Available Commands

1. exec <plan.json> [ <name> ]

Purpose: Submit a workflow execution request.

Details:

• <plan.json>: Path to a logical plan file.
• [<name>] (optional): Custom execution name; defaults to the plan’s filename (without extension).

Format auto-detection:

• If the JSON text contains an "operatorPositions" field, it’s treated as exported JSON format and then it will be converted.
• Otherwise it’s loaded directly as converted JSON format.

Example:

> exec my_plan.json

---

2. page <operatorId> <size> <pageIndex> [ --export <dir> ]

Purpose: Fetch a page of results for a given operator.

Positional arguments:

• <operatorId>: Operator identifier string.
• <size>: Number of rows per page.
• <pageIndex>: 1-based page index.

Optional --export <dir> (-e <dir>): Directory where a JSON Lines file will be written.

• File name: <operatorId>_<size>_<pageIndex>.jsonl
• Content: Each line in the file is one JSON object from the result rows.

Example:

> page CSVFileScan-1 100 2 --export ./exports

This will:

1. Send a ResultPaginationRequest for page 2, size 100.
2. Upon receiving the PaginatedResultEvent, write each row as a JSON line into ./exports/CSVFileScan-1_100_2.jsonl.

---

3. kill

Purpose: Abort the currently running workflow execution.

Example:

> kill

Sends a WorkflowKillRequest immediately.