kennknowles commented on a change in pull request #4153: [BEAM-3219]
DataflowRunner: delegate @Setup and @Teardown in stateful ParDo
URL: https://github.com/apache/beam/pull/4153#discussion_r153360051
##########
File path:
sdks/java/core/src/test/java/org/apache/beam/sdk/transforms/ParDoLifecycleTest.java
##########
@@ -151,6 +156,21 @@ public void testFnCallSequenceMulti() {
p.run();
}
+ @Test
+ @Category({ValidatesRunner.class, UsesStatefulParDo.class})
+ public void testFnCallSequenceStateful() {
+ PCollectionList.of(p.apply("Impolite", Create.of(KV.of("a", 1), KV.of("b",
2), KV.of("a", 4))))
+ .and(
+ p.apply(
+ "Polite", Create.of(KV.of("b", 3), KV.of("a", 5), KV.of("c",
6), KV.of("c", 7))))
+ .apply(Flatten.<KV<String, Integer>>pCollections())
Review comment:
I believe this adds coverage for flatten unzipping. I just carried it over
from the other test, since this is actually a somewhat disjoint codepath.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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
