The one caveat is that a Clock isn't particularly complex and it may be that the other implementation is focusing on stuff we don't care about. I don't have a particular stake here and haven't evaluated the specifics. Certainly if you do bring this in, I'd like you to make sure to replace what we have in place so far. The worst of both worlds is to have 2 slightly diverging implementations.
John =:-> On Wed, Aug 5, 2015 at 4:00 AM, Andrew Wilkins <[email protected] > wrote: > On Wed, Aug 5, 2015 at 9:47 AM Tim Penhey <[email protected]> > wrote: > >> On 04/08/15 17:54, Andrew Wilkins wrote: >> > Hi, >> > >> > I'd like to add a new third-party dependency for testing: >> > github.com/benbjohnson/clock <http://github.com/benbjohnson/clock>. I >> > intend to use this for testing forthcoming retry scheduling logic in the >> > storage provisioner. Any objections? >> >> I now that William has been working on mocking time a different way for >> some of the leadership tests. >> >> We should make sure we have consistency across this. >> >> Tim >> > > The approach William has taken is actually pretty much identical to what > I'm doing. The primary difference is that he has implemented a mock Clock > himself. The mock does not implement After(), which I do need in my code. I > can implement this myself, but I'd rather we just used something off the > shelf. > > Cheers, > Andrew > > -- > Juju-dev mailing list > [email protected] > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/juju-dev > >
-- Juju-dev mailing list [email protected] Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/juju-dev
