Re: Exp state on top of dead

[Mark D. Baushke]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

CARTER-HITCHIN, David, GBM <address@hidden> writes:

> > This is not a problem with validate_repo. A
> > dead revision followed by a live revision is a
> > perfectly valid state for a repository file,
> > and in no way (in general) does it indicate a
> > problem.
> 
> For branch versions, perhaps, but if I had
> 'dead' at 1.5 and 'Exp' at 1.6 that would be a
> problem, no?

Not at all. It is in fact how a file is
resurrected when it has been removed from the main
trunk at one point in time (at which point it is
also moved into the Attic as an optimization) and
later added back to the main trunk (at which point
it is moved up from the Attic).

> The other people on this list seemed to think so
> a while ago.

I believe you are talking about the fact that
someone managed to have a live revision (where the
state was NOT 'dead') in the Attic of its
directory. That is/was an observed bug in that a
checkout of the main trunk would not find that
particular file correctly.

If your 1.6 revision is the most recent revision
on the trunk and the ,v file is in the Attic
subdirectory of its module, then that IS a
problem. Otherwise, there is nothing wrong.

        -- Mark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.4 (FreeBSD)

iD8DBQFE/ZU2Cg7APGsDnFERAnyQAKDH0OXJXAU24mbyJ1uc3X7yXW6CGACfTRu0
lg+IQB9BwUEXXcthtJJMM3U=
=UTaX
-----END PGP SIGNATURE-----