TisonKun commented on a change in pull request #10313: [FLINK-14840] Use
Executor interface in SQL cli
URL: https://github.com/apache/flink/pull/10313#discussion_r351079362
##########
File path:
flink-table/flink-sql-client/src/main/java/org/apache/flink/table/client/gateway/local/LocalExecutor.java
##########
@@ -523,12 +525,12 @@ public void stop(SessionContext session) {
}
// store the result with a unique id (the job id for now)
- final String resultId = jobGraph.getJobID().toString();
Review comment:
> Still using a `JobID` here is a bit weird
If a result is associated with a job, why not using `JobID` as key? I'm ok
with `UUID` in this patch but I think such a `UUID` is less expressive than a
`JobID` if we're able to ensure the `JobID` to be the one of the actual job.
What do you think we let `Pipeline` take an optional JobID i.e. `@Nullable
JobID getJobID()`? As mentioned above
> A Pipeline corresponds to a job and is able to hold a job id.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
