Hi Anirban,

Is ISSUE-1 anything close to the scenario described here: http://forums.infragistics.com/forums/t/64875.aspx ? Note that as I have already described in that forum thread, add/delete nodes is not supported directly in the tree, but you can be doing it in the data source.

About ISSUE-2. Loading XML directly by the browser and using it as a data source is not a very common scenario as it is not consistently (if at all) supported by all browsers. Some browsers like IE throw security warnings, other directly block it. That is the reason for us to have better support for XML loaded on the server and delivered to the client in the form of JSON (this is the MVC XML scenarion sampled here: http://samples.infragistics.com/jquery/tree/binding-in-asp-net-mvc).

Is it a must for your purpose to have XML directly loaded on the client? XMLs are usually large and this is slow even in terms of loading time.

Looking forward to your reply!

ISSUE-1 : Before I even add a node, how do I get to show the nodes in the tree when I do not know the level. The tree can have upto nth levvel. In that case what will be my approach to show the data. In all the examples shown, I see the child level is always known. In the example discussed http://forums.infragistics.com/forums/t/64875.aspx, what if I have a node like this : 

• Node 100
  • Node 010
    • Node 001
    • Node 002
    • Node 003
      • Node003_1
        • Node003_1_1
        • …
          • …..
  • Node 020
• Node 200
• Node 300
• ..

Now to show the data on on demand basis, what will be my approach to create the nodes on an on-demand basis ? 

 I am now returning Json data back to the client.  i.e. initially I load Node 100, Node 200 & Node 300. Now when user expands Node 100, I need to go and fetch the nodes Node001, Node002 & Node003. Again when user expands Node 003, since it has child nodes, I need to fetch Node003_1 and so on. Here, I do not know upto what level this can go. 

Please note, the structure, I am returning on to the client, the parent node has a property where it always  knows whether it has a child node or not. 

I understand your point how to add Nodes to the level but that is coming at a later point. 

Hope I am able to put my scenario. Please let me know your thoughts on this.

The documentation for the igTree is located here: http://help.infragistics.com/NetAdvantage/jQuery/Current/CLR4.0

To explain the request parameters when data is requested:

Consider the following example from the samples page:
http://samples.infragistics.local/jquery/tree/TreeGetData?controller=LoadOnDemand&sampleid=90d937ec-1b9e-44a8-ab0a-83de0fc8be31&path=ProductCategoryID:1/@ProductSubcategories&binding=textKey:Name,primaryKey:ProductSubcategoryID,childDataProperty:Products&depth=0&_=1323938304689

The first parameter is the path which contains of the primary key of the currently being expanded node and the child data field that we're requesting. They can be split by '/' and the first part of the string can be split by ':' in order to get the name/value of the primary key. The @Attribute at the second part of the string is the field we're looking for in the table.

In this example you can see the path to the node, including the property name of where children are located, the binding and the depth at which the node is located. The ProductCategoryID is the primary key and 1 is its value. ProductSubcategories is the name of the field to be extracted for child data, so the Linq query produced is SELECT ProductSubcategories FROM Context WHERE ProductCategoryID == 1. ProductSubcategories are needed in the request string because the table may contain other child level data references such as ProductInventories and we should be able to distinguish between them.
 

I hope this helps!

I found something very similar which I would like to share so that the same can be incorporated in the igTree . http://wwwendt.de/tech/dynatree/doc/samples.html

Thanks,

Anirban