Bill Allombert writes ("Re: Automated testing - politics, information, and
Ubuntu's plans"):
> To avoid any confusion: what kind of test are you speaking about ?
Well, after installing the package, does it work ? The idea is that
eventually every package will come with at least a basic couple of
tests to see if its key binaries run. (Obviously this is harder for
some packages than others.)
Bugs that tend to regress could be added to the test suite as well as
to the code.
For packages where it's plausible, we could even arrange to run
upstream's tests on the installed program.
> We are already doing a lot of test even if it is not centralized.
Right. But it's difficult to do it automatically and systematically
because there's no standard way to find the tests for a package.
Ian.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
