.com/certifi/python-certifi |
| cffi | 2.0.0 | MIT | https://cffi.readthedocs.io/en/latest/whatsnew.html |
| cfgv | 3.5.0 | MIT | https://github.com/asottile/cfgv |
| charset-normalizer | 3.4.4 | MIT | https://github.com/jawah/charset_normalizer/blob/master/CHANGELOG.md |
| click | 8.3.1 | BSD-3-Clause | https://github.com/pallets/click/ |
| cohere | 5.20.4 | MIT License | https://github.com/cohere-ai/cohere-python |
| colorama | 0.4.6 | BSD License | https://github.com/tartley/colorama |
| colpali_engine | 0.3.13 | MIT License | https://github.com/illuin-tech/colpali |
| comm | 0.2.3 | BSD License | https://github.com/ipython/comm |
| cryptography | 46.0.4 | Apache-2.0 OR BSD-3-Clause | https://github.com/pyca/cryptography |
| datafusion | 51.0.0 | Apache Software License | https://datafusion.apache.org/python |
| debugpy | 1.8.20 | MIT License | https://aka.ms/debugpy |
| decorator | 5.2.1 | BSD License | UNKNOWN |
| defusedxml | 0.7.1 | Python Software Foundation License | https://github.com/tiran/defusedxml |
| deprecation | 2.1.0 | Apache Software License | http://deprecation.readthedocs.io/ |
| distlib | 0.4.0 | Python Software Foundation License | https://github.com/pypa/distlib |
| distro | 1.9.0 | Apache Software License | https://github.com/python-distro/distro |
| docutils | 0.19 | BSD License; GNU General Public License (GPL); Public Domain; Python Software Foundation License | https://docutils.sourceforge.io/ |
| duckdb | 1.4.4 | MIT License | https://github.com/duckdb/duckdb-python |
| executing | 2.2.1 | MIT License | https://github.com/alexmojaki/executing |
| fastavro | 1.12.1 | MIT | https://github.com/fastavro/fastavro |
| fastjsonschema | 2.21.2 | BSD License | https://github.com/horejsek/python-fastjsonschema |
| filelock | 3.20.3 | Unlicense | https://github.com/tox-dev/py-filelock |
| frozenlist | 1.8.0 | Apache-2.0 | https://github.com/aio-libs/frozenlist |
| fsspec | 2026.2.0 | BSD-3-Clause | https://github.com/fsspec/filesystem_spec |
| ftfy | 6.3.1 | Apache-2.0 | https://ftfy.readthedocs.io/en/latest/ |
| ghp-import | 2.1.0 | Apache Software License | https://github.com/c-w/ghp-import |
| google-ai-generativelanguage | 0.6.15 | Apache Software License | https://github.com/googleapis/google-cloud-python/tree/main/packages/google-ai-generativelanguage |
| google-api-core | 2.25.2 | Apache Software License | https://github.com/googleapis/python-api-core |
| google-api-python-client | 2.189.0 | Apache Software License | https://github.com/googleapis/google-api-python-client/ |
| google-auth | 2.48.0 | Apache Software License | https://github.com/googleapis/google-auth-library-python |
| google-auth-httplib2 | 0.3.0 | Apache Software License | https://github.com/GoogleCloudPlatform/google-auth-library-python-httplib2 |
| google-generativeai | 0.8.6 | Apache Software License | https://github.com/google/generative-ai-python |
| googleapis-common-protos | 1.72.0 | Apache Software License | https://github.com/googleapis/google-cloud-python/tree/main/packages/googleapis-common-protos |
| griffe | 2.0.0 | ISC | https://mkdocstrings.github.io/griffe |
| griffecli | 2.0.0 | ISC | UNKNOWN |
| griffelib | 2.0.0 | ISC | UNKNOWN |
| grpcio | 1.78.0 | Apache-2.0 | https://grpc.io |
| grpcio-status | 1.71.2 | Apache Software License | https://grpc.io |
| h11 | 0.16.0 | MIT License | https://github.com/python-hyper/h11 |
| hf-xet | 1.2.0 | Apache-2.0 | https://github.com/huggingface/xet-core |
| httpcore | 1.0.9 | BSD-3-Clause | https://www.encode.io/httpcore/ |
| httplib2 | 0.31.2 | MIT License | https://github.com/httplib2/httplib2 |
| httpx | 0.28.1 | BSD License | https://github.com/encode/httpx |
| huggingface_hub | 0.36.2 | Apache Software License | https://github.com/huggingface/huggingface_hub |
| ibm-cos-sdk | 2.14.3 | Apache Software License | https://github.com/ibm/ibm-cos-sdk-python |
| ibm-cos-sdk-core | 2.14.3 | Apache Software License | https://github.com/ibm/ibm-cos-sdk-python-core |
| ibm-cos-sdk-s3transfer | 2.14.3 | Apache Software License | https://github.com/IBM/ibm-cos-sdk-python-s3transfer |
| ibm_watsonx_ai | 1.5.1 | BSD License | https://ibm.github.io/watsonx-ai-python-sdk/changelog.html |
| identify | 2.6.16 | MIT | https://github.com/pre-commit/identify |
| idna | 3.11 | BSD-3-Clause | https://github.com/kjd/idna |
| iniconfig | 2.3.0 | MIT | https://github.com/pytest-dev/iniconfig |
| ipykernel | 6.31.0 | BSD-3-Clause | https://ipython.org |
| ipython | 9.10.0 | BSD-3-Clause | https://ipython.org |
| ipython_pygments_lexers | 1.1.1 | BSD License | https://github.com/ipython/ipython-pygments-lexers |
| isodate | 0.7.2 | BSD License | https://github.com/gweis/isodate/ |
| jedi | 0.19.2 | MIT License | https://github.com/davidhalter/jedi |
| jiter | 0.13.0 | MIT License | https://github.com/pydantic/jiter/ |
| jmespath | 1.0.1 | MIT License | https://github.com/jmespath/jmespath.py |
| joblib | 1.5.3 | BSD-3-Clause | https://joblib.readthedocs.io |
| jsonschema | 4.26.0 | MIT | https://github.com/python-jsonschema/jsonschema |
| jsonschema-specifications | 2025.9.1 | MIT | https://github.com/python-jsonschema/jsonschema-specifications |
| jupyter_client | 8.8.0 | BSD License | https://jupyter.org |
| jupyter_core | 5.9.1 | BSD-3-Clause | https://jupyter.org |
| jupyterlab_pygments | 0.3.0 | BSD License | https://github.com/jupyterlab/jupyterlab_pygments |
| jupytext | 1.19.1 | MIT License | https://github.com/mwouts/jupytext |
| lance-namespace | 0.4.5 | Apache-2.0 | https://github.com/lance-format/lance-namespace |
| lance-namespace-urllib3-client | 0.4.5 | Apache-2.0 | https://github.com/lance-format/lance-namespace |
| lancedb | 0.29.2 | Apache Software License | https://github.com/lancedb/lancedb |
| lomond | 0.3.3 | BSD License | https://github.com/wildfoundry/dataplicity-lomond |
| markdown-it-py | 4.0.0 | MIT License | https://github.com/executablebooks/markdown-it-py |
| matplotlib-inline | 0.2.1 | UNKNOWN | https://github.com/ipython/matplotlib-inline |
| mdit-py-plugins | 0.5.0 | MIT License | https://github.com/executablebooks/mdit-py-plugins |
| mdurl | 0.1.2 | MIT License | https://github.com/executablebooks/mdurl |
| mergedeep | 1.3.4 | MIT License | https://github.com/clarketm/mergedeep |
| mistune | 3.2.0 | BSD License | https://github.com/lepture/mistune |
| mkdocs | 1.6.1 | BSD-2-Clause | https://github.com/mkdocs/mkdocs |
| mkdocs-autorefs | 1.4.3 | ISC | https://mkdocstrings.github.io/autorefs |
| mkdocs-get-deps | 0.2.0 | MIT | https://github.com/mkdocs/get-deps |
| mkdocs-jupyter | 0.25.1 | Apache-2.0 | https://github.com/danielfrg/mkdocs-jupyter |
| mkdocs-material | 9.7.1 | MIT | https://github.com/squidfunk/mkdocs-material |
| mkdocs-material-extensions | 1.3.1 | MIT | https://github.com/facelessuser/mkdocs-material-extensions |
| mkdocstrings | 1.0.3 | ISC | https://mkdocstrings.github.io |
| mkdocstrings-python | 2.0.2 | ISC | https://mkdocstrings.github.io/python |
| mpmath | 1.3.0 | BSD License | http://mpmath.org/ |
| msal | 1.34.0 | MIT License | https://github.com/AzureAD/microsoft-authentication-library-for-python |
| msal-extensions | 1.3.1 | MIT License | https://github.com/AzureAD/microsoft-authentication-extensions-for-python/releases |
| multidict | 6.7.1 | Apache License 2.0 | https://github.com/aio-libs/multidict |
| nbclient | 0.10.4 | BSD License | https://jupyter.org |
| nbconvert | 7.17.0 | BSD License | https://jupyter.org |
| nbformat | 5.10.4 | BSD License | https://jupyter.org |
| nest-asyncio | 1.6.0 | BSD License | https://github.com/erdewit/nest_asyncio |
| networkx | 3.6.1 | BSD-3-Clause | https://networkx.org/ |
| nodeenv | 1.10.0 | BSD License | https://github.com/ekalinin/nodeenv |
| numpy | 2.4.2 | BSD-3-Clause AND 0BSD AND MIT AND Zlib AND CC0-1.0 | https://numpy.org |
| ollama | 0.6.1 | MIT | https://ollama.com |
| open_clip_torch | 3.2.0 | MIT License | https://github.com/mlfoundations/open_clip |
| openai | 2.18.0 | Apache Software License | https://github.com/openai/openai-python |
| packaging | 26.0 | Apache-2.0 OR BSD-2-Clause | https://github.com/pypa/packaging |
| paginate | 0.5.7 | MIT License | https://github.com/Signum/paginate |
| pandas | 2.3.3 | BSD License | https://pandas.pydata.org |
| pandocfilters | 1.5.1 | BSD License | http://github.com/jgm/pandocfilters |
| parso | 0.8.6 | MIT License | https://github.com/davidhalter/parso |
| pathspec | 1.0.4 | Mozilla Public License 2.0 (MPL 2.0) | UNKNOWN |
| peft | 0.17.1 | Apache Software License | https://github.com/huggingface/peft |
| pexpect | 4.9.0 | ISC License (ISCL) | https://pexpect.readthedocs.io/ |
| pillow | 12.1.0 | MIT-CMU | https://python-pillow.github.io |
| platformdirs | 4.5.1 | MIT | https://github.com/tox-dev/platformdirs |
| pluggy | 1.6.0 | MIT License | UNKNOWN |
| polars | 1.3.0 | MIT License | https://www.pola.rs/ |
| pre_commit | 4.5.1 | MIT | https://github.com/pre-commit/pre-commit |
| prompt_toolkit | 3.0.52 | BSD License | https://github.com/prompt-toolkit/python-prompt-toolkit |
| propcache | 0.4.1 | Apache Software License | https://github.com/aio-libs/propcache |
| proto-plus | 1.27.1 | Apache Software License | https://github.com/googleapis/proto-plus-python |
| protobuf | 5.29.6 | 3-Clause BSD License | https://developers.google.com/protocol-buffers/ |
| psutil | 7.2.2 | BSD-3-Clause | https://github.com/giampaolo/psutil |
| ptyprocess | 0.7.0 | ISC License (ISCL) | https://github.com/pexpect/ptyprocess |
| pure_eval | 0.2.3 | MIT License | http://github.com/alexmojaki/pure_eval |
| pyarrow | 23.0.0 | Apache-2.0 | https://arrow.apache.org/ |
| pyarrow-stubs | 20.0.0.20251215 | BSD-2-Clause | https://github.com/zen-xu/pyarrow-stubs |
| pyasn1 | 0.6.2 | BSD-2-Clause | https://github.com/pyasn1/pyasn1 |
| pyasn1_modules | 0.4.2 | BSD License | https://github.com/pyasn1/pyasn1-modules |
| pycparser | 3.0 | BSD-3-Clause | https://github.com/eliben/pycparser |
| pydantic | 2.12.5 | MIT | https://github.com/pydantic/pydantic |
| pydantic_core | 2.41.5 | MIT | https://github.com/pydantic/pydantic-core |
| pylance | 2.0.0 | Apache Software License | UNKNOWN |
| pymdown-extensions | 10.20.1 | MIT | https://github.com/facelessuser/pymdown-extensions |
| pyparsing | 3.3.2 | MIT | https://github.com/pyparsing/pyparsing/ |
| pyright | 1.1.408 | MIT | https://github.com/RobertCraigie/pyright-python |
| pytest | 9.0.2 | MIT | https://docs.pytest.org/en/latest/ |
| pytest-asyncio | 1.3.0 | Apache-2.0 | https://github.com/pytest-dev/pytest-asyncio |
| pytest-mock | 3.15.1 | MIT License | https://github.com/pytest-dev/pytest-mock/ |
| python-dateutil | 2.9.0.post0 | Apache Software License; BSD License | https://github.com/dateutil/dateutil |
