This is the mail archive of the
mailing list for the GCC project.
Re: ObjC/ObjC++ - fixed Objective-C 2.0 dot-syntax with 'self' and 'super'
- From: Mike Stump <mikestump at comcast dot net>
- To: Nicola Pero <nicola dot pero at meta-innovation dot com>
- Cc: "GCC Patches" <gcc-patches at gcc dot gnu dot org>
- Date: Fri, 5 Nov 2010 12:51:23 -0700
- Subject: Re: ObjC/ObjC++ - fixed Objective-C 2.0 dot-syntax with 'self' and 'super'
- References: <email@example.com>
On Nov 5, 2010, at 12:18 PM, Nicola Pero wrote:
> As a final note, we have a testcase (property-neg-4.m) with a class and an attempt to use a class accessor which matched
> an instance accessor of the same class but no class accessor; the testcase was expected to produce an error. But, in
> the testcase the class was a root class, which means that assuming we're looking up accessors in the normal way that methods
> are looked up (which is what I implemented), the testcase would pass and run because if a class method is not found,
> the instance methods of the root class are looked up. I was tempted to fix the testcase to have it pass, but then I
> checked the testcases for Apple's llvm-gcc and they still have the same testcase, so I assume it was intended to work
> that way (and it's not just a bug) and I changed our implementation to match that behaviour.
As time wears one, you'll want to try the code on clang for A/B comparisons and checking your understanding of Apple's understanding of the spec...
> Ok to commit to trunk ?