[cfe-dev] Better type-specifier representation

[Ted Kremenek]

On Sep 12, 2008, at 4:17 PM, Argiris Kirtzidis wrote:

> These are good points, I agree. Putting type-spec info in the type  
> is the wrong way to go.
>
> Daniel Dunbar wrote:
>> If we want to explicitly represent the type specifier, what is  
>> wrong with
>> actually making this a separate node (not a Type and not a Decl).  
>> We can
>> probably present the abstraction that the AST has TypeSpecifier nodes
>> without actually creating them (by encoding in DeclGroup or other  
>> tricks).
>>
>> This also gives a clear place to attach source range, makes clear  
>> what the
>> representation of sizeof(type) is (a SizeofAlignOfTypeExpr which  
>> points to
>> a TypeSpecifier which in turn points to the type), and should make  
>> ownership
>> clear.
>
> This is a great idea!
>
> How about using something like this:
>
> -In the AST, Decls do not have Types, they have TypeSpecifiers.  
> Exprs have Types.
> -TypeSpecifiers can own RecordDecls
> -DeclGroups are used for declaration lists:
>     int x,y,z; -> DeclGroup
> but not for parameters. If the ParmVarDecl has a TypeSpecifier, we  
> don't have to use a DeclGroup in its place.
>
>
> Let me know what you think.
>
> -Argiris

This sounds like the right approach to me!

The one outstanding grossness that I can think of right now is VLAs.   
How should we represent:

   int x[foo()][bar()][...], y;

Right now, the type for the VLA owns the expressions foo(), bar(), and  
so on.  Both DeclGroups and TypeSpecifiers don't solve this problem (I  
believe).