I have three tabs and each has a listbox with different types of files.
When I right-click on an item in the listbox, I want a ContextMenu with "New, Edit and Delete" as Item headers.
I guess I could have a ContextMenu for each listbox, and then have a seperate method for each header, such as:
<ListBox.ContextMenu>
<ContextMenu x:Name="NewEditDeleteAdvCalcFileContextMenu">
<MenuItem Name="NewAdv" Header="New" Click="NewAdv_Click" />
<MenuItem Name="EditAdv" Header="Edit" Click="EditAdv_Click"/>
<MenuItem Name="DeleteAdv" Header="Delete" Click="DeleteAdv_Click"/>
</ContextMenu>
</ListBox.ContextMenu>
But really, I hope there is a better way.
I saw this post which shows the ContextMenu as Static Resource
and this seems to be something I would like to do.
In the same thread it is suggested to use commands:
ContextMenu with Commands
and with that I'm hoping I can get the type of the ListBoxItem that was clicked, because I need that. A new file type B must be handled differently than a new file type C, but I don't want a gazillion contextmenus and New/Edit/Delete methods.
So, currently I have this higher up in my xaml file:
<UserControl.Resources>
<ContextMenu x:Key="NewEditDeleteContextMenu">
<MenuItem Header="New"
Command="{Binding Path=NewFileCommand}"
CommandTarget="{Binding RelativeSource={RelativeSource FindAncestor, AncestorType={x:Type ListBoxItem}}}"/>
<MenuItem Header="Edit"
Command="{Binding Path=EditFileCommand}"
CommandTarget="{Binding RelativeSource={RelativeSource FindAncestor, AncestorType={x:Type ListBoxItem}}}"/>
<MenuItem Header="Delete"
Command="{Binding Path=DeleteFileCommand}"
CommandTarget="{Binding RelativeSource={RelativeSource FindAncestor, AncestorType={x:Type ListBoxItem}}}"/>
</ContextMenu>
</UserControl.Resources>
And then a listbox in the tabItem:
<ListBox Name="CalcFilesListBox"
Margin="20" ItemsSource="{Binding CalcFilesList}"
PreviewMouseRightButtonUp="ListBox_PreviewMouseRightButtonUp"
ContextMenu="{StaticResource NewEditDeleteContextMenu}">
<ListBox.ItemTemplate>
<DataTemplate>
<StackPanel Orientation="Horizontal">
<TextBlock Text="{Binding Path=Name}" />
</StackPanel>
</DataTemplate>
</ListBox.ItemTemplate>
<ListBox.ItemContainerStyle>
<Style TargetType="{x:Type ListBoxItem}" BasedOn="{StaticResource {x:Type ListBoxItem}}">
<EventSetter Event="MouseDoubleClick" Handler="CalcFileListBox_MouseDoubleClick"/>
</Style>
</ListBox.ItemContainerStyle>
</ListBox>
Question #1
How do I get the rightclick of a ListBoxItem to show the ContextMenu, which is now a static resource?
Because in my xaml.cs I had this:
private void ListBox_PreviewMouseRightButtonUp(object sender, System.Windows.Input.MouseButtonEventArgs e)
{
// SelectItemOnRightClick(e);
NewEditDeleteContextMenu.PlacementTarget = sender as UIElement;
NewEditDeleteContextMenu.IsOpen = true;
}
But now I have an error saying:
The name 'NewEditDeleteContextMenu' does not exist in the current context.
because originally I had the contextmenu as part of the ListBox such as:
<ListBox.ContextMenu>
...
But as far as I could see that would mean a separate ContextMenu for each ListBox.
Question #2
Is the correct way to use a command, let's say NewFileCommand for the New item header in the ContextMenu (shown in the UserControl.Resources block of code) to do the following:
In my ViewModel:
public RelayCommand<string> NewFileCommand { get; private set; }
and then in the ViewModel's constructor:
public CalcViewModel()
{
NewFileCommand = new RelayCommand<object>(NewFile);
}
public void NewFile(object sender)
{
//Determine the type of file, based on the ListBoxItem's DataContext.
That is, supposing the ListBoxItem is the object being passed as the sender.
}
Basically, I want one ContextMenu bound to the different ListBox components, and this should pop up on a rightclick, and when for instance the New item is chosen on the ContextMenu, I want to determine the type of the file that has been bound to the ListBox.
E.g.: ListBox 1 is bound to a collection of file type B. ListBox 2 is bound to a collection of file type C. When I rightclick on an item in ListBox 2, and choose New, I need to make a new file of type C.
Question #3
This isn't a very intricate View. I haven't used a MVVM framework because so far I haven't thought that the time it would take me to learn one would be worth it, but considering this scenario, and a simpler case for a double-click on the ListBoxItems that can be seen in one of the blocks of code, would you recommend the use of a framework?
You're going in the right direction, you code just needs a bit of updating. First, don't need any right-click handlers -- if a control has a ContextMenu set, right-clicking will invoke that ContextMenu. Having a ContextMenu as a StaticResource and attaching it to multiple controls creates a bit of a problem because of a bug in .NET where a ContextMenu doesn't update its DataContext after initially setting it. That means if you first invoke the menu on listbox #2, you'll get the selected item in that listbox... but if you then invoke it on listbox #3, you'll still get the selected item in listbox #2. But there's a way around this.
First, let's look at the context menu and how it's bound to a list box:
<ContextMenu x:Key="contextMenu" DataContext="{Binding PlacementTarget, RelativeSource={RelativeSource Self}}">
<MenuItem Header="New" Command="{Binding DataContext.NewFileCommand}" CommandParameter="{Binding}"/>
<MenuItem Header="Delete" Command="{Binding DataContext.DeleteFileCommand}" CommandParameter="{Binding SelectedItem}"/>
</ContextMenu>
...
<ListBox Margin="10" ItemsSource="{Binding Files1}" ContextMenu="{StaticResource contextMenu}"/>
PlacementTarget is the control the ContextMenu is attached to. Explicitly binding the menu's data context to PlacementTarget ensures it's pointing to the correct ListBox every time it's invoked. Commands like "Edit" and "Delete" that deal with list items are then easy: Just bind the CommandParameter (not the CommandTarget as you did) to the ListBox's SelectedItem. The item you want to edit or delete will then be given as a parameter to the command.
Since you used RelayCommand I'm assuming you used GalaSoft's MVVM framework. In that case here's how your "Delete" command might look:
public RelayCommand<object> DeleteFileCommand { get; } = new RelayCommand<object>( DeleteFile_Executed, DeleteFile_CanExecute );
private static bool DeleteFile_CanExecute( object file )
{
return file != null;
}
private static void DeleteFile_Executed( object file )
{
var filetype = file.GetType();
System.Diagnostics.Debug.WriteLine( string.Format( "Deleting file {0} of type {1}", file, file.GetType() ) );
// if( filetype == typeof( FileTypeA ) ) DeleteFileTypeA( file as FileTypeA );
// else if( filetype == typeof( FileTypeB ) ) DeleteFileTypeB( file as FileTypeB );
// etc...
}
The "New" command will be a bit tricker because you want to be able to create a new item whether an item is selected or not. So we'll bind the CommandParameter to the ListBox itself. Unfortunately there's not a good way to get the type of item the ListBox contains. It could contain multiple types of items, or no items at all. You could give it an x:Name then look at the name in your command handler, but what I choose to do is put the type of item this ListBox handles as the Tag parameter of the ListBox. Tag is a bit of extra data you can use for whatever purpose you like:
<ListBox Margin="10" ItemsSource="{Binding Files1}" ContextMenu="{StaticResource contextMenu}" Tag="{x:Type local:FileTypeA}"/>
Now we can define our "New" command handlers like this:
private static bool NewFile_CanExecute( ListBox listbox ) { return true; }
private static void NewFile_Executed( ListBox listbox )
{
var filetype = listbox.Tag as Type;
System.Diagnostics.Debug.WriteLine( string.Format( "Creating new file of type {0}", filetype ) );
// if( filetype == typeof( FileTypeA ) ) CreateNewFileTypeA();
// else if( filetype == typeof( FileTypeB ) ) CreateNewFileTypeB();
// etc...
