UI quickstart

This walkthrough uses the bundled csv-to-sqlite task. It reads CSV records, builds a full_name, and upserts rows into a SQLite table.

You need an admin or superadmin account because the workflow creates connectors and starts a run.

1. Prepare a CSV

Create a file named users.csv:

id,first_name,last_name,email,status
1,Ada,Lovelace,ada@example.com,active
2,Grace,Hopper,grace@example.com,active

2. Create the CSV connector

1. Open Connectors and select New Connector.
2. Choose CSV.
3. Name it Quickstart CSV.
4. For a standard Docker deployment, set Directory to /app/data.
5. Save the connector.
6. Upload the file at users/users.csv.

The task reads the resource named users; the CSV connector resolves that to CSV files under the users directory.

3. Create the SQLite connector

1. Create another connector and choose SQLite.
2. Name it Quickstart SQLite.
3. Set Path to /app/dbs/quickstart.db.
4. Run Test Connection.
5. Save the connector when the test passes.

Use an absolute path under /app/dbs in the hardened Docker topology.

4. Locate the task

Open Tasks and select csv-to-sqlite. Bundled tasks are imported on first boot unless seed import was disabled for the deployment.

If the task is not present, import the supplied tasks/csv-to-sqlite.yaml file through Tasks → Import.

5. Run the task

1. Select Run.
2. Map csv_source to Quickstart CSV.
3. Map db to Quickstart SQLite.
4. Confirm the run.

Envoy creates a run record and dispatches the task to the executor. The detail page updates while the run is active.

6. Inspect the result

On the run detail page, check:

• status is success;
• the step list reached log_result;
• the logs do not contain a connector or filesystem error;
• the item count matches the CSV rows.

Then open Data Explorer, choose Quickstart SQLite, select the users table, and run a query. The rows should include full_name values such as Ada Lovelace.

What happened

The task did not contain either path. Its YAML declared two typed connector slots:

connectors:
  csv_source:
    type: csv
  db:
    type: sqlite

The run mapping supplied the actual connector records. This separation is what makes the same task reusable in another environment.

Read Tasks and runs for versioning, reruns, stopping, and log handling. For exact task YAML, continue to Task structure.