GooglePubSub

Google Pub/Sub sink connector

Description

Publishes each incoming SeaTunnel row as one message to a Google Pub/Sub topic.

Support Those Engines

Spark
Flink
SeaTunnel Zeta

Key Features

• batch
• stream
• exactly-once
• cdc
• support multiple table write
• timer flush

Options

name	type	required	default value
project_id	string	yes	-
topic	string	yes	-
credentials_path	string	no	-
emulator_host	string	no	-
format	enum	no	json
field_delimiter	string	no	,
common-options		no	-

project_id [string]

Google Cloud project ID that owns the target topic.

topic [string]

Target Pub/Sub topic ID. The topic must exist before the job starts.

credentials_path [string]

Path to a Google Cloud service account JSON key file. If this option is not set, the connector uses Application Default Credentials.

emulator_host [string]

Pub/Sub emulator host and port, for example pubsub-emulator:8085. When set, the connector uses a plaintext connection without credentials. Do not use this option for a production Pub/Sub endpoint.

format [enum]

Message payload format. Supported values:

• json: writes the row as a JSON object.
• text: joins row fields with field_delimiter.

field_delimiter [string]

Field delimiter used when format = text. The default is ,.

common options

Sink plugin common parameters, please refer to Sink Common Options for details.

Delivery Semantics

The connector uses the Google Pub/Sub publisher's batching, flow-control, and retry behavior. It waits for all accepted publish operations during checkpoints and shutdown and fails the task when an asynchronous publish fails.

Pub/Sub publishing is at-least-once from the SeaTunnel job's perspective. A task retry can publish a message again, so downstream consumers should tolerate duplicates when required by the use case.

This first sink implementation publishes only the serialized row payload. Message attributes, ordering keys, and per-row topic routing are not supported.

Task Example

Application Default Credentials

env {
  parallelism = 1
  job.mode = "BATCH"
}

source {
  FakeSource {
    row.num = 10
    schema = {
      fields {
        event_id = string
        event_type = string
      }
    }
  }
}

sink {
  GooglePubSub {
    project_id = "my-gcp-project"
    topic = "events"
    format = json
  }
}

Service Account Key File

sink {
  GooglePubSub {
    project_id = "my-gcp-project"
    topic = "events"
    credentials_path = "/secrets/service-account.json"
    format = text
    field_delimiter = "|"
  }
}

Pub/Sub Emulator

sink {
  GooglePubSub {
    project_id = "local-project"
    topic = "events"
    emulator_host = "pubsub-emulator:8085"
  }
}

Changelog

Change Log

Change	Commit	Version
[Feature][Connector-V2] Add Google Pub/Sub sink connector	-	Next