Chroma

Since Camel 4.17

Only producer is supported

The Chroma Component provides support for interacting with the Chroma Vector Database.

Chroma is an open-source embedding database designed for AI applications. It allows you to store, search, and retrieve embeddings along with their associated metadata and documents.

URI format

chroma:collection[?options]

Where collection represents a named collection in your Chroma database.

Configuring Options

Camel components are configured on two separate levels:

  • component level

  • endpoint level

Configuring Component Options

At the component level, you set general and shared configurations that are, then, inherited by the endpoints. It is the highest configuration level.

For example, a component may have security settings, credentials for authentication, urls for network connection and so forth.

Some components only have a few options, and others may have many. Because components typically have pre-configured defaults that are commonly used, then you may often only need to configure a few options on a component; or none at all.

You can configure components using:

  • the Component DSL.

  • in a configuration file (application.properties, *.yaml files, etc).

  • directly in the Java code.

Configuring Endpoint Options

You usually spend more time setting up endpoints because they have many options. These options help you customize what you want the endpoint to do. The options are also categorized into whether the endpoint is used as a consumer (from), as a producer (to), or both.

Configuring endpoints is most often done directly in the endpoint URI as path and query parameters. You can also use the Endpoint DSL and DataFormat DSL as a type safe way of configuring endpoints and data formats in Java.

A good practice when configuring options is to use Property Placeholders.

Property placeholders provide a few benefits:

  • They help prevent using hardcoded urls, port numbers, sensitive information, and other settings.

  • They allow externalizing the configuration from the code.

  • They help the code to become more flexible and reusable.

The following two sections list all the options, firstly for the component followed by the endpoint.

Component Options

The Chroma component supports 5 options, which are listed below.

Name Description Default Type

configuration (producer)

The configuration;.

ChromaConfiguration

host (producer)

The Chroma server host URL.

String

lazyStartProducer (producer)

Whether the producer should be started lazy (on the first message). By starting lazy you can use this to allow CamelContext and routes to startup in situations where a producer may otherwise fail during starting and cause the route to fail being started. By deferring this startup to be lazy then the startup failure can be handled during routing messages via Camel’s routing error handlers. Beware that when the first message is processed then creating and starting the producer may take a little time and prolong the total processing time of the processing.

false

boolean

maxResults (producer)

Max results for similarity search.

10

int

autowiredEnabled (advanced)

Whether autowiring is enabled. This is used for automatic autowiring options (the option must be marked as autowired) by looking up in the registry to find if there is a single instance of matching type, which then gets configured on the component. This can be used for automatic configuring JDBC data sources, JMS connection factories, AWS Clients, etc.

true

boolean

Endpoint Options

The Chroma endpoint is configured using URI syntax:

chroma:collection

With the following path and query parameters:

Path Parameters (1 parameters)

Name Description Default Type

collection (producer)

Required The collection name.

String

Query Parameters (3 parameters)

Name Description Default Type

host (producer)

The Chroma server host URL.

String

maxResults (producer)

Max results for similarity search.

10

int

lazyStartProducer (producer (advanced))

Whether the producer should be started lazy (on the first message). By starting lazy you can use this to allow CamelContext and routes to startup in situations where a producer may otherwise fail during starting and cause the route to fail being started. By deferring this startup to be lazy then the startup failure can be handled during routing messages via Camel’s routing error handlers. Beware that when the first message is processed then creating and starting the producer may take a little time and prolong the total processing time of the processing.

false

boolean

Message Headers

The Chroma component supports 12 message header(s), which is/are listed below:

Name Description Default Type

CamelChromaAction (producer)

Constant: ACTION

The action to be performed.

Enum values:

  • CREATE_COLLECTION

  • DELETE_COLLECTION

  • GET_COLLECTION

  • ADD

  • QUERY

  • GET

  • UPDATE

  • UPSERT

  • DELETE

String

CamelChromaCollectionName (producer)

Constant: COLLECTION_NAME

The collection name.

String

CamelChromaIds (producer)

Constant: IDS

The embedding IDs.

List

CamelChromaEmbeddings (producer)

Constant: EMBEDDINGS

The embeddings.

List

CamelChromaMetadatas (producer)

Constant: METADATAS

The metadata for embeddings.

List

CamelChromaDocuments (producer)

Constant: DOCUMENTS

The documents for embeddings.

List

CamelChromaQueryEmbeddings (producer)

Constant: QUERY_EMBEDDINGS

The query embeddings for similarity search.

List

CamelChromaNResults (producer)

Constant: N_RESULTS

The number of results to return.

10

Integer

CamelChromaWhere (producer)

Constant: WHERE

Chroma where filter.

