[cfe-commits] r134302 - /cfe/trunk/lib/Sema/AnalysisBasedWarnings.cpp

[Matthieu Monrocq]

> From: Douglas Gregor <dgregor at apple.com>
> Subject: Re: [cfe-commits] r134302 -
>        /cfe/trunk/lib/Sema/AnalysisBasedWarnings.cpp
> To: Jordy Rose <jediknil at belkadan.com>
> Cc: cfe-commits at cs.uiuc.edu
> Message-ID: <B5725724-0AA7-49C1-8BB5-F09412D5AD1C at apple.com>
> Content-Type: text/plain; CHARSET=US-ASCII
>
>
> On Jul 1, 2011, at 6:06 PM, Jordy Rose wrote:
>
> > Just to be silly, if we're in an ObjC context and 'NULL' is defined, but
> 'nil' isn't, is 'NULL' preferable to 0? Or is that a contrived situation we
> don't care about?
>
> It feels contrived to me; 'nil' is generally available in any real ObjC
> code.
>
> > (Also, do we care about 'Nil', the null value for ObjC 'Class' type, and
> 'false' for C99 _Bool?)
>
>
> I, personally, don't care much about those particular cases, but we could
> hack them in.
>
>        - Doug
>

Also, in C++0x, I think "nullptr" would be better, or is that Clang does not
support it yet ?

-- Matthieu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20110702/87fefb28/attachment.html>