## Issue
<!-- Please specify the ID of the issue this PR is addressing. -->
N/A - routine dependency maintenance prompted by the new Azure Java SDK
release.
## Change
Bumps `com.azure:azure-sdk-bom` from `1.3.5` to `1.3.7` (latest). Most
managed libraries are minor/patch bumps, but the new BOM pulls two
changes that require work:
- **`azure-search-documents` 11.8.1 -> 12.0.0** (major release) used by
`langchain4j-azure-ai-search`.
- **`azure-core` 1.57.1 -> 1.58.0**, which shifts the required
Netty/Reactor stack.
**azure-search-documents 12.0.0 migration**
(`langchain4j-azure-ai-search`). The public langchain4j API of the
module is unchanged; only the internal calls into the Azure SDK were
rewritten:
- `search(text, options, Context)` ->
`search(options.setSearchText(text))`
- `SearchDocument` / `getDocument(...)` removed -> read fields via
`SearchResult.getAdditionalProperties()` (a `Map`)
- `uploadDocuments` / `deleteDocuments` removed ->
`indexDocuments(IndexDocumentsBatch)` built from `IndexAction`s (new
private `toUploadBatch` / `toSearchDocument` helpers)
- `VectorSearchOptions` / `SemanticSearchOptions` removed ->
`setVectorQueries(...)` and `setSemanticConfigurationName(...)`
- `getSemanticSearch().getRerankerScore()` -> `getRerankerScore()`;
`setKNearestNeighborsCount` -> `setKNearestNeighbors`; `new
SearchIndex(name, fields)` constructor
**Dependency re-alignment.** The bump to `azure-core 1.58.0` raises the
minimum Netty/Reactor versions, so the local `dependencyManagement`
workaround pins in `github-models`, `azure-cosmos-nosql`,
`document-loader-azure-storage-blob`, and
`code-execution-engine-azure-acads` were updated (`reactor-core 3.7.17`,
`reactor-netty-http 1.2.16`, `azure-core-http-netty 1.16.4` and
siblings, `netty-bom 4.1.132.Final`) so the `maven-enforcer`
`RequireUpperBoundDeps` rule keeps passing.
Note for reviewers: this is a major version jump of
`azure-search-documents`, so please give the rewritten index/read/write
paths in `AbstractAzureAiSearchEmbeddingStore` a careful look. End-user
behaviour of the module is intended to be unchanged. Integration tests
require live Azure credentials and were not run; all unit tests pass
locally across every affected Azure module.
## General checklist
- [x] There are no breaking changes (API, behaviour)
- [x] I have added unit and/or integration tests for my change
- [x] The tests cover both positive and negative cases
- [x] I have manually run all the unit and integration tests in the
module I have added/changed, and they are all green
- [x] I have manually run all the unit and integration tests in the
[core](https://github.com/langchain4j/langchain4j/tree/main/langchain4j-core)
and
[main](https://github.com/langchain4j/langchain4j/tree/main/langchain4j)
modules, and they are all green
- [ ] I have added/updated the
[documentation](https://github.com/langchain4j/langchain4j/tree/main/docs/docs)
- [ ] I have added an example in the [examples
repo](https://github.com/langchain4j/langchain4j-examples) (only for
"big" features)
- [ ] I have added/updated [Spring Boot
starter(s)](https://github.com/langchain4j/langchain4j-spring) (if
applicable)
## Checklist for adding new maven module
N/A - no new module.
## Checklist for adding new embedding store integration
N/A - no new embedding store integration.
## Checklist for changing existing embedding store integration
- [ ] I have manually verified that the `AzureAiSearchEmbeddingStore`
works correctly with the data persisted using the latest released
version of LangChain4j
<!-- Integration tests require live Azure resources; the existing IT
compiles but was not executed in this change. -->
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
I want to deprecate the `langchain4j-github-models` module (which I
created), in favor of the `langchain4j-openai-official` module (which I
also created).
Here is why:
- The `langchain4j-github-models` uses the Azure Inference API which
isn't as well maintained as the OpenAi Java SDK, and is limited to Azure
(the OpenAI Java SDK works everywhere, including on competing platforms,
which is better for our end-users)
- If you are using Microsoft Foundry (the new name for Azure OpenAi) or
OpenAi (which is operated by Microsoft), underneath it's the exact same
thing as GitHub Models: we might as well have the same SDK to connect to
all of those.
- I would also like to focus on less modules, this would allow better
quality from my side.
