How to dispose Gluon PopupView? - gluon

I have a ChoiceFieldComponent which shows a PopupView when it is clicked on. The content of the PopupView is a custom ChoiceMenu showing a list of cocktail ingredients. As the number of ingredients for a recipe varies, it is possible to add / remove a ChoiceField.
When a ChoiceField is removed from its parent View, there seems to remain a strong reference, which prevents the ChoiceField and the ChoiceMenu from getting garbage collected.
I think this is happening because PopupView holds a reference to both nodes:
popup = new PopupView(getSkinnable(), choiceMenu);
I have tried to remove the reference to choiceMenu by:
BorderPane container = (BorderPane) popup.lookup(".container");
container.setCenter(null);
But there is still the reference to PopupView.ownerNode (ChoiceField), which I can not clear.
Because the number of PopupViews is increasing with every ChoiceField added, I wonder if there is a way to remove a PopupView.

Related

reached limit cannot create any more controls for this form vb6

I am not able to create a new menu in my mdi form.As already so many existing menus are there.
Its giving error "reached limit cannot create any more controls for this form".
Please help me to know that how to add new menu with this error.
You can use control arrays for your menus to overcome 256 controls per form limit. In Menu Editor you have to set Index property to an unique integer value to create control arrays of entries with same Names.
A common strategy is to designate mnuMain name for a control array with top menus i.e. first mnuMain(1) would be "File", then mnuMain(2) would be "Edit", etc.
Then in form's code declare an enum like this
Private Enum MenuIndexesEnum
idxFile = 1
idxEdit
idxTool
....
End Enum
and use it throughout code like mnuMain(idxFile) etc.
For "File" sub-menu designate mnuFile control array with unique indexes starting from 1 for "New", "Open", "Print", etc. "Exit" and extend the MenuIndexesEnum enum like this
Private Enum MenuIndexesEnum
idxFile = 1
idxEdit
idxTool
....
idxNew = 1
idxOpen
idxPrint
idxExit = 99
...
End Enum
Then continue with mnuEdit for "Edit" sub-menu, etc.
As #wqw said, the problem is that you've reached the 256 controls-per-form limit, and the solution is to start wrapping them up into control arrays. However, the control you're having an issue with (a menu) isn't necessarily the one you need to make into a control array.
I find that the most insidious, yet easiest to solve, control "consumers" are the label controls sprinkled everywhere on a typical form. To turn those into a control array, I just adopted the practice of copying and pasting an existing label anytime I need a new one; the first time, VB will ask if you want to create a control array (say "Yes"), and thereafter it will automatically increment the index for you every time you create a new copy of the label.
For me, label controls are the most convenient to make into an array, because there's usually no code associated with them, and hence no need to worry about the index at all.

How to tell if an element is a Widget? (CKEditor)

