Correct, it doesn’t look like I use the XMLNode reference returned by the simpleXMLEncoder.encodeValue() method.
I haven’t looked at this code in a long time, but here’s what I think is happening:
1) The first line creates a new XMLDocument object.
2) The second line creates a new SimpleXMLEncoder object and passes the XMLDocument object created in on the first line. I believe the SimpleXMLEncoder constructor sets the empty XMLDocument object as the root XML node.
3) The third line calls calls the encodeValue() method on the SimpleXMLEncoder object which “encodes an ActionScript object to XML using default serialization”. According to the comments in the source code, the first parameter (obj) is “the ActionScript object to encode”, the second parameter (qName) is “the qualified name of the child node”, and the third parameter (xmlDocument) is “an XMLNode under which to put the encoded value”.
4) The fourth line converts the XMLDocument object to an XML object.

So, if you only used the first and last line, you’d always have an empty XML object. The second line creates a SimpleXMLEncoder instance which gets used in the third line to covert the specified Object and insert it into the xmlDocument instance. I’m guessing the magic that is happening is that the SimpleXMLEncoder is directly modifying the XMLDocument object, which is why I didn’t need to use the returned XMLNode object from line 3.

If you want to see exactly what happens in the SimpleXMLEncoder class, you can check out the source code in the installed SDKs /frameworks/projects/rpc/src/mx/rpc/xml/SimpleXMLEncoder.as file.

Peter

var xmlDocument:XMLDocument = new XMLDocument();
var simpleXMLEncoder:SimpleXMLEncoder = new SimpleXMLEncoder(xmlDocument);
var xmlNode:XMLNode = simpleXMLEncoder.encodeValue(obj, qName, xmlDocument);
var xml:XML = new XML(xmlDocument.toString());

xmlDocument is in the first line, and gets referred to in the last line, but the objects simpleXMLEncoder and xmlNode do not get used. Why not simply use the first line and the last line? What additional work is happening in the middle two lines?

Thanks. You can file Flex bugs/enhancement requests at: http://bugs.adobe.com/flex/
If you post the bug number here, perhaps a few of us can subscribe and/or vote on the issue.

Peter

I am thinking this is a bug and I will attempt to submit to Adobe.

    else if (typeType == CustomXMLEncoder.ARRAY_TYPE) {
        var numMembers:uint = obj.length;
        // var itemQName:QName = new QName("", "item");

        for (var i:uint = 0; i  0) {
            var className:String = classNameArray[classNameArray.length -1];
            itemQName = new QName("", className.toLowerCase());
        }
        else itemQName = new QName("", "item");

        encodeValue(obj[i], itemQName, myElement);
    }
}

I am using the SimpleXMLEncoder class to covert objects to XML. All is well except when dealing with types Array Collections.

If I have an array of objects who themselves have their own types, the XML that the encoder generates refers to them as items, not the type of the object within the array. Here is an example.
Here is what I want….

<campuses>
<campuse_choice>
     properties here......
</campuse_choice>
</campuses>

What i get is…..

<campuses>
     <list>
       <source>
           <item>
           </item>
       </source>
     </list>
</campuses>

I looked in the SimpleXMLEncoder class and I see the line (152) that converts them to items. any way to have it pick up the object types within the ArrayCollection?

The above example that I gave will output

1.A
1.B

What’s up with the double item? How can this be removed?

Andrew.

…this worked for me:

arrColl[i]=[
{
c1: “1.A”,
c2: “1.B”
}
];

How would we provide titles inside the Array (similar to your mx:object c=”?”) so it doesn’t simple output “item” for each XML tag.

Thanks again for the great example.

Can you please log a bug with a simple, reproducible test case at http://bugs.adobe.com/flex/.

Also, if you post the bug number here, I can vote on the issue and subscribe.

Thanks,
Peter