Just a quick follow-up on my previous message.  After a little more testing, I find the reported problem with the out-of-order colModel:cellattr function parameters occurs when adding rows.  The problem does not occur when sorting rows!! 

That is, after calling 'addRowData', the parameters passed to the colModel:cellattr function have 'rowId' as the third parameter and 'rawObject' as the first parameter...

After clicking on a grid column header to sort a column, the parameters passed to the colModel:cellattr function are in the order specified in the on-line documentation - 'rowId' as the first parameter and 'rawObject' as the third parameter.

Please note: in both cases, adding rows and sorting columns, the 'rdata' parameter is a boolean value.  This value does not appear to agree with the on-line documentation's description of the parameter.

Hope this helps and thanks again...

Best,

Brian

Many thanks for your prompt response.  Please find attached a short description of the issue and, as 'proof' an image of the Visual Studio 2010 debugger showing the function parameter values...

Thanks again to you and your associates for a terrific job on jqGrid!!

Best,

Brian

Re: jqGrid v4.1.1 - 2011-06-19

On-line documentation for colModel option: cellattr lists:

 - rowId as the first parameter

 - rawObject as the third parameter

 - rdata as the last parameter and describes it as 'the data row which will be inserted in the row'

When I make a jqGrid and debug the parameters passed to the cellattr function:

 - the first parameters is a JavaScript object

 - the third parameter is a rowId

 - the last parameter is the boolean value: true

Which is correct, the documentation or the v4.1.1 code?  If the documentation is correct is a code fix available in a later jqGrid release or is a fix scheduled for release?

Many thanks in advance for your help!!

Brian

Sorry for my mystake. I have deleted not the right post.

Also, please post it again.

Sorry.

Tony

P.S. I appeciate your description – Thank you, but just to know that only speaking that some feature work not as you expected

should be proof

Just added a topic (with the same name as this one) to the 'Discussion' forum.  Since it describes what might be a bug it probably belongs in the 'Bugs' forum.

Sorry for the confusion...

Brian