| pytz | 2025.2 | MIT License | http://pythonhosted.org/pytz |
| pyyaml_env_tag | 1.1 | MIT | https://github.com/waylan/pyyaml-env-tag |
| pyzmq | 27.1.0 | BSD License | https://pyzmq.readthedocs.org |
| referencing | 0.37.0 | MIT | https://github.com/python-jsonschema/referencing |
| regex | 2026.1.15 | Apache-2.0 AND CNRI-Python | https://github.com/mrabarnett/mrab-regex |
| requests | 2.32.5 | Apache Software License | https://requests.readthedocs.io |
| rpds-py | 0.30.0 | MIT | https://github.com/crate-py/rpds |
| rsa | 4.7.2 | Apache Software License | https://stuvel.eu/rsa |
| ruff | 0.15.0 | MIT License | https://docs.astral.sh/ruff |
| s3transfer | 0.16.0 | Apache Software License | https://github.com/boto/s3transfer |
| safetensors | 0.7.0 | Apache Software License | https://github.com/huggingface/safetensors |
| scikit-learn | 1.8.0 | BSD-3-Clause | https://scikit-learn.org |
| scipy | 1.17.0 | BSD License | https://scipy.org/ |
| sentence-transformers | 5.2.2 | Apache Software License | https://www.SBERT.net |
| sentencepiece | 0.2.1 | UNKNOWN | https://github.com/google/sentencepiece |
| six | 1.17.0 | MIT License | https://github.com/benjaminp/six |
| sniffio | 1.3.1 | Apache Software License; MIT License | https://github.com/python-trio/sniffio |
| soupsieve | 2.8.3 | MIT | https://github.com/facelessuser/soupsieve |
| stack-data | 0.6.3 | MIT License | http://github.com/alexmojaki/stack_data |
| sympy | 1.14.0 | BSD License | https://sympy.org |
| tabulate | 0.9.0 | MIT License | https://github.com/astanin/python-tabulate |
| tantivy | 0.25.1 | UNKNOWN | UNKNOWN |
| threadpoolctl | 3.6.0 | BSD License | https://github.com/joblib/threadpoolctl |
| timm | 1.0.24 | Apache Software License | https://github.com/huggingface/pytorch-image-models |
| tinycss2 | 1.4.0 | BSD License | https://www.courtbouillon.org/tinycss2 |
| tokenizers | 0.22.2 | Apache Software License | https://github.com/huggingface/tokenizers |
| torch | 2.8.0 | BSD License | https://pytorch.org/ |
| torchvision | 0.23.0 | BSD | https://github.com/pytorch/vision |
| tornado | 6.5.4 | Apache Software License | http://www.tornadoweb.org/ |
| tqdm | 4.67.3 | MPL-2.0 AND MIT | https://tqdm.github.io |
| traitlets | 5.14.3 | BSD License | https://github.com/ipython/traitlets |
| transformers | 4.57.6 | Apache Software License | https://github.com/huggingface/transformers |
| types-requests | 2.32.4.20260107 | Apache-2.0 | https://github.com/python/typeshed |
| typing-inspection | 0.4.2 | MIT | https://github.com/pydantic/typing-inspection |
| typing_extensions | 4.15.0 | PSF-2.0 | https://github.com/python/typing_extensions |
| tzdata | 2025.3 | Apache-2.0 | https://github.com/python/tzdata |
| uritemplate | 4.2.0 | BSD 3-Clause OR Apache-2.0 | https://uritemplate.readthedocs.org |
| urllib3 | 2.6.3 | MIT | https://github.com/urllib3/urllib3/blob/main/CHANGES.rst |
| virtualenv | 20.36.1 | MIT | https://github.com/pypa/virtualenv |
| watchdog | 6.0.0 | Apache Software License | https://github.com/gorakhargosh/watchdog |
| webencodings | 0.5.1 | BSD License | https://github.com/SimonSapin/python-webencodings |
| yarl | 1.22.0 | Apache Software License | https://github.com/aio-libs/yarl |
File diff suppressed because it is too large Load Diff
+40 -15
View File
@@ -9,7 +9,7 @@ import json
from ._lancedb import async_permutation_builder, PermutationReader from ._lancedb import async_permutation_builder, PermutationReader
from .table import LanceTable from .table import LanceTable
from .background_loop import LOOP from .background_loop import LOOP
from .util import batch_to_tensor from .util import batch_to_tensor, batch_to_tensor_rows
from typing import Any, Callable, Iterator, Literal, Optional, TYPE_CHECKING, Union from typing import Any, Callable, Iterator, Literal, Optional, TYPE_CHECKING, Union
if TYPE_CHECKING: if TYPE_CHECKING:
@@ -333,7 +333,11 @@ class Transforms:
""" """
@staticmethod @staticmethod
def arrow2python(batch: pa.RecordBatch) -> dict[str, list[Any]]: def arrow2python(batch: pa.RecordBatch) -> list[dict[str, Any]]:
return batch.to_pylist()
@staticmethod
def arrow2pythoncol(batch: pa.RecordBatch) -> dict[str, list[Any]]:
return batch.to_pydict() return batch.to_pydict()
@staticmethod @staticmethod
@@ -687,7 +691,17 @@ class Permutation:
return return
def with_format( def with_format(
self, format: Literal["numpy", "python", "pandas", "arrow", "torch", "polars"] self,
format: Literal[
"numpy",
"python",
"python_col",
"pandas",
"arrow",
"torch",
"torch_col",
"polars",
],
) -> "Permutation": ) -> "Permutation":
""" """
Set the format for batches Set the format for batches
@@ -696,16 +710,18 @@ class Permutation:
The format can be one of: The format can be one of:
- "numpy" - the batch will be a dict of numpy arrays (one per column) - "numpy" - the batch will be a dict of numpy arrays (one per column)
- "python" - the batch will be a dict of lists (one per column) - "python" - the batch will be a list of dicts (one per row)
- "python_col" - the batch will be a dict of lists (one entry per column)
- "pandas" - the batch will be a pandas DataFrame - "pandas" - the batch will be a pandas DataFrame
- "arrow" - the batch will be a pyarrow RecordBatch - "arrow" - the batch will be a pyarrow RecordBatch
- "torch" - the batch will be a two dimensional torch tensor - "torch" - the batch will be a list of tensors, one per row
- "torch_col" - the batch will be a 2D torch tensor (first dim indexes columns)
- "polars" - the batch will be a polars DataFrame - "polars" - the batch will be a polars DataFrame
Conversion may or may not involve a data copy. Lance uses Arrow internally Conversion may or may not involve a data copy. Lance uses Arrow internally
and so it is able to zero-copy to the arrow and polars. and so it is able to zero-copy to the arrow and polars formats.
Conversion to torch will be zero-copy but will only support a subset of data Conversion to torch_col will be zero-copy but will only support a subset of data
types (numeric types). types (numeric types).
Conversion to numpy and/or pandas will typically be zero-copy for numeric Conversion to numpy and/or pandas will typically be zero-copy for numeric
@@ -718,6 +734,8 @@ class Permutation:
assert format is not None, "format is required" assert format is not None, "format is required"
if format == "python": if format == "python":
return self.with_transform(Transforms.arrow2python) return self.with_transform(Transforms.arrow2python)
if format == "python_col":
return self.with_transform(Transforms.arrow2pythoncol)
elif format == "numpy": elif format == "numpy":
return self.with_transform(Transforms.arrow2numpy) return self.with_transform(Transforms.arrow2numpy)
elif format == "pandas": elif format == "pandas":
@@ -725,6 +743,8 @@ class Permutation:
elif format == "arrow": elif format == "arrow":
return self.with_transform(Transforms.arrow2arrow) return self.with_transform(Transforms.arrow2arrow)
elif format == "torch": elif format == "torch":
return self.with_transform(batch_to_tensor_rows)
elif format == "torch_col":
return self.with_transform(batch_to_tensor) return self.with_transform(batch_to_tensor)
elif format == "polars": elif format == "polars":
return self.with_transform(Transforms.arrow2polars()) return self.with_transform(Transforms.arrow2polars())
@@ -746,15 +766,20 @@ class Permutation:
def __getitem__(self, index: int) -> Any: def __getitem__(self, index: int) -> Any:
""" """
Return a single row from the permutation Returns a single row from the permutation by offset
The output will always be a python dictionary regardless of the format.
This method is mostly useful for debugging and exploration. For actual
processing use [iter](#iter) or a torch data loader to perform batched
processing.
""" """
pass return self.__getitems__([index])
def __getitems__(self, indices: list[int]) -> Any:
"""
Returns rows from the permutation by offset
"""
async def do_getitems():
return await self.reader.take_offsets(indices, selection=self.selection)
batch = LOOP.run(do_getitems())
return self.transform_fn(batch)
@deprecated(details="Use with_skip instead") @deprecated(details="Use with_skip instead")
def skip(self, skip: int) -> "Permutation": def skip(self, skip: int) -> "Permutation":
+29 -34
View File
@@ -2118,19 +2118,17 @@ class LanceHybridQueryBuilder(LanceQueryBuilder):
""" # noqa: E501 """ # noqa: E501
self._create_query_builders() self._create_query_builders()
results = ["Vector Search Plan:"] reranker_label = str(self._reranker) if self._reranker else "No reranker"
results.append( vector_plan = self._table._explain_plan(
self._table._explain_plan( self._vector_query.to_query_object(), verbose=verbose
self._vector_query.to_query_object(), verbose=verbose
)
) )
results.append("FTS Search Plan:") fts_plan = self._table._explain_plan(
results.append( self._fts_query.to_query_object(), verbose=verbose
self._table._explain_plan(
self._fts_query.to_query_object(), verbose=verbose
)
) )
return "\n".join(results) # Indent sub-plans under the reranker
indented_vector = "\n".join(" " + line for line in vector_plan.splitlines())
indented_fts = "\n".join(" " + line for line in fts_plan.splitlines())
return f"{reranker_label}\n {indented_vector}\n {indented_fts}"
def analyze_plan(self): def analyze_plan(self):
"""Execute the query and display with runtime metrics. """Execute the query and display with runtime metrics.
@@ -3164,23 +3162,20 @@ class AsyncHybridQuery(AsyncStandardQuery, AsyncVectorQueryBase):
... plan = await table.query().nearest_to([1.0, 2.0]).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) ... print(plan)
>>> asyncio.run(doctest_example()) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE >>> asyncio.run(doctest_example()) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
Vector Search Plan: RRFReranker(K=60)
ProjectionExec: expr=[vector@0 as vector, text@3 as text, _distance@2 as _distance] ProjectionExec: expr=[vector@0 as vector, text@3 as text, _distance@2 as _distance]
Take: columns="vector, _rowid, _distance, (text)" Take: columns="vector, _rowid, _distance, (text)"
CoalesceBatchesExec: target_batch_size=1024 CoalesceBatchesExec: target_batch_size=1024
GlobalLimitExec: skip=0, fetch=10 GlobalLimitExec: skip=0, fetch=10
FilterExec: _distance@2 IS NOT NULL FilterExec: _distance@2 IS NOT NULL
SortExec: TopK(fetch=10), expr=[_distance@2 ASC NULLS LAST, _rowid@1 ASC NULLS LAST], preserve_partitioning=[false] SortExec: TopK(fetch=10), expr=[_distance@2 ASC NULLS LAST, _rowid@1 ASC NULLS LAST], preserve_partitioning=[false]
KNNVectorDistance: metric=l2 KNNVectorDistance: metric=l2
LanceRead: uri=..., projection=[vector], ... LanceRead: uri=..., projection=[vector], ...