Map

CamelChromaWhereDocument (producer)

Constant: WHERE_DOCUMENT

Chroma where document filter.

Map

CamelChromaInclude (producer)

Constant: INCLUDE

The fields to include in the result.

List

CamelChromaOperationStatus (producer)

Constant: OPERATION_STATUS

The operation status.

String

Embedding Function

Chroma requires an EmbeddingFunction to convert documents into vector embeddings. You must configure an embedding function for the component to work properly.

The chromadb-java-client provides several embedding function implementations:

  • DefaultEmbeddingFunction - Uses ONNX runtime for local embeddings

  • OpenAIEmbeddingFunction - Uses OpenAI API

  • CohereEmbeddingFunction - Uses Cohere API

  • OllamaEmbeddingFunction - Uses Ollama for local embeddings

  • HuggingFaceEmbeddingFunction - Uses Hugging Face API

You can also implement your own EmbeddingFunction interface.

Configuring Embedding Function

  • Java

// Using OpenAI embeddings
OpenAIEmbeddingFunction embeddingFunction = new OpenAIEmbeddingFunction(
    System.getenv("OPENAI_API_KEY"),
    "text-embedding-ada-002"
);

ChromaComponent component = context.getComponent("chroma", ChromaComponent.class);
component.getConfiguration().setHost("http://localhost:8000");
component.getConfiguration().setEmbeddingFunction(embeddingFunction);

Examples

Collection Examples

Create Collection

In the route below, we use the chroma component to create a collection named myCollection:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.CREATE_COLLECTION)
    .to("chroma:myCollection");

Get Collection

In the route below, we use the chroma component to get information about a collection named myCollection:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.GET_COLLECTION)
    .to("chroma:myCollection")
    .process(exchange -> {
        Collection collection = exchange.getMessage().getBody(Collection.class);
        log.info("Collection name: {}", collection.getName());
    });

Delete Collection

In the route below, we use the chroma component to delete a collection named myCollection:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.DELETE_COLLECTION)
    .to("chroma:myCollection");

Document Examples

Add Documents

In the route below, we use the chroma component to add documents to a collection. The embedding function will automatically generate embeddings for the documents:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.ADD)
    .setHeader(ChromaHeaders.IDS)
        .constant(Arrays.asList("doc1", "doc2", "doc3"))
    .setBody()
        .constant(Arrays.asList(
            "Camel is an open source integration framework",
            "Chroma is a vector database for AI applications",
            "Apache Camel supports many components"
        ))
    .to("chroma:myCollection");

Add Documents with Metadata

You can also add metadata to documents for filtering during queries:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.ADD)
    .setHeader(ChromaHeaders.IDS)
        .constant(Arrays.asList("doc1", "doc2"))
    .setHeader(ChromaHeaders.METADATAS)
        .constant(Arrays.asList(
            Map.of("source", "wiki", "category", "integration"),
            Map.of("source", "docs", "category", "database")
        ))
    .setBody()
        .constant(Arrays.asList(
            "Camel is an open source integration framework",
            "Chroma is a vector database"
        ))
    .to("chroma:myCollection");

Add Documents with Pre-computed Embeddings

If you have pre-computed embeddings, you can pass them directly via headers:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.ADD)
    .setHeader(ChromaHeaders.IDS)
        .constant(Arrays.asList("doc1"))
    .setHeader(ChromaHeaders.EMBEDDINGS)
        .constant(Arrays.asList(
            Arrays.asList(0.1f, 0.2f, 0.3f, 0.4f)  // Your embedding vector
        ))
    .setBody()
        .constant(Arrays.asList("Document text"))
    .to("chroma:myCollection");

Query Examples

Query Documents

In the route below, we use the chroma component to perform a semantic search query:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.QUERY)
    .setHeader(ChromaHeaders.N_RESULTS)
        .constant(5)
    .setBody()
        .constant(Collections.singletonList("integration framework"))
    .to("chroma:myCollection")
    .process(exchange -> {
        Collection.QueryResponse response = exchange.getMessage()
            .getBody(Collection.QueryResponse.class);
        // Process query results
    });

Query with Metadata Filter

You can filter query results using metadata:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.QUERY)
    .setHeader(ChromaHeaders.N_RESULTS)
        .constant(10)
    .setHeader(ChromaHeaders.WHERE)
        .constant(Map.of("category", "integration"))
    .setBody()
        .constant(Collections.singletonList("apache camel"))
    .to("chroma:myCollection");

Query with Include Options

You can specify which fields to include in the response:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.QUERY)
    .setHeader(ChromaHeaders.N_RESULTS)
        .constant(5)
    .setHeader(ChromaHeaders.INCLUDE)
        .constant(Arrays.asList("documents", "metadatas", "distances"))
    .setBody()
        .constant(Collections.singletonList("search query"))
    .to("chroma:myCollection");

