"Justin Erenkrantz" <justin_at_erenkrantz.com> writes:
> On Feb 4, 2008 9:27 PM, Karl Fogel <kfogel_at_red-bean.com> wrote:
>> In the issue, Mike Pilato proposes this solution: we let (m)ine
>> and (t)heirs only make determinations on conflicting hunks, and
>> reserve (M)ine and (T)heirs for mine-in-full and theirs-in-full.
>
> I don't like case-sensitivity being meaningful.
>
> How about adding ! - that is m! means mine-in-full and t! means
> theirs-in-full. (Or do + or something else after the m/t
> combination.)
>
> But 'M' and 'm' meaning something different seems error-prone. -- justin
Hmmm. I see your point, but it would mean we have to rewrite the
prompts in all sorts of ways, because (in English) we can no longer
just put parens around the first letter of the word. But maybe this
isn't a problem, since the non-English translations are presumably
already in that circumstance and we haven't heard a revolt yet.
Unrelatedly to the above, in r29251 I changed "all" to "full", because
of David Glasser's point that "all" might be mistaken for "all the
remaining files".
I'd prefer 'm+'/'t+' or even 'mf'/'tf', rather than 'm!'/'t!', because
in some editors (ahem, Emacs), '!' can be a command (e.g., in
query-replace) that means "answer yes to all the remaining instances,
without prompting". That's not the association we want here. I think
the mnemonic value of having 'f' match "full" makes it somewhat
preferable to '+' or any other special char anyway.
I'll wait a bit before making this tweak, so people have a chance to
comment. (That also means I'll wait a bit before proposing this
cluster for backport to 1.5.x.)
-Karl
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: dev-help_at_subversion.tigris.org
Received on 2008-02-09 19:27:57 CET