JiriOndrusek opened a new issue, #8966:
URL: https://github.com/apache/camel-quarkus/issues/8966

   When Camel routes ingest documents into an EmbeddingStore (via 
langchain4j-embeddingstore component), the retrieved knowledge should be 
available to Quarkus LangChain4j `@RegisterAiService` for RAG-augmented chat. 
This adds auto-production of `RetrievalAugmentor` CDI beans that bridge the 
two, so they share the same EmbeddingStore instance without explicit user 
wiring.
   
   ## Covered scenarios
   
   - Single store + model, no existing RAG: auto-produce a `defaultBean()` 
RetrievalAugmentor backed by the `@Default` store (zero config)
   - Single store + model, existing RAG (Easy RAG or user-provided): 
`defaultBean()` yields, no conflict
   - Multiple stores, no config: auto-produce a default RAG backed by the 
`@Default` store
   - Multiple stores with config: named `RetrievalAugmentor` beans via 
`quarkus.camel.langchain4j.rag.augmentors.<name>.embedding-store-name`
   
   ## New files
   
   - `RagBridgeConfig` — `@ConfigRoot(BUILD_AND_RUN_TIME_FIXED)` for named 
augmentor config
   - `DefaultRetrievalAugmentorSupplier` — lazily resolves EmbeddingStore + 
EmbeddingModel from CDI
   - Build steps in `SupportQuarkusLangchain4jProcessor`
   - Integration test module `langchain4j-rag-bridge-ql4j`
   
   ## Related
   
   - Benefits from (but does not require) 
[CAMEL-24366](https://issues.apache.org/jira/browse/CAMEL-24366) which adds 
auto-embed to the embeddingstore component


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to