Per CKEditor, initialize widget added with insertElement, we are doing an insertElement() and then initializing with initOn(). The problem is that some of the elements we are inserting are not supposed to be widgets and initOn() makes them widgets and the context menu doesn't work right. I am having trouble finding any properties inside the item/element to tell if something is/is not a widget so I can then call initOn().
Cross-posted downstream on Drupal.org here https://www.drupal.org/node/2466297
First of all - which element do you mean?
(Note: In this section I am assuming that a widget was correctly and fully initialised.)
Widget element
A widget can obviously consists of many elements. One of them is called the "widget element" and this is the element which you "upcasted" and which you can later access through widget.element.
Since CKEditor 4.5.0 there will be such method available:
Widget.isDomWidgetElement = function( node ) {
return node.type == CKEDITOR.NODE_ELEMENT && node.hasAttribute( 'data-widget' );
};
You can of course already use this code to check if a given node is a widget element.
Widget wrapper
Second important element is the widget's wrapper. It is created during data processing if a widget element was marked to be upcasted or when initOn() is called if the widget element wasn't wrapped yet. You can access this element through the widget.wrapper property.
Since CKEditor 4.5.0 there will be a following method available:
Widget.isDomWidgetWrapper = function( node ) {
return node.type == CKEDITOR.NODE_ELEMENT && node.hasAttribute( 'data-cke-widget-wrapper' );
};
And again - you can use this code already.
Important note here - since you mention insertElemet() in your question. As I explained in CKEditor, initialize widget added with insertElement editor#insertElement() does not trigger data processing. Therefore, element that you insert is inserted as is. This means that the widget wrapper is not created during insertion and will be created once you call initOn().
Finding widgets by any element
Many times you want to find a widget instance by some element that you have (any element that can be inside a widget). There's a useful method for that: getByElement().
What should become a widget? Aka - how to deal with editor.insertElement()?
You mentioned that you use editor.insertElement() and that you don't know which elements are supposed to be widgets. This should never happen. editor.insertElement() is a quite low level method which will not do all the data processing and upcasting magic which editor.insertHtml() does. It means that it is supposed to be used in a different case - when you want to insert exactly the element that you have.
For instance, your table plugin is building a table structure to be inserted into editor. You know that the table is empty, so you control every bit of it (other plugins should not interfere here). It is also important that it's the table's plugin decision, not e.g. a template's plugin decision. The table's plugin control the table feature, while the template plugin only uses tables. So in such case, when you have a full control, you can use editor.insertElement(). Then you always know what you insert and what is supposed to become a widget.
In all other scenarios you should use editor.insertHtml(), so the whole data processing layer is triggered. Thanks to it other features like the widgets system, the link plugin (which turns empty anchors into fake objects), etc. can prepare the data that you insert to be fully editable and integrated.
Tl;dr
If your plugin knows what it does, it can use editor.insertElement(), but since it knows what it does it will know which inserted element must become a widget.
If your plugin does not fully control the situation, then you should use the editor.isertHtml() method which is far more automated and will turn proper elements into widgets based on the upcast callbacks.

EmberJS: programmatically adding a child view to a collection view without changing underlying content

My EmberJS app is confusing me a lot at the moment. I have a collection view, that in turn defines an itemViewClass of a custom view I have defined in my code. Something like:
App.CarouselView = Ember.CollectionView.extend({
itemViewClass: App.SlideView.extend(),
});
And this CarouselView is rendered inside a template that has a dynamic segment backing it (I hope this makes sense?) . The controller for these dynamic segment is an array controller because the model for these dynamic segments is a collection :) (more confusion, please let me know)
App.SlidesController = Ember.ArrayController.extend();
By now all of you have figured that I am basically rendering a bunch of slides inside of a carousel. And these are dynamically backed in the collectionView by setting the property
contentBinding:'controller' // property set in CarouselView, controller corresponds to SlidesController
The confusion begins now. I want to add a slide to the existing set of slides. So I provide a button with an action : 'add' target='view'
In the SlidesView,
actions:{
add: function(){
var carouselView = this.get('childViews')[0];
// SlidesView has carouselView and another view as it's child, hence this.get('childViews')[0] is carouselView
var newCard = carouselView.createChildView(App.SlideView.extend());
carouselView.get('childViews').pushObject(newCard);
}
}
The above piece of code sucks and hurts me bad. I want to basically add a new SlideView to my CarouselView collection programmatically upon a button trigger. But apparently Ember recommends that childViews should not be manipulated directly and I ought to instead change the underlying content.
It states in my console.log that manipulating childViews is deprecated etc.
So essentially I need to add something to my content to my SlidesController content ? However, I don't want to add something to the content, this is just a soft add, that is providing the user with a slide so that he may choose to edit or add something if he wants to. He can always discard it. A hard add that will require me to persist the new slide to the DB will come once the user decides to save this information.

