Коммиты

This reverts commit d030b16ecbec032fc39e6a1878374387f7344f80.

This reverts commit 3af1fe5ea22eefa1884c0fdd7e49d753bc96396a.

This is very similar to what is done for OpenAI and it will directly enable [this](https://github.com/quarkiverse/quarkus-langchain4j/issues/371) cc @langchain4j <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Introduced a new client for interacting with Mistral AI services, enabling chat completions, streaming chat completions, embeddings, and model listings. - **Refactor** - Updated `MistralAiClient` to an abstract class with enhanced functionality and a new builder method for configuration. - **Chores** - Changed access modifiers of several classes and methods from package-private to public, improving their usability across the codebase. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

This will give visibility to the LangChain4J project to Azure OpenAI and OpenAI. _WARNING_ I had to update OpenAI4j version to 0.15 and I don't know about the other impacts of this upgrade. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Enhanced API interactions by setting a default user agent across various models. This improves compatibility and monitoring with external services. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Introduced classes and interfaces to facilitate chat interactions with the Anthropic API, enabling chat completion functionalities. - Developed a client class for seamless interaction with the AnthropicAI API, including authentication and request handling. - Implemented utility methods for message conversion and managing token usage in chat interactions. - Defined an enum to distinguish between user and assistant roles in chat scenarios. - Added logging interceptors for HTTP requests and responses to enhance debugging capabilities. - Created a model class for generating AI responses from chat messages using the Anthropic API. - Added a request model class for creating messages in the Anthropic system. - Introduced a class for representing image content with type and source details. - Included integration tests for the `AnthropicChatModel` class covering various functionalities. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

- Add Gemini function calling support, in both streaming and non-streaming models - Upgrade the Gemini Java SDK <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Introduced functionality for converting tool execution requests into function calls and vice versa, enhancing interaction capabilities. - Expanded message handling in chat models to support different message types and tool execution results. - Added support for generating chat messages based on tool specifications, allowing for more dynamic conversations. - **Enhancements** - Improved type conversion and message construction processes for better efficiency and reliability. - **Tests** - Added comprehensive test coverage for new functionalities, ensuring robustness and reliability of conversions and message handling. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

It's possible to customize different kinds of tasks for embedding: - `RETRIEVAL_QUERY` - `RETRIEVAL_DOCUMENT` - `SEMANTIC_SIMILARITY` - `CLASSIFICATION, CLUSTERING` <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Introduced `TaskType` enum for specifying task types in embedding instances, enhancing customization for retrieval, similarity, classification, and clustering tasks. - Added `taskType` parameter to the constructor for setting task type in embedding instances. - **Tests** - Added tests for semantic similarity, text classification, and document retrieval embeddings using different task types. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

Fixes the problem described in https://github.com/langchain4j/langchain4j/pull/710#issuecomment-1985891002 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Refactor** - Improved consistency in setting retriever-related configurations in AI services. - Updated test methods in `AiServicesBuilderTest.java` to use `Mockito.when` for throwing exceptions. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

It is now possible for Qwen models to return messages with content when finishReason == STOP (but not always). compatible with this situation. When finishReason != STOP, if the returned message has no content, NullPointerException will still be thrown. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Enhanced conversation handling in the application with new methods for checking and extracting answers. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

ZhipuAI is a large model focusing on Chinese cognition <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Introduced serialization and deserialization for assistant messages. - Enhanced utility methods for data conversion in Zhipu AI processing. - Implemented JSON serialization/deserialization support. - Defined an interface for Zhipu AI service interactions. - Introduced classes for handling chat completions and embedding requests with Zhipu AI. - Provided structure for chat messages, choices, models, requests, and responses. - Added classes for function calls, parameters, tool interactions, and web searches within chats. - Established data structures for embedding information and requests. - Implemented builders for chat and embedding model instances. - **Tests** - Added integration tests for chat model, embedding model, and streaming chat model functionalities. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Documentation** - Updated the AI services tutorial to reflect changes in message annotation semantics within the chat system. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

