This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: ObjC improvements and ObjC++
- From: Mark Mitchell <mark at codesourcery dot com>
- To: Stan Shebs <shebs at apple dot com>
- Cc: gcc at gcc dot gnu dot org
- Date: 08 Jul 2003 10:51:11 -0700
- Subject: Re: ObjC improvements and ObjC++
- References: <3F0A22C2.6050601@apple.com>
On Mon, 2003-07-07 at 18:47, Stan Shebs wrote:
> Now that Apple's developers are happily banging away on Panther
> and Xcode and GCC 3.3, we have an opportunity to get local work
> on Objective-C into FSF GCC 3.4, and to get started on Obj-C++
> as well.
Thanks for presenting the plan!
The overall plan makes sense to me -- but I don't think 3.4 is a
realistic target. Stage 1 for GCC 3.4 is already closed, and this is
inevitably going to be a major change. I'm also going to be staring at
the C++ changes very carefully before they go in; the basic "mutually
oblivious" strategy makes sense, but the devil is in those dang
details. (Overload resolution for Objective-C classes? Template
deduction? Etc.)
If those details are not specified clearly -- as well as implemented
tidily -- I'm likely to oppose the patch. The C++ front end is already
complicated enough and a lot of the complexity comes from having to
implement very complex semantics. The only hope we have is that the C++
standard (mostly) specifies those semantics. I do not fancy trying to
figure out why the overloading code has some weird Objective-C thing in
there in five years with no specification at which to look.
You already said you're going to work up a detailed specification, so I
think you're fully aware of these issues; I'm just reiterating. :-)
(Frankly, I'd rather see C and C++ get combined into a single front end
before doing Objective-C++, but that's so much work that I don't think
it's reasonable to even request it.)
--
Mark Mitchell
CodeSourcery, LLC
mark@codesourcery.com