CAMEL-21540: Add PGVector component for PostgreSQL vector database

[gnodet]

Summary

Implements CAMEL-21540: Vector Database capabilities for PostgreSQL.

• New camel-pgvector component under components/camel-ai/ that provides vector similarity search capabilities using the PostgreSQL pgvector extension
• Supports actions: CREATE_TABLE, DROP_TABLE, UPSERT, DELETE, SIMILARITY_SEARCH
• Uses JDBC with the com.pgvector:pgvector library for pgvector type support
• Configurable distance types: cosine (default), euclidean, inner product
• LangChain4j data type transformers (pgvector:embeddings and pgvector:rag) for RAG pipeline integration
• Integration tests using testcontainers pgvector image
• LangChain4j embeddings integration test with AllMiniLmL6V2 embedding model

Test plan

• PgVectorComponentIT (7 tests) - CRUD operations and similarity search
• LangChain4jEmbeddingsComponentPgVectorTargetIT (4 tests) - end-to-end LangChain4j embedding + pgvector integration
• Code formatted and imports sorted
• Generated files committed

[github-actions]

🌟 Thank you for your contribution to the Apache Camel project! 🌟
🤖 CI automation will test this PR automatically.

🐫 Apache Camel Committers, please review the following items:

• First-time contributors require MANUAL approval for the GitHub Actions to run
• You can use the command /component-test (camel-)component-name1 (camel-)component-name2.. to request a test from the test bot although they are normally detected and executed by CI.
• You can label PRs using build-all, build-dependents, skip-tests and test-dependents to fine-tune the checks executed by this PR.
• Build and test logs are available in the summary page. Only Apache Camel committers have access to the summary.

⚠️ Be careful when sharing logs. Review their contents before sharing them publicly.

[github-actions]

🧪 CI tested the following changed modules:

• bom/camel-bom
• catalog/camel-allcomponents
• catalog/camel-catalog
• components/camel-ai
• components/camel-ai/camel-langchain4j-embeddings
• components/camel-ai/camel-pgvector
• core/camel-main
• docs
• dsl/camel-componentdsl
• dsl/camel-endpointdsl
• dsl/camel-kamelet-main
• parent
• tooling/maven/camel-package-maven-plugin

ℹ️ Dependent modules were not tested because the total number of affected modules exceeded the threshold (50). Use the test-dependents label to force testing all dependents.

Build reactor — dependencies compiled but only changed modules were tested (26 modules)

• Camel :: AI :: LangChain4j :: Embedding
• Camel :: AI :: LangChain4j :: Embedding [jar]
• Camel :: AI :: PGVector
• Camel :: AI :: PGVector [jar]
• Camel :: AI :: Parent
• Camel :: AI :: Parent [pom]
• Camel :: All Components Sync point
• Camel :: All Components Sync point [pom]
• Camel :: BOM
• Camel :: BOM [pom]
• Camel :: Catalog :: Camel Catalog
• Camel :: Catalog :: Camel Catalog [jar]
• Camel :: Component DSL
• Camel :: Component DSL [jar]
• Camel :: Docs
• Camel :: Docs [pom]
• Camel :: Endpoint DSL
• Camel :: Endpoint DSL [jar]
• Camel :: Kamelet Main
• Camel :: Kamelet Main [jar]
• Camel :: Main
• Camel :: Main [jar]
• Camel :: Maven Plugins :: Camel Maven Package
• Camel :: Maven Plugins :: Camel Maven Package [maven-plugin]
• Camel :: Parent
• Camel :: Parent [pom]

[apupier]

What is the reason for a different postgres component specific for the vectors? Why not including these capabilities directly in the existing Postgres component?

[gnodet]

Claude Code on behalf of Guillaume Nodet

Good question! There are several reasons for keeping pgvector as a separate component under camel-ai/ rather than adding it to the existing PostgreSQL components:

1. Consistency with other vector DB components. All vector database components live under camel-ai/: Milvus, Qdrant, Pinecone, Weaviate, Neo4j. pgvector serves the same purpose — vector similarity search for AI/RAG pipelines — it just happens to use PostgreSQL as its storage engine. Users looking for vector database components would naturally look in the AI group.

2. Different use case and audience. The existing PG components (camel-pgevent for LISTEN/NOTIFY, camel-pg-replication-slot for WAL replication) are PostgreSQL-specific integration utilities. pgvector targets a completely different use case: embeddings storage, similarity search, and RAG pipelines. These serve different personas.

3. Additional dependencies. pgvector brings in com.pgvector:pgvector (for vector type support) and LangChain4j data type transformers — AI-oriented dependencies that would be out of place in a general PostgreSQL component.

4. Separate lifecycle. Keeping it as its own component allows it to evolve independently without affecting the existing PostgreSQL components.

This is the same approach taken by other frameworks — for example, LangChain and LlamaIndex treat pgvector as a vector store alongside Pinecone/Qdrant/etc., not as a PostgreSQL utility.