Get Examples

Get Documents by IDs

In the route below, we retrieve specific documents by their IDs:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.GET)
    .setHeader(ChromaHeaders.IDS)
        .constant(Arrays.asList("doc1", "doc2"))
    .to("chroma:myCollection")
    .process(exchange -> {
        Collection.GetResult result = exchange.getMessage()
            .getBody(Collection.GetResult.class);
        List<String> documents = result.getDocuments();
        List<String> ids = result.getIds();
    });

Get Documents with Filter

You can filter documents using metadata:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.GET)
    .setHeader(ChromaHeaders.WHERE)
        .constant(Map.of("source", "wiki"))
    .to("chroma:myCollection");

Update Examples

Update Documents

In the route below, we update existing documents:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.UPDATE)
    .setHeader(ChromaHeaders.IDS)
        .constant(Arrays.asList("doc1"))
    .setHeader(ChromaHeaders.METADATAS)
        .constant(Arrays.asList(
            Map.of("source", "wiki", "updated", "true")
        ))
    .setBody()
        .constant(Arrays.asList("Updated document content"))
    .to("chroma:myCollection");

Upsert Examples

Upsert Documents

Upsert will insert new documents or update existing ones:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.UPSERT)
    .setHeader(ChromaHeaders.IDS)
        .constant(Arrays.asList("doc1", "doc4"))
    .setBody()
        .constant(Arrays.asList(
            "Updated content for doc1",
            "New document doc4"
        ))
    .to("chroma:myCollection");

Delete Examples

Delete Documents by IDs

In the route below, we delete specific documents by their IDs:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.DELETE)
    .setHeader(ChromaHeaders.IDS)
        .constant(Arrays.asList("doc1", "doc2"))
    .to("chroma:myCollection");

Delete Documents with Filter

You can delete documents matching a metadata filter:

  • Java

from("direct:in")
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.DELETE)
    .setHeader(ChromaHeaders.WHERE)
        .constant(Map.of("source", "temporary"))
    .to("chroma:myCollection");

RAG (Retrieval-Augmented Generation) Example

Here’s an example of using Chroma for RAG with a language model:

  • Java

from("direct:ask")
    // First, query Chroma for relevant documents
    .setHeader(ChromaHeaders.ACTION)
        .constant(ChromaAction.QUERY)
    .setHeader(ChromaHeaders.N_RESULTS)
        .constant(3)
    .to("chroma:knowledgeBase")
    // Build context from retrieved documents
    .process(exchange -> {
        Collection.QueryResponse response = exchange.getMessage()
            .getBody(Collection.QueryResponse.class);
        String context = String.join("\n", response.getDocuments().get(0));
        String question = exchange.getProperty("question", String.class);
        String prompt = "Context:\n" + context + "\n\nQuestion: " + question;
        exchange.getMessage().setBody(prompt);
    })
    // Send to LLM for answer generation
    .to("langchain4j-chat:myModel");

Spring Boot Auto-Configuration

When using chroma with Spring Boot make sure to use the following Maven dependency to have support for auto configuration:

<dependency>
  <groupId>org.apache.camel.springboot</groupId>
  <artifactId>camel-chroma-starter</artifactId>
  <version>x.x.x</version>
  <!-- use the same version as your Camel core version -->
</dependency>

The component supports 6 options, which are listed below.

Name Description Default Type

camel.component.chroma.autowired-enabled

Whether autowiring is enabled. This is used for automatic autowiring options (the option must be marked as autowired) by looking up in the registry to find if there is a single instance of matching type, which then gets configured on the component. This can be used for automatic configuring JDBC data sources, JMS connection factories, AWS Clients, etc.

true

Boolean

camel.component.chroma.configuration

The configuration;. The option is a org.apache.camel.component.chroma.ChromaConfiguration type.

ChromaConfiguration

camel.component.chroma.enabled

Whether to enable auto configuration of the chroma component. This is enabled by default.

Boolean

camel.component.chroma.host

The Chroma server host URL.

String

camel.component.chroma.lazy-start-producer

Whether the producer should be started lazy (on the first message). By starting lazy you can use this to allow CamelContext and routes to startup in situations where a producer may otherwise fail during starting and cause the route to fail being started. By deferring this startup to be lazy then the startup failure can be handled during routing messages via Camel’s routing error handlers. Beware that when the first message is processed then creating and starting the producer may take a little time and prolong the total processing time of the processing.

false

Boolean

camel.component.chroma.max-results

Max results for similarity search.

10

Integer
Back to top