On 6/13/07, Giulio Troccoli <Giulio.Troccoli@uk.linedata.com> wrote:
> > If something actually commits, you'll get a string returned indicating
> > what revision was committed. If there's no commit, you'll get either a
> > blank response (exit 0 = no errors, nothing to do), or a string
> > indicating why the commit didn't happen (exit != 1, error occurred).
>
> No, that it the problem Andy: if there is no commit I get an returned
> code != 0, but that's it. If I have a returned code of 0 it means that
> something was actually committed.
>
> As you said later, it's the client that stops and empty commit, so there
> is not way to detect it in a "common" place for all clients (that's why
> I thought of the server).
There's really no control like you're looking for that can be enforced
at the server end, especially since in this scenario you're never
talking to the server.
> As Mark suggested I should check that there is actually something to
> commit if I do that in a script. For the users, I suppose I could write
> a script to commit changes and instruct them to use the script and not
> svn commit. Not the best solution maybe, but an acceptable one, unless
> someone else have a better idea :-)
If I'm understanding you properly, your issue boils down to the following:
"If nothing is reported back to the user, the user doesn't understand
that nothing was done."
Is that a fair assessment?
> > As far as I know, if the client doesn't see anything to commit (read:
> > it detects no changes in the WC), then it'll stop right there. Why
> > bother with a server trip at that point?
>
>
> Linedata Services (UK) Ltd
> Registered Office: Bishopsgate Court, 4-12 Norton Folgate, London, E1 6DB
> Registered in England and Wales No 3027851 VAT Reg No 778499447
>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Wed Jun 13 16:27:31 2007