Google Pub/Sub input connector

note

This page describes configuration options specific to the Google Pub/Sub connector. See top-level connector documentation for general information about configuring input and output connectors.

The Google Pub/Sub connector is used to consume a stream of changes to a SQL table from a Pub/Sub subscription.

Configuration options

This connector supports four groups of configuration options:

1. Authentication options control Google Cloud authentication
2. Subscription options specify an existing Pub/Sub subscription from which the connector will pull messages
3. Connectivity options configure gRPC connection to the Pub/Sub service
4. Emulator options are used to connect to a Pub/Sub emulator instead of the real Google Pub/Sub service

The only required option that must be present in any valid Pub/Sub connector configuration is subscription.

Authentication

• credentials - The content of a Google Cloud credentials JSON file.

  In order to access a Pub/Sub topic, the connector must authenticate with Google Cloud. When Feldera runs in an environment with Application Default Credentials (ADC) configured, the connector will pickup the credentials automatically from the GOOGLE_APPLICATION_CREDENTIALS environment variable or from the $HOME/.config/gcloud/application_default_credentials.json file.

  When running Feldera in an environment where ADC are not configured,   e.g., a Docker container, use the credentials option to ship Google Cloud credentials from another environment. For example, if you use the gcloud auth application-default login command for authentication in your local development environment, ADC are stored in the $HOME/.config/gcloud/application_default_credentials.json file.

  Note that authentication is not required when running against a Pub/Sub emulator.

  See example below for more details on configuring credentials.

• project_id - Google Cloud project_id. When not specified, the connector will use the project id associated with the authenticated account.

Subscribtion

• subscription - Pub/Sub subscription name. The subscription must exist.

• snapshot - Reset subscription's backlog to a given snapshot on startup, using the Pub/Sub Seek API. This option is mutually exclusive with the timestamp option.

• timestamp - Reset subscription's backlog to a given timestamp on startup, using the Pub/Sub Seek API. The value of this field is an ISO 8601-encoded UTC time, e.g., "2024-08-17T16:39:57-08:00". This option is mutually exclusive with the snapshot option.

Connectivity

These options configure gRPC connection to the Pub/Sub service.

• endpoint - Overrides the default service endpoint 'pubsub.googleapis.com'.

• pool_size - gRPC channel pool size.

• timeout_seconds - gRPC request timeout.

• connect_timeout_seconds - gRPC connection timeout.

Emulator

• emulator - Pub/Sub emulator URL in the 'host:port' format. Use this option to connect to a Pub/Sub emulator instead of the production service.

Example

In this section, we provide a step-by-step guide to configuring and using a Feldera Pub/Sub connector. Please ensure you have a Google Cloud account before proceeding.

1. Create Application Default Credentials in your local development environment.

   Install gcloud, the Google Cloud CLI

   # Opens a Web browser for authentication.
   gcloud init
   gcloud auth application-default login

   The second command will store Application Default Credentials (ADC) in the $HOME/.config/gcloud/application_default_credentials.json file.

2. Use gcloud CLI to create a test topic and a subscription.

   gcloud pubsub topics create test-topic
   gcloud pubsub subscriptions create test-subscription --retain-acked-messages --topic test-topic

3. Publish some messages to the topic

   for i in {1..10}; do gcloud pubsub topics publish test-topic --message "{\"id\": $i}"; done

4. Ingest messages in Feldera.

   Copy the following code to the Feldera WebConsole, replacing the credentials field with escaped content of $HOME/.config/gcloud/application_default_credentials.json.

   CREATE TABLE pub_sub_input (
     id int
   ) WITH (
       'connectors' = '[
       {
         "transport": {
             "name": "pub_sub_input",
             "config": {
                 "subscription": "test-subscription",
                 "timestamp": "2024-08-17T00:00:00-00:00",
                 "credentials": "{\"account\": \"\",\"client_id\": \"<CLIENT_ID>\",\"client_secret\": \"<CLIENT_SECRET>\", \"quota_project_id\": \"feldera-test\", \"refresh_token\": \"<REFRESH_TOKEN>\",  \"type\": \"authorized_user\",  \"universe_domain\": \"googleapis.com\"}",
             }
         },
         "format": {
             "name": "json",
             "config": {
                 "update_format": "raw"
             }
         }
     }]'
   );

   Select the pub_sub_input table in the "Changes stream" tab and run the pipeline. You should see the ten messages you publishe in Step 3 in the change stream of the pipeline.

Additional resources

For more information, see:

• Top-level connector documentation
• Supported data formats