Re: RFC: reviewer's statement of oversight

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Monday, 8 October 2007 21:26, Scott Preece wrote:
> On 10/8/07, J. Bruce Fields <bfields@fieldses.org> wrote:
> > On Mon, Oct 08, 2007 at 08:34:47PM +0200, Stefan Richter wrote:
> ...
> > > So, putting a Tested-by into the changelog is only useful if the
> > > necessary testing is rather simple (i.e. "fixed the bug which I was
> > > always able to reproduce before") or if the tester is known to have
> > > performed rigorous and sufficiently broad tests.
> >
> > Well, you can still include those test-method details in the body of the
> > message in addition to adding the "Tested-by:".
> >
> > Does "Tested-by" just mean they ran some relevant test on the final
> > version of the patch?  The really hard part is often the initial work
> > required to find a good reproduceable test case, capture the right error
> > report, or bisect to the right commit.  I think that also counts as
> > "testing".  And it'd be nice to have a tag for those sorts of
> > contributions, partly just as another way to ackowledge them.
> ---
> 
> Tested-by should, at the very least, have a description of the test
> environment in the body (suggesting that the change compiled and ran
> in that environment). Preferably it should also have a description of
> the test or test suite run and whether that test failed on an
> unpatched system.

Tested-by: is sort of trivial for a fix patch, for example, if a bug reporter
confirms that the proposed patch actually fixes the issue.  IMHO it wouldn't
be practical to complicate that.

Greetings,
Rafael