PATCH: Add two ObjC++ callback functions to cp/decl.c

Ziemowit Laski zlaski@apple.com
Sun Sep 12 22:19:00 GMT 2004


On 10 Sep 2004, at 20.05, Gabriel Dos Reis wrote:

> Ziemowit Laski <zlaski@apple.com> writes:
>
> [...]
>
> | > The SC specifically made Objective-C++ a second-class citizen.  So,
> | > I'm unsympathetic to your counter-argument that changes to the C++
> | > front end might affect Objective-C++.  That was explicitly part of
> | > the deal.
> |
> | Fair enough; but by the same token, if the proverbial ObjC++ 
> fragments
> | suddenly stop compiling, you can just '#if 0' them out and tell me to
> | fix them.
>
> I think I fully share Mark's concerns here.
> It looks to me that you want to modify the C++ front-end to work
> ObjC++ without letting C++ maintainers to have any say.

No, that's nonsense, Gaby, and you know it. :-)  Note that most of my
patches do ask for comments and approval.  The misunderstanding here
is that I was a bit surprised why you'd want to have a say for patches
that do not affect how the C++ front-end functions.  However, since
you apparently _do_ want to have a say, I'll ask for approval for
all my C++-area patches from now on.

>  For example, you made a change to the scope
> handling stuff -- that you described as obvious -- and when suggested
> a diffferent way and asked you to elaborate, you declined the
> suggestion arguing that once the change is in place, one'd look for
> alternatives.  I definitely disagree with that line.

And I definitely disagree with the "preemption by vaporware" strategy
you tried to use there.  You essentially said, "you shouldn't do it
this way because we may rewrite this part of the code soon and you
stuff will no longer work."  What am I supposed to do at that point?
I cannot adjust my ObjC++ stuff, since I don't know what _your_ stuff
will look like.  And, I don't believe that you sent me suggestions as
to how to rewrite my stuff.  Thus, my only option is to grind to a
halt and wait until you commit your rewrite who knows when.

--Zem



More information about the Gcc-patches mailing list