Camel Components

Pinecone

Since Camel 4.6

Only producer is supported

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

URI format

pinecone:collection[?options]

Where collection represents a named set of points (vectors with a payload) defined in your 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 Pinecone component supports the following options which are listed below.

Name Description Default Type

action (producer)

Action to perform.

Enum values:

  • CREATE_COLLECTION

  • CREATE_SERVERLESS_INDEX

  • CREATE_POD_INDEX

  • FETCH

  • UPSERT

  • DELETE_INDEX

  • DELETE_COLLECTION

  • QUERY

  • QUERY_BY_ID

  • UPDATE

  • DELETE_BY_ID

PineconeVectorDbAction

cloud (producer)

Sets the cloud type to use (aws/gcp/azure).

Enum values:

  • aws

  • azure

  • gcp

String

cloudRegion (producer)

Sets the cloud region.

String

collectionDimension (producer)

Sets the Collection Dimension to use (1-1536).

1536

Integer

collectionSimilarityMetric (producer)

Sets the Collection Similarity Metric to use (cosine/euclidean/dotproduct).

String

configuration (producer)

The configuration;.

PineconeVectorDbConfiguration

host (producer)

Sets a custom host URL to connect to.

String

indexName (producer)

Sets the index name to use.

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

proxyHost (producer)

Set the proxy host.

String

proxyPort (producer)

Set the proxy port.

Integer

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

tls (security)

Whether the client uses Transport Layer Security (TLS) to secure communications.

false

boolean

token (security)

Sets the API key to use for authentication.

String

Endpoint Options

The Pinecone endpoint is configured using URI syntax:

pinecone:collection

With the following path and query parameters:

Path Parameters

Name Description Default Type

collection (producer)

Required The collection Name. (Only used by some actions).

String

Query Parameters

Name Description Default Type

action (producer)

Action to perform.

Enum values:

  • CREATE_COLLECTION

  • CREATE_SERVERLESS_INDEX

  • CREATE_POD_INDEX

  • FETCH

  • UPSERT

  • DELETE_INDEX

  • DELETE_COLLECTION

  • QUERY

  • QUERY_BY_ID

  • UPDATE

  • DELETE_BY_ID

PineconeVectorDbAction

cloud (producer)

Sets the cloud type to use (aws/gcp/azure).

Enum values:

  • aws

  • azure

  • gcp

String

cloudRegion (producer)

Sets the cloud region.

String

collectionDimension (producer)

Sets the Collection Dimension to use (1-1536).

1536

Integer

collectionSimilarityMetric (producer)

Sets the Collection Similarity Metric to use (cosine/euclidean/dotproduct).

String

host (producer)

Sets a custom host URL to connect to.

String

indexName (producer)

Sets the index name to use.

String

proxyHost (producer)

Set the proxy host.

String

proxyPort (producer)

Set the proxy port.

Integer

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

tls (security)

Whether the client uses Transport Layer Security (TLS) to secure communications.

false

boolean

token (security)

Sets the API key to use for authentication.

String

Message Headers

The Pinecone component supports the following message header(s), which is/are listed below:

Name Description Default Type

CamelPineconeAction (producer)

Constant: ACTION

The action to be performed.

Enum values:

  • CREATE_COLLECTION

  • CREATE_SERVERLESS_INDEX

  • CREATE_POD_INDEX

  • FETCH

  • UPSERT

  • DELETE_INDEX

  • DELETE_COLLECTION

  • QUERY

  • QUERY_BY_ID

  • UPDATE

  • DELETE_BY_ID

String

CamelPineconeTextFieldName (producer)

Constant: TEXT_FIELD_NAME

Text Field Name for Insert/Upsert operation.

String

CamelPineconeVectorFieldName (producer)

Constant: VECTOR_FIELD_NAME

Vector Field Name for Insert/Upsert operation.

String

CamelPineconeIndexName (producer)

Constant: INDEX_NAME

Index Name.

String

CamelPineconeIndexPodType (producer)

Constant: INDEX_POD_TYPE

Index Pod Type.

String

CamelPineconeIndexPodEnvironment (producer)

Constant: INDEX_POD_ENVIRONMENT

Index Pod Environment.

String

CamelPineconeCollectionName (producer)

Constant: COLLECTION_NAME

Collection Name for Insert/Upsert operation.

String

CamelPineconeCollectionSimilarityMetric (producer)

Constant: COLLECTION_SIMILARITY_METRIC

Collection Similarity Metric.

Enum values:

  • cosine

  • euclidean

  • dotproduct

String

CamelPineconeCollectionDimension (producer)

Constant: COLLECTION_DIMENSION

Collection Dimension.

int

CamelPineconeCollectionCloud (producer)

Constant: COLLECTION_CLOUD

Collection Cloud Vendor.

Enum values:

  • aws

  • gcp

  • azure

String

CamelPineconeCollectionCloudRegion (producer)

Constant: COLLECTION_CLOUD_REGION

Collection Cloud Vendor Region.

Enum values:

  • aws

  • gcp

  • azure

String

CamelPineconeIndexId (producer)

Constant: INDEX_ID

The id of the vector for UPSERT, UPDATE and QUERY_BY_ID. Without it, UPSERT generates a random UUID and sets this header to it.

String

CamelPineconeQueryTopK (producer)

Constant: QUERY_TOP_K

The number of matches QUERY and QUERY_BY_ID return.

10

Integer

CamelPineconeNamespace (producer)

Constant: NAMESPACE

Namespace for actions (query/upsert/etc).