<BLANKLINE> ProjectionExec: expr=[vector@2 as vector, text@3 as text, _score@1 as _score]
FTS Search Plan: Take: columns="_rowid, _score, (vector), (text)"
ProjectionExec: expr=[vector@2 as vector, text@3 as text, _score@1 as _score] CoalesceBatchesExec: target_batch_size=1024
Take: columns="_rowid, _score, (vector), (text)" GlobalLimitExec: skip=0, fetch=10
CoalesceBatchesExec: target_batch_size=1024 MatchQuery: column=text, query=hello
GlobalLimitExec: skip=0, fetch=10
MatchQuery: column=text, query=hello
<BLANKLINE>
Parameters Parameters
---------- ----------
@@ -3192,12 +3187,12 @@ class AsyncHybridQuery(AsyncStandardQuery, AsyncVectorQueryBase):
plan : str plan : str
""" # noqa: E501 """ # noqa: E501
results = ["Vector Search Plan:"] vector_plan = await self._inner.to_vector_query().explain_plan(verbose)
results.append(await self._inner.to_vector_query().explain_plan(verbose)) fts_plan = await self._inner.to_fts_query().explain_plan(verbose)
results.append("FTS Search Plan:") # Indent sub-plans under the reranker
results.append(await self._inner.to_fts_query().explain_plan(verbose)) indented_vector = "\n".join(" " + line for line in vector_plan.splitlines())
indented_fts = "\n".join(" " + line for line in fts_plan.splitlines())
return "\n".join(results) return f"{self._reranker}\n {indented_vector}\n {indented_fts}"
async def analyze_plan(self): async def analyze_plan(self):
""" """
@@ -42,10 +42,18 @@ class AnswerdotaiRerankers(Reranker):
rerankers = attempt_import_or_raise( rerankers = attempt_import_or_raise(
"rerankers" "rerankers"
) # import here for faster ops later ) # import here for faster ops later
self.model_name = model_name
self.model_type = model_type
self.reranker = rerankers.Reranker( self.reranker = rerankers.Reranker(
model_name=model_name, model_type=model_type, **kwargs model_name=model_name, model_type=model_type, **kwargs
) )
def __str__(self):
return (
f"AnswerdotaiRerankers(model_type={self.model_type}, "
f"model_name={self.model_name})"
)
def _rerank(self, result_set: pa.Table, query: str): def _rerank(self, result_set: pa.Table, query: str):
result_set = self._handle_empty_results(result_set) result_set = self._handle_empty_results(result_set)
if len(result_set) == 0: if len(result_set) == 0:
+3
View File
@@ -40,6 +40,9 @@ class Reranker(ABC):
if ARROW_VERSION.major <= 13: if ARROW_VERSION.major <= 13:
self._concat_tables_args = {"promote": True} self._concat_tables_args = {"promote": True}
def __str__(self):
return self.__class__.__name__
def rerank_vector( def rerank_vector(
self, self,
query: str, query: str,
@@ -44,6 +44,9 @@ class CohereReranker(Reranker):
self.top_n = top_n self.top_n = top_n
self.api_key = api_key self.api_key = api_key
def __str__(self):
return f"CohereReranker(model_name={self.model_name})"
@cached_property @cached_property
def _client(self): def _client(self):
cohere = attempt_import_or_raise("cohere") cohere = attempt_import_or_raise("cohere")
@@ -50,6 +50,9 @@ class CrossEncoderReranker(Reranker):
if self.device is None: if self.device is None:
self.device = "cuda" if torch.cuda.is_available() else "cpu" self.device = "cuda" if torch.cuda.is_available() else "cpu"
def __str__(self):
return f"CrossEncoderReranker(model_name={self.model_name})"
@cached_property @cached_property
def model(self): def model(self):
sbert = attempt_import_or_raise("sentence_transformers") sbert = attempt_import_or_raise("sentence_transformers")
@@ -45,6 +45,9 @@ class JinaReranker(Reranker):
self.top_n = top_n self.top_n = top_n
self.api_key = api_key self.api_key = api_key
def __str__(self):
return f"JinaReranker(model_name={self.model_name})"
@cached_property @cached_property
def _client(self): def _client(self):
import requests import requests
@@ -38,6 +38,9 @@ class LinearCombinationReranker(Reranker):
self.weight = weight self.weight = weight
self.fill = fill self.fill = fill
def __str__(self):
return f"LinearCombinationReranker(weight={self.weight}, fill={self.fill})"
def rerank_hybrid( def rerank_hybrid(
self, self,
query: str, # noqa: F821 query: str, # noqa: F821
+6
View File
@@ -54,6 +54,12 @@ class MRRReranker(Reranker):
self.weight_vector = weight_vector self.weight_vector = weight_vector
self.weight_fts = weight_fts self.weight_fts = weight_fts
def __str__(self):
return (
f"MRRReranker(weight_vector={self.weight_vector}, "
f"weight_fts={self.weight_fts})"
)
def rerank_hybrid( def rerank_hybrid(
self, self,
query: str, # noqa: F821 query: str, # noqa: F821
@@ -43,6 +43,9 @@ class OpenaiReranker(Reranker):
self.column = column self.column = column
self.api_key = api_key self.api_key = api_key
def __str__(self):
return f"OpenaiReranker(model_name={self.model_name})"
def _rerank(self, result_set: pa.Table, query: str): def _rerank(self, result_set: pa.Table, query: str):
result_set = self._handle_empty_results(result_set) result_set = self._handle_empty_results(result_set)
if len(result_set) == 0: if len(result_set) == 0:
+3
View File
@@ -36,6 +36,9 @@ class RRFReranker(Reranker):
super().__init__(return_score) super().__init__(return_score)
self.K = K self.K = K
def __str__(self):
return f"RRFReranker(K={self.K})"
def rerank_hybrid( def rerank_hybrid(
self, self,
query: str, # noqa: F821 query: str, # noqa: F821
@@ -52,6 +52,9 @@ class VoyageAIReranker(Reranker):
self.api_key = api_key self.api_key = api_key
self.truncation = truncation self.truncation = truncation
def __str__(self):
return f"VoyageAIReranker(model_name={self.model_name})"
@cached_property @cached_property
def _client(self): def _client(self):
voyageai = attempt_import_or_raise("voyageai") voyageai = attempt_import_or_raise("voyageai")
+8 -2
View File
@@ -904,7 +904,9 @@ class Table(ABC):
---------- ----------
field_names: str or list of str field_names: str or list of str
The name(s) of the field to index. The name(s) of the field to index.
can be only str if use_tantivy=True for now. If ``use_tantivy`` is False (default), only a single field name
(str) is supported. To index multiple fields, create a separate
FTS index for each field.
replace: bool, default False replace: bool, default False
If True, replace the existing index if it exists. Note that this is If True, replace the existing index if it exists. Note that this is
not yet an atomic operation; the index will be temporarily not yet an atomic operation; the index will be temporarily
@@ -2298,7 +2300,11 @@ class LanceTable(Table):
): ):
if not use_tantivy: if not use_tantivy:
if not isinstance(field_names, str): if not isinstance(field_names, str):
raise ValueError("field_names must be a string when use_tantivy=False") raise ValueError(
"Native FTS indexes can only be created on a single field "
"at a time. To search over multiple text fields, create a "
"separate FTS index for each field."
)
if tokenizer_name is None: if tokenizer_name is None:
tokenizer_configs = { tokenizer_configs = {
+19
View File
@@ -419,3 +419,22 @@ def batch_to_tensor(batch: pa.RecordBatch):
""" """
torch = attempt_import_or_raise("torch", "torch") torch = attempt_import_or_raise("torch", "torch")
return torch.stack([torch.from_dlpack(col) for col in batch.columns]) return torch.stack([torch.from_dlpack(col) for col in batch.columns])
def batch_to_tensor_rows(batch: pa.RecordBatch):
"""
Convert a PyArrow RecordBatch to a list of PyTorch Tensor, one per row
Each column is converted to a tensor (using zero-copy via DLPack)
and the columns are then stacked into a single tensor. The 2D tensor
is then converted to a list of tensors, one per row
Fails if torch or numpy is not installed.
Fails if a column's data type is not supported by PyTorch.
