W3C home > Mailing lists > Public > xml-dist-app@w3.org > September 2001

Re: sparse arrays - too complex?

From: Pete Hendry <peter.hendry@capeclear.com>
Date: Fri, 21 Sep 2001 15:11:48 +0100
Message-ID: <3BAB4AA4.5A771204@capeclear.com>
To: xml-dist-app@w3.org
CC: Andrew Layman <andrewl@microsoft.com>

> Per the spec, omitting an accessor and having it with xsi:nil="true" MAY
> or MAY NOT be equivalent.  The spec leave open all possibilities.  Some
> implementations of structure serialization evidently choose to treat
> them as equivalent representations, but neither the SOAP nor the XML
> Schemas specification requires this.

The problem this produces is, when a test like echoStringArray or echoSparseArray or whatever is created, then there has
to be a contract between the client and server to agree upon what missing elements mean. If the client has some special
value for a missing element which is not the same as its representation when nil="true", and the server treats a missing
element as null then the value of the array echoed back may contain elements with nil=true that were missing elements in
the request. The client may then fault because it got back a different value.

I am in agreement with those that think sparse arrays should be kept out of the low level SOAP spec and should be
implemented on top of it. This is not because they are particularly hard to implement, but they are hard to define. The
SOAP layer should provide only the basic upon which more complex behaviour can be built. It should avoid as much
complexity as possible.

This looks like a rerun of a soapbuilders discussion which never seemed to reach agreement.

Received on Friday, 21 September 2001 10:12:18 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 22:01:15 UTC