db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Knut Anders Hatlen <Knut.Hat...@Sun.COM>
Subject Re: Usage of "Test" component in JIRA
Date Tue, 30 Jun 2009 14:06:00 GMT
Dag.Wanvik@Sun.COM (Dag H. Wanvik) writes:

> Dag.Wanvik@Sun.COM (Dag H. Wanvik) writes:
>> This left many issues without component. I have added components for
>> open issues..
> I have added component "Test" (as an a priori component) for those
> issues which had no other component when the "Regression Test Failure"
> was removed. But this started me thinking and I have a question on the
> recommended usage of the "Test" component. 
> Normally, when we fix a Derby bug, we add a new test case to the
> regression tests. This means that a patch for an issue will touch
> both, say, the "Store" component and the "Test" component.
> I don't think we have had a clear practice on whether to mark such
> issues with the additional "Test" component or not. Since the focus of
> the issue would be the bug or modification in Derby itself, the "Test"
> component is often not given. Do we want to tighten this up and
> recommend to use the "Test" component in addition if the test code is
> at all modified by the patch? Or would it be better to reserve the
> "Test" component usage for cases where the focus is test or test
> infrastructure?

I think I'd prefer that we only use the Test component if there's a bug
in the test code or if the main focus of the issue is to improve the
test code. Otherwise, almost all issues would have the Test component,
since most improvements and bug fixes are accompanied by a regression
test. Limiting the Test component to pure test issues would make it more
usable in queries, in my opinion.

Knut Anders

View raw message