[cfe-commits] r61994 - in /cfe/trunk: Driver/ASTConsumers.cpp Driver/RewriteObjC.cpp include/clang/AST/Decl.h include/clang/AST/DeclBase.h include/clang/AST/DeclObjC.h lib/AST/ASTContext.cpp lib/AST/DeclObjC.cpp lib/CodeGen/CGCXX.cpp lib/CodeGen/CGCall.cpp lib/CodeGen/CGObjCGNU.cpp lib/CodeGen/CGObjCMac.cpp lib/CodeGen/CodeGenTypes.cpp lib/Sema/SemaDeclObjC.cpp lib/Sema/SemaExprObjC.cpp lib/Sema/SemaType.cpp

[Douglas Gregor]

On Jan 21, 2009, at 10:31 PM, Chris Lattner wrote:

>
> On Jan 9, 2009, at 9:18 AM, Douglas Gregor wrote:
>
>> Author: dgregor
>> Date: Fri Jan  9 11:18:27 2009
>> New Revision: 61994
>>
>> URL: http://llvm.org/viewvc/llvm-project?rev=61994&view=rev
>> Log:
>> Provide a new kind of iterator, the specific_decl_iterator, that
>> filters the decls seen by decl_iterator with two criteria: the  
>> dynamic
>> type of the declaration and a run-time predicate described by a  
>> member
>> function. This simplifies EnumDecl, RecordDecl, and ObjCContainerDecl
>> considerably. It has no measurable performance impact.
>
> Very nice!
>
> Two random questions:
>
>> +  template<typename SpecificDecl>
>> +  class specific_decl_iterator {
>> +    /// Current - The current, underlying declaration iterator,  
>> which
>> +    /// will either be the same as End or will point to a  
>> declaration of
>> +    /// type SpecificDecl.
>> +    DeclContext::decl_iterator Current;
>> +
>> +    /// End - One past the last declaration within the DeclContext.
>> +    DeclContext::decl_iterator End;
>
> Is there a sentinel that identifies the end of list (e.g. null)  
> which could be used instead of storing the End iterator here?

There is now! (But there wasn't when the original patch went in <g>)

>> +    /// Acceptable - If non-NULL, points to a member function that
>> +    /// will determine if a particular declaration of type
>> +    /// SpecificDecl should be visited by the iteration.
>> +    bool (SpecificDecl::*Acceptable)() const;
>
> Could this be a template argument?

Yep, but I thought I'd get skewered for such fancy C++ :)

Anyway, it means that we now have two iterator classes  
(specific_decl_iterator and filtered_decl_iterator) to cope with, but  
they're only one pointer large (yay).

Thanks for the suggestions!

	- Doug