echauchot commented on code in PR #643: URL: https://github.com/apache/flink-web/pull/643#discussion_r1192157171
########## docs/content/posts/howto-test-batch-source.md: ########## @@ -0,0 +1,202 @@ +--- +title: "Howto test a batch source with the new Source framework" +date: "2023-04-14T08:00:00.000Z" +authors: +- echauchot: + name: "Etienne Chauchot" + twitter: "echauchot" + +--- + +## Introduction + +The Flink community has +designed [a new Source framework](https://nightlies.apache.org/flink/flink-docs-release-1.16/docs/dev/datastream/sources/) +based +on [FLIP-27](https://cwiki.apache.org/confluence/display/FLINK/FLIP-27%3A+Refactor+Source+Interface) +lately. This article is the +continuation of +the [howto create a batch source with the new Source framework article](https://flink.apache.org/2023/04/14/howto-create-batch-source/) +. Now it is +time to test the created source ! As the previous article, this one was built while implementing the +[Flink batch source](https://github.com/apache/flink-connector-cassandra/commit/72e3bef1fb9ee6042955b5e9871a9f70a8837cca) +for [Cassandra](https://cassandra.apache.org/_/index.html). + +## Unit testing the source + +### Testing the serializers + +[example Cassandra SplitSerializer](https://github.com/apache/flink-connector-cassandra/blob/d92dc8d891098a9ca6a7de6062b4630079beaaef/flink-connector-cassandra/src/main/java/org/apache/flink/connector/cassandra/source/split/CassandraSplitSerializer.java) +and [SplitEnumeratorStateSerializer](https://github.com/apache/flink-connector-cassandra/blob/d92dc8d891098a9ca6a7de6062b4630079beaaef/flink-connector-cassandra/src/main/java/org/apache/flink/connector/cassandra/source/enumerator/CassandraEnumeratorStateSerializer.java) + +In the previous article, we +created [serializers](https://flink.apache.org/2023/04/14/howto-create-batch-source/#serializers) +for Split and SplitEnumeratorState. We should now test them in unit tests. As usual, to test serde +we just create an object, serialize it using the serializer and then deserialize it using the same +serializer and finally assert on the equality of the two objects. Thus, hascode() and equals() need +to be implemented for the serialized objects. + +### Other unit tests + +Of course, we also need to unit test low level processing such as query building for example or any +processing that does not require a running backend. + +## Integration testing the source + +[example Cassandra SourceITCase +](https://github.com/apache/flink-connector-cassandra/blob/d92dc8d891098a9ca6a7de6062b4630079beaaef/flink-connector-cassandra/src/test/java/org/apache/flink/connector/cassandra/source/CassandraSourceITCase.java) + +For tests that require a running backend, Flink provides a JUnit5 source test framework. To use it +we create an *ITCase named class that +extends [SourceTestSuiteBase](https://nightlies.apache.org/flink/flink-docs-master/api/java/org/apache/flink/connector/testframe/testsuites/SourceTestSuiteBase.html) +. This test suite provides all +the necessary tests already (single split, multiple splits, idle reader, etc...). It is targeted for +batch and streaming sources, so for our batch source case here, the tests below need to be disabled +as they are targeted for streaming sources. They can be disabled by overriding them in the ITCase +and annotating them with @Disabled: + +* testSourceMetrics +* testSavepoint +* testScaleUp +* testScaleDown +* testTaskManagerFailure + +Of course we can add our own integration tests cases for example tests on limits, tests on low level +splitting or any test that requires a running backend. But for most cases we only need to provide +Flink test environment classes to configure the ITCase: + +### Flink runtime environment + +We add this annotated field to our ITCase and we're done + +`@TestEnv +MiniClusterTestEnvironment flinkTestEnvironment = new MiniClusterTestEnvironment(); +` + +### Backend runtime environment Review Comment: ok fair enough, in that case I'll also remove "runtime" from "Flink runtime environment" title for coherence. -- 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: issues-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
