aljoscha commented on a change in pull request #10311: [FLINK-14762][client]
Enrich JobClient API
URL: https://github.com/apache/flink/pull/10311#discussion_r351690850
##########
File path:
flink-clients/src/main/java/org/apache/flink/client/deployment/ClusterClientJobClientAdapter.java
##########
@@ -48,25 +54,45 @@
private final JobID jobID;
- private final Thread shutdownHook;
+ private final AtomicBoolean running = new AtomicBoolean(true);
- public ClusterClientJobClientAdapter(final ClusterClient<ClusterID>
clusterClient, final JobID jobID, final boolean withShutdownHook) {
+ public ClusterClientJobClientAdapter(
+ final ClusterClient<ClusterID> clusterClient,
+ final JobID jobID) {
this.jobID = checkNotNull(jobID);
this.clusterClient = checkNotNull(clusterClient);
-
- if (withShutdownHook) {
- shutdownHook = ShutdownHookUtil.addShutdownHook(
- clusterClient::shutDownCluster,
clusterClient.getClass().getSimpleName(), LOG);
- } else {
- shutdownHook = null;
- }
}
@Override
public JobID getJobID() {
return jobID;
}
+ @Override
+ public CompletableFuture<JobStatus> getJobStatus() {
+ return
clusterClient.getJobStatus(jobID).thenApply(org.apache.flink.runtime.jobgraph.JobStatus::toCoreJobStatus);
+ }
+
+ @Override
+ public CompletableFuture<Map<String, OptionalFailure<Object>>>
getAccumulators(ClassLoader classLoader) {
Review comment:
Actually, I think it's better to just return a
`CompletableFuture<Map<String, Object>>`. We can fail the future if there is
any failure in the map but otherwise just return a complete map. I don't think
it's useful for the user to have to do all the unpacking. What do you think?
----------------------------------------------------------------
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
