On 9/20/07, C. Michael Pilato <cmpilato@collab.net> wrote:
> Erik Huelsmann wrote:
> > This is why I say, "let's support -c 40-45" which would mean exactly
> > what people can find in the mergeinfo property *and* as the output
> > from the commandline client. That way, there's little room for
> > confusion, since we'll just start telling people to use -c with diff
> > and merge. -r is for looking at the nodes, -c for the edges, meaning
> > that -r makes sense for log, update, switch, copy etc. Just not for
> > diff, merge and mergetracking output.
>
> Where, in your mind, does "blame" fall in all of this? Which syntax makes
> sense for it?
In my mind -r would make sense for blame: it annotates with data *in a
given state*. I.e. in my mind, the output from blame says "this line
has been here (unchanged) since state r40" (for example), much like
the -r parameter for log works.
bye,
Erik.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Sep 20 14:21:16 2007