Is there any plan to provide WEB API server wrap? If so, we can use it in web and mobile at the same time.

Now that jsbin supports Typescript, putting Kendo's typescript definitions in a public place would be helpful.  Demonstrating the power of Typescript and Kendo together in jsbin is a powerful combination for answering questions and teaching!

Add new attribute that the Validator will detect and skip validation on that element/widget.

Eg; skip-validation="true"
<input type="text" required />

and then at run time you could toggle the skip-validation attribute.

<input type="text" required skip-validation="true" />

This would be perfect for complex forms where you are togging the visibility of div container of controls depending on what options that user selects in the form. 

The problem is at the moment, the validation will validate controls that are not visible.  Having a skip-validation attribute would solve this problem, as when you hide the div container you cold also toggle the skip-validation attribute on it's input controls.

Regards,

Scott

At least, including:

1. Tab stop for mobilebutton, mobilebuttongroup, mobileswitch.
2. Selection support for mobilebutton (spacebar), mobileswitch (spacebar/left/right), mobilebuttongroup (left/right/up/down)

MobileButtonGroup and MobileSwitch are elegant replacements for radio and checkbox inputs, but they need keyboard support to improve desktop accessibility.

In Kendo Web - Mobile as ListView - search field should not be scrollable, only listings should be scrollable.

If user scrolls down couple of pages down and now he wants improvise the search, then again he has to scroll back to type in search field. If only listing scrollable then search field is always visible where user can type whenever wanted.
I dont know how this is missed in the current release. This is very basic.

I need homogenous data items for multilevels in your tree control.
Most data structures coming from our services are not homogenous.
We use the following to create a homogenous structure for the tree.
Maybe you could incorporate something like this - note that we use angular, and I use their $parse service in my implementation:

// CALLER:

          var levelDescriptors = [
            { item: "functionGroup", items: "functions" },
            { item: "this", items: null }];
          itemHierarchy = Portal.buildItemHierarchy(
            functionGroups,
            levelDescriptors,
            $parse);

// UTIL CODE:

    buildItemHierarchy: function(hierarchicalItems, levelDescriptors, $parse) {
      var hierarchicalDataItems = [];
      if(Portal.isDefined(hierarchicalItems)) {
        hierarchicalDataItems = Portal.createHierarchicalDataItems(0, hierarchicalItems, levelDescriptors, $parse);
      }
      return hierarchicalDataItems;
    },

    createHierarchicalDataItems: function(level, levelItems, levelDescriptors, $parse) {
      var levelDataItems = [];
      if(Portal.isDefined(levelItems)) {
        if(Portal.isDefined(levelDescriptors)) {
          if(level < levelDescriptors.length) {
            for(var i=0; i<levelItems.length; i++) {
              // extract datasource for current level.
              // create an associated node data item.
              var levelItem = levelItems[i];
              var levelDescriptor = levelDescriptors[level];
              var item = levelItem;
              if(levelDescriptor.item !== "this") {
                item = $parse(levelDescriptor.item)(levelItem);
              }
              if(Portal.isDefined(item)) {
                var levelDataItem = Portal.createNodeDataItem(item);
                levelDataItems.push(levelDataItem);
                // extract children for current level.
                // create associated (child) node data items.
                var items = $parse(levelDescriptor.items)(levelItem);
                if(Portal.isDefined(items)) {
                  var itemsLevel = level + 1;
                  levelDataItem.items = Portal.createHierarchicalDataItems(
                    itemsLevel,
                    items,
                    levelDescriptors,
                    $parse);
                }
              }
            }
          }
        }
      }
      return levelDataItems;
    },

    createNodeDataItem: function(item) {
      return {
        id: item.id,
        text: item.name,
        dataSource: item,
        expanded: false,
        items: []
      };
    },

I'm creating a FAQ for a client and I need the Question and Answers to be collabsable inside a listview but I also need to filter the list by keyword...  

I would like something similar to the Collapsible Lists in jQuery Mobile:  http://jquerymobile.com/demos/1.2.0/docs/lists/lists-collapsible.html 

Currently the actions in an ActionSheet can only execute functions defined in the global space as stated in the documentation here: http://docs.telerik.com/kendo-ui/mobile/actionsheet#executing-actions

"The callback can be either a function, or a method of a JavaScript object in the global scope."

I want to be able to specify a data-model attribute so that the callback for my actions can be functions defined in a kendo observable (more specifically the kendo observable used by my view).

This functionality would be similar to how data-model attributes work with modal views.

As it is I don't want to be defining functions in the global scope - I consider it sloppy and messy. Does it not makes sense to use the scope of the containing view's model?

For setters in general this shouldn't cause any trouble to return the widget so I could perform more things with it.

For the get-set functions (like value() etc..), if I get a parameter this means I'm using it as a setter hence I might want to do something else with the widget later on... "Chainability" jQuery style.

A small code snippet to illustrate:
//the datePicker is already in place and has been initialized somewhere else
//I would like to be able to do this:
datePicker = $("#aDatePicker") //jQuery Selector
  .getKendoDatePicker() //retrieve the picker
  .setOptions(dpOptions) //set new options
  .value(aDate); //set the current date

//Instead of this:
datepicker = $("#aDatePicker").getKendoDatePicker();
datepicker.setOptions(dpOptions);
datepicker.value(aDate);

Regards
Carlos A Luengas

When using inline/batch/popup editing for grids, there is a standard way to handle server-side validation errors (from the ModelState) on the client : http://blogs.telerik.com/kendoui/posts/13-08-29/handling-server-side-validation-errors-in-your-kendo-ui-grid

When using the Upload control, the proposed solution is to use XMLHttpRequest.status and XMLHttpRequest.statusText, which is currently not very reliable, and complicated to handle. It's also a non-standard way in ASP.Net MVC, and breaks existing best-practices for error handling.