String

CamelPineconeQueryFilter (producer)

Constant: QUERY_FILTER

Query Filter.

Struct

CamelPineconeQueryIncludeValues (producer)

Constant: QUERY_INCLUDE_VALUES

Query Include Values.

boolean

CamelPineconeQueryIncludeMetadata (producer)

Constant: QUERY_INCLUDE_METADATA

Query Include Metadata.

boolean

CamelPineconeMetadata (producer)

Constant: METADATA

The metadata UPSERT stores with the vector, and the fields of the metadata UPDATE sets. The pinecone:embeddings data type transformer puts the text of the embedding in its text_segment field, which the pinecone:rag data type transformer reads.

Struct

Message Body

The message body at each side of the Pinecone component: Consumer is the body of the messages the consumer creates, Consumer reply the bodies a consumer that answers its caller can reply with, Producer input the bodies the producer can send or write, and Producer output the body after the producer.

Side Body Description

Producer input

list or null

By the action in the action option or the CamelPineconeAction header: the vector as a List of numbers, such as the List of Double a JSON array unmarshals to, or a float or double array for UPSERT, UPDATE and QUERY, with the id in the CamelPineconeIndexId header (UPSERT generates one without it), and the ids as a List of String for FETCH and DELETE_BY_ID. The other actions do not read the body. Another body fails: the pinecone:embeddings data type transformer builds the vector from the output of langchain4j-embeddings.

  • With action=UPSERT: list

  • With action=UPDATE: list

  • With action=QUERY: list

  • With action=FETCH: list

  • With action=DELETE_BY_ID: list

  • With action=QUERY_BY_ID: any

  • With action=CREATE_COLLECTION: any

  • With action=CREATE_SERVERLESS_INDEX: any

  • With action=CREATE_POD_INDEX: any

  • With action=DELETE_INDEX: any

  • With action=DELETE_COLLECTION: any

Producer output

pojo or unchanged

By the action: the response of the Pinecone SDK, such as an UpsertResponse, a FetchResponse or a QueryResponseWithUnsignedIndices with the matches. DELETE_INDEX and DELETE_COLLECTION keep the body.

  • With action=UPSERT: pojo (io.pinecone.proto.UpsertResponse)

  • With action=UPDATE: pojo (io.pinecone.proto.UpdateResponse)

  • With action=QUERY: pojo (io.pinecone.unsigned_indices_model.QueryResponseWithUnsignedIndices)

  • With action=QUERY_BY_ID: pojo (io.pinecone.unsigned_indices_model.QueryResponseWithUnsignedIndices)

  • With action=FETCH: pojo (io.pinecone.proto.FetchResponse)

  • With action=DELETE_BY_ID: pojo (io.pinecone.proto.DeleteResponse)

  • With action=CREATE_COLLECTION: pojo (org.openapitools.db_control.client.model.CollectionModel)

  • With action=CREATE_SERVERLESS_INDEX: pojo (org.openapitools.db_control.client.model.IndexModel)

  • With action=CREATE_POD_INDEX: pojo (org.openapitools.db_control.client.model.IndexModel)

  • With action=DELETE_INDEX: unchanged

  • With action=DELETE_COLLECTION: unchanged

The bodies are the ones with the default options. A side that is not listed is not known, and tooling does not guess it. The carriers, the sides and the tools that use them are described in Message Body Types.

Usage

The action option, or the CamelPineconeAction header, selects the operation. The first route below creates a serverless index. The second route queries the index with a vector, and the CamelPineconeQueryTopK header sets how many matches it returns (10 without it):

- route:
    from:
      uri: direct:createIndex
      steps:
        - to:
            uri: pinecone:my-collection
            parameters:
              action: CREATE_SERVERLESS_INDEX
              token: "{{pinecone.token}}"
              indexName: my-index
              collectionSimilarityMetric: cosine
              collectionDimension: 3
              cloud: aws
              cloudRegion: us-east-1
- route:
    from:
      uri: direct:query
      steps:
        - setHeader:
            name: CamelPineconeQueryTopK
            expression:
              constant:
                expression: "3"
        - setBody:
            expression:
              groovy:
                expression: "[0.1f, 0.2f, 0.3f]"
        - to:
            uri: pinecone:my-collection
            parameters:
              action: QUERY
              token: "{{pinecone.token}}"
              indexName: my-index
        - log:
            message: "${body}"

The vector of UPSERT, UPDATE and QUERY is the message body: a List of numbers, such as the List of Double a JSON array unmarshals to, or a float[] or a double[]; each number is converted to a float. Another body fails, and so does a List with an element that is not a number. UPSERT stores the vector under the id in the CamelPineconeIndexId header; without the header, it generates a random UUID and sets the header to it.

UPSERT stores the CamelPineconeMetadata header, a com.google.protobuf.Struct, as the metadata of the vector, and UPDATE sets the fields of that header in the metadata of the vector.

The pinecone:embeddings data type transformer turns the embedding of the CamelLangChain4jEmbeddingsVector header, which the langchain4j-embeddings component sets, into the vector of the message body, and fails when the header is missing. It puts the text of a LangChain4j TextSegment or String body in the text_segment field of the CamelPineconeMetadata header, and keeps the other fields of that header. Without the CamelPineconeIndexName and CamelPineconeIndexId headers, it sets them to embeddings and to a random UUID.

The pinecone:rag data type transformer turns the result of a QUERY or QUERY_BY_ID into the List of the texts of its matches, read from their text_segment metadata, which the query returns when the CamelPineconeQueryIncludeMetadata header is true.