macdoor opened a new issue, #18689:
URL: https://github.com/apache/dolphinscheduler/issues/18689
## Feature proposal
### What problem does this feature solve?
Add a new task type **FLINK_SGW** that submits SQL scripts to a **Flink SQL
Gateway** via its JDBC driver (`flink-sql-jdbc-driver`), so batch SQL workloads
can be scheduled by DolphinScheduler without a local Flink client installation.
Main capabilities:
- Main SQL script source: inline script **or** resource-center file
(`SqlSourceType.SCRIPT` / `FILE`)
- Optional init script (same two sources), executed before the main script
(e.g. `SET pipeline.name=...` and DDL)
- Configurable statement separator (default `;`), statements are split and
submitted one by one
- Leading line comments (`--`) and blank lines are stripped per statement so
the Flink parser receives clean statements
- JDBC connection properties are configurable; `flinkJdbcUrl` supports
parameter placeholders (e.g. `${flink_sqlgateway_url}`)
- Optional query-result row printing (`maxPrintRows`) to avoid log explosion
### Why the short name FLINK_SGW?
The natural candidate `FLINK_SQL_GATEWAY` is too long as a task type label:
it wraps/overflows the DAG node icon and the task-type picker layout.
`FLINK_SGW` (SQL Gateway) keeps the node readable; the DAG node reuses the
Flink family icon.
### Is this a regression?
No, this is a new task plugin.
### Additional context
The plugin has been running in an internal production environment
(schedule-only batch workloads through a dedicated Flink SQL Gateway) since
2026-03, which is where the naming and layout motivation comes from.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
[email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
