Re: [Gnu-arch-users] Re: Which cscvs to use?

[Robert Collins]

On Tue, 2003-09-16 at 00:20, John Goerzen wrote:
> Robert Collins <address@hidden> writes:
> 
> >> which results in patch-1 to patch-300-odd with patch logs reading
> >> "initial import", each adding one file.
> 
> I believe I will suffer the same problem, except for the tree I'm
> working with, it would be more like 16,000 commits.

in one branch? whoa.

> > Ah yeah. So what you want to do is 'import' the first changeset, and
> > then 'commit' the remaining ones on the branch.
> 
> Can you give a recipe to do that?  And how does this ensure that the
> first changeset does include the full tree and not just one file?

The first changeset will be dependent on what went into CVS. If the CVS
repository was created with one file, then thats what you get.

cscvs cache -b
cscvs totla -i 1 <path to tla tree, that is ready for import, but with
no files>
cscvs totla -c 1:: <path to the same tree>

I've branched cduffy's cscvs--experimental--1.1 into
address@hidden/cscvs--experimental--1.0.

I've pulled across my usability tweaks. However, it is still very rough
around the edges. The performance is reasonable though. (And the squid
CVS repo has ~ 10K changesets in it).

Rob

-- 
GPG key available at: <http://members.aardvark.net.au/lifeless/keys.txt>.

signature.asc
Description: This is a digitally signed message part