[cfe-commits] r91535 - in /cfe/trunk: include/clang/Analysis/PathSensitive/GRExprEngine.h include/clang/Analysis/PathSensitive/MemRegion.h include/clang/Analysis/PathSensitive/Store.h lib/Analysis/GRExprEngine.cpp lib/Analysis/MemRegion.cpp lib/A

[Ted Kremenek]

Note that I think it's fine to add subclasses of SymbolicRegion if it  
makes doing specific things easier.  My main thought on using  
SymbolicRegions is that we already have a general lazy-abstraction in  
RegionStoreManager for SymbolicRegions.  Does a distinct  
CXXObjectRegion really add anything?

On Dec 21, 2009, at 10:43 AM, Ted Kremenek <kremenek at apple.com> wrote:

> Could we use a SymbolicRegion for temporaries as well?  We currently  
> associate many symbols with specific expressions, and can  
> distinguish different symbolic values for the same expression along  
> different loop iterations.
>
> On Dec 20, 2009, at 10:41 PM, Zhongxing Xu <xuzhongxing at gmail.com>  
> wrote:
>
>> 2009/12/17 Ted Kremenek <kremenek at apple.com>:
>>> Hi Zhongxing,
>>>
>>> How do we plan on using CXXObjectRegion?
>>>
>>> For stack allocated objects, we will use VarRegions.  For heap  
>>> allocated (e.g., 'new'), we will use SymbolicRegions.  For fields  
>>> that are C++ objects we will use FieldRegions.
>>
>>
>> I got another potential usage for CXXObjectRegion: for
>> CXXTemporaryObjectExpr. It's not a varregion, not a symbolic region.