Best practice for moving between multiple editors?

RoyceSimpson

I've got three feature layers (point, line and polygon) that I'd like to split out into three different editors so that I can have each editor in it's own tab in a Tab Navigator.  I've implemented the following mxml:


<mx:TabNavigator id="tabNavigator" width="251" height="347" x="10" y="10" change="setEditor(event)"
      creationPolicy="auto">
  <s:NavigatorContent id="pointsTab" label="Points" width="100%" height="100%">
   <esri:Editor id="pointsEditor" height="100%" width="100%" map="{map}" toolbarVisible="true" featureLayers="{[map.getLayer('TypePointsLayer') as FeatureLayer]}"/>
  </s:NavigatorContent>
  <s:NavigatorContent id="linesTab" label="Lines" width="100%" height="100%" >
   <esri:Editor id="linesEditor" height="100%" width="100%" map="{map}" toolbarVisible="true" featureLayers="{[map.getLayer('TypeLinesLayer') as FeatureLayer]}"/>
  </s:NavigatorContent>
  <s:NavigatorContent id="areasTab" label="Areas" width="100%" height="100%" >
   <esri:Editor id="areasEditor" height="100%" width="100%" map="{map}" toolbarVisible="true" featureLayers="{[map.getLayer('TypePolysLayer') as FeatureLayer]}"
       geometryService="{new GeometryService(_geometryURL)}"/>
  </s:NavigatorContent>
</mx:TabNavigator>

The problem is, as I go from one tab to the next, each editor becomes active and editing mayhem kicks in.  So, I added a "change event" to the tab navigator with the following code:

   private function setEditor(event:IndexChangedEvent):void
   {
    for each (var layer:Layer in map.layers)
    {
     if (layer is FeatureLayer)
     {
      var fl:FeatureLayer = layer as FeatureLayer;
      if (fl.selectedFeatures.length > 0)
      {
       fl.clearSelection();
      }
     }
    }
   
    var i:int = event.target.selectedIndex;   
    switch(i)
    {
     case 0:
     {
      if(linesEditor)
      {
       linesEditor.editTool.deactivate();
       linesEditor.templatePicker.clearSelection();
      }
      if(areasEditor)
      {
       areasEditor.editTool.deactivate();
       areasEditor.templatePicker.clearSelection();
      }
      break;
     }
     case 1:
     {
      if(pointsEditor)
      {
       pointsEditor.editTool.deactivate();
       pointsEditor.templatePicker.clearSelection();
      
      }
     
      if(areasEditor)
      {
       areasEditor.editTool.deactivate();
       areasEditor.templatePicker.clearSelection();
      }
      break;
     }
     case 2:
     {
      if(pointsEditor)
      {
       pointsEditor.editTool.deactivate();
       pointsEditor.templatePicker.clearSelection();
      }
      if(linesEditor)
      {
       linesEditor.editTool.deactivate();
       linesEditor.templatePicker.clearSelection();
      }
      break;
     }
    }
   }

This doesn't work for some reason.  The "deactivate" method on the editTool for the editors that I'd like to "disable" don't seem to do any deactivating.  For example if I have the "line editor" tab active, I'd like to deactivate the "point editor edittool" and "poly editor edittool" so that the user can't click/select/edit existing points or polys.

Maybe there is a better approach to doing this.

Thanks for any advice.
-r

SarthakDatt

Yes it gives you a more fine grain control...however maintaining several instances of the editor would always be a pain. I know, in real world an app wont have that many featureLayers...but still.

You could follow the skinning approach. An example would be when you are working with one feature Layer...you could disable/hide the other or show a message so that users could not/wont use it. The default skin(TemplatePickerSkin) does something like that now if a featureLayer is invisible/out of scale range.

Having said that, from an API perspective, I think it would be better if we had something like a clear() or destroy() on the editor, so that users wont have to worry about cleaning it up themselves.

--Sarthak Datt