Re: RPC Style Issues Part (2) - spec updated

Hi Umit,

> The following rules are missing from the definition of the rpc style [1] :
>
>     * The complex type that defines the body of an input or an output
>       element must not contain any attributes.

The current wording says:

    Input/output elements contain only local element children.
                                  ^^^^
The use of "only" does imply no attributes. I will put your additional
condition in if you insist, but I believe the current text covers it
and your extra clause is redundant.

(I did <emph> "only" - maybe that will help.)

>     * An element that designates an in/out parameter must be declared by
>       using the same type value both in the input element and the output
>       element.

I update the in/out parameter bullet as follows:

    <item><p> If there exists foo such that there are
    child elements named foo declared to be of the
    same type in both input and output elements, then that
    represents an in/out parameter.</p></item>

Is that sufficient?

>     * Each element designates a single parameter  in a function
>       signature irrespective of its cardinality.

Added.

>     *  The parameter order of the function is designated by the order of
>       elements in the sequence as specified in the schema.

Added because Jonathan said this email is now status quo, but of
course we are still talking about this topic.

Sanjiva.

Received on Sunday, 2 November 2003 05:17:52 UTC