Ryan Schmidt wrote:
> You do not need to create a separate repository to use externals.
> Externals can reference items in the same repository.
My situation is the following: I have a file (a bibliography list)
that I would like to commit to two repositories. The repositories have
restricted access, so I am the only person who has access to both of
them. I can't easily get a third repository of my own, nor would I like
to do it to store just a single file.
Am I right in thinking, that externals cannot help in this case?
> When you say "from your hard drive", do you mean that the file is
> unversioned, or is it in fact in a repository somewhere? If
> unversioned then I can't suggest a solution, but as long as the file
> is in a repository, then externals could be the answer.
Yes, the files I want to reuse are not versioned in a public
repository, nor do I see any reason for them to be.
Les Mikesell wrote:
> Normally you expect version control systems to be able to be able to
> reproduce what you committed. If you commit a symlink you should
> expect to get a symlink back in a checkout, shouldn't you?
Given that symlinks are not portable and might point outside of the
repository, I'd expect to have control over what is done about them.
> When you describe the behavior of "almost all unix programs" you are
> conveniently ignoring the programs intended to do backup/restore or
> literal copies (tar, cpio, dump, etc.) which would be the the obvious
> counterpart for the expected action.
I am familiar with two backup programs: rsync and unison. Both leave
the choice to you: either they try to reproduce the links, or they
follow them and copy their targets instead. I guess tar also has the
option to follow the links, but might be wrong there.
> But, why don't you include the 'real' copy in one of the
> places you want it, and after committing it, reference the directory
> holding it as an external in all those other places you want the same
> things to appear?
Partly because of access restriction mentioned above: people who have
access to repository A might not have access to B and vice versa.
Also, I wouldn't like the health of the repository A to depend on the
health of B.
To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].
Received on 2009-07-21 14:40:25 CEST