Re: [Cvs-cvs] ccvs/src ChangeLog rcs.c [cvs1-11-x-branch]

[Larry Jones]

Mark D. Baushke writes:
> 
> Good catch. I'll back it out.

As a side note to you and/or Kelly, the fact that the original code
passes the sanity tests implies that we need a more robust test for
correct branch numbers added to the test suite.

> Larry Jones <address@hidden> writes:
> >
> > I think you may have jummped the gun on this one, Mark.  As implemented,
> > this function finds the highest rev across *all* branches with the right
> > number of dots, not just branches rooted at the same revision.  So, for
> > example, if you have a number of branches off 1.1 -- 1.1.0.2, 1.1.0.4,
> > and 1.1.0.6, for example -- and then create a new branch off of 1.2,
> > you'll get 1.2.0.8 instead of 1.2.0.2 like you should.  In fact, I see
> > now that it checks *all* tags with the right number of dots, not just
> > magic branch tags, so plain revision tags will pollute the branch
> > number, too.

-Larry Jones

Oh, now YOU'RE going to start in on me TOO, huh? -- Calvin