"""
torch = attempt_import_or_raise("torch", "torch")
numpy = attempt_import_or_raise("numpy", "numpy")
columns = [col.to_numpy(zero_copy_only=False) for col in batch.columns]
stacked = torch.tensor(numpy.column_stack(columns))
rows = list(stacked.unbind(dim=0))
return rows
-2
View File
@@ -163,9 +163,7 @@ async def test_explain_plan(table: AsyncTable):
table.query().nearest_to_text("dog").nearest_to([0.1, 0.1]).explain_plan(True) table.query().nearest_to_text("dog").nearest_to([0.1, 0.1]).explain_plan(True)
) )
assert "Vector Search Plan" in plan
assert "KNNVectorDistance" in plan assert "KNNVectorDistance" in plan
assert "FTS Search Plan" in plan
assert "LanceRead" in plan assert "LanceRead" in plan
+141 -35
View File
@@ -664,23 +664,20 @@ def test_iter_basic(some_permutation: Permutation):
expected_batches = (950 + batch_size - 1) // batch_size # ceiling division expected_batches = (950 + batch_size - 1) // batch_size # ceiling division
assert len(batches) == expected_batches assert len(batches) == expected_batches
# Check that all batches are dicts (default python format) # Check that all batches are lists of dicts (default python format)
assert all(isinstance(batch, dict) for batch in batches) assert all(isinstance(batch, list) for batch in batches)
# Check that batches have the correct structure # Check that batches have the correct structure
for batch in batches: for batch in batches:
assert "id" in batch assert "id" in batch[0]
assert "value" in batch assert "value" in batch[0]
assert isinstance(batch["id"], list)
assert isinstance(batch["value"], list)
# Check that all batches except the last have the correct size # Check that all batches except the last have the correct size
for batch in batches[:-1]: for batch in batches[:-1]:
assert len(batch["id"]) == batch_size assert len(batch) == batch_size
assert len(batch["value"]) == batch_size
# Last batch might be smaller # Last batch might be smaller
assert len(batches[-1]["id"]) <= batch_size assert len(batches[-1]) <= batch_size
def test_iter_skip_last_batch(some_permutation: Permutation): def test_iter_skip_last_batch(some_permutation: Permutation):
@@ -699,11 +696,11 @@ def test_iter_skip_last_batch(some_permutation: Permutation):
if 950 % batch_size != 0: if 950 % batch_size != 0:
assert len(batches_without_skip) == num_full_batches + 1 assert len(batches_without_skip) == num_full_batches + 1
# Last batch should be smaller # Last batch should be smaller
assert len(batches_without_skip[-1]["id"]) == 950 % batch_size assert len(batches_without_skip[-1]) == 950 % batch_size
# All batches with skip_last_batch should be full size # All batches with skip_last_batch should be full size
for batch in batches_with_skip: for batch in batches_with_skip:
assert len(batch["id"]) == batch_size assert len(batch) == batch_size
def test_iter_different_batch_sizes(some_permutation: Permutation): def test_iter_different_batch_sizes(some_permutation: Permutation):
@@ -720,12 +717,12 @@ def test_iter_different_batch_sizes(some_permutation: Permutation):
# Test with batch size equal to total rows # Test with batch size equal to total rows
single_batch = list(some_permutation.iter(950, skip_last_batch=False)) single_batch = list(some_permutation.iter(950, skip_last_batch=False))
assert len(single_batch) == 1 assert len(single_batch) == 1
assert len(single_batch[0]["id"]) == 950 assert len(single_batch[0]) == 950
# Test with batch size larger than total rows # Test with batch size larger than total rows
oversized_batch = list(some_permutation.iter(10000, skip_last_batch=False)) oversized_batch = list(some_permutation.iter(10000, skip_last_batch=False))
assert len(oversized_batch) == 1 assert len(oversized_batch) == 1
assert len(oversized_batch[0]["id"]) == 950 assert len(oversized_batch[0]) == 950
def test_dunder_iter(some_permutation: Permutation): def test_dunder_iter(some_permutation: Permutation):
@@ -738,15 +735,13 @@ def test_dunder_iter(some_permutation: Permutation):
# All batches should be full size # All batches should be full size
for batch in batches: for batch in batches:
assert len(batch["id"]) == 100 assert len(batch) == 100
assert len(batch["value"]) == 100
some_permutation = some_permutation.with_batch_size(400) some_permutation = some_permutation.with_batch_size(400)
batches = list(some_permutation) batches = list(some_permutation)
assert len(batches) == 2 # floor(950 / 400) since skip_last_batch=True assert len(batches) == 2 # floor(950 / 400) since skip_last_batch=True
for batch in batches: for batch in batches:
assert len(batch["id"]) == 400 assert len(batch) == 400
assert len(batch["value"]) == 400
def test_iter_with_different_formats(some_permutation: Permutation): def test_iter_with_different_formats(some_permutation: Permutation):
@@ -761,7 +756,7 @@ def test_iter_with_different_formats(some_permutation: Permutation):
# Test with python format (default) # Test with python format (default)
python_perm = some_permutation.with_format("python") python_perm = some_permutation.with_format("python")
python_batches = list(python_perm.iter(batch_size, skip_last_batch=False)) python_batches = list(python_perm.iter(batch_size, skip_last_batch=False))
assert all(isinstance(batch, dict) for batch in python_batches) assert all(isinstance(batch, list) for batch in python_batches)
# Test with pandas format # Test with pandas format
pandas_perm = some_permutation.with_format("pandas") pandas_perm = some_permutation.with_format("pandas")
@@ -780,8 +775,8 @@ def test_iter_with_column_selection(some_permutation: Permutation):
# Check that batches only contain the id column # Check that batches only contain the id column
for batch in batches: for batch in batches:
assert "id" in batch assert "id" in batch[0]
assert "value" not in batch assert "value" not in batch[0]
def test_iter_with_column_rename(some_permutation: Permutation): def test_iter_with_column_rename(some_permutation: Permutation):
@@ -791,9 +786,9 @@ def test_iter_with_column_rename(some_permutation: Permutation):
# Check that batches have the renamed column # Check that batches have the renamed column
for batch in batches: for batch in batches:
assert "id" in batch assert "id" in batch[0]
assert "data" in batch assert "data" in batch[0]
assert "value" not in batch assert "value" not in batch[0]
def test_iter_with_limit_offset(some_permutation: Permutation): def test_iter_with_limit_offset(some_permutation: Permutation):
@@ -812,14 +807,14 @@ def test_iter_with_limit_offset(some_permutation: Permutation):
assert len(limit_batches) == 5 assert len(limit_batches) == 5
no_skip = some_permutation.iter(101, skip_last_batch=False) no_skip = some_permutation.iter(101, skip_last_batch=False)
row_100 = next(no_skip)["id"][100] row_100 = next(no_skip)[100]["id"]
# Test with both limit and offset # Test with both limit and offset
limited_perm = some_permutation.with_skip(100).with_take(300) limited_perm = some_permutation.with_skip(100).with_take(300)
limited_batches = list(limited_perm.iter(100, skip_last_batch=False)) limited_batches = list(limited_perm.iter(100, skip_last_batch=False))
# Should have 3 batches (300 / 100) # Should have 3 batches (300 / 100)
assert len(limited_batches) == 3 assert len(limited_batches) == 3
assert limited_batches[0]["id"][0] == row_100 assert limited_batches[0][0]["id"] == row_100
def test_iter_empty_permutation(mem_db): def test_iter_empty_permutation(mem_db):
@@ -842,7 +837,7 @@ def test_iter_single_row(mem_db):
# With skip_last_batch=False, should get one batch # With skip_last_batch=False, should get one batch
batches = list(perm.iter(10, skip_last_batch=False)) batches = list(perm.iter(10, skip_last_batch=False))
assert len(batches) == 1 assert len(batches) == 1
assert len(batches[0]["id"]) == 1 assert len(batches[0]) == 1
# With skip_last_batch=True, should skip the single row (since it's < batch_size) # With skip_last_batch=True, should skip the single row (since it's < batch_size)
batches_skip = list(perm.iter(10, skip_last_batch=True)) batches_skip = list(perm.iter(10, skip_last_batch=True))
@@ -860,8 +855,7 @@ def test_identity_permutation(mem_db):
batches = list(permutation.iter(10, skip_last_batch=False)) batches = list(permutation.iter(10, skip_last_batch=False))
assert len(batches) == 1 assert len(batches) == 1
assert len(batches[0]["id"]) == 10 assert len(batches[0]) == 10
assert len(batches[0]["value"]) == 10
permutation = permutation.remove_columns(["value"]) permutation = permutation.remove_columns(["value"])
assert permutation.num_columns == 1 assert permutation.num_columns == 1
@@ -904,10 +898,10 @@ def test_transform_fn(mem_db):
py_result = list(permutation.with_format("python").iter(10, skip_last_batch=False))[ py_result = list(permutation.with_format("python").iter(10, skip_last_batch=False))[
0 0
] ]
assert len(py_result) == 2 assert len(py_result) == 10
assert len(py_result["id"]) == 10 assert "id" in py_result[0]
assert len(py_result["value"]) == 10 assert "value" in py_result[0]
assert isinstance(py_result, dict) assert isinstance(py_result, list)
try: try:
import torch import torch
@@ -915,9 +909,11 @@ def test_transform_fn(mem_db):
torch_result = list( torch_result = list(
permutation.with_format("torch").iter(10, skip_last_batch=False) permutation.with_format("torch").iter(10, skip_last_batch=False)
)[0] )[0]
assert torch_result.shape == (2, 10) assert isinstance(torch_result, list)
assert torch_result.dtype == torch.int64 assert len(torch_result) == 10
assert isinstance(torch_result, torch.Tensor) assert isinstance(torch_result[0], torch.Tensor)
assert torch_result[0].shape == (2,)
assert torch_result[0].dtype == torch.int64
except ImportError: except ImportError:
# Skip check if torch is not installed # Skip check if torch is not installed
pass pass
@@ -945,3 +941,113 @@ def test_custom_transform(mem_db):
batch = batches[0] batch = batches[0]
assert batch == pa.record_batch([range(10)], ["id"]) assert batch == pa.record_batch([range(10)], ["id"])
def test_getitems_basic(some_permutation: Permutation):
"""Test __getitems__ returns correct rows by offset."""
result = some_permutation.__getitems__([0, 1, 2])
assert isinstance(result, list)
assert "id" in result[0]
assert "value" in result[0]
assert len(result) == 3
def test_getitems_single_index(some_permutation: Permutation):
"""Test __getitems__ with a single index."""
result = some_permutation.__getitems__([0])
assert len(result) == 1
def test_getitems_preserves_order(some_permutation: Permutation):
"""Test __getitems__ returns rows in the requested order."""
# Get rows in forward order
forward = some_permutation.__getitems__([0, 1, 2, 3, 4])
# Get the same rows in reverse order
reverse = some_permutation.__getitems__([4, 3, 2, 1, 0])
assert [r["id"] for r in forward] == list(reversed([r["id"] for r in reverse]))
assert [r["value"] for r in forward] == list(
reversed([r["value"] for r in reverse])
)
def test_getitems_non_contiguous(some_permutation: Permutation):
"""Test __getitems__ with non-contiguous indices."""
result = some_permutation.__getitems__([0, 10, 50, 100, 500])
assert len(result) == 5
# Each id/value pair should match what we'd get individually
for i, offset in enumerate([0, 10, 50, 100, 500]):
single = some_permutation.__getitems__([offset])
assert result[i]["id"] == single[0]["id"]
assert result[i]["value"] == single[0]["value"]
def test_getitems_with_column_selection(some_permutation: Permutation):
"""Test __getitems__ respects column selection."""
id_only = some_permutation.select_columns(["id"])
result = id_only.__getitems__([0, 1, 2])
assert "id" in result[0]
assert "value" not in result[0]
assert len(result) == 3
def test_getitems_with_column_rename(some_permutation: Permutation):
"""Test __getitems__ respects column renames."""
renamed = some_permutation.rename_column("value", "data")
result = renamed.__getitems__([0, 1])
assert "data" in result[0]
assert "value" not in result[0]
assert len(result) == 2
def test_getitems_with_format(some_permutation: Permutation):
"""Test __getitems__ applies the transform function."""
arrow_perm = some_permutation.with_format("arrow")
result = arrow_perm.__getitems__([0, 1, 2])
assert isinstance(result, pa.RecordBatch)
assert result.num_rows == 3
def test_getitems_with_custom_transform(some_permutation: Permutation):
"""Test __getitems__ with a custom transform."""
def transform(batch: pa.RecordBatch) -> list:
return batch.column("id").to_pylist()
custom = some_permutation.with_transform(transform)
result = custom.__getitems__([0, 1, 2])
assert isinstance(result, list)
assert len(result) == 3
def test_getitems_identity_permutation(mem_db):
"""Test __getitems__ on an identity permutation."""
tbl = mem_db.create_table(
"test_table", pa.table({"id": range(10), "value": range(10)})
)
perm = Permutation.identity(tbl)
result = perm.__getitems__([0, 5, 9])
assert [r["id"] for r in result] == [0, 5, 9]
assert [r["value"] for r in result] == [0, 5, 9]
def test_getitems_with_limit_offset(some_permutation: Permutation):
"""Test __getitems__ on a permutation with skip/take applied."""
limited = some_permutation.with_skip(100).with_take(200)
# Should be able to access offsets within the limited range
result = limited.__getitems__([0, 1, 199])
assert len(result) == 3
# The first item of the limited permutation should match offset 100 of original
full_result = some_permutation.__getitems__([100])
limited_result = limited.__getitems__([0])
assert limited_result[0]["id"] == full_result[0]["id"]
def test_getitems_invalid_offset(some_permutation: Permutation):
"""Test __getitems__ with an out-of-range offset raises an error."""
