From: David Abrahams (dave_at_[hidden])
Date: 2008-07-16 10:14:15
on Wed Jul 16 2008, "Peter Dimov" <pdimov-AT-pdimov.com> wrote:
> David Abrahams:
>> That was one of the kinds of cases I had in mind when wondering if it
>> was a good idea to rely on dependent libraries' tests to discover
>> problems in a library.
> Having dependent libraries act as regression tests is a good thing. It
> increases the test coverage and helps us uncover regressions before
> our users.
Sure, sure. The more testing the better, and we have to test those
libraries anyway. The point I was making was that *relying* on these
dependent tests to find out about interface changes in dependencies is
not very satisfying. It's at best an indirect way to discover such
changes, with poor coverage of the dependency's interface, and we ought
to have a more direct and reliable way to address the problem if we can
think of one.
-- Dave Abrahams BoostPro Computing http://www.boostpro.com
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk