[cfe-dev] Adding type source info to ImplicitValueInitExpr.

[John McCall]

On Feb 13, 2010, at 5:18 AM, Enea Zaffanella wrote:

> John McCall wrote:
>> On Jan 18, 2010, at 5:24 PM, Abramo Bagnara wrote:
>>> Il 18/01/2010 22:33, John McCall ha scritto:
>>>> On Jan 18, 2010, at 3:21 AM, Enea Zaffanella wrote:
>>>>> The attached patch adds (optional) type source info to class
>>>>> ImplicitValueInitExpr: the type source info will have a valid value
>>>>> if the ImplicitValueInitExpr results from __builtin_offsetof
>>>>> constructs.
>>>> I think we'd rather not take this.  The primary use of
>>>> ImplicitValueInitExpr does not involve a type written in the source;
>>>> in fact, it involves no source code at all. This patch is useful
>>>> solely because of the current implementation of __builtin_offsetof.
>>>> That implementation is terrible, and it's long past time for
>>>> __builtin_offsetof to get its own AST class. When that happens, this
>>>> patch will be unnecessary.
> 
> 
> Hello.
> 
> A student of mine would like to have a try fixing the thing above
> (a.k.a., PR 5390), unless there already is someone working on it.
> 
> The approach we would follows is as follows:
> 
> 1) Take out OffsetOf from UnaryOperator::OpCode and have instead a dedicated class inheriting from Expr (OffsetOfExpr).
> 
> 2) Create a new class OffsetOfBaseExpr that will replace current (ab-)uses of ImplicitValueInitExpr in the context of the new class above: this new class will embed a TypeSourceInfo object, so as to be able to provide suitable source location info.

I don't understand why you need this second expression class;  the base of an offsetof is a type, not an expression.  You should be able to get by with just having a TypeSourceInfo as a field in the OffsetOfExpr.

Otherwise this sounds excellent;  by all means, go ahead.  I don't know of anyone else doing this.

John.