}
As for whether this scenario warrants an MVVM or not, you can certainly put your three file lists in a ViewModel, along with code that actually creates, edits, and deletes the files, and have your commands in the Window invoke the code in the ViewModel. I usually don't, though, until the scenario becomes more complicated.
Related
I am using WPF and I try to follow MVVM.
So I have a VM which has my Model as a property. The Model has a property which is a list of model1, which implements a command.
The reason why I don't want to move the command to the ViewModel, so I would have access to the view is that I don't know how many elements my list will have and I want to be sure that my command access its model1.
This commands also does some processing and I want the window, which holds the button binded to the command, to hide, during this processing.
How can I achieve this? Where should I look?
In cases like this, it is best if your ViewModel has the command, and it takes a parameter. That way, you will be passed the item that the user is trying to modify. So if you have a ItemsControl:
<ItemsControl ItemsSource="{Binding MyItems}">
<ItemsControl.ItemTemplate>
<DataTemplate>
<Button Content="{Binding Name, StringFormat=Push {0}}"
Command="{Binding DataContext.ItemPushedCommand, RelativeSource={RelativeSource AncestorType={x:Type ItemsControl}}}"
CommandParameter="{Binding}" />
</DataTemplate>
</ItemsControl.ItemTemplate>
</ItemsControl>
In your ViewModel, you would define your command like so (I'm using the DelegateCommand from prism, you can use whatever command you are comfortable with):
private readonly DelegateCommand<Model> itemPushedCommand;
public ICommand ItemPushedCommand { get { return itemPushedCommand; } }
public MyViewModel()
{
itemPushedCommand = new DelegateCommand<Model>(OnItemPushed);
}
private void OnItemPushed(Model item)
{
// your item has been pushed!
}
Is there a convention when using MVVM to bind the items of a ListBox to a ViewModel?
In the below XAML, I'm creating a ListBox of buttons. The ListBox is bound to an observable collection from my ViewModel. I then want to bind the button's Command property to an ICommand. The problem is that when I add that binding, I'm binding against the data object, not the ViewModel.
Do I just change the MyListOfDataObjects property to be a list of ViewModels? If so, where do I instantiate those new objects? I'd prefer to use dependency injection since they will have several dependencies. Do I change the GetData lambda?
In general: what's considered good practice here? I wasn't able to find any examples for this situation, although I assume it is rather common.
I'm using the MVVMLight framework, but I'm willing to look at any other frameworks.
<Window x:Class="KeyMaster.MainWindow"
DataContext="{Binding Main, Source={StaticResource Locator}}">
<Window.Resources>
<ResourceDictionary>
<DataTemplate x:Key="MyDataTemplate">
<Button Command="{Binding ButtonPressedCommand}"
CommandParameter="{Binding .}"
Content="{Binding Name}" />
</DataTemplate>
</ResourceDictionary>
</Window.Resources>
<Grid x:Name="LayoutRoot">
<ListBox ItemsSource="{Binding MyListOfDataObjects}"
ItemTemplate="{StaticResource MyDataTemplate}">
<ItemsControl.ItemsPanel>
<ItemsPanelTemplate>
<StackPanel Orientation="Horizontal"
IsItemsHost="True" />
</ItemsPanelTemplate>
</ItemsControl.ItemsPanel>
</ListBox>
</Grid>
</Window>
I'm using the standard MVVMLight ViewModel:
using GalaSoft.MvvmLight;
using KeyMaster.Model;
using System.Collections.ObjectModel;
namespace KeyMaster.ViewModel
{
public class MainViewModel : ViewModelBase
{
private readonly IDataService _dataService;
private ObservableCollection<MyData> _myListOfDataObjects;
public MainViewModel(IDataService dataService)
{
_dataService = dataService;
_dataService.GetData(
(item, error) =>
{
if (error != null)
{
return;
}
MyListOfDataObjects = new ObservableCollection<MyData>(item);
});
}
public ObservableCollection<MyData> MyListOfDataObjects
{
get { return _myListOfDataObjects; }
set
{
if (_myListOfDataObjects == value) return;
_myListOfDataObjects = value;
RaisePropertyChanged(() => MyListOfDataObjects);
}
}
}
}
Thanks.
