[cfe-commits] r152137 - in /cfe/trunk: include/clang/AST/ include/clang/Basic/ include/clang/Driver/ include/clang/Frontend/ include/clang/Parse/ include/clang/Sema/ include/clang/Serialization/ lib/AST/ lib/CodeGen/ lib/Driver/ lib/Lex/ lib/Pars

Nico Weber thakis at chromium.org
Thu Mar 15 11:33:06 PDT 2012


On Thu, Mar 15, 2012 at 11:31 AM, Ted Kremenek <kremenek at apple.com> wrote:
> On Mar 15, 2012, at 11:01 AM, Nico Weber <thakis at chromium.org> wrote:
>
> It seems to work fine on 10.5 and 10.6 (except for subscripting). As
> far as I can tell, literals are just converted into ObjC method calls
> that exist on 10.5 – I guess what I'm asking is if you expect literals
> to be compiled into something else soonish (say, within the next
> year). If you plan to support 10.7 and the 10.7 ObjC runtime doesn't
> support "compiled literals" already, that's probably not the case?
>
>
> We try not to break existing code like that once we roll out a language
> feature.  We'd only change how this feature is compiled if the minimum
> deployment target was set to something where we could break backward
> compatibility because we could change the runtime.  That change would be
> transparent to developers.
>
> (What I really want to know if we can use this in chromium, which is
> still compiled for 10.5 :-) )
>
>
> If you find that it works for you, that is something you can rely on.  My
> main point is that we haven't tested it on 10.5, and if there are bugs
> because of subtle runtime differences, etc., we're not likely to invest time
> in fixing them.

Ok, thanks for the response. We'll give it a try and see how it works for us.

Nico




More information about the cfe-commits mailing list