This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: PROPOSED PATCH: Robustify cp_fold_obj_type_ref() for ObjC++ use
On Sep 21, 2004, at 12:29 AM, Gabriel Dos Reis wrote:
That is becoming ridiculous. Why do Objective-C++ people want to
modify the C++ front-end with C++ maintainers having no say?
Huh, I am not a Objective-C++ maintainer as much as you are a C++
maintainer. Yes I am the maintainer of the support library
for Objective-C but that is just like you are a maintainer for
libstdc++ nothing more. Also this was asked with approval and not
just committed like your name-lookup changes which to me looks
like a violation of trust (also you broke powerpc-darwin with
those changes IIRC as I had to fix them twice).
Why is
it that when it comes to Objective-C++, do some otherwise respectable
people at Apple come with those games?
Huh????????? Really if we want to get along in the the GCC
world, I think we need to understand each other and not say
"why are people are playing these games". I am not playing a
game unless you think questioning maintainers is a game when it
is like questioning the police why they rejected something. I
am just arguing my side of story to which you call games.
It is still my side of the story and you can disagree with it
but don't call this playing games as it is not.
Another thing I am writing this as a non-Apple employee (I am no
longer an intern at Apple anymore). But really the C++ front-end
people are looking more and more dictators than maintainers.
Making the wrong decision. The only reason I got involved with
this patch in the first place is because Mark did not understand
what Objective-C was using OBJ_TYPE_REF in the first place. The
next reason why I replied was because Mark is speeding up the
"compiler" so if we introduce a new language hook in the
Objective-C++ front-end and call the C++ one, we just slowed
down C++ code compiled with the Objective-C++ compiler a bigger
amount than we would add in the patch which Zem or I have proposed.
Again if Mark really cares about compile time, it should be for
all languages and not just C++.
After "vapor ware", "brutalize", what is next?
dictators (that was added to keep you happy:) ).
This is my last message on this subject unless the C front-end
removes all of the C++ code too then but I know this cannot happen
for 4.0.0 like it should (but it is too late for that change) as
that would make your and Mark's argument more possible. But since
again the C front-end still checks the dialects, why again can we
not check the dialect in the C++ front-end, except maybe to say
Objective-C++ is secondary language when it should not be as Zem
made it work on at least one other target than darwin, x86-linux
which is where most of the people want the support for it?
I should also say that the GNUStep people have been waiting 3 to
4 releases of GCC to get support for Objective-C++ why keep
a GNU project waiting so long.
-- Pinski