## New EmbeddingStore (metadata) `Filter` API Many embedding stores, such as [Pinecone](https://docs.pinecone.io/docs/metadata-filtering) and [Milvus](https://milvus.io/docs/boolean.md) support strict filtering (think of an SQL "WHERE" clause) during similarity search. So, if one has an embedding store with movies, for example, one could search not only for the most semantically similar movies to the given user query but also apply strict filtering by metadata fields like year, genre, rating, etc. In this case, the similarity search will be performed only on those movies that match the filter expression. Since LangChain4j supports (and abstracts away) many embedding stores, there needs to be an embedding-store-agnostic way for users to define the filter expression. This PR introduces a `Filter` interface, which can represent both simple (e.g., `type = "documentation"`) and composite (e.g., `type in ("documentation", "tutorial") AND year > 2020`) filter expressions in an embedding-store-agnostic manner. `Filter` currently supports the following operations: - Comparison: - `IsEqualTo` - `IsNotEqualTo` - `IsGreaterThan` - `IsGreaterThanOrEqualTo` - `IsLessThan` - `IsLessThanOrEqualTo` - `IsIn` - `IsNotIn` - Logical: - `And` - `Not` - `Or` These operations are supported by most embedding stores and serve as a good starting point. However, the list of operations will expand over time to include other operations (e.g., `Contains`) supported by embedding stores. Currently, the DSL looks like this: ```java Filter onlyDocs = metadataKey("type").isEqualTo("documentation"); Filter docsAndTutorialsAfter2020 = metadataKey("type").isIn("documentation", "tutorial").and(metadataKey("year").isGreaterThan(2020)); // or Filter docsAndTutorialsAfter2020 = and( metadataKey("type").isIn("documentation", "tutorial"), metadataKey("year").isGreaterThan(2020) ); ``` ## Filter expression as a `String` Filter expression can also be specified as a `String`. This might be necessary, for example, if the filter expression is generated dynamically by the application or by the LLM (as in [self querying](https://python.langchain.com/docs/modules/data_connection/retrievers/self_query/)). This PR introduces a `FilterParser` interface with a simple `Filter parse(String)` API, allowing for future support of multiple syntaxes (if this will be required). For the out-of-the-box filter syntax, ANSI SQL's `WHERE` clause is proposed as a suitable candidate for several reasons: - SQL is well-known among Java developers - There is extensive tooling available for SQL (e.g., parsers) - LLMs are pretty good at generating valid SQL, as there are tons of SQL queries on the internet, which are included in the LLM training datasets. There are also specialized LLMs that are trained for text-to-SQL task, such as [SQLCoder](https://huggingface.co/defog). The downside is that SQL's `WHERE` clause might not support all operations and data types that could be supported in the future by various embedding stores. In such case, we could extend it to a superset of ANSI SQL `WHERE` syntax and/or provide an option to express filters in the native syntax of the store. An out-of-the-box implementation of the SQL `FilterParser` is provided as a `SqlFilterParser` in a separate module `langchain4j-embedding-store-filter-parser-sql`, using [JSqlParser](https://github.com/JSQLParser/JSqlParser) under the hood. `SqlFilterParser` can parse SQL "SELECT" (or just "WHERE" clause) statement into a `Filter` object: - `SELECT * FROM fake_table WHERE userId = '123-456'` -> `metadataKey("userId").isEqualTo("123-456")` - `userId = '123-456'` -> `metadataKey("userId").isEqualTo("123-456")` It can also resolve `CURDATE()` and `CURRENT_DATE`/`CURRENT_TIME`/`CURRENT_TIMESTAMP`: `SELECT * FROM fake_table WHERE year = EXTRACT(YEAR FROM CURRENT_DATE` -> `metadataKey("year").isEqualTo(LocalDate.now().getYear())` ## Changes in `Metadata` API Until now, `Metadata` supported only `String` values. This PR expands the list of supported value types to `Integer`, `Long`, `Float` and `Double`. In the future, more types may be added (if needed). The method `String get(String key)` will be deprecated later in favor of: - `String getString(String key)` - `Integer getInteger(String key)` - `Long getLong(String key)` - etc New overloaded `put(key, value)` methods are introduced to support more value types: - `put(String key, int value)` - `put(String key, long value)` - etc ## Changes in `EmbeddingStore` API New method `search` is added that will become the main entry point for search in the future. All `findRelevant` methods will be deprecated later. New `search` method accepts `EmbeddingSearchRequest` and returns `EmbeddingSearchResult`. `EmbeddingSearchRequest` contains all search criteria (e.g. `maxResults`, `minScore`), including new `Filter`. `EmbeddingSearchResult` contains a list of `EmbeddingMatch`. ```java EmbeddingSearchResult search(EmbeddingSearchRequest request); ``` ## Changes in `EmbeddingStoreContentRetriever` API `EmbeddingStoreContentRetriever` can now be configured with a static `filter` as well as dynamic `dynamicMaxResults`, `dynamicMinScore` and `dynamicFilter` in the builder: ```java ContentRetriever contentRetriever = EmbeddingStoreContentRetriever.builder() .embeddingStore(embeddingStore) .embeddingModel(embeddingModel) ... .maxResults(3) // or .dynamicMaxResults(query -> 3) // You can define maxResults dynamically. The value could, for example, depend on the query or the user associated with the query. ... .minScore(0.3) // or .dynamicMinScore(query -> 0.3) ... .filter(metadataKey("userId").isEqualTo("123-456")) // Assuming your TextSegments contain Metadata with key "userId" // or .dynamicFilter(query -> metadataKey("userId").isEqualTo(query.metadata().chatMemoryId().toString())) ... .build(); ``` So now you can define `maxResults`, `minScore` and `filter` both statically and dynamically (they can depend on the query, user, etc.). These values will be propagated to the underlying `EmbeddingStore`. ## ["Self-querying"](https://python.langchain.com/docs/modules/data_connection/retrievers/self_query/) This PR also introduces `LanguageModelSqlFilterBuilder` in `langchain4j-embedding-store-filter-parser-sql` module which can be used with `EmbeddingStoreContentRetriever`'s `dynamicFilter` to automatically build a `Filter` object from the `Query` using language model and `SqlFilterParser`. For example: ```java TextSegment groundhogDay = TextSegment.from("Groundhog Day", new Metadata().put("genre", "comedy").put("year", 1993)); TextSegment forrestGump = TextSegment.from("Forrest Gump", new Metadata().put("genre", "drama").put("year", 1994)); TextSegment dieHard = TextSegment.from("Die Hard", new Metadata().put("genre", "action").put("year", 1998)); // describe metadata keys as if they were columns in the SQL table TableDefinition tableDefinition = TableDefinition.builder() .name("movies") .addColumn("genre", "VARCHAR", "one of [comedy, drama, action]") .addColumn("year", "INT") .build(); LanguageModelSqlFilterBuilder sqlFilterBuilder = new LanguageModelSqlFilterBuilder(model, tableDefinition); ContentRetriever contentRetriever = EmbeddingStoreContentRetriever.builder() .embeddingStore(embeddingStore) .embeddingModel(embeddingModel) .dynamicFilter(sqlFilterBuilder::build) .build(); String answer = assistant.answer("Recommend me a good drama from 90s"); // Forrest Gump ``` ## Which embedding store integrations will support `Filter`? In the long run, all (provided the embedding store itself supports it). In the first iteration, I aim to add support to just a few: - `InMemoryEmbeddingStore` - Elasticsearch - Milvus <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit ## Summary by CodeRabbit - **New Features** - Introduced filters for checking key's value existence in a collection for improved data handling. - **Enhancements** - Updated `InMemoryEmbeddingStoreTest` to extend a different class for improved testing coverage and added a new test method. - **Refactor** - Made minor formatting adjustments in the assertion block for better readability. - **Documentation** - Updated class hierarchy information for clarity. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Enhanced `FileSystemDocumentLoader` with capabilities for recursive loading and filtering documents in directories. - **Tests** - Improved test suite for `FileSystemDocumentLoader` including parameterized tests and updated logic to match new functionalities. - **Documentation** - Added a new document type to test resources, expanding the variety of document types handled by the system. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Enhanced message eviction logic in chat memory to maintain capacity by intelligently handling different message types, including AI messages and system messages. - **Refactor** - Improved handling of token estimations and message evictions in chat memory tests for more accurate testing scenarios. - **Tests** - Added comprehensive tests for new eviction logic, covering various scenarios including the presence of AI messages with tool execution requests and system messages. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

HuggingFace -> Hugging Face <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Documentation** - Updated brand name "HuggingFace" to "Hugging Face" for consistency across the documentation. <!-- end of auto-generated comment: release notes by coderabbit.ai -->

Provide instructions on creating Google Cloud Platform account and establishing a new project with use of Vertex AI API. Add basic code samples for using google's PaLM2, Gemini and Embedding models. List available model names. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Documentation** - Updated the guide on Google Vertex AI Embedding Models with dependencies, example code, and available models. - Added a link to an article about image generation with Imagen. - Expanded the documentation for Google Gemini with setup instructions, authentication strategies, example uses, and early access information. - Detailed instructions for getting started with Google Vertex AI PaLM 2, including dependencies, example code, and model information. <!-- end of auto-generated comment: release notes by coderabbit.ai --> Co-authored-by: Lize Raes <49833622+LizeRaes@users.noreply.github.com>

…entor are set As we discussed in https://github.com/quarkiverse/quarkus-langchain4j/pull/353#discussion_r1516119442 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Enhanced AI services with new validation rules to ensure exclusive setting of components, improving system stability and predictability. <!-- end of auto-generated comment: release notes by coderabbit.ai -->