In MVVM, there is a clear seperation between the raw data (also known as the Model) and the ViewModel. The ViewModel is the one who is in charge of parsing the data and even modifying it to whatever form it wishes, before passing it to the View.
A simple example is having the Model as XML and having the ViewModel parse it, take only a specific property (for example a "Name") from each element and add them to a list. Only this list will be shown in the View.
That said, I guess you can see where I'm going - the Command should be in the ViewModel not in the Model. As you stated by yourself, you should keep as much of the UI logic out of both the VM and the Model.
If you have a specific command that does something specific on a certain type of data, you can want it in a more "general" type of ViewModel, you can use the CanExectue to only allow this command in specific cases. But still, the command should sit in the ViewModel.
In your specific case, I don't see a problem having the command in the ViewModel, and when raised it will do whatever you need on your data. You don't need a list of ViewModels, you need only one.
I would say it'd depend where you want the functionality of the button-press. If it is always related to the MyData object then (if possible) would it be so out of place to put the Command in the MyData object? (ps. I wouldn't call your MyData object ViewModels just because you're adding a command property to them, as they're not associated with a view)
Alternatively if you want the command in the VM then you could try bind the command using the datacontext of the window. ie something like;
<Button Command="{Binding RelativeSource={RelativeSource Mode=FindAncestor, AncestorType={x:Type Window}}, Path=DataContext.ButtonPressedCommand}"
CommandParameter="{Binding .}"
Content="{Binding Name}" />
Although I've had trouble in the past with that and went with adding the command to the individual objects.
I'm working on a WPF TabControl whose last item is always a button to add a new tab, similar to Firefox:
The TabControl's ItemSource is bound to an ObservableCollection, and adding an item to the collection via this "+" button works very well. The only problem I'm having is that, after having clicked the "+" tab, I cannot for the life of me set the newly created (or any other existing tab) to focus, and so when a tab is added, the UI looks like this:
To explain a bit how I'm achieving this "special" tab behavior, the TabControl is templated and its NewButtonHeaderTemplate has a control (Image in my case) which calls the AddListener Command in the view-model (only relevant code is shown):
<Window x:Class="AIS2.PortListener.MainWindow"
xmlns="http://schemas.microsoft.com/winfx/2006/xaml/presentation"
xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml"
xmlns:ais="http://www.leica-geosystems.com/xaml"
xmlns:l="clr-namespace:AIS2.PortListener"
xmlns:i="clr-namespace:System.Windows.Interactivity;assembly=System.Windows.Interactivity"
xmlns:cmd="clr-namespace:GalaSoft.MvvmLight.Command;assembly=GalaSoft.MvvmLight.Extras.WPF4"
DataContext="{Binding Source={StaticResource Locator}>
<Window.Resources>
<ResourceDictionary>
<DataTemplate x:Key="newTabButtonHeaderTemplate">
<Grid>
<Image Source="..\Images\add.png" Height="16" Width="16">
</Image>
<i:Interaction.Triggers>
<i:EventTrigger EventName="MouseLeftButtonDown">
<cmd:EventToCommand
Command="{Binding Source={StaticResource Locator},
Path=PortListenerVM.AddListenerCommand}"/>
</i:EventTrigger>
</i:Interaction.Triggers>
</Grid>
</DataTemplate>
<DataTemplate x:Key="newTabButtonContentTemplate"/>
<DataTemplate x:Key="itemHeaderTemplate">
<TextBlock Text="{Binding Name}"/>
</DataTemplate>
<DataTemplate x:Key="itemContentTemplate">
<l:ListenerControl></l:ListenerControl>
</DataTemplate>
<l:ItemHeaderTemplateSelector x:Key="headerTemplateSelector"
NewButtonHeaderTemplate="{StaticResource newTabButtonHeaderTemplate}"
ItemHeaderTemplate="{StaticResource itemHeaderTemplate}"/>
<l:ItemContentTemplateSelector x:Key="contentTemplateSelector"
NewButtonContentTemplate="{StaticResource newTabButtonContentTemplate}"
ItemContentTemplate="{StaticResource itemContentTemplate}"/>
</ResourceDictionary>
</Window.Resources>
<TabControl Name="MainTab" Grid.Row="2" ItemsSource="{Binding Listeners}"
