BigQuery

List BigQuery datasets and tables, and run read-only GoogleSQL queries.

Connect it in the console under Integrations > Connectors. See Connectors for how connectors work and who can use them.

Create a read-only credential

In the Google Cloud console, create a service account and give it BigQuery Data Viewer and BigQuery Job User on the project. Create a JSON key for it (Keys > Add key > JSON) and paste the whole file into VAIF Studio.

Settings

Setting		Notes
Project ID	Required	For example my-gcp-project.
Service account key (JSON)	Required	Grant it BigQuery Data Viewer and BigQuery Job User. Stored encrypted and never shown again.
Location	Optional	For example US.
Max bytes billed per query	Optional	A query that would scan more fails instead of costing money (default 1 GB). Default 1000000000.

Tables

Table	Filters	Returns
datasets	page_token	Datasets in the project.
tables	datasetId (required), page_token	Tables in one dataset.
sql	query (required)	Run one read-only GoogleSQL query (capped by max bytes billed). Needs the developer role.

Query from the API

curl -s -X POST "https://api.vaif.studio/database/connector/$VAIF_PROJECT_ID/query" \
  -H "Authorization: Bearer $VAIF_TOKEN" -H "Content-Type: application/json" \
  -d '{"wrapperId":"bigquery_fdw","table":"sql","filters":{"query":"SELECT country, COUNT(*) AS users FROM `my-project.app.users` GROUP BY country"},"limit":50}'

$VAIF_TOKEN is a console access token. See Connectors for the routes.

How it stays read-only

VAIF Studio signs in with the bigquery.readonly scope, so BigQuery itself refuses writes, and it refuses statements that write before sending them. Queries are billed to your Google Cloud project; the bytes-billed cap limits each one.