[ 
https://issues.apache.org/jira/browse/DRILL-5752?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16163937#comment-16163937
 ] 

ASF GitHub Bot commented on DRILL-5752:
---------------------------------------

Github user ilooner commented on a diff in the pull request:

    https://github.com/apache/drill/pull/940#discussion_r138499857
  
    --- Diff: 
exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/writer/TestParquetWriter.java
 ---
    @@ -54,11 +56,13 @@
     import org.junit.Ignore;
     import org.junit.Rule;
     import org.junit.Test;
    +import org.junit.experimental.categories.Category;
     import org.junit.rules.TemporaryFolder;
     import org.junit.runner.RunWith;
     import org.junit.runners.Parameterized;
     
     @RunWith(Parameterized.class)
    +@Category({SecondaryTest.class, ParquetTest.class})
    --- End diff --
    
    To me the main distinction between smoke and secondary tests should be 
runtime. It's hard in my mind to distinguish based on importance, because the 
relevance of a test depends on the change being done. And you would end up 
running all the tests before finishing your change no matter what. The workflow 
I had in mind was the following:
    
      1. Make a change
      1. Run Fast tests for my category of interest (e.g. **OperatorTest**). 
Oops there's a failure
      1. Fix the failure
      1. Run Fast tests for category **OperatorTest** again. Oops another 
failure
      1. Fix the failure
      1. Run Fast tests for category **OperatorTest** again. Yay they pass
      1. Run all the tests
      1. Done
    
    Let me know what you think.


> Speed Up Unit Tests
> -------------------
>
>                 Key: DRILL-5752
>                 URL: https://issues.apache.org/jira/browse/DRILL-5752
>             Project: Apache Drill
>          Issue Type: Improvement
>            Reporter: Timothy Farkas
>            Assignee: Timothy Farkas
>
> Tests can be split into categories.
> High-level categories:
> * Fast
> * Slow
> Low-level categories:
> * Vector
> * WebUI
> * Planner
> * Operator
> * Storage
> * Hive
> * JDBC
> * Kudu
> * Mongo
> * Hbase
> After the tests are categorized the Travis build can just run the fast tests.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Reply via email to