Re: vss2cvs migration of shared files

[Derek R. Price]

Laine Stump wrote:

> equivalent functionality in CVS. Some people emulate it by symlinking
> the ,v file in the repository into multiple directories, but doing so
> is begging for trouble, since CVS sets mutex locks on a per-directory
> basis to prevent multiple processes simultaneously writing to the same
> ,v file during commits and tags. Having a ,v file existent in two
> directories breaks the data model that is assumed for proper operation
> of these locks, and will eventually lead to a corrupted file.

Even aside from the locking issues, which may not be too hard to fix
since CVS must be tracing the symlinks to do the final 'mv' of the
temporary RCS file it writes to correctly, CVS _isn't_ tracing things
like the Attic in regards to symlinks.  If the file is removed from the
trunk, all the symlinks to it will break when the archive is moved to the
attic.  If a symlink is moved to the Attic its parent and other symlinks
won't be (I don't think this one is fatal).  To avoid this cheaply, a CVS
interface would have to be added for adding shared files ('cvs radd'?)
and metadata maintained.

There's a less important issue where tagging can attempt to tag a file
multiple times when a symlink exists.

I might be missing some issues as well.  Anyhow, right now it is messy
and it all sounds like an awful lot of work to fix.

Derek

--
Derek Price                      CVS Solutions Architect ( http://CVSHome.org )
mailto:address@hidden         CollabNet ( http://collab.net )
--
We shall have our follies without doubt.  Some one or more of them will always
be afloat.  But ours will be the follies of enthusiasm, not of bigotry, not of
Jesuitism.  Bigotry is the disease of ignorance, of morbid minds; enthusiasm of
the free and buoyant.  Education and free discussion are the antidotes of both.
We are destined to be a barrier against the return of ignorance and barbarism.

                        - Thomas Jefferson to John Adams, 1816