On Mon, Oct 02, 2006 at 09:09:27PM +0200, Tommy Pettersson wrote:
> On Mon, Oct 02, 2006 at 11:26:08AM -0700, David Roundy wrote:
> > wrt the idea of extracting flag names from their DarcsFlag, I think I'm
> > probably with Eric on this, in that it seems like a good idea.  I can
> > see Tommy's point that in general it's a bug to change the name of any
> > flags, but making things "automatically right" is just a good thing.
> > You could
> 
> Then I have one more objection. My point of view is the readability of
> the source code. What would the name of the option identifier be but
> nearly the same as the string representation of the option? Anything else
> would be confusing.  So this "automatically right thing" would from my
> point of view do the wrong thing, by not updating the name of the option
> in the source code (instead hiding it under a level of abstraction for no
> good).

I'm willing to flip-flop back with Eric on this.  Perhaps it's better not
to add an abstraction layer here, although in general I like the idea of
auto-generated docs and error messages, as they're often easier to
refactor.  But I'm thinking now that perhaps this is not the level at which
we want an abstraction (i.e. it's more helpful to abstract things where you
would otherwise end up duplicating more content, as is the case with the
descriptions of the command actions).
-- 
David Roundy
http://www.darcs.net

_______________________________________________
darcs-devel mailing list
[email protected]
http://www.abridgegame.org/cgi-bin/mailman/listinfo/darcs-devel

Reply via email to