Re: import inconsistency

[Paul Edwards]

"Stefan Monnier" <monnier+gnu.cvs.bug/news/@flint.cs.yale.edu> wrote in message 
5lptlk9ps8.fsf@rum.cs.yale.edu">news:5lptlk9ps8.fsf@rum.cs.yale.edu...
> > If revision 1.1 has a state of dead, and there's no 1.2, and it
> > isn't in the Attic, that sounds pretty dead to me still.
>
> My understand of CVS's code is that such a situation is not considered
> as "dead" but as a bug.  The redundancy between the "in the Attic" bit
> and the "dead head" bit is due to history.

Ok, now I understand.  In that case, I suggest that that no longer
be considered a bug, that either dead head, or in the Attic, be
considered dead.

Because otherwise it will not be possible to do away with the
Attic, unless you make the file system no longer upwardly
compatible.

> >> so the file location is mostly irrelevant.
> > You just said "iff" (if and only if) (or was it a typo?), meaning
> > that it is the exact opposite of irrelevant.
>
> Well, .... it's irrelevant in that moving the file is not going to
> resurrect the file.  If moving the file out of the Attic is not done to
> resurrect it, then what is it for ?

It is for consistency, so that the different paths that lead to the
same result, don't end up with one of them weird and the other
clean.

> I'd understand if you just threw
> away the whole Attic thing, but then you have to do it wholesale.

Potentially.

BFN.  Paul.