pvary opened a new issue #1826:
URL: https://github.com/apache/iceberg/issues/1826


   On #1607 @rdblue suggested that we should put the Hive specific test to an 
integration suit like @jackye1995 did with Glue, so if there are changes only 
in spark or flink code then we do not have to run the whole Hive suite to test 
the changes.
   
   My understanding of @jackye1995's solution is that we should create a new 
source folder for the integration tests to separate out the ones we do not want 
to run every time.
   ```
     sourceSets {
       integration {
         java.srcDir "$projectDir/src/integration/java"
         resources.srcDir "$projectDir/src/integration/resources"
         compileClasspath += main.output + test.output
         runtimeClasspath += main.output + test.output
       }
     }
   
     configurations {
       integrationImplementation.extendsFrom testImplementation
       integrationRuntime.extendsFrom testRuntime
     }
   
     task integrationTest(type: Test) {
       testClassesDirs = sourceSets.integration.output.classesDirs
       classpath = sourceSets.integration.runtimeClasspath
     }
   ```
   
   What I have not seen or understand is the automation around which would:
   1. Decide when we need to run the integration tests
   2. Run the actual tests on the CI
   
   Did I miss something? Or this is what should be implemented?
   If so, then how should we decide when to run the integration tests? I have 
found this 
[thread](https://discuss.gradle.org/t/incremental-testing-support/24785) about 
it on the gradle forums. And I have seen this kotlin 
[code](https://medium.com/swlh/incremental-testing-for-gradle-multi-projects-2ad6cacd8e51)
 as an actual implementation.
   This is what we are looking for? Is Iceberg tests are organized on a way 
that the subproject compile hierarchy also defines which test should be run? Or 
we should just stick to the specific spark/flink only change should not start 
Hive tests?
   
   Thanks,
   Peter


----------------------------------------------------------------
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]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to