with pytest.raises(Exception):
some_permutation.__getitems__([999999])
+24
View File
@@ -4,6 +4,7 @@
import pyarrow as pa import pyarrow as pa
import pytest import pytest
from lancedb.util import tbl_to_tensor from lancedb.util import tbl_to_tensor
from lancedb.permutation import Permutation
torch = pytest.importorskip("torch") torch = pytest.importorskip("torch")
@@ -16,3 +17,26 @@ def test_table_dataloader(mem_db):
for batch in dataloader: for batch in dataloader:
assert batch.size(0) == 1 assert batch.size(0) == 1
assert batch.size(1) == 10 assert batch.size(1) == 10
def test_permutation_dataloader(mem_db):
table = mem_db.create_table("test_table", pa.table({"a": range(1000)}))
permutation = Permutation.identity(table)
dataloader = torch.utils.data.DataLoader(permutation, batch_size=10, shuffle=True)
for batch in dataloader:
assert batch["a"].size(0) == 10
permutation = permutation.with_format("torch")
dataloader = torch.utils.data.DataLoader(permutation, batch_size=10, shuffle=True)
for batch in dataloader:
assert batch.size(0) == 10
assert batch.size(1) == 1
permutation = permutation.with_format("torch_col")
dataloader = torch.utils.data.DataLoader(
permutation, collate_fn=lambda x: x, batch_size=10, shuffle=True
)
for batch in dataloader:
assert batch.size(0) == 1
assert batch.size(1) == 10
+18 -1
View File
@@ -6,7 +6,7 @@ use std::sync::{Arc, Mutex};
use crate::{ use crate::{
arrow::RecordBatchStream, connection::Connection, error::PythonErrorExt, table::Table, arrow::RecordBatchStream, connection::Connection, error::PythonErrorExt, table::Table,
}; };
use arrow::pyarrow::ToPyArrow; use arrow::pyarrow::{PyArrowType, ToPyArrow};
use lancedb::{ use lancedb::{
dataloader::permutation::{ dataloader::permutation::{
builder::{PermutationBuilder as LancePermutationBuilder, ShuffleStrategy}, builder::{PermutationBuilder as LancePermutationBuilder, ShuffleStrategy},
@@ -328,4 +328,21 @@ impl PyPermutationReader {
Ok(RecordBatchStream::new(stream)) Ok(RecordBatchStream::new(stream))
}) })
} }
#[pyo3(signature = (indices, *, selection=None))]
pub fn take_offsets<'py>(
slf: PyRef<'py, Self>,
indices: Vec<u64>,
selection: Option<Bound<'py, PyAny>>,
) -> PyResult<Bound<'py, PyAny>> {
let selection = Self::parse_selection(selection)?;
let reader = slf.reader.clone();
future_into_py(slf.py(), async move {
let batch = reader
.take_offsets(&indices, selection)
.await
.infer_error()?;
Ok(PyArrowType(batch))
})
}
} }
+5349
View File
File diff suppressed because it is too large Load Diff
+1 -1
View File
@@ -1,6 +1,6 @@
[package] [package]
name = "lancedb" name = "lancedb"
version = "0.26.1" version = "0.26.2"
edition.workspace = true edition.workspace = true
description = "LanceDB: A serverless, low-latency vector database for AI applications" description = "LanceDB: A serverless, low-latency vector database for AI applications"
license.workspace = true license.workspace = true
+9 -2
View File
@@ -921,8 +921,15 @@ pub struct ConnectBuilder {
} }
#[cfg(feature = "remote")] #[cfg(feature = "remote")]
const ENV_VARS_TO_STORAGE_OPTS: [(&str, &str); 1] = const ENV_VARS_TO_STORAGE_OPTS: [(&str, &str); 4] = [
[("AZURE_STORAGE_ACCOUNT_NAME", "azure_storage_account_name")]; ("AZURE_STORAGE_ACCOUNT_NAME", "azure_storage_account_name"),
("AZURE_TENANT_ID", "azure_tenant_id"),
("AZURE_CLIENT_ID", "azure_client_id"),
(
"AZURE_FEDERATED_TOKEN_FILE",
"azure_federated_token_file",
),
];
impl ConnectBuilder { impl ConnectBuilder {
/// Create a new [`ConnectOptions`] with the given database URI. /// Create a new [`ConnectOptions`] with the given database URI.
@@ -39,6 +39,9 @@ pub struct PermutationReader {
limit: Option<u64>, limit: Option<u64>,
available_rows: u64, available_rows: u64,
split: u64, split: u64,
// Cached map of offset to row id for the split
#[allow(clippy::type_complexity)]
offset_map: Arc<tokio::sync::Mutex<Option<Arc<HashMap<u64, u64>>>>>,
} }
impl std::fmt::Debug for PermutationReader { impl std::fmt::Debug for PermutationReader {
@@ -72,6 +75,7 @@ impl PermutationReader {
limit: None, limit: None,
available_rows: 0, available_rows: 0,
split, split,
offset_map: Arc::new(tokio::sync::Mutex::new(None)),
}; };
slf.validate().await?; slf.validate().await?;
// Calculate the number of available rows // Calculate the number of available rows
@@ -157,6 +161,7 @@ impl PermutationReader {
let available_rows = self.verify_limit_offset(self.limit, Some(offset)).await?; let available_rows = self.verify_limit_offset(self.limit, Some(offset)).await?;
self.offset = Some(offset); self.offset = Some(offset);
self.available_rows = available_rows; self.available_rows = available_rows;
self.offset_map = Arc::new(tokio::sync::Mutex::new(None));
Ok(self) Ok(self)
} }
@@ -164,6 +169,7 @@ impl PermutationReader {
let available_rows = self.verify_limit_offset(Some(limit), self.offset).await?; let available_rows = self.verify_limit_offset(Some(limit), self.offset).await?;
self.available_rows = available_rows; self.available_rows = available_rows;
self.limit = Some(limit); self.limit = Some(limit);
self.offset_map = Arc::new(tokio::sync::Mutex::new(None));
Ok(self) Ok(self)
} }
@@ -180,8 +186,9 @@ impl PermutationReader {
base_table: &Arc<dyn BaseTable>, base_table: &Arc<dyn BaseTable>,
row_ids: RecordBatch, row_ids: RecordBatch,
selection: Select, selection: Select,
has_row_id: bool,
) -> Result<RecordBatch> { ) -> Result<RecordBatch> {
let has_row_id = Self::has_row_id(&selection)?;
let num_rows = row_ids.num_rows(); let num_rows = row_ids.num_rows();
let row_ids = row_ids let row_ids = row_ids
.column(0) .column(0)
@@ -282,14 +289,13 @@ impl PermutationReader {
row_ids: DatasetRecordBatchStream, row_ids: DatasetRecordBatchStream,
selection: Select, selection: Select,
) -> Result<SendableRecordBatchStream> { ) -> Result<SendableRecordBatchStream> {
let has_row_id = Self::has_row_id(&selection)?;
let mut stream = row_ids let mut stream = row_ids
.map_err(Error::from) .map_err(Error::from)
.try_filter_map(move |batch| { .try_filter_map(move |batch| {
let selection = selection.clone(); let selection = selection.clone();
let base_table = base_table.clone(); let base_table = base_table.clone();
async move { async move {
Self::load_batch(&base_table, batch, selection, has_row_id) Self::load_batch(&base_table, batch, selection)
.await .await
.map(Some) .map(Some)
} }
@@ -397,6 +403,82 @@ impl PermutationReader {
Self::row_ids_to_batches(self.base_table.clone(), row_ids, selection).await Self::row_ids_to_batches(self.base_table.clone(), row_ids, selection).await
} }
/// If we are going to use `take` then we load the offset -> row id map once for the split and cache it
///
/// This method fetches the map with find-or-create semantics.
async fn get_offset_map(
&self,
permutation_table: &Arc<dyn BaseTable>,
) -> Result<Arc<HashMap<u64, u64>>> {
let mut offset_map_ref = self.offset_map.lock().await;
if let Some(offset_map) = &*offset_map_ref {
return Ok(offset_map.clone());
}
let mut offset_map = HashMap::new();
let mut row_ids_query = Table::from(permutation_table.clone())
.query()
.select(Select::Columns(vec![SRC_ROW_ID_COL.to_string()]))
.only_if(format!("{} = {}", SPLIT_ID_COLUMN, self.split));
if let Some(offset) = self.offset {
row_ids_query = row_ids_query.offset(offset as usize);
}
if let Some(limit) = self.limit {
row_ids_query = row_ids_query.limit(limit as usize);
}
let mut row_ids = row_ids_query.execute().await?;
while let Some(batch) = row_ids.try_next().await? {
let row_ids = batch
.column(0)
.as_primitive::<UInt64Type>()
.values()
.to_vec();
for (i, row_id) in row_ids.iter().enumerate() {
offset_map.insert(i as u64, *row_id);
}
}
let offset_map = Arc::new(offset_map);
*offset_map_ref = Some(offset_map.clone());
Ok(offset_map)
}
pub async fn take_offsets(&self, offsets: &[u64], selection: Select) -> Result<RecordBatch> {
if let Some(permutation_table) = &self.permutation_table {
let offset_map = self.get_offset_map(permutation_table).await?;
let row_ids = offsets
.iter()
.map(|o| offset_map.get(o).copied().expect_ok().map_err(Error::from))
.collect::<Result<Vec<_>>>()?;
let row_ids = RecordBatch::try_new(
Arc::new(arrow_schema::Schema::new(vec![arrow_schema::Field::new(
"row_id",
arrow_schema::DataType::UInt64,
false,
)])),
vec![Arc::new(UInt64Array::from(row_ids))],
)?;
Self::load_batch(&self.base_table, row_ids, selection).await
} else {
let table = Table::from(self.base_table.clone());
let batches = table
.take_offsets(offsets.to_vec())
.select(selection.clone())
.execute()
.await?
.try_collect::<Vec<_>>()
.await?;
if let Some(first_batch) = batches.first() {
let schema = first_batch.schema();
let batch = arrow::compute::concat_batches(&schema, &batches)?;
Ok(batch)
} else {
Ok(RecordBatch::try_new(
self.output_schema(selection).await?,
vec![],
)?)
}
}
}
pub async fn output_schema(&self, selection: Select) -> Result<SchemaRef> { pub async fn output_schema(&self, selection: Select) -> Result<SchemaRef> {
let table = Table::from(self.base_table.clone()); let table = Table::from(self.base_table.clone());
table.query().select(selection).output_schema().await table.query().select(selection).output_schema().await
@@ -543,4 +625,224 @@ mod tests {
check_batch(&mut stream, &row_ids[7..9]).await; check_batch(&mut stream, &row_ids[7..9]).await;
assert!(stream.try_next().await.unwrap().is_none()); assert!(stream.try_next().await.unwrap().is_none());
} }
/// Helper to create a base table and permutation table for take_offsets tests.
/// Returns (base_table, row_ids_table, shuffled_row_ids).
