Bean Providers¶

Forage automatically creates and registers beans in the Camel registry based on your configuration. You never write Java code to instantiate a datasource, connection factory, or AI model — you just declare what you need in a properties file.

Named Beans¶

Every Forage bean has a name. The name you choose in the properties file becomes the bean name in the Camel registry:

# "ordersDb" is the bean name
forage.ordersDb.jdbc.db.kind=postgresql
forage.ordersDb.jdbc.url=jdbc:postgresql://localhost:5432/orders
forage.ordersDb.jdbc.username=admin
forage.ordersDb.jdbc.password=secret

Reference it in your routes with #ordersDb:

- to:
    uri: sql
    parameters:
      query: select * from orders
      dataSource: "#ordersDb"

You can create as many beans as you need — just use different names:

# Two databases, two beans
forage.ordersDb.jdbc.db.kind=postgresql
forage.ordersDb.jdbc.url=jdbc:postgresql://localhost:5432/orders

forage.inventoryDb.jdbc.db.kind=mysql
forage.inventoryDb.jdbc.url=jdbc:mysql://localhost:3306/inventory

This pattern is the same for all Forage modules — JDBC, JMS, AI agents, vector databases, and more.

How Discovery Works¶

When Forage starts, it:

1. Scans properties for Forage configuration entries (e.g., forage.*.jdbc.*, forage.*.agent.*)
2. Identifies the provider based on the configuration (e.g., db.kind=postgresql selects the PostgreSQL provider)
3. Creates the bean using the provider — a pooled datasource, a chat model, a connection factory, etc.
4. Registers the bean in the Camel registry under the name you chose

This discovery uses Java's ServiceLoader mechanism. When you add a Forage module to your classpath (e.g., forage-jdbc-postgresql), its provider is automatically available — no manual wiring required.

Provider Types¶

Forage includes providers for several categories of beans:

Each category has multiple implementations. For example, JDBC supports PostgreSQL, MySQL, H2, and more. The kind or model.kind property selects the implementation:

# Selects the PostgreSQL provider
forage.myDb.jdbc.db.kind=postgresql

# Selects the Ollama model provider
forage.myAgent.agent.model.kind=ollama

Conditional Beans¶

Some providers create additional beans based on your configuration. For example, when you enable transactions on a JDBC datasource:

forage.myDb.jdbc.db.kind=postgresql
forage.myDb.jdbc.url=jdbc:postgresql://localhost:5432/mydb
forage.myDb.jdbc.transaction.enabled=true

Forage automatically registers transaction policy beans (PROPAGATION_REQUIRED, MANDATORY, etc.) that you can use in transacted routes:

- route:
    from:
      uri: jms:queue:orders
      steps:
        - policy:
            ref: PROPAGATION_REQUIRED
        - to:
            uri: sql
            parameters:
              query: "INSERT INTO orders ..."
              dataSource: "#myDb"

Similarly, you can declare aggregation repositories and idempotent repositories directly in your datasource configuration:

forage.myDb.jdbc.aggregation.repository.name=myAggRepo
forage.myDb.jdbc.idempotent.repository.name=myIdempotentRepo

These are registered as separate beans (#myAggRepo, #myIdempotentRepo) backed by the same datasource.