Re: adding on branch

[Stefan Monnier]

>> Not at all.  I just want the above example to give me the same behavior
>> independently from the timing of 2 and 3.  That probably means that
>> if the import see a dead-revision 1.1 it should still set the default
>> branch to 1.1.1, as it does when the RCS didn't exist at all beforehand.

> Hmmm.... one problem is to figure out how things should be setup to do
> merge operations between the mainline trunk and your development branch.

I think the first problem is to get things working as I think they should.
The merging case comes afterwards.  And honestly, CVS is not good at finding
the ancestor on its own anyway, so it wouldn't be the first time that the
ancestor needs to be given explicitly in order to get a good merge.

> Sure. Actually, there has been some talk of just removing the Attic
> optimization for the mainline. It was needed before cvs was taught about
> the dead state, but it is not really needed these days.

Whatever works...  The main question is: do you now agree that the current
behavior is conceptually wrong (regarding the timing-sensitivity for steps
2 and 3 of my example) ?
 
> It is for dealing with merger issues... that is still the hardest
> problem. Finding the correct greatest common ancestor version.

Easy: the ancestor is provided on the command-line by the user anyway.


        Stefan