async fn setup_permutation_tables(num_rows: usize) -> (Table, Table, Vec<u64>) {
let base_table = lance_datagen::gen_batch()
.col("idx", lance_datagen::array::step::<Int32Type>())
.col("other_col", lance_datagen::array::step::<UInt64Type>())
.into_mem_table("tbl", RowCount::from(num_rows as u64), BatchCount::from(1))
.await;
let mut row_ids = collect_column::<UInt64Type>(&base_table, "_rowid").await;
row_ids.shuffle(&mut rand::rng());
let split_ids = UInt64Array::from_iter_values(std::iter::repeat_n(0u64, row_ids.len()));
let permutation_batch = RecordBatch::try_new(
Arc::new(Schema::new(vec![
Field::new("row_id", DataType::UInt64, false),
Field::new(SPLIT_ID_COLUMN, DataType::UInt64, false),
])),
vec![
Arc::new(UInt64Array::from(row_ids.clone())),
Arc::new(split_ids),
],
)
.unwrap();
let row_ids_table = virtual_table("row_ids", &permutation_batch).await;
(base_table, row_ids_table, row_ids)
}
#[tokio::test]
async fn test_take_offsets_with_permutation_table() {
let (base_table, row_ids_table, row_ids) = setup_permutation_tables(10).await;
let reader = PermutationReader::try_from_tables(
base_table.base_table().clone(),
row_ids_table.base_table().clone(),
0,
)
.await
.unwrap();
// Take specific offsets and verify the returned rows match the permutation order
let offsets = vec![0, 2, 4];
let batch = reader.take_offsets(&offsets, Select::All).await.unwrap();
assert_eq!(batch.num_rows(), 3);
let idx_values = batch
.column(0)
.as_primitive::<Int32Type>()
.values()
.to_vec();
let expected: Vec<i32> = offsets
.iter()
.map(|&o| row_ids[o as usize] as i32)
.collect();
assert_eq!(idx_values, expected);
}
#[tokio::test]
async fn test_take_offsets_preserves_order() {
let (base_table, row_ids_table, row_ids) = setup_permutation_tables(10).await;
let reader = PermutationReader::try_from_tables(
base_table.base_table().clone(),
row_ids_table.base_table().clone(),
0,
)
.await
.unwrap();
// Take offsets in reverse order and verify returned rows match that order
let offsets = vec![5, 3, 1, 0];
let batch = reader.take_offsets(&offsets, Select::All).await.unwrap();
assert_eq!(batch.num_rows(), 4);
let idx_values = batch
.column(0)
.as_primitive::<Int32Type>()
.values()
.to_vec();
let expected: Vec<i32> = offsets
.iter()
.map(|&o| row_ids[o as usize] as i32)
.collect();
assert_eq!(idx_values, expected);
}
#[tokio::test]
async fn test_take_offsets_with_column_selection() {
let (base_table, row_ids_table, row_ids) = setup_permutation_tables(10).await;
let reader = PermutationReader::try_from_tables(
base_table.base_table().clone(),
row_ids_table.base_table().clone(),
0,
)
.await
.unwrap();
let offsets = vec![1, 3];
let batch = reader
.take_offsets(&offsets, Select::Columns(vec!["idx".to_string()]))
.await
.unwrap();
assert_eq!(batch.num_rows(), 2);
assert_eq!(batch.num_columns(), 1);
assert_eq!(batch.schema().field(0).name(), "idx");
let idx_values = batch
.column(0)
.as_primitive::<Int32Type>()
.values()
.to_vec();
let expected: Vec<i32> = offsets
.iter()
.map(|&o| row_ids[o as usize] as i32)
.collect();
assert_eq!(idx_values, expected);
}
#[tokio::test]
async fn test_take_offsets_invalid_offset() {
let (base_table, row_ids_table, _) = setup_permutation_tables(5).await;
let reader = PermutationReader::try_from_tables(
base_table.base_table().clone(),
row_ids_table.base_table().clone(),
0,
)
.await
.unwrap();
// Offset 999 doesn't exist in the offset map
let result = reader.take_offsets(&[0, 999], Select::All).await;
assert!(result.is_err());
}
#[tokio::test]
async fn test_take_offsets_identity_reader() {
let base_table = lance_datagen::gen_batch()
.col("idx", lance_datagen::array::step::<Int32Type>())
.into_mem_table("tbl", RowCount::from(10), BatchCount::from(1))
.await;
let reader = PermutationReader::identity(base_table.base_table().clone()).await;
// With no permutation table, take_offsets uses the base table directly
let offsets = vec![0, 2, 4, 6];
let batch = reader.take_offsets(&offsets, Select::All).await.unwrap();
assert_eq!(batch.num_rows(), 4);
let idx_values = batch
.column(0)
.as_primitive::<Int32Type>()
.values()
.to_vec();
assert_eq!(idx_values, vec![0, 2, 4, 6]);
}
#[tokio::test]
async fn test_take_offsets_caches_offset_map() {
let (base_table, row_ids_table, row_ids) = setup_permutation_tables(10).await;
let reader = PermutationReader::try_from_tables(
base_table.base_table().clone(),
row_ids_table.base_table().clone(),
0,
)
.await
.unwrap();
// First call populates the cache
let batch1 = reader.take_offsets(&[0, 1], Select::All).await.unwrap();
// Second call should use the cached offset map and produce consistent results
let batch2 = reader.take_offsets(&[0, 1], Select::All).await.unwrap();
let values1 = batch1
.column(0)
.as_primitive::<Int32Type>()
.values()
.to_vec();
let values2 = batch2
.column(0)
.as_primitive::<Int32Type>()
.values()
.to_vec();
assert_eq!(values1, values2);
let expected: Vec<i32> = vec![row_ids[0] as i32, row_ids[1] as i32];
assert_eq!(values1, expected);
}
#[tokio::test]
async fn test_take_offsets_single_offset() {
let (base_table, row_ids_table, row_ids) = setup_permutation_tables(5).await;
let reader = PermutationReader::try_from_tables(
base_table.base_table().clone(),
row_ids_table.base_table().clone(),
0,
)
.await
.unwrap();
let batch = reader.take_offsets(&[2], Select::All).await.unwrap();
assert_eq!(batch.num_rows(), 1);
let idx_values = batch
.column(0)
.as_primitive::<Int32Type>()
.values()
.to_vec();
assert_eq!(idx_values, vec![row_ids[2] as i32]);
}
} }
+20 -15
View File
@@ -438,21 +438,26 @@ impl<S: HttpSend> RestfulLanceDbClient<S> {
); );
} }
if let Some(v) = options.0.get("account_name") { // Map storage options to HTTP headers for Azure configuration.
headers.insert( const OPTION_TO_HEADER: &[(&str, &str)] = &[
HeaderName::from_static("x-azure-storage-account-name"), ("account_name", "x-azure-storage-account-name"),
HeaderValue::from_str(v).map_err(|_| Error::InvalidInput { ("azure_storage_account_name", "x-azure-storage-account-name"),
message: format!("non-ascii storage account name '{}' provided", db_name), ("azure_tenant_id", "x-azure-tenant-id"),
})?, ("azure_client_id", "x-azure-client-id"),
); (
} "azure_federated_token_file",
if let Some(v) = options.0.get("azure_storage_account_name") { "x-azure-federated-token-file",
headers.insert( ),
HeaderName::from_static("x-azure-storage-account-name"), ];
HeaderValue::from_str(v).map_err(|_| Error::InvalidInput { for (opt_key, header_name) in OPTION_TO_HEADER {
message: format!("non-ascii storage account name '{}' provided", db_name), if let Some(v) = options.get(opt_key) {
})?, headers.insert(
); HeaderName::from_static(header_name),
HeaderValue::from_str(v).map_err(|_| Error::InvalidInput {
message: format!("non-ascii value for '{}' provided", opt_key),
})?,
);
}
} }
for (key, value) in &config.extra_headers { for (key, value) in &config.extra_headers {
+11 -1
View File
@@ -792,11 +792,21 @@ impl RemoteOptions {
pub fn new(options: HashMap<String, String>) -> Self { pub fn new(options: HashMap<String, String>) -> Self {
Self(options) Self(options)
} }
pub fn get(&self, key: &str) -> Option<&String> {
self.0.get(key)
}
} }
impl From<StorageOptions> for RemoteOptions { impl From<StorageOptions> for RemoteOptions {
fn from(options: StorageOptions) -> Self { fn from(options: StorageOptions) -> Self {
let supported_opts = vec!["account_name", "azure_storage_account_name"]; let supported_opts = vec![
"account_name",
"azure_storage_account_name",
"azure_tenant_id",
"azure_client_id",
"azure_federated_token_file",
];
let mut filtered = HashMap::new(); let mut filtered = HashMap::new();
for opt in supported_opts { for opt in supported_opts {
if let Some(v) = options.0.get(opt) { if let Some(v) = options.0.get(opt) {
+4 -183
View File
@@ -16,8 +16,6 @@ use datafusion_physical_plan::union::UnionExec;
use datafusion_physical_plan::ExecutionPlan; use datafusion_physical_plan::ExecutionPlan;
use futures::{FutureExt, StreamExt, TryFutureExt}; use futures::{FutureExt, StreamExt, TryFutureExt};
use lance::dataset::builder::DatasetBuilder; use lance::dataset::builder::DatasetBuilder;
use lance::dataset::cleanup::RemovalStats;
use lance::dataset::optimize::{compact_files, CompactionMetrics, IndexRemapperOptions};
use lance::dataset::scanner::Scanner; use lance::dataset::scanner::Scanner;
pub use lance::dataset::ColumnAlteration; pub use lance::dataset::ColumnAlteration;
pub use lance::dataset::NewColumnTransform; pub use lance::dataset::NewColumnTransform;
@@ -46,7 +44,6 @@ use lance_namespace::models::{
use lance_namespace::LanceNamespace; use lance_namespace::LanceNamespace;
use lance_table::format::Manifest; use lance_table::format::Manifest;
use lance_table::io::commit::ManifestNamingScheme; use lance_table::io::commit::ManifestNamingScheme;
use log::info;
use serde::{Deserialize, Serialize}; use serde::{Deserialize, Serialize};
use std::collections::HashMap; use std::collections::HashMap;
use std::format; use std::format;
@@ -79,6 +76,7 @@ pub mod datafusion;
pub(crate) mod dataset; pub(crate) mod dataset;
pub mod delete; pub mod delete;
pub mod merge; pub mod merge;
pub mod optimize;
pub mod schema_evolution; pub mod schema_evolution;
pub mod update; pub mod update;
@@ -86,12 +84,12 @@ use crate::index::waiter::wait_for_index;
pub use chrono::Duration; pub use chrono::Duration;
pub use delete::DeleteResult; pub use delete::DeleteResult;
use futures::future::{join_all, Either}; use futures::future::{join_all, Either};
pub use lance::dataset::optimize::CompactionOptions;
pub use lance::dataset::refs::{TagContents, Tags as LanceTags}; pub use lance::dataset::refs::{TagContents, Tags as LanceTags};
pub use lance::dataset::scanner::DatasetRecordBatchStream; pub use lance::dataset::scanner::DatasetRecordBatchStream;
use lance::dataset::statistics::DatasetStatisticsExt; use lance::dataset::statistics::DatasetStatisticsExt;
use lance_index::frag_reuse::FRAG_REUSE_INDEX_NAME; use lance_index::frag_reuse::FRAG_REUSE_INDEX_NAME;
pub use lance_index::optimize::OptimizeOptions; pub use lance_index::optimize::OptimizeOptions;
pub use optimize::{CompactionOptions, OptimizeAction, OptimizeStats};
pub use schema_evolution::{AddColumnsResult, AlterColumnsResult, DropColumnsResult}; pub use schema_evolution::{AddColumnsResult, AlterColumnsResult, DropColumnsResult};
use serde_with::skip_serializing_none; use serde_with::skip_serializing_none;
pub use update::{UpdateBuilder, UpdateResult}; pub use update::{UpdateBuilder, UpdateResult};
@@ -169,85 +167,6 @@ impl TableDefinition {
} }
} }
/// Optimize the dataset.
///
/// Similar to `VACUUM` in PostgreSQL, it offers different options to
/// optimize different parts of the table on disk.
///
/// By default, it optimizes everything, as [`OptimizeAction::All`].
pub enum OptimizeAction {
/// Run all optimizations with default values
All,
/// Compacts files in the dataset
///
/// LanceDb uses a readonly filesystem for performance and safe concurrency. Every time
/// new data is added it will be added into new files. Small files
/// can hurt both read and write performance. Compaction will merge small files
/// into larger ones.
///
/// All operations that modify data (add, delete, update, merge insert, etc.) will create
/// new files. If these operations are run frequently then compaction should run frequently.
///
/// If these operations are never run (search only) then compaction is not necessary.
Compact {
options: CompactionOptions,
remap_options: Option<Arc<dyn IndexRemapperOptions>>,
},
/// Prune old version of datasets
///
/// Every change in LanceDb is additive. When data is removed from a dataset a new version is
/// created that doesn't contain the removed data. However, the old version, which does contain
/// the removed data, is left in place. This is necessary for consistency and concurrency and
/// also enables time travel functionality like the ability to checkout an older version of the
/// dataset to undo changes.
///
/// Over time, these old versions can consume a lot of disk space. The prune operation will
/// remove versions of the dataset that are older than a certain age. This will free up the
/// space used by that old data.
///
/// Once a version is pruned it can no longer be checked out.
Prune {
/// The duration of time to keep versions of the dataset.
older_than: Option<Duration>,
/// Because they may be part of an in-progress transaction, files newer than 7 days old are not deleted by default.
/// If you are sure that there are no in-progress transactions, then you can set this to True to delete all files older than `older_than`.
delete_unverified: Option<bool>,
/// If true, an error will be returned if there are any old versions that are still tagged.
error_if_tagged_old_versions: Option<bool>,
},
/// Optimize the indices
///
/// This operation optimizes all indices in the table. When new data is added to LanceDb
/// it is not added to the indices. However, it can still turn up in searches because the search
/// function will scan both the indexed data and the unindexed data in parallel. Over time, the
/// unindexed data can become large enough that the search performance is slow. This operation
/// will add the unindexed data to the indices without rerunning the full index creation process.
///
/// Optimizing an index is faster than re-training the index but it does not typically adjust the
/// underlying model relied upon by the index. This can eventually lead to poor search accuracy
/// and so users may still want to occasionally retrain the index after adding a large amount of
/// data.
///
/// For example, when using IVF, an index will create clusters. Optimizing an index assigns unindexed
/// data to the existing clusters, but it does not move the clusters or create new clusters.
Index(OptimizeOptions),
}
impl Default for OptimizeAction {
fn default() -> Self {
Self::All
}
}
/// Statistics about the optimization.
pub struct OptimizeStats {
/// Stats of the file compaction.
pub compaction: Option<CompactionMetrics>,
/// Stats of the version pruning
pub prune: Option<RemovalStats>,
}
/// Describes what happens when a vector either contains NaN or /// Describes what happens when a vector either contains NaN or
/// does not have enough values /// does not have enough values
#[derive(Clone, Debug, Default)] #[derive(Clone, Debug, Default)]
@@ -1859,16 +1778,6 @@ impl NativeTable {
}) })
} }
async fn optimize_indices(&self, options: &OptimizeOptions) -> Result<()> {
info!("LanceDB: optimizing indices: {:?}", options);
self.dataset
.get_mut()
.await?
