- From: Glen Daniels <gdaniels@macromedia.com>
- Date: Mon, 24 Mar 2003 13:14:58 -0500
- To: "'Martin Gudgin'" <mgudgin@microsoft.com>, "'David Fallside'" <fallside@us.ibm.com>, "'Alex DeJarnatt'" <alexdej@microsoft.com>, "'Jeffrey Schlimmer'" <jeffsch@windows.microsoft.com>, "'John Koropchak'" <johnko@microsoft.com>, "'akarmark@oracle.com'" <akarmark@oracle.com>, "'cunnings@lectrosonics.com'" <cunnings@lectrosonics.com>, "'Davanum Srinivas'" <dims@yahoo.com>, "'david.orchard@bea.com'" <david.orchard@bea.com>, "'Don Mullen'" <donmullen@tibco.com>, Glen Daniels <gdaniels@macromedia.com>, "'Jacek Kopecky'" <jacek@systinet.com>, "'jbaysdon@tibco.com'" <jbaysdon@tibco.com>, "'manoj.cheenath@bea.com'" <manoj.cheenath@bea.com>, "'Paul Kulchenko'" <paulclinger@yahoo.com>, "'Sam Ruby'" <rubys@us.ibm.com>, "'xan@tibco.com'" <xan@tibco.com>, "'Yasser Shohoud'" <yassers@microsoft.com>
- Cc: "'xml-dist-app@w3.org'" <xml-dist-app@w3.org>, "'axis-dev@ws.apache.org'" <axis-dev@ws.apache.org>
[cc'ed to axis-dev] Apologies for the ridiculously long delay in getting back to you on this, Gudge. I think the answer for Axis is 'No', we don't particularly plan to implement nodeType support. The rationale goes something like this: NodeType seems to be there to help deserialize encoded SOAP object graphs when there's no type information around. Primarily, it seems useful when there are child elements of a given node, no comprehensible xsi:type value, and no itemType/arraySize attributes. In the face of this situation, there would be no good way to tell if the child elements were array items or structure fields. This situation doesn't seem all that common, however (I would guess most implementations will send an itemType/arraySize for arrays). Also, since Axis is a Java-based toolkit, and our deserialization mechanisms generally rely on a specific Java class being available (a JavaBean with accessors for structural fields, for instance) to map to a given type in XML, we don't have a strong need to do "dynamic" deserialization of random SOAP-encoded values. That's much more something you'd want in a scripting language / typeless environment, I think. So the end result is that I don't think we have a very strong need for this feature. The other reason is that open-source development happens when someone with coding ability needs to "scratch an itch". If someone gets enthused and writes support for this, we will certainly be happy to have it, but I don't know of anyone currently clamoring for it. Thanks, --Glen > -----Original Message----- > From: Martin Gudgin [mailto:mgudgin@microsoft.com] > Sent: Thursday, March 06, 2003 10:07 PM > To: David Fallside; Alex DeJarnatt; Jeffrey Schlimmer; John Koropchak; > akarmark@oracle.com; cunnings@lectrosonics.com; Davanum Srinivas; > david.orchard@bea.com; Don Mullen; gdaniels@macromedia.com; Jacek > Kopecky; jbaysdon@tibco.com; manoj.cheenath@bea.com; Paul > Kulchenko; Sam > Ruby; xan@tibco.com; Yasser Shohoud > Cc: xml-dist-app@w3.org > Subject: Request for information: nodeType in SOAP 1.2 Implementations > > > > This is a request from the W3C XMLP WG for information about status of > implementations of nodeType functionality. The WG would like to know > answers to the following questions: > > A) Are you going to implement nodeType? > > B1) If the answer to A) is 'Yes', can you please give a rough idea > of when this feature would be implemented. > > B2) If the answer to A) is 'No', can you please give a rationale for > why you do not intend to implement this feature? > > Regards > > Martin Gudgin > For the XML Protocol Working Group >
Received on Monday, 24 March 2003 13:15:33 UTC