Re: Ping [PATCH] c++/42121 - diagnose invalid flexible array members

[Bernd Schmidt <bschmidt at redhat dot com>]

> The patch should bring C++ support for flexible array members closer
> to C (most of the same constructs should be accepted and rejected).
> The only C change in this patch is to include the size of excessively
> large types in diagnostics (I found knowing the size helpful when
> adding tests and I think it might be helpful to others as well).

Can't really comment on the C++ parts, but I spotted some formatting issues.

> +      && TREE_CODE (type) != ERROR_MARK
> +      && (DECL_NAME (fld) || RECORD_OR_UNION_TYPE_P (type)))
> +    {
> +
> +      return TYPE_SIZE (type)
> +	&& (TREE_CODE (TYPE_SIZE (type)) != INTEGER_CST
> +	    || !tree_int_cst_equal (size_zero_node, TYPE_SIZE (type)));

Unnecessary blank line. Multi-line expressions should be wrapped in 
parentheses for indentation. Lose braces around single statements.


> +   a flexible array member or a zero-size array.
> +*/

Comment terminators go at the end of the line.

> +  tree size_max_node =
> +    int_const_binop (MINUS_EXPR, size_zero_node, size_one_node);

The = operator should start the line.

> +	  tree flexarray =
> +	    check_flexarrays (t, TYPE_FIELDS (fldtype), seen_field);

Here too.

> +  {
> +    bool dummy = false;
> +    check_flexarrays (t, TYPE_FIELDS (t), &dummy);
> +  }

No reason to wrap this in braces.

> +  if (NULL_TREE == size)
> +    return build_index_type (NULL_TREE);

Don't know whether the conventions in cp/ are different, but usually 
this is "size == NULL_TREE".

> +	    pedwarn (input_location, OPT_Wpedantic,
> +		     "ISO C++ forbids zero-size arrays");
>  	}
> +
>      }

Extra blank line.

> @@ -11082,6 +11094,10 @@ grokdeclarator (const cp_declarator *declarator,
>  		     || !COMPLETE_TYPE_P (TREE_TYPE (type))
>  		     || initialized == 0))
>  	  {
> +	    if (TREE_CODE (type) != ARRAY_TYPE
> +		|| !COMPLETE_TYPE_P (TREE_TYPE (type)))
> +	      {
> +
>  		if (unqualified_id)

Here too.


Bernd