ItemTemplateSelector="{StaticResource headerTemplateSelector}"
ContentTemplateSelector="{StaticResource contentTemplateSelector}"
SelectedItem="{Binding SelectedListener}">
</TabControl>
The AddListener command simply adds an item to the ObservableCollection which has for effect to update the TabControl's ItemSource and add a new tab:
private ObservableCollection<Listener> _Listeners;
public ObservableCollection<Listener> Listeners
{
get { return _Listeners; }
}
private object _SelectedListener;
public object SelectedListener
{
get { return _SelectedListener; }
set
{
_SelectedListener = value;
OnPropertyChanged("SelectedListener");
}
}
public PortListenerViewModel()
{
// Place the "+" tab at the end of the tab control
var itemsView = (IEditableCollectionView)CollectionViewSource.GetDefaultView(_Listeners);
itemsView.NewItemPlaceholderPosition = NewItemPlaceholderPosition.AtEnd;
}
private RelayCommand _AddListenerCommand;
public RelayCommand AddListenerCommand
{
get
{
if (_AddListenerCommand == null)
_AddListenerCommand = new RelayCommand(param => this.AddListener());
return _AddListenerCommand;
}
}
public void AddListener()
{
var newListener = new TCPListener(0, "New listener");
this.Listeners.Add(newListener);
// The following two lines update the property, but the focus does not change
//this.SelectedListener = newListener;
//this.SelectedListener = this.Listeners[0];
}
But setting the SelectedListener property does not work, even though the TabControl's SelectedItem is bound to it. It must have something to do with the order in which things get updated in WPF, because if I set a breakpoint in the SelectedListener's set I can see the following happening:
this.Listeners.Add(newListener);
this.SelectedListener = newListener;
SelectedListener set gets called with correct Listener object
SelectedListener set gets called with NewItemPlaceholder object (of type MS.Internal.NamedObject according to the debugger)
Is there a way that I can work around this issue? Do I have the wrong approach?
I think you are triggering two events when you click the new tab: MouseLeftButtonDown and TabControl.SelectionChanged
I think they're both getting queued, then processing one at a time.
So your item is getting added, set as selected, and then before the re-draw occurs the SelectionChanged event occurs to change the selection to the [+] tab.
Perhaps try using the Dispatcher to set the SelectedItem so it occurs after the TabControl changes it's selection. Or make it so if the user tries to switch to the NewTab, it cancels the SelectionChanged event so the selected tab doesn't actually change (of course, the SelectedTab will be your NewItem since the MouseDown event will have occurred)
When I did something like this in the past, I actually overwrote the TabControl Template to create the AddTab button as a Button, not as a TabItem. I want to suggest doing that instead of using the NewItemPlaceholder in the first place, but I've never tried working with the NewItemPlaceholder so don't really know if it's better or worse than overwriting the Template.
Take a look at this post regarding sentinel objects: WPF Sentinel objects and how to check for an internal type
There are several ways to work around issues with them, that post offers one of them.
This is regards to MVVM & WPF -showing dynamic tooltip based on mouse location on a WPF Treeviewitems. Let say when the user hovers over a node which has got some business data based on that it should display tooltip.
Example: let say If i hover over manager items..should say -"locked" and for others "Ready to edit"..etc. Depends on the mouse over items data..tooltip text should guide the user to do further action. I'm trying to do some VM setting of tooltip text with the help of Tooltip opening event of TreeViewitem and trying to update the tooltip text..but facing some issue.
What are all the best and simple way to do it. Tried behavior and end up with some error.
System.Windows.Markup.XamlParseException occurred
Message="Cannot add content of type 'x.x.Views.CustomTreeView.TreeTooltipBehavior' to an object of type 'System.Windows.Interactivity.BehaviorCollection'. Error at object 'x.x.x.Views.CustomTreeView.TreeTooltipBehavior' in markup file 'x.x.x.Views;component/mypage.xaml' Line 72 Position 53."