- Update to the Azure SDK 1.3.3, which fixes several transitive
dependencies vulnerabilities (specifically with Azure Identity)
- Update the OpenAI Official SDK to its latest version
- Started to modify the naming from Azure OpenAi to Microsoft Foundry
(the new official name)
- Replaced Dall-e 3.0 (which will be discontinued soon) by
gpt-image-1-mini
## Change
- Added `maven-flatten-plugin` to `langchain4j-parent` and
`langchain4j-bom`
- Removed integration-specific dependencies from `langchain4j-parent`'s
`dependencyManagement` section and moved them to the modules where these
dependencies are used
- Explicitly added missing implicit dependencies
- Removed redundant `<maven.compiler.release>` for cassandra, infinispan
and opensearch modules
- Removed redundant license declarations and outdated properties
## General checklist
- [ ] There are no breaking changes
- [ ] I have added unit and/or integration tests for my change
- [ ] The tests cover both positive and negative cases
- [x] I have manually run all the unit and integration tests in the
module I have added/changed, and they are all green
- [x] I have manually run all the unit and integration tests in the
[core](https://github.com/langchain4j/langchain4j/tree/main/langchain4j-core)
and
[main](https://github.com/langchain4j/langchain4j/tree/main/langchain4j)
modules, and they are all green
- [ ] I have added/updated the
[documentation](https://github.com/langchain4j/langchain4j/tree/main/docs/docs)
- [ ] I have added an example in the [examples
repo](https://github.com/langchain4j/langchain4j-examples) (only for
"big" features)
- [ ] I have added/updated [Spring Boot
starter(s)](https://github.com/langchain4j/langchain4j-spring) (if
applicable)
## Issue
Streamline maven dependencies
Should fix [this
issue](https://github.com/langchain4j/langchain4j/actions/runs/13666001227/job/38207360909?pr=2640#step:5:16630)
```
2025/03/05 00:52:59,050 3922 [INFO ] [main] netty.implementation.NettyUtility - {"az.sdk.message":"The following Netty versions were found on the classpath and have a mismatch with the versions used by azure-core-http-netty. If your application runs without issue this message can be ignored, otherwise please align the Netty versions used in your application. For more information, see https://aka.ms/azsdk/java/dependency/troubleshoot.","azure-netty-version":"4.1.118.Final","azure-netty-native-version":"2.0.70.Final","classpath-netty-version-io.netty:netty-common":"4.1.115.Final","classpath-netty-version-io.netty:netty-handler":"4.1.110.Final","classpath-netty-version-io.netty:netty-handler-proxy":"4.1.110.Final","classpath-netty-version-io.netty:netty-buffer":"4.1.110.Final","classpath-netty-version-io.netty:netty-codec":"4.1.110.Final","classpath-netty-version-io.netty:netty-codec-http":"4.1.110.Final","classpath-netty-version-io.netty:netty-codec-http2":"4.1.110.Final","classpath-netty-version-io.netty:netty-transport-native-unix-common":"4.1.110.Final","classpath-netty-version-io.netty:netty-transport-native-epoll":"4.1.110.Final","classpath-netty-version-io.netty:netty-transport-native-kqueue":"4.1.118.Final","classpath-native-netty-version-io.netty:netty-tcnative-boringssl-static":"2.0.70.Final"}
```
## Change
This pull request includes the following changes to improve project
dependency management:
- Removed unnecessary and redundant Netty dependencies and exclusions.
- Updated and reorganized dependency versions for consistency, including
new BOM imports for Azure SDK, Netty, and Reactor.
- Eliminated AWS SDK v1 dependencies, transitioning to
`software.amazon.awssdk` BOM for better compatibility and reduced
clutter.
- Removed redundant test dependencies like `assertj-core` and `mockito`
across modules, referring to the parent POM.
- Adjusted `kotlinx-coroutines-test` to use JVM-specific artifact and
standardized dependency version variables.
- Refined POM XML formatting for consistency and clarity (`mvn
spotless:apply`)
## General checklist
<!-- Please double-check the following points and mark them like this:
[X] -->
- [ ] There are no breaking changes
- [ ] I have added unit and/or integration tests for my change
- [ ] The tests cover both positive and negative cases
- [ ] I have manually run all the unit and integration tests in the
module I have added/changed, and they are all green
- [ ] I have manually run all the unit and integration tests in the
[core](https://github.com/langchain4j/langchain4j/tree/main/langchain4j-core)
and
[main](https://github.com/langchain4j/langchain4j/tree/main/langchain4j)
modules, and they are all green
<!-- Before adding documentation and example(s) (below), please wait
until the PR is reviewed and approved. -->
- [ ] I have added/updated the
[documentation](https://github.com/langchain4j/langchain4j/tree/main/docs/docs)
- [ ] I have added an example in the [examples
repo](https://github.com/langchain4j/langchain4j-examples) (only for
"big" features)
- [ ] I have added/updated [Spring Boot
starter(s)](https://github.com/langchain4j/langchain4j-spring) (if
applicable)
## Checklist for adding new maven module
<!-- Please double-check the following points and mark them like this:
[X] -->
- [ ] I have added my new module in the root `pom.xml` and
`langchain4j-bom/pom.xml`
## Checklist for adding new embedding store integration
<!-- Please double-check the following points and mark them like this:
[X] -->
- [ ] I have added a `{NameOfIntegration}EmbeddingStoreIT` that extends
from either `EmbeddingStoreIT` or `EmbeddingStoreWithFilteringIT`
- [ ] I have added a `{NameOfIntegration}EmbeddingStoreRemovalIT` that
extends from `EmbeddingStoreWithRemovalIT`
## Checklist for changing existing embedding store integration
<!-- Please double-check the following points and mark them like this:
[X] -->
- [ ] I have manually verified that the
`{NameOfIntegration}EmbeddingStore` works correctly with the data
persisted using the latest released version of LangChain4j
---------
Co-authored-by: kpavlov <{ID}+{username}@users.noreply.github.com>
Co-authored-by: Dmytro Liubarskyi <ljubarskij@gmail.com>
github-actions[bot]