MvvmCross : databinding programmatically

I think I need to bind data programmatically to solve my problem.
I use a TabHost which hosts 2 Tabs.
I need to load the MvxBindableListView in the second tab when TabHost appears and keep the first tab as default tab.
What I'm doing is starting the second tab activity without problem because I check the process using this code:
protected override void OnViewModelSet()
{
SetContentView(Resource.Layout.Page_ActivityView);
System.Diagnostics.Debug.WriteLine("activityView started");
MvxBindableListView mvxBindableListView = FindViewById<MvxBindableListView>(Resource.Id.mvxBindableListView);
mvxBindableListView.ChildViewAdded += new System.EventHandler<Android.Views.ViewGroup.ChildViewAddedEventArgs>(mvxBindableListView_ChildViewAdded);
}
"activityView started" is displayed in output debugger, but MvxBindableListView.ChildViewAdded event isn't raised, only when I click the second tab.
So I suppose that MvxBindableListView is not databound.
Thanks in advance to help me loading my second tab programmatically.
I think ChildViewAdded is an event that occurs when the ListView is rendered - at that time when it needs to 'draw list items', then it will ask its adapter for child Views for the screen. As you scroll down the list, it will then ask for more child views - but it will also reuse views - so for a simple list you should only ever (hopefully) get N+1 calls on ChildViewAdded for a list which shows N items at one time.
So it's perfectly possible for a list to be databound but never to call ChildViewAdded - that won't get called until the list is 'drawn'
Sadly the Xamarin docs aren't helpful here - http://docs.mono-android.net/monodoc.ashx?link=E%3AAndroid.Views.ViewGroup.ChildViewAdded
Note: if you do actually want to bind programatically, then you can do this to - using Bind() methods and extension methods. However, I haven't expanded on that here - as it doesn't sound like that's actually what you need!

wp7: navigating dynamic pivotitems in a pivot

I have a pivot page, with one fixed pivotitem, and depending on data, a dynamic number of additional pivotitems. The fixed pivotitem keeps a hyperlink list of the new pivotitems created, and the idea is to click any of the links and navigate to that pivotitem.
It would look something like this:
FixedItem | DynamicItem1 | DynamicItem2
link: DynamicItem1
link: DynamicItem2
The problem I am facing is with the hyperlink click, it doesn't take me to the respective pivotitem, but instead takes me to the pivot page with no dynamic pivotitems. I am using the following code for navigation:
hyperlink.NavigateUri = new Uri("/MainPage.xaml?name=" + p.name, UriKind.Relative);
where p.name is the name of the pivotitem. I am not sure if this is the right syntax, but what's confusing is that all the created pivotitems get lost, leaving only the fixeditem - as if it was opening a new instance of the pivot page.
Then, in the OnNavigatedTo I tried the following:
if (NavigationContext.QueryString.ContainsKey("name"))
{
// URI is '/page?name=PivotItemToSelect'.
string selectedPivotItem = e.Uri.Query.Split('=').Last();
// Match PivotItemToSelect with the PivotItem's Name.
PivotItem pivotItemToShow = pivot.Items.Cast<PivotItem>().Single(i => i.Name == selectedPivotItem);
pivot.SelectedItem = pivotItemToShow;
base.OnNavigatedTo(e);
}
On the first line, I get an exception that this operation cannot be done on a "relative" URI. Any other way I could that information?
If I change the pivotitem name to a number, say i, and pass that i as the index, I got the following code to work as the _click method of the hyperlinkbutton - but in real usage the name will most likely be a text string:
pivot.SelectedIndex = int.Parse(i);
I am not sure how totally off the track I am, and would appreciate any pointers in the right direction.
Thanks.
The correct way to access the QueryString in your OnNavigatedTo event is like this:
string selectedPivotItem = this.NavigationContext.QueryString["name"];
Using a Pivot control for your navigation like this seems very strange, but since I know nothing about the context of your app, I'll assume you have chosen it for a reason.
If you are designing an app with a dynamic list of items to navigate to, it would be better to have a single page for your link list and a single detail page that can use the query string to bind to the specific object you are wanting. This will perform better and make more sense to your users. But again, I don't know the context of your app or your reasoning, so that's up to you! :)

Resources