I am running into some strange behaviors when using the dnn.xml and dnn.xml.jsparse namespaces.  Let me first start with the dnn.xml.  In my javascript, this is the code:

var oDoc = new dnn.xml;
oDoc.init(); 
oDoc.loadXml(sXml);

 When the init function runs, it determines the browser type and loads the appropriate parsing type (i.e. MSXML, DOMParse, or JsParse).  The challenge I run into is that with IE, the parserName is "ActiveX", so it will load the default JsParse and error out because the dnn.xml.jsparse namespace has not been loaded.  By writing this, it has got me thinking again.  This maybe associated with the order of loading the dnn.xmlhttp and the dnn.xml namespace.  So, I will continue the next strange behavior I am experiencing.

When using the dnn.xml.jsparse namespace with Opera browser, I am not getting the data out of the nodes as expected.  For example, with IE and Firefox, I can use this javascript to retrieve data from a xml document.  Here is the code:

//this is the xml string
<name>
   <test>
      <1>One</1>
   </test>
</name>

// this is the javascript
var oDoc = new dnn.xml;
oDoc.loadXml(sXml);

var myvariable = oDoc.childNodes(0).childNodes(0).getXml();

//results of code
The result of myvariable is "One" for IE and Firefox.  For Opera, whom uses the jsparse, returns "".  After futher investigation, I have discoverd the problem.  This is the structure of the xml doc:

<name>
   <test>
      <1></1>
      One
   </test>
</name>

So, my impression that the jsparse is not populating the oDoc object in the same manner that the ActiveX or DomParser is performing.  I have started rewriting the jsparse, but then I stopped to get input if I am barking up the wrong tree.  I haven't seen any talk with regards to the actual namespaces and actual usage.  I am hoping this note could spawn some dialog. 

Thanks,

Phebous...

Jon,

Thank you for your time and input.   I apologized about the post and some of the appalling mistakes I made.  I was writing from memory and not while sitting in front of the code.  I will wait to make any further comments about the code until I am in front of it.  I am glad to hear that the order was taken into account when loading in the namespaces. 

You have made one comment that has me worried.  It was: "you should note that the xml you are testing is not valid as well."  I am wondering if you were referencing the this:

//this is the xml string
<name>
   <test>
      <1>One</1>
   </test>
</name>

Or the mangled representation of the Opera oDoc objects that you are referencing?  If this testing xml above is not compliant, can you tell me what is wrong?

I will also perform some testing on some un-nested elements to determine if it is a nesting of the elements is the problem.  I figured that, since you have done some beautiful recursion, it would deal with the nesting of elements.  I was, however, a bit puzzled though with dealing of the xml in a character by character manner.  I would have used a regular expression more line of something like this:  <(.*)>(.*)<\/\1>.   I am sure that if I were a bit more versed in xml and all the rules around it, I could craft a regular expression to deal with it in a much easier manner.  I do know that this worked for me since I wrote my own parser to work around my parsing problem with the api. 

Regards,

Phebous...

P.S.  Thanks for all your hard work on DNN!