kfogel@collab.net writes:
> "C. Michael Pilato" <cmpilato@collab.net> writes:
> > It's not exactly a bug in the sense yoy mean. --revprop is for
> > revision props, and -t isn't a revision, so I would expect the
> > combination of the two not to work. I would, however, expect that the
> > program would tell you about the illegal combination of options.
> > Additionally, it's just a pointless limitation. The option should be
> > called --rev-or-txn-prop or something, and should work with either -r
> > or -t.
>
> Yeah, there is some language weirdness going on here, but we have sort
> of been using "revprop" to mean "the unversioned properties attached
> to a revision or transaction". I'm happy to deprecate the old option
> name and use a new one if you want, but in any case, it should work
> with txns as well as revisions (which it now does, since r15203).
No, I thought more about this, and decided the change was unnecessary.
If anything, perhaps --unversioned-props would be appropriate and
consistent with other communications (like the Book, etc.) But your
tweak to the option's usage string is *plenty* helpful to users.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Jun 30 22:15:03 2005