There should be a way to handle ModelState errors on upload.

Hi there,

I would be great if you could pass the event name into event handlers, as we can't pass any additional args in, like we can with jquery events, so we can write generic event handlers that can detect the event type and handle it.

Regards,

Scott

It would be wonderful to have code highlighting and Intellisense for script blocks using the text/x-kendo-tmpl type in Visual Studio.

Currently you can change the type to text/html to get partial html syntax highlighting, but it does not highlight Kendo specific elements inside #: #. 

When a chart is exported from an application, the source is not very clear. It can be written in the author property, but that is not visible. 
I would like to be able to create page header/footer.
I can export a PDF page with custom layout using a DOM selector, but in that case the output does not includes paged data in for example a Gantt chart

There are times when it is not obvious that the Listview contains more records than are displayed. It would be nice to have an optional indicator that there are more records above or below the current scrolled view. Another option is to keep the scrollbar turned on.

When the chart is rendered, it would be nice if the various chart elements (SVG groups/paths/shapes/etc) were assigned class names so that we can further tweak the chart using javascript, d3.js, drawing api, etc. 

The addition of visual styles helps, but they are not available on all elements,

We need the ability to disable the keyboard or input read only for the combo box widget.  When we don't want a user to input their own data on the combo box currently there is no option to disable this function.

When setting the 'useNativeScrolling' to true in 'app.js', the rendering of a drawer is effected.  Additionally, trying to scroll on the drawer actually scrolls the view (which will be partially off the screen).

Some views will not need native scrolling so it would be beneficial to have the option to choose which views use native scrolling and which do not.

I am aware there is the data-use-native-scrolling attribute but this seems to have the same problem as 'useNavtiveScrolling' in 'app.js'.

The mobile application exposes an 'init' event when the application has initialised. This would be useful on all UI widgets, especially in circumstances when MVVM is not being used, for example to check whether other widgets are also initialised and perform actions accordingly.

When you have a ListView that can show thousands of records (presumably using a remote dataSource), it would be very useful to be able to start at a given point in the list, rather than having to start at the top of the list and scroll down thousands of records to get to the point you want.

This is currently possible to do when you have a ListView with set number of records that are all displayed at once by using the view's .scrollTo() function, but not possible when endlessScroll is on on the ListView.

Here is my original support ticket and the response I received regarding this matter:

<BEGIN MY SUPPORT TICKET>
I have a ListView which has a remote dataSource.  EndlessScroll is on in the ListView.  That part works as expected.  However, at times I need to programmatically "scrollTo" a particular point in the ListView.  To do this, I call .scrollTo() on the view's scroller (e.g. $('#MyView').data('kendoMobileView').scroller.scrollTo(0, -725)).  If I .scrollTo() a point that is too far down the page, sometimes the data doesn't show or only a couple of records/rows show and the "loading" icon is beneath them.  Here is an example of what I mean.

Given that we have a ListView with endlessScrolling on to which is attached a dataSource that reads from a remote url and has a pageSize of 50.  If we .scrollTo() about 25% down the list (approximately 13 rows down), everything works fine (i.e. after the .scrollTo() completes, we are at the row we want and all the records are there).  However, if instead of 25% we .scrollTo() about 60% down (approximately row 30), none or only some of the records with a spinning "loading" icon will be visible.

Based off of what I read here http://docs.telerik.com/kendo-ui/getting-started/mobile/listview#press-to-load-more-/-endless-scrolling, I believe what is happening is that since the ListView "adds" more of the records from the page that was retrieved from the dataSource as you scroll down the list when in virtual mode, if we .scrollTo() too far down the page, the ListView isn't smart enough to know to "add more" rows to the DOM and also make a call to the next page of the dataSource if it is necessary.  Basically, I believe ".scrollTo()" needs to trigger the ListView to add to the DOM and make a call to the next page of the dataSource the same way that scrolling with your finger would do.

I hope there is a solution/fix/hack to this because I have spent many hours on this.
<END MY SUPPORT TICKET>


<BEGIN TELERIK'S RESPONSE>
Thank you very much for the detailed information.

I am afraid that when the ListView is used in "endless scrolling" mode the scrollTo() method will not work as expected. The main idea behind the endless scrolling functionality is to load records on demand and load only the elements that the user wants to see, this is why the elements displayed are constantly changed as new items are loaded.

Currently there is not a workaround that I can offer. If you really want to see this functionality added in the future I would suggest you to check our uservoice section, where you can post this as a feature request, so it will be considered for a future implementation.
<END TELERIK'S RESPONSE>

In followup to my last statement about adding data to the validation events, Kendo UI's validator would be greatly, greatly, GREATLY augmented with an extremely simple function...

onErrorDisplay(target, message, invalid) {

}


The sole purpose of this function is to allow users to hook into it, and custom handle error display more finely. It is fairly simple to add, look at line 11924 of the kendo source file; At Line 11948, there is the code to show the error message. The onErrorDisplay function would allow for catching that and interpreting it differently if requested.