when the page is first loaded, by default the sorting is in ascending order and when we select the row in first table same id is passed to get the data for below table. Now when we click on the header to display the records in descending order (now the order of the record has changed) and then click on same record the id that is passed to the below table to fetch the values is different, it is the id of the record that was there at that position before sorting (before click on header). This means the order of the record is changed just to display them on the screen while the id associated with them are not changing.

Please find the implementation

function doSelectRow(event, target) {
YAHOO.mydata.myDataTable.onEventSelectRow(event, target);
var selectedRows = YAHOO.mydata.myDataTable.getSelectedRows();
var record = YAHOO.mydata.myDataTable.getRecord(selectedRows[0]);

Dom.get(“assignedobjects”).innerHTML = “Please wait. Searching…“;
getDataforBelowTable(record.getData(“myid”));

}

Not all the fields read by the DataSource need to be listed in the columns definitions, you can list one of the fields to be combined and not the other. In the one listed, you can assign a formatter function which has access to the Record object for that row and can read any value of any field read in by the DataSource, regardless whether it is listed in the Column Defs or not.

For JSON data, as used in the example, the DataSource locates the node containing the repeating items, the records, through the responseSchema.resultsList property. Anything in other branches of the JSON data structure will be ignored.

Once at that node, it will expect to find an array and it will loop through its contents. For each item it finds, it will look for the values of the properties you name in the responseSchema.fields array.

For JSON data you have to be quite explicit in how to reach each piece of data giving the full path to each value. For XML data, it expects a single name for each value in the fields array and it will locate any value inside a tag of that name no matter how deep or, failing that, an attribute with that name. For HTML, plain arrays or text formats, it will pick fields in the order in which the fields are listed.

For the example using SalesDatabase, all the information is a nested property of that item – the metadata and the actual results.

But for your examples, you don’t give a top-level property. You say this:

DataSource will also completely ignore the replyCode and replyText properties: It will go straight for the data. Nevertheless, if it doesn’t find the data element, it will trigger a responseParseEvent event, where you should be able to get to the full response and find out why did it fail.

Does the DataSource look for the first array that is a value of a property, or MUST you specify what the top level property is, and then have the actual data be the value of a sub-property?

Hope this makes sense, I’m not stating this too well. It may be I don’t understand how JSON is structured enough to know the answer.

This is a fine place to discuss conceptual issues but when asking for specific solutions to particular problems, the YAHOO Developers Network – JavaScript list at: http://tech.groups.yahoo.com/group/ydn-javascript/ is a much better place to deal with them.

The DataTable does not keep any reference to the original source of data, whatever it might have been. All changes are done on its internal RecordSet. It is up to you to send those updates elsewhere. The DataTable will only help in notifying you that changes have been made and telling you what they are. How you process them is up to you.