W3C home > Mailing lists > Public > www-ws-desc@w3.org > October 2003

Re: parameter order

From: Roberto Chinnici <Roberto.Chinnici@Sun.COM>
Date: Tue, 28 Oct 2003 13:25:52 -0800
To: Sanjiva Weerawarana <sanjiva@watson.ibm.com>
Cc: "'WS Description List'" <www-ws-desc@w3.org>
Message-id: <3F9EDEE0.8080705@sun.com>

Sanjiva Weerawarana wrote:
>>Same messages, different parameter orders.
> 
> 
> While your interpretation of parameterOrder="" seems reasonable,
> I can't find anything in the WSDL 1.1 spec that implies this
> usage. Of course, WSDL 1.1 left a lot for the imagination ;-).

Well, in 2.4.6 it says "the return value part is not present in the
list". Of course the spec then says that this information serves
as a "hint", double-quoted in the original -- how is a "hint"
different from a hint?  ;-)

> If its parameterOrder we desire let's make that proposal. After
> all, its well-known already (well hated IMO, but that's my view)!

I'm not proposing that. I would observe though that the argument
we're currently having on distinguishing the following two signatures:

   void f([out] int x)
   int f()

may well be followed by another one over these ones:

   void f([in] int x, [out] double y)
   void f([out] double y, [in] int x)

I don't quite understand where the RPC crowd is going to draw the line,
and why.

Roberto

-- 
Roberto Chinnici
Java Web Services
Sun Microsystems, Inc.
roberto.chinnici@sun.com
Received on Tuesday, 28 October 2003 16:24:40 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:27 GMT