[Axis2] Question about Axis 2 SAAJ implementation

[R J Scheuerle Jr]

Consider the following code from the SOAPElementTest:

final String value = "foo";
soapElem.addTextNode(value);

I should be able to get the javax.xml.soap.Node using several different accessors (getChildElements(), getChildNodes(), getFirstChild()), etc.

------------------------------
Problem 1:
Object x = soapElem.getChildElements().next();
Object y = soapElem.getChildElements().next();

In this case, x and y are both javax.xml.soap.Node objects (good); however they are not the same object. The iterator is constructing new objects each time it is invoked. This is not efficient and incorrect.

-------------------------------
Problem 2
Object z1 = soapElem.getChildNodes().item(0);
Object z2 = soapElem.getFirstChild();

In both cases, the returned object is not an SAAJ javax.xml.soap.Node. This also seems to be a violation of the specification.

===========================
The proposed solution is to clearly separate the SAAJ implementation from the "backing" axis2.om.impl.dom tree.
The SAAJ layer is a semantic view of the underlying DOM model. Thus it should always return SAAJ objects when its methods are invoked.


Comments ?

Rich Scheuerle
Senior Developer
IBM WebSphere Web Services Engine & Tooling
512-838-5115 (IBM TL 678-5115)