Source="PresentationFramework"
Code:
<MyMyControls:ExtendedTreeView x:Name="MyTreeView" ItemsSource="{Binding MyDriveCollection}"
ItemContainerStyle="{StaticResource TVStyleTemplate}">
<MyMyControls:ExtendedTreeView.ItemTemplate>
<HierarchicalDataTemplate DataType="{x:Type NavModel:TreeDataItem}" ItemsSource="{Binding MyDriveCollection}">
<MyControls:SimpleEditableTextBlock x:Name="TabLabel" Text="{Binding Path=MenuText, Mode=TwoWay}"
ToolTip="{Binding MenuText,Mode=TwoWay}">
</MyControls:SimpleEditableTextBlock>
</HierarchicalDataTemplate>
</MyControls:ExtendedTreeView.ItemTemplate>
<MyControls:ExtendedTreeView.ContextMenu>
<ContextMenu ItemsSource="{Binding ContextMenuItems}">
<ContextMenu.ItemTemplate>
<DataTemplate>
<MenuItem Header="{Binding MenuText}"
CommandParameter="{Binding RelativeSource={RelativeSource FindAncestor,
AncestorType={x:Type TreeView}},Path=SelectedItem}" Command="{Binding Command}">
</MenuItem>
</DataTemplate>
</ContextMenu.ItemTemplate>
</ContextMenu>
</MyControls:ExtendedTreeView.ContextMenu>
<i:Interaction.Behaviors>
<CustomTreeView:TreeTooltipBehavior CustomTreeView:ToolTipOpeningCommand="{Binding ToolTipOpeningCommand,Mode=TwoWay,diag:PresentationTraceSources.TraceLevel=High}" />
<CustomTreeView:WorkspaceTreeViewContextMenuBehavior ContextMenuOpeningCommand="{Binding ContextMenuOpeningCommand}"/>
</i:Interaction.Behaviors>
</MyControls:ExtendedTreeView>
TreeTooltipBehavior.cs
public class TreeTooltipBehavior : Behavior<ExtendedTreeViewItem>
{
protected override void OnAttached()
{
base.OnAttached();
this.AssociatedObject.ToolTipOpening += new ToolTipEventHandler(AssociatedObject_ToolTipOpening);
}
void AssociatedObject_ToolTipOpening(object sender, ToolTipEventArgs e)
{
if (sender != null)
{
TreeDataItem hit = ((TreeDataItem) (((FrameworkElement) (sender)).DataContext));
if (ToolTipOpeningCommand != null)
{
ToolTipOpeningCommand.Execute(hit);
}
}
}
public static readonly DependencyProperty ToolTipOpeningCommandProperty = DependencyProperty.Register(
"ToolTipOpeningCommand",
typeof(ICommand),
typeof(TreeTooltipBehavior),
new PropertyMetadata(null));
public ICommand ToolTipOpeningCommand
{
get { return (ICommand)GetValue(ToolTipOpeningCommandProperty); }
set { SetValue(ToolTipOpeningCommandProperty, value); }
}
}
In my view model I'm expecting to handle the ToolTipOpeningCommand and declared enough to get the event via Behavior class.
interestingly the contextmenu behavior works fine but tooltip behavior throws xaml parser exception..
1) Am i defined at the right place (for behavior) ?
2) If Contextmenu behavior works then why not tooltipbehavior ?
3) Any clue from the exception pasted at the top ?
I'm expecting to,
(Xaml)-tooltip behavior -> to invoke tooltipopening event in the (behavior class) -> which in turns invoke the command defined in the ViewModel. I tried this similar way for context menu and works fine.
Pls provide some hint about fixing this issue.
The XAML parser error is because you are trying to attach a Behavior that only applies to ExtendedTreeViewItem to an ExtendedTreeView element. In other words, if you change Behavior<ExtendedTreeViewItem> to Behavior<ExtendedTreeView>, you will fix the parse error.
Although we cannot see from the code you presented, the reason the ContextMenu works is probably because WorkspaceTreeViewContextMenuBehavior derives from a Behavior with a generic type parameter that is compatible with ExtendedTreeView, such as FrameworkElement.
