On Feb 25, 2005, at 5:16 PM, Ben Collins-Sussman wrote:
> It's not just about the protocol. If I checkout a working copy using
> JavaSVN, can I then use the regular 'svn' C program against that
> working copy with no problems? What about the opposite scenario?
Aye, I totally agree: there are two different and important protocols.
The WC representation as being a 'standard Subversion WC as the svn
client expects' vs the client/server protocol. One doesn't necessarily
imply the other or vice-versa.
svk I guess is a good example of something that does use the
client/server protocol (via the Subversion C libs and associated
bindings, but I imagine that it could in theory be done with an
alternate pure Java implementation like JavaSVN) but it doesn't employ
the standard Subversion WC in which users might try a different client
program such as 'svn' (I hope I got that right; I haven't used it yet).
So, for a "rogue" :-) like JavaSVN, there are two protocols to test for
a 'svn' replacement:
- client/server
- svn WC
Not all Subversion-related activities would necessarily require
proficiency at both. (Obviously, I've been thinking about a few such
things since I started the "Transaction building with bindings, do I
need a WC on disk" thread on this list just two days ago.)
-Travis
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sat Feb 26 01:29:29 2005