Mark J. Nelson wrote: > > And what's the level of churn in the integrated Mercurial bits, and > their level of dependence on ON changes? Is it worth ON PIT resources > to detect what will likely never be a problem, or is the real > requirement that the tests pass prior to integration?
Since we are integrating new versions of Mercurial on a regular basis, it's a good idea to test it before and after integration. I think we can get by with just one sparc and one x64 run per build. > Providing the package (as Danek suggests) might be reasonable, but the > bar for making it worthwhile in PIT is likely higher. Soon Mercurial will be used by all Solaris developers, and it is already integrated into Solaris. I think it is time to test it regularly. If we had a teamware test I would suggest that it be tested too. Cheers, Jim