I want to build a simple application with the MVVM pattern.
This application will have two main parts:
menu on top
content below
The navigation will be simple:
each menu item (e.g. "Manage Customers" or "View Reports") will fill the content area with a new page that has some particular functionality
I have done this before with code behind where the code-behind event-handler for menu items had all pages loaded and the one that should be displayed was loaded in as a child of a StackPanel. This, however, will not work in MVVM since you don't want to be manually filling a StackPanel but displaying e.g. a "PageItem" object with a DataTemplate, etc.
So those of you who have made a simple click-menu application like this with MVVM, what was your basic application structure? I'm thinking along these lines:
MainView.xaml:
<DockPanel LastChildFill="False">
<Menu
ItemsSource="{Binding PageItemsMainMenu}"
ItemTemplate="{StaticResource MainMenuStyle}"/>
<ContentControl
Content="{Binding SelectedPageItem}"/>
</DockPanel>
where the Menu is filled with a collection of "PageItems" and the DataTemplate displays the Title of each "PageItem object" as the Header of each MenuItem.
And the ContentControl will be filled with a View/ViewModel pair which has full functionality, but am not sure on this.
First, I think you should keep the code-behind event handler, there's no point in changing a simple 2 line event handler to a complex command driven monster for no practical reason (and don't say testebility, this is the main menu, it will be tested every time you run the app).
Now, if you do want to go the pure MVVM route, all you have to do it to make your menu fire a command, first, in some resource section add this style:
<Style x:Key="MenuItemStyle" TargetType="MenuItem">
<Setter Property="Command"
Value="{Binding DataContext.SwitchViewCommand,
RelativeSource={RelativeSource FindAncestor, AncestorType={x:Type Menu}}}"/>
<Setter Property="CommandParameter"
Value="{Binding}"/>
</Style>
This style will make the menu item fire a the SwitchViewCommand on the attached view model with the MenuItem's DataContext as the command parameter.
The actual view is the same as your code with an additional reference to that style as the ItemContainerStyle (so it applies to the menu item and not the content of the DataTemplate):
<DockPanel LastChildFill="False">
<Menu DockPanel.Dock="Top"
ItemsSource="{Binding PageItemsMainMenu}"
ItemTemplate="{StaticResource MainMenuStyle}"
ItemContainerStyle="{StaticResource MenuItemStyle}"/>
<ContentControl
Content="{Binding SelectedPageItem}"/>
</DockPanel>
Now in the view model you need (I used strings because I don't have your PageItem code):
private string _selectedViewItem;
public List<string> PageItemsMainMenu { get; set; }
public string SelectedPageItem
{
get { return _selectedViewItem; }
set { _selectedViewItem = value; OnNotifyPropertyChanged("SelectedPageItem"); }
}
public ICommand SwitchViewCommand { get; set; }
And use whatever command class you use to make the command call this code:
private void DoSwitchViewCommand(object parameter)
{
SelectedPageItem = (string)parameter;
}
Now, when the user clicks a menu item the menu item will call the SwitchViewCommand with the page item as the parameter.
The command will call the DoSwitchViewCommand that will set the SelectedPageItem property
The property will raise the NotifyPropertyChanged that will make the UI update via data binding.
Or, you can write a 2 line event handler, your choice
i could imagine an ObservableCollection in the VM, that holds all the pages to be callable from the menu.
Then bind an ItemsControl And the ContentControl to it to make the ContentControl always show the CurrentItem from that List.
Of course, the menu will only bind to some Title property
whereas the ContentControl will adopt the whole item and plug in some appropriate view according to the type.
Another option is to use a ListBox instead of a menu, style the ListBox to look like a menu and then you can bind to the selected value, like this:
<DockPanel LastChildFill="False">
<ListBox
ItemsSource="{Binding PageItemsMainMenu}"
ItemTemplate="{StaticResource MainMenuStyle}"
IsSynchronizedWithCurrentItem="True"/>
<ContentControl
Content="{Binding PageItemsMainMenu/}"/>
</DockPanel>
Note the IsSynchronizedWithCurrentItem="True" to set the selected item and the {Binding PageItemsMainMenu/} with the trailing slash to use it.