[ 
https://issues.apache.org/jira/browse/LUCENE-5622?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Robert Muir updated LUCENE-5622:
--------------------------------

    Attachment: LUCENE-5622.patch

Done.

THere are problems with the fancy nested tests:
{noformat}
   [junit4] Tests with failures:
   [junit4]   - 
org.apache.lucene.util.junitcompat.TestExceptionInBeforeClassHooks.testExceptionWithinTestFailsTheTest
   [junit4]   - 
org.apache.lucene.util.junitcompat.TestExceptionInBeforeClassHooks.testExceptionWithinBefore
   [junit4]   - 
org.apache.lucene.util.junitcompat.TestExceptionInBeforeClassHooks.testExceptionInBeforeClassFailsTheTest
   [junit4]   - 
org.apache.lucene.util.junitcompat.TestSeedFromUncaught.testUncaughtDumpsSeed
{noformat}

Also, I think we should allow output when a sysprop (tests.noisy?) is set. This 
is important to automatically add when using -Dtestcase, some developers like 
to debug with SOP. Often times its the case where you need to use a 
shell-script to beast a test, etc. So these workflows will still work, we just 
want to fail when running the whole suite.

> Fail tests if they print, and tests.verbose is not set
> ------------------------------------------------------
>
>                 Key: LUCENE-5622
>                 URL: https://issues.apache.org/jira/browse/LUCENE-5622
>             Project: Lucene - Core
>          Issue Type: Bug
>            Reporter: Robert Muir
>            Assignee: Dawid Weiss
>         Attachments: LUCENE-5622.patch, LUCENE-5622.patch, LUCENE-5622.patch
>
>
> Some tests print so much stuff they are now undebuggable (see LUCENE-5612).
> I think its bad that the testrunner hides this stuff, we used to stay on top 
> of it. Instead, whne tests.verbose is false, we should install a printstreams 
> (system.out/err) that fail the test instantly because they are noisy. 
> This will ensure that our tests don't go out of control.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

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

Reply via email to