Because you are using server-side processing. With server-side processing, whenever DataTables does a draw, it will always make a request to the server (or whatever override function is used) for the data with which to draw the page.
I missed the fact that you need the serverSide option.
What's the difference between implementing the ajax in the client and using a server ajax with the serverSide=false, why do we need our table to be redrawn?
Why can't we mark the serverSide as false?
That is the default - you should be able to mark it as false (although it is redundant since that is the default). In the configuration shown above you have serverSide:true. I'm not sure if you need server-side processing or not. Normally you would only need it if you are working with tens of thousands of rows. Full details about it are in the manual.
The ajax option should be called (as a function) on first loading of the table, if it is provided. It would only then be called again with ajax.reload() (assuming client-side processing now).
you are correct of course, we had an exception because of our ajax implementation.
Now we have a strange behavior that after receiving the answer from the server (containing the row that was updated) and using it as the data for the successCallback
successCallback({data:result.aaData});
The updated row disappears from the table, any idea as to what can cause this?
This is in Editor's success callback now is it? The edited row would disappear if the JSON response doesn't include the information about that row. See the client / server communication for Editor.
we just need to pass data back from the server, and that's what we do.
I debugged it a bit and row.any() returned true so row.remove(); was called in the editor commit.
action = "edit"
identifier = {"sizzle1470139851858":{"parentNode":[3415,134,false]}}
store={"rowIds":["5"]}
data=our data.
Were we suppose the row in some way? are we suppose to return something in the data that should do it?
Can you help me understanding what to check?
Ok! I think the problem is that we are missing the DT_RowId from the returning value.
If we put the value of the edited line in the DT_RowId of the data we get a new phenomena.
We are using interceptors to set the cells renderFunc like so:
renderFunc.function = function (data, type, row, meta)
In the regular way we get information in the data, but when using the above concept we get data=null, do we need to change the way we work or are we missing something in our data when calling the successCallback.
Answers
Because you are using server-side processing. With server-side processing, whenever DataTables does a draw, it will always make a request to the server (or whatever override function is used) for the data with which to draw the page.
Allan
Hi Allan,
Sorry, but I don't understand what's the difference with this example.
In that example, you have
Which is a server side processing, but you don't rerender the entire table, do you?
Thanks,
Yishay
It doesn't use server-side processing (the
serverSideoption is not used in that example).It is used here, and you'll be able to see after an edit, all of the rows in the
tbodyare redrawn.Allan
Hi allan,
I missed the fact that you need the serverSide option.
What's the difference between implementing the ajax in the client and using a server ajax with the serverSide=false, why do we need our table to be redrawn?
Why can't we mark the serverSide as false?
Thanks,
Yishay
That is the default - you should be able to mark it as false (although it is redundant since that is the default). In the configuration shown above you have
serverSide:true. I'm not sure if you need server-side processing or not. Normally you would only need it if you are working with tens of thousands of rows. Full details about it are in the manual.Allan
Hi Allan,
Sorry, I wasn't clear.
We marked the field as false, but than our ajax function was not called.
Thanks,
Yishay
The
ajaxoption should be called (as a function) on first loading of the table, if it is provided. It would only then be called again withajax.reload()(assuming client-side processing now).Allan
Hi Allan,
you are correct of course, we had an exception because of our ajax implementation.
Now we have a strange behavior that after receiving the answer from the server (containing the row that was updated) and using it as the data for the successCallback
The updated row disappears from the table, any idea as to what can cause this?
Many many thanks,
Yishay
This is in Editor's success callback now is it? The edited row would disappear if the JSON response doesn't include the information about that row. See the client / server communication for Editor.
Allan
Hi Allan,
we just need to pass data back from the server, and that's what we do.
I debugged it a bit and row.any() returned true so row.remove(); was called in the editor commit.
action = "edit"
identifier = {"sizzle1470139851858":{"parentNode":[3415,134,false]}}
store={"rowIds":["5"]}
data=our data.
Were we suppose the row in some way? are we suppose to return something in the data that should do it?
Can you help me understanding what to check?
Many thanks,
Yishay
Ok! I think the problem is that we are missing the DT_RowId from the returning value.
If we put the value of the edited line in the DT_RowId of the data we get a new phenomena.
We are using interceptors to set the cells renderFunc like so:
renderFunc.function = function (data, type, row, meta)
In the regular way we get information in the data, but when using the above concept we get data=null, do we need to change the way we work or are we missing something in our data when calling the successCallback.
Thanks,
Yishay
Hi Allan,
We found the solutions to our problems.
Many thanks for help along the way,
Yishay
Can you show me the data that you are passing to the success callback please?
Allan
Hi Allan,
sure, did you see that I wrote we overcome the issues we had?
Thanks,
Yishay
Sorry I missed that - I had the thread open in a tab and we must of overlapped.
Great to hear you've got it up and running now!
Regards,
Allan