[cfe-commits] r44682 - /cfe/trunk/Sema/SemaExpr.cpp

[Steve Naroff]

On Dec 8, 2007, at 11:58 AM, Chris Lattner wrote:

>> URL: http://llvm.org/viewvc/llvm-project?rev=44682&view=rev
>> Log:
>>
>> Lookup methods in the global pool even when a statically typed  
>> object's class interface isn't in scope!
>> As the comment in the code indicates, I'm not fond of this.  
>> Nevertheless, gcc compat is a goal.
>> Here is the case I'm talking about...
>>
>> #import <Foundation/Foundation.h>
>>
>> @interface AnyClass : NSObject
>> - (NSRect)rect;
>> @end
>>
>> @class Helicopter;
>>
>> static void func(Helicopter *obj) {
>> NSRect r = [obj rect];
>> }
>>
>> ..before this patch, we would warn/error. With this patch,  
>> everything "just works".
>
> Yuck :).  Please add this to the testsuite.
>

Will do.

>> = 
>> = 
>> = 
>> = 
>> = 
>> = 
>> = 
>> = 
>> = 
>> =====================================================================
>> --- cfe/trunk/Sema/SemaExpr.cpp (original)
>> +++ cfe/trunk/Sema/SemaExpr.cpp Fri Dec  7 14:41:14 2007
>> @@ -2245,6 +2245,10 @@
>>      if (ObjcImplementationDecl *ImpDecl =
>>            ObjcImplementations[ClassDecl->getIdentifier()])
>>        Method = ImpDecl->lookupInstanceMethod(Sel);
>> +	  // If we still haven't found a method, look in the global pool.
>> +	  // I am not fond of this behavior, however we conform to what  
>> gcc does.
>> +	  if (!Method)
>> +	    Method = InstanceMethodPool[Sel].Method;
>
> Generally, please avoid the first person in comments: "This behavior  
> isn't very desirable, however we need it for GCC compatibility."
>

Agreed.

> Thanks!
>
> -Chris