Re: [C++ PATCH] Fix 11614

[Nathan Sidwell <nathan at codesourcery dot com>]

Jason Merrill wrote:
> I think it's simpler to defer handling this until we've parsed the whole
> declarator.  Here's a patch I haven't gotten around to testing yet:

>         }
>       else if (decl_context == FIELD)
>         {
> + 	/* The C99 flexible array extension.  */
> + 	if (!staticp && TREE_CODE (type) == ARRAY_TYPE
> + 	    && TYPE_DOMAIN (type) == NULL_TREE)
> + 	  {
> + 	    tree itype = compute_array_index_type (dname, integer_zero_node);
> + 	    type = build_cplus_array_type (TREE_TYPE (type), itype);
> + 	  }
> + 
>   	if (type == error_mark_node)
>   	  {
>   	    /* Happens when declaring arrays of sizes which
What about
	typedef int ary_t[];
	struct { ary_t s; };
? I think you'll be allowing that.

nathan
--
Nathan Sidwell    ::   http://www.codesourcery.com   ::     CodeSourcery LLC
         The voices in my head said this was stupid too
nathan@codesourcery.com    ::     http://www.planetfall.pwp.blueyonder.co.uk