.optimize_indices(options)
.await?;
Ok(())
}
/// Merge new data into this table. /// Merge new data into this table.
pub async fn merge( pub async fn merge(
&mut self, &mut self,
@@ -1884,47 +1793,6 @@ impl NativeTable {
Ok(()) Ok(())
} }
/// Remove old versions of the dataset from disk.
///
/// # Arguments
/// * `older_than` - The duration of time to keep versions of the dataset.
/// * `delete_unverified` - Because they may be part of an in-progress
/// transaction, files newer than 7 days old are not deleted by default.
/// If you are sure that there are no in-progress transactions, then you
/// can set this to True to delete all files older than `older_than`.
///
/// This calls into [lance::dataset::Dataset::cleanup_old_versions] and
/// returns the result.
async fn cleanup_old_versions(
&self,
older_than: Duration,
delete_unverified: Option<bool>,
error_if_tagged_old_versions: Option<bool>,
) -> Result<RemovalStats> {
Ok(self
.dataset
.get_mut()
.await?
.cleanup_old_versions(older_than, delete_unverified, error_if_tagged_old_versions)
.await?)
}
/// Compact files in the dataset.
///
/// This can be run after making several small appends to optimize the table
/// for faster reads.
///
/// This calls into [lance::dataset::optimize::compact_files].
async fn compact_files(
&self,
options: CompactionOptions,
remap_options: Option<Arc<dyn IndexRemapperOptions>>,
) -> Result<CompactionMetrics> {
let mut dataset_mut = self.dataset.get_mut().await?;
let metrics = compact_files(&mut dataset_mut, options, remap_options).await?;
Ok(metrics)
}
// TODO: why are these individual methods and not some single "get_stats" method? // TODO: why are these individual methods and not some single "get_stats" method?
pub async fn count_fragments(&self) -> Result<usize> { pub async fn count_fragments(&self) -> Result<usize> {
Ok(self.dataset.get().await?.count_fragments()) Ok(self.dataset.get().await?.count_fragments())
@@ -3010,55 +2878,8 @@ impl BaseTable for NativeTable {
} }
async fn optimize(&self, action: OptimizeAction) -> Result<OptimizeStats> { async fn optimize(&self, action: OptimizeAction) -> Result<OptimizeStats> {
let mut stats = OptimizeStats { // Delegate to the submodule implementation
compaction: None, optimize::execute_optimize(self, action).await
prune: None,
};
match action {
OptimizeAction::All => {
stats.compaction = self
.optimize(OptimizeAction::Compact {
options: CompactionOptions::default(),
remap_options: None,
})
.await?
.compaction;
stats.prune = self
.optimize(OptimizeAction::Prune {
older_than: None,
delete_unverified: None,
error_if_tagged_old_versions: None,
})
.await?
.prune;
self.optimize(OptimizeAction::Index(OptimizeOptions::default()))
.await?;
}
OptimizeAction::Compact {
options,
remap_options,
} => {
stats.compaction = Some(self.compact_files(options, remap_options).await?);
}
OptimizeAction::Prune {
older_than,
delete_unverified,
error_if_tagged_old_versions,
} => {
stats.prune = Some(
self.cleanup_old_versions(
older_than.unwrap_or(Duration::try_days(7).expect("valid delta")),
delete_unverified,
error_if_tagged_old_versions,
)
.await?,
);
}
OptimizeAction::Index(options) => {
self.optimize_indices(&options).await?;
}
}
Ok(stats)
} }
async fn add_columns( async fn add_columns(
+781
View File
@@ -0,0 +1,781 @@
// SPDX-License-Identifier: Apache-2.0
// SPDX-FileCopyrightText: Copyright The LanceDB Authors
//! Table optimization operations for compaction, pruning, and index optimization.
//!
//! This module contains the implementation of optimization operations that help
//! maintain good performance for LanceDB tables.
use std::sync::Arc;
use lance::dataset::cleanup::RemovalStats;
use lance::dataset::optimize::{compact_files, CompactionMetrics, IndexRemapperOptions};
use lance_index::optimize::OptimizeOptions;
use lance_index::DatasetIndexExt;
use log::info;
pub use chrono::Duration;
pub use lance::dataset::optimize::CompactionOptions;
use super::NativeTable;
use crate::error::Result;
/// Optimize the dataset.
///
/// Similar to `VACUUM` in PostgreSQL, it offers different options to
/// optimize different parts of the table on disk.
///
/// By default, it optimizes everything, as [`OptimizeAction::All`].
pub enum OptimizeAction {
/// Run all optimizations with default values
All,
/// Compacts files in the dataset
///
/// LanceDb uses a readonly filesystem for performance and safe concurrency. Every time
/// new data is added it will be added into new files. Small files
/// can hurt both read and write performance. Compaction will merge small files
/// into larger ones.
///
/// All operations that modify data (add, delete, update, merge insert, etc.) will create
/// new files. If these operations are run frequently then compaction should run frequently.
///
/// If these operations are never run (search only) then compaction is not necessary.
Compact {
options: CompactionOptions,
remap_options: Option<Arc<dyn IndexRemapperOptions>>,
},
/// Prune old version of datasets
///
/// Every change in LanceDb is additive. When data is removed from a dataset a new version is
/// created that doesn't contain the removed data. However, the old version, which does contain
/// the removed data, is left in place. This is necessary for consistency and concurrency and
/// also enables time travel functionality like the ability to checkout an older version of the
/// dataset to undo changes.
///
/// Over time, these old versions can consume a lot of disk space. The prune operation will
/// remove versions of the dataset that are older than a certain age. This will free up the
/// space used by that old data.
///
/// Once a version is pruned it can no longer be checked out.
Prune {
/// The duration of time to keep versions of the dataset.
older_than: Option<Duration>,
/// Because they may be part of an in-progress transaction, files newer than 7 days old are not deleted by default.
/// If you are sure that there are no in-progress transactions, then you can set this to True to delete all files older than `older_than`.
delete_unverified: Option<bool>,
/// If true, an error will be returned if there are any old versions that are still tagged.
error_if_tagged_old_versions: Option<bool>,
},
/// Optimize the indices
///
/// This operation optimizes all indices in the table. When new data is added to LanceDb
/// it is not added to the indices. However, it can still turn up in searches because the search
/// function will scan both the indexed data and the unindexed data in parallel. Over time, the
/// unindexed data can become large enough that the search performance is slow. This operation
/// will add the unindexed data to the indices without rerunning the full index creation process.
///
/// Optimizing an index is faster than re-training the index but it does not typically adjust the
/// underlying model relied upon by the index. This can eventually lead to poor search accuracy
/// and so users may still want to occasionally retrain the index after adding a large amount of
/// data.
///
/// For example, when using IVF, an index will create clusters. Optimizing an index assigns unindexed
/// data to the existing clusters, but it does not move the clusters or create new clusters.
Index(OptimizeOptions),
}
impl Default for OptimizeAction {
fn default() -> Self {
Self::All
}
}
/// Statistics about the optimization.
#[derive(Debug, Default)]
pub struct OptimizeStats {
/// Stats of the file compaction.
pub compaction: Option<CompactionMetrics>,
/// Stats of the version pruning
pub prune: Option<RemovalStats>,
}
/// Internal implementation of optimize_indices
///
/// This logic was moved from NativeTable to keep table.rs clean.
pub(crate) async fn optimize_indices(table: &NativeTable, options: &OptimizeOptions) -> Result<()> {
info!("LanceDB: optimizing indices: {:?}", options);
table
.dataset
.get_mut()
.await?
.optimize_indices(options)
.await?;
Ok(())
}
/// Remove old versions of the dataset from disk.
///
/// # Arguments
/// * `older_than` - The duration of time to keep versions of the dataset.
/// * `delete_unverified` - Because they may be part of an in-progress
/// transaction, files newer than 7 days old are not deleted by default.
/// If you are sure that there are no in-progress transactions, then you
/// can set this to True to delete all files older than `older_than`.
///
/// This calls into [lance::dataset::Dataset::cleanup_old_versions] and
/// returns the result.
pub(crate) async fn cleanup_old_versions(
table: &NativeTable,
older_than: Duration,
delete_unverified: Option<bool>,
error_if_tagged_old_versions: Option<bool>,
) -> Result<RemovalStats> {
Ok(table
.dataset
.get_mut()
.await?
.cleanup_old_versions(older_than, delete_unverified, error_if_tagged_old_versions)
.await?)
}
/// Compact files in the dataset.
///
/// This can be run after making several small appends to optimize the table
/// for faster reads.
///
/// This calls into [lance::dataset::optimize::compact_files].
pub(crate) async fn compact_files_impl(
table: &NativeTable,
options: CompactionOptions,
remap_options: Option<Arc<dyn IndexRemapperOptions>>,
) -> Result<CompactionMetrics> {
let mut dataset_mut = table.dataset.get_mut().await?;
let metrics = compact_files(&mut dataset_mut, options, remap_options).await?;
Ok(metrics)
}
/// Execute the optimize operation on the table.
///
/// This is the main entry point for all optimization operations.
