Re: [PATCH] cvsimport: rewrite to use cvsps 3.x to fix major bugs

From: Eric S. Raymond <esr@thyrsus.com>
Date: Jan 12, 2013, 16:11 UTC
Message-ID: <20130112161105.GB3270@thyrsus.com>
In-Reply-To: <50F17DB0.2050802@alum.mit.edu>

Michael Haggerty <mhagger@alum.mit.edu>:

Show 7 quoted lines

> Otherwise, how do we know that cvsps currently works with git-cvsimport?
> (OK, you claim that it does, but in the next breath you admit that
> there is a new failure in "one pathological tagging case".)  How can we
> understand its strengths/weaknesses?  How can we gain confidence that it
> works on different platforms?  How will we find out if a future versions
> of cvsps stops working (e.g., because of a breakage or a
> non-backwards-compatible change)?

You can't. But in practice the git crew was going to lose that capability anyway simply because the new wrapper will support three engines rather than just one. It's not practical for the git tests to handle that many variant external dependencies.

However, there is a solution.

The solution is for git to test that the wrapper is *generating the expected commands*. So what the git tree ends up with is conditional assurance; the wrapper will do the right thing if the engine it calls is working correctly. I think that's really all the git-tree tests can hope for.

Michael, the engines are my problem and yours - it's *our* responsibility to develop a (hopefully shared) test suite to verify that they convert repos correctly. I'm working my end as fast as I can; I hope to have the test suite factored out of cvsps and ready to check multiple engines by around Wednesday. I still need to convert t9604, too.

I have parsecvs working since yesterday, so we really are up to three engines.

I have two minor features I need to merge into parsecvs before I can start on splitting out the test suite.

-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>