Julian Foad writes:
> Peter N. Lundblad wrote:
> > Stefan Küng writes:
> > > I'd expect the blame result to contain the locally modified lines
> > > (either with an empty author or maybe '(local)'), but actually it
> > > ignores the local modifications completely. It's as if it blames not
> > > against the working copy but the working BASE.
> >
> > Without actually digging up the code, I'm pretty confident this is a correct
> > analyses.
> >
> > > So I think this is either not intended and might be a bug, or the
> > > documentation for svn_client_blame3() should mention that
> > > svn_opt_revision_working blames against the BASE.
> >
> > I'd say the latter. Does including the local mods give you anything?
>
> Eww, no, don't let "_working" mean "_base": that's both silently doing the
> wrong thing, and getting in the way of us implementing the correct behaviour later.
Yeah, I didn't read careful enough. Sorry. We should document that working
doesn't work and possibly return an error until we implement working support.
> Yes, including the local mods gives us something useful. Often, I'm working on
> a file and want to see the "blame" info for the portion I'm working on; at
> present, I have to get a "blame" of the base version and then manually find the
> corresponding portion of it, whereas with this implemented I would be able to
> list the blame info alongside my working file and the lines would automatically
> match up.
OK, in the cmdline client, we will need -rWC (or -rWORKING) to support
this, but that might be useful for other commands as well.
Thanks,
//Peter
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sun May 7 22:21:41 2006