RFR: 8284535 : Fix PrintLatinCJKTest.java test that is failing with Parse Exception [v8]

Alexey Ivanov aivanov at openjdk.java.net
Tue Apr 12 13:52:46 UTC 2022


On Mon, 11 Apr 2022 18:17:34 GMT, lawrence.andrews <duke at openjdk.java.net> wrote:

>> We need a common manual test framework code that can be shared across all the client manual test. This framework class should have the following
>> 1) Frame which contains test instruction .
>> 2) Pass & Fail button so that user can mark the test pass or fail
>> 3) Upon failing the test user should be able to elaborate why the test failed and this can be added to the test failure. 
>> 4) Disposing of all the frames.
>> 
>> @aivanov-jdk
>
> lawrence.andrews has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Mark info as final

> > An interesting idea. However, this would require updating all the existing tests including those that use its custom UI for run/manual.
> 
> Isn't this a goal of this fix:
> 
> > _We need a common manual test framework code that can be shared across all the client manual test._
> > ?

Doesn't this accomplish this goal?

No, it doesn't replace the (outdated) built-in machinery in jtreg for `manual=yesno`. At the same time, it provides a building block to update those tests.

I like this approach better, in which a test can be easily run without the need for jtreg, I think it greatly simplifies debugging and updating the test.

However, you're right that `applet/manual=yesno` in jtreg can be reimplemented using this new class `PassFailJFrame` even though it's not a goal at this very time. Initially I interpreted your suggestion as replacing all `manual` tests, which doesn't seem a right thing to do because it will break all existing manual tests.

-------------

PR: https://git.openjdk.java.net/jdk/pull/7966



More information about the client-libs-dev mailing list