pub(crate) async fn execute_optimize(
table: &NativeTable,
action: OptimizeAction,
) -> Result<OptimizeStats> {
let mut stats = OptimizeStats {
compaction: None,
prune: None,
};
match action {
OptimizeAction::All => {
// Call helper functions directly to avoid async recursion issues
stats.compaction =
Some(compact_files_impl(table, CompactionOptions::default(), None).await?);
stats.prune = Some(
cleanup_old_versions(
table,
Duration::try_days(7).expect("valid delta"),
None,
None,
)
.await?,
);
optimize_indices(table, &OptimizeOptions::default()).await?;
}
OptimizeAction::Compact {
options,
remap_options,
} => {
stats.compaction = Some(compact_files_impl(table, options, remap_options).await?);
}
OptimizeAction::Prune {
older_than,
delete_unverified,
error_if_tagged_old_versions,
} => {
stats.prune = Some(
cleanup_old_versions(
table,
older_than.unwrap_or(Duration::try_days(7).expect("valid delta")),
delete_unverified,
error_if_tagged_old_versions,
)
.await?,
);
}
OptimizeAction::Index(options) => {
optimize_indices(table, &options).await?;
}
}
Ok(stats)
}
#[cfg(test)]
mod tests {
use arrow_array::{Int32Array, RecordBatch, RecordBatchIterator, StringArray};
use arrow_schema::{DataType, Field, Schema};
use rstest::rstest;
use std::sync::Arc;
use crate::connect;
use crate::index::{scalar::BTreeIndexBuilder, Index};
use crate::query::ExecutableQuery;
use crate::table::{CompactionOptions, OptimizeAction, OptimizeStats};
use futures::TryStreamExt;
#[tokio::test]
async fn test_optimize_compact_simple() {
let conn = connect("memory://").execute().await.unwrap();
// Create a table with initial data
let schema = Arc::new(Schema::new(vec![Field::new("i", DataType::Int32, false)]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from_iter_values(0..100))],
)
.unwrap();
let table = conn
.create_table(
"test_compact",
RecordBatchIterator::new(vec![Ok(batch)], schema.clone()),
)
.execute()
.await
.unwrap();
// Add more data to create multiple fragments
for i in 0..5 {
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from_iter_values(
(i * 100 + 100)..((i + 1) * 100 + 100),
))],
)
.unwrap();
table
.add(RecordBatchIterator::new(vec![Ok(batch)], schema.clone()))
.execute()
.await
.unwrap();
}
// Verify we have multiple fragments before compaction
let initial_row_count = table.count_rows(None).await.unwrap();
assert_eq!(initial_row_count, 600);
// Run compaction
let stats = table
.optimize(OptimizeAction::Compact {
options: CompactionOptions {
target_rows_per_fragment: 1000,
..Default::default()
},
remap_options: None,
})
.await
.unwrap();
// Verify compaction occurred
assert!(stats.compaction.is_some());
let compaction_metrics = stats.compaction.unwrap();
assert!(compaction_metrics.fragments_removed > 0);
// Verify data integrity after compaction
let final_row_count = table.count_rows(None).await.unwrap();
assert_eq!(final_row_count, 600);
// Verify data content is correct
let batches = table
.query()
.execute()
.await
.unwrap()
.try_collect::<Vec<_>>()
.await
.unwrap();
let total_rows: usize = batches.iter().map(|b| b.num_rows()).sum();
assert_eq!(total_rows, 600);
// Verify the values are as expected
let mut all_values: Vec<i32> = Vec::new();
for batch in &batches {
let array = batch["i"].as_any().downcast_ref::<Int32Array>().unwrap();
all_values.extend(array.values().iter().copied());
}
all_values.sort();
let expected: Vec<i32> = (0..600).collect();
assert_eq!(all_values, expected);
}
#[tokio::test]
async fn test_optimize_prune_versions() {
let conn = connect("memory://").execute().await.unwrap();
// Create a table
let schema = Arc::new(Schema::new(vec![Field::new("i", DataType::Int32, false)]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from_iter_values(0..10))],
)
.unwrap();
let table = conn
.create_table(
"test_prune",
RecordBatchIterator::new(vec![Ok(batch)], schema.clone()),
)
.execute()
.await
.unwrap();
// Make several modifications to create versions
for i in 0..5 {
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from_iter_values(
(i * 10 + 10)..((i + 1) * 10 + 10),
))],
)
.unwrap();
table
.add(RecordBatchIterator::new(vec![Ok(batch)], schema.clone()))
.execute()
.await
.unwrap();
}
// Verify multiple versions exist
let versions = table.list_versions().await.unwrap();
assert!(versions.len() > 1);
// Run prune with a very old cutoff (won't delete recent versions)
let stats = table
.optimize(OptimizeAction::Prune {
older_than: Some(chrono::Duration::try_days(0).unwrap()),
delete_unverified: Some(true),
error_if_tagged_old_versions: None,
})
.await
.unwrap();
// Prune-only operation should not have compaction stats
assert!(stats.compaction.is_none());
// Verify prune stats
let prune_stats = stats.prune.unwrap();
assert!(prune_stats.bytes_removed > 0);
assert_eq!(prune_stats.old_versions, 5);
// Verify data is still intact
let final_row_count = table.count_rows(None).await.unwrap();
assert_eq!(final_row_count, 60);
// Verify data content is correct
let batches = table
.query()
.execute()
.await
.unwrap()
.try_collect::<Vec<_>>()
.await
.unwrap();
let mut all_values: Vec<i32> = Vec::new();
for batch in &batches {
let array = batch["i"].as_any().downcast_ref::<Int32Array>().unwrap();
all_values.extend(array.values().iter().copied());
}
all_values.sort();
let expected: Vec<i32> = (0..60).collect();
assert_eq!(all_values, expected);
}
#[tokio::test]
async fn test_optimize_index() {
let conn = connect("memory://").execute().await.unwrap();
// Create a table with data
let schema = Arc::new(Schema::new(vec![Field::new("i", DataType::Int32, false)]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from_iter_values(0..100))],
)
.unwrap();
let table = conn
.create_table(
"test_index_optimize",
RecordBatchIterator::new(vec![Ok(batch)], schema.clone()),
)
.execute()
.await
.unwrap();
// Create an index
table
.create_index(&["i"], Index::BTree(BTreeIndexBuilder::default()))
.execute()
.await
.unwrap();
// Add more data (unindexed)
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from_iter_values(100..200))],
)
.unwrap();
table
.add(RecordBatchIterator::new(vec![Ok(batch)], schema.clone()))
.execute()
.await
.unwrap();
// Verify index stats before optimization
let indices = table.list_indices().await.unwrap();
assert_eq!(indices.len(), 1);
let index_name = indices[0].name.clone();
let stats_before = table.index_stats(&index_name).await.unwrap().unwrap();
assert_eq!(stats_before.num_indexed_rows, 100);
assert_eq!(stats_before.num_unindexed_rows, 100);
// Run index optimization
let stats = table
.optimize(OptimizeAction::Index(Default::default()))
.await
.unwrap();
// For index optimization, compaction and prune stats should be None
assert!(stats.compaction.is_none());
assert!(stats.prune.is_none());
// Verify index stats after optimization
let stats_after = table.index_stats(&index_name).await.unwrap().unwrap();
assert_eq!(stats_after.num_indexed_rows, 200);
assert_eq!(stats_after.num_unindexed_rows, 0);
assert!(stats_after.num_indices.is_some());
// Verify data integrity
let final_row_count = table.count_rows(None).await.unwrap();
assert_eq!(final_row_count, 200);
}
#[tokio::test]
async fn test_optimize_all() {
let conn = connect("memory://").execute().await.unwrap();
// Create a table with data
let schema = Arc::new(Schema::new(vec![Field::new("i", DataType::Int32, false)]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from_iter_values(0..100))],
)
.unwrap();
let table = conn
.create_table(
"test_optimize_all",
RecordBatchIterator::new(vec![Ok(batch)], schema.clone()),
)
.execute()
.await
.unwrap();
// Add more data
for i in 0..3 {
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from_iter_values(
(i * 100 + 100)..((i + 1) * 100 + 100),
))],
)
.unwrap();
table
.add(RecordBatchIterator::new(vec![Ok(batch)], schema.clone()))
.execute()
.await
.unwrap();
}
// Run all optimizations
let stats = table.optimize(OptimizeAction::All).await.unwrap();
// Verify stats from both compaction and prune
assert!(stats.compaction.is_some());
assert!(stats.prune.is_some());
// Verify data integrity
let final_row_count = table.count_rows(None).await.unwrap();
assert_eq!(final_row_count, 400);
// Verify data content
let batches = table
.query()
.execute()
.await
.unwrap()
.try_collect::<Vec<_>>()
.await
.unwrap();
let mut all_values: Vec<i32> = Vec::new();
for batch in &batches {
let array = batch["i"].as_any().downcast_ref::<Int32Array>().unwrap();
all_values.extend(array.values().iter().copied());
}
all_values.sort();
let expected: Vec<i32> = (0..400).collect();
assert_eq!(all_values, expected);
}
#[tokio::test]
async fn test_optimize_default_action() {
// Verify that default action is All
let action: OptimizeAction = Default::default();
assert!(matches!(action, OptimizeAction::All));
}
#[tokio::test]
async fn test_optimize_stats_default() {
// Verify OptimizeStats default values
let stats: OptimizeStats = Default::default();
assert!(stats.compaction.is_none());
assert!(stats.prune.is_none());
}
#[tokio::test]
async fn test_compact_with_deferred_index_remap() {
// Smoke test: verifies compaction with deferred index remap doesn't error.
// We don't currently assert that remap is actually deferred.
let conn = connect("memory://").execute().await.unwrap();
// Create a table with data
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_iter_values(0..100))],
)
.unwrap();
let table = conn
.create_table(
"test_deferred_remap",
RecordBatchIterator::new(vec![Ok(batch.clone())], schema.clone()),
)
.execute()
.await
.unwrap();
// Add more data
table
.add(RecordBatchIterator::new(vec![Ok(batch)], schema.clone()))
.execute()
.await
.unwrap();
// Create an index
table
.create_index(&["id"], Index::BTree(BTreeIndexBuilder::default()))
.execute()
.await
.unwrap();
// Run compaction with deferred index remap
let stats = table
.optimize(OptimizeAction::Compact {
options: CompactionOptions {
target_rows_per_fragment: 2000,
defer_index_remap: true,
..Default::default()
},
remap_options: None,
})
.await
.unwrap();
assert!(stats.compaction.is_some());
// Verify data integrity after compaction
let final_row_count = table.count_rows(None).await.unwrap();
assert_eq!(final_row_count, 200);
// Verify data content is correct
let batches = table
.query()
.execute()
.await
.unwrap()
.try_collect::<Vec<_>>()
.await
.unwrap();
let mut all_values: Vec<i32> = Vec::new();
for batch in &batches {
let array = batch["id"].as_any().downcast_ref::<Int32Array>().unwrap();
all_values.extend(array.values().iter().copied());
}
all_values.sort();
// Since we added the same data twice (0..100 twice), we expect 200 values
// with values 0-99 appearing twice
let mut expected: Vec<i32> = (0..100).chain(0..100).collect();
expected.sort();
assert_eq!(all_values, expected);
}
#[tokio::test]
async fn test_compaction_preserves_schema() {
let conn = connect("memory://").execute().await.unwrap();
// Create a table with multiple columns
let schema = Arc::new(Schema::new(vec![
Field::new("id", DataType::Int32, false),
Field::new("name", DataType::Utf8, true),
]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![
Arc::new(Int32Array::from_iter_values(0..10)),
Arc::new(StringArray::from(
(0..10).map(|i| format!("name_{}", i)).collect::<Vec<_>>(),
)),
],
)
.unwrap();
let original_schema = batch.schema();
let table = conn
.create_table(
"test_schema_preserved",
RecordBatchIterator::new(vec![Ok(batch.clone())], schema.clone()),
)
.execute()
.await
.unwrap();
// Add more data
table
.add(RecordBatchIterator::new(vec![Ok(batch)], schema.clone()))
.execute()
.await
.unwrap();
// Run compaction
table
.optimize(OptimizeAction::Compact {
options: CompactionOptions::default(),
remap_options: None,
})
.await
.unwrap();
// Verify schema is preserved
let current_schema = table.schema().await.unwrap();
assert_eq!(current_schema, original_schema);
// Verify data is intact and correct
let batches = table
.query()
.execute()
.await
.unwrap()
.try_collect::<Vec<_>>()
.await
.unwrap();
let total_rows: usize = batches.iter().map(|b| b.num_rows()).sum();
assert_eq!(total_rows, 20);
}
#[tokio::test]
async fn test_optimize_empty_table() {
let conn = connect("memory://").execute().await.unwrap();
// Create a table and delete all data
let schema = Arc::new(Schema::new(vec![Field::new("i", DataType::Int32, false)]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from_iter_values(0..10))],
)
.unwrap();
let table = conn
.create_table(
"test_empty_optimize",
RecordBatchIterator::new(vec![Ok(batch)], schema.clone()),
)
.execute()
.await
.unwrap();
// Delete all rows
table.delete("true").await.unwrap();
// Verify table is empty
assert_eq!(table.count_rows(None).await.unwrap(), 0);
// Optimize should work on empty table
let stats = table.optimize(OptimizeAction::All).await.unwrap();
assert!(stats.compaction.is_some());
assert!(stats.prune.is_some());
// Verify table is still empty but schema is preserved
assert_eq!(table.count_rows(None).await.unwrap(), 0);
let current_schema = table.schema().await.unwrap();
assert_eq!(current_schema, schema);
}
#[rstest]
#[case::all(OptimizeAction::All)]
#[case::compact(OptimizeAction::Compact {
options: CompactionOptions::default(),
remap_options: None,
})]
#[case::prune(OptimizeAction::Prune {
older_than: Some(chrono::Duration::try_days(0).unwrap()),
delete_unverified: Some(true),
error_if_tagged_old_versions: None,
})]
#[case::index(OptimizeAction::Index(Default::default()))]
#[tokio::test]
async fn test_optimize_fails_on_checked_out_table(#[case] action: OptimizeAction) {
let conn = connect("memory://").execute().await.unwrap();
let schema = Arc::new(Schema::new(vec![Field::new("i", DataType::Int32, false)]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(Int32Array::from_iter_values(0..10))],
)
.unwrap();
let table = conn
.create_table(
"test_checkout_optimize",
RecordBatchIterator::new(vec![Ok(batch.clone())], schema.clone()),
)
.execute()
.await
.unwrap();
table
.add(RecordBatchIterator::new(vec![Ok(batch)], schema.clone()))
.execute()
.await
.unwrap();
table.checkout(1).await.unwrap();
let result = table.optimize(action).await;
assert!(result.is_err());
let err_msg = result.unwrap_err().to_string();
assert!(
err_msg.contains("cannot be modified when a specific version is checked out"),
"Expected error message about checked out table, got: {}",
err_msg
);
}
}