ContentControl - WPF Discussion Boards

9-Nov-11 12:59

For my application I need to have different context menus for different text boxes. One of the features I wanted to add to the context menu was the ability to add words to a custom dictionary as the application will be used in a technical field where a lot of words won't be in a standard dictionary. I found a snippet of code online that showed how to do this, however it is in C# and uses a lambda expression that I'm having a very hard time converting. The specific line of code I'm having trouble with is this:

mnuAddToDictionary.Click += (object o, RoutedEventArgs rea) =>
                {
this.AddToDictionary(this.txtEdit.SelectedText);
                };

mnuAddToDictionary is a menu item and AddToDictionary is the method that actually does work of adding the selected misspelled word to the custom dictionary.

The AddToDictionary method is set up thus:

Public Sub AddToDictionary(ByVal entry as string)

End Sub

The closest I was able to get was AddHandler AddToDictionary.Click, AddressOf AddToDictionary, however I'm getting an error message because the delegate and the method do not have the same signature.

I pretty new to WPF and I'm not familiar with lambda expression yet because I never needed them. Any help would be greatly appreciated as I'm at a loss of what to do now. Any help would be greatly appreciated and thanks in advance.

SledgeHammer019-Nov-11 13:19

SledgeHammer01

9-Nov-11 13:19

The Click handler has to have a signature of object, RoutedEventArgs. You'd have to have an intermediate function with that signature that then calls your AddToDictionary function. Probably why they used a lambda Smile | :)

. Vb.net supports lambdas too. I don't know vb.net very well, but I think the syntax is something like:

mnuAddToDictionary.Click = Sub(o, e)
AddToDictionary(txtEdit.SelectedText)
End Sub

Or you can just point it to an intermediate function as I said.

Dominick Marciano9-Nov-11 13:30

9-Nov-11 13:30

I was trying something along those lines however when I try to create the lambda expression I get this error message:

Public Event Click(sender As Object, e As System.Windows.RoutedEventArgs) is an event, and cannot be called directly. Use a 'RaiseEvent' statement to raise an event.

That's why I was trying to use the AddHandler method because I can't figure out how to hook into the Click event. I think the lambda expression is correct because I don't get any error messages (although I had to change the 'e' to 'rea' because the variable 'e' hides in an enclosing block.

I think I will go with the immediate function but if someone can explain how I can hook into the click event I would appreciate it since that seems to be a neater why to do it over using an intermediate function. Thanks!

Dominick Marciano9-Nov-11 13:35

9-Nov-11 13:35

Actually though I think there might be a problem with the intermediate function. I have several text boxes that have this custom context menu so I created a independent function. When you right-click on any of these text boxes, they call this function which know by the sender parameter which text box called it. In that function is where the AddToDictionary menu item is added. If I use AddHandle mnuAddToDictionary, AddressOf AddEntryToDictionary, then when the user clicks on 'Add To Dictionary' how does the AddEntryToDictionary method know which text box the context menu is being currently displayed on? I was thinking I could use a public variable to save the current text box, but I don't think that is a very efficient way.

Dominick Marciano9-Nov-11 13:46

HorizontalContentAlignment=Left and still stretch the contentcontrol

9-Nov-11 13:46

Okay...so I finally figured out how to do it. It is actually a combination of the AddHandler method and a lambda expression. The line of code that worked was:

AddHandler AddToDictionary.Click, Sub()
                                      AddEntryToDictionary(selectedText)
                                  End Sub

Originally I was trying to use sender.SelectedText where sender was a ByRef parameter that was the text box that was calling this method, however ByRef parameters apparently can't be used in lambda expressions so I had to do add this line Dim selectedText As String = sender.SelectedText just before the AddHandler line.

Hope this helps others.

indian1439-Nov-11 10:36

9-Nov-11 10:36

Hi,

Can I set HorizontalContentAlignment=Left and still stretch the contentcontrol to max remaining width of the browser?

Can nybody please help me in this?

Thanks in advance.

Thanks & Regards,

Abdul Aleem Mohammad
St Louis MO - USA

ContentControl

indian1438-Nov-11 13:25

8-Nov-11 13:25

Hi All,

How can we make contentcontrol aligned left and stretched to the max of the remaining browser size.

Let us suppose I have a content control in which there are few other user controls are loaded. Now can I make the content control to stretch full remaining browser size and still the user control in it be left aligned.

When I make the content control left aligned, the user control is stretching within small area. If I make the control control center aligned, the user control stretching very well to the max size of the remaining size, but coming the center.

Is there any way to stretch it and still remaining as left alignment?

I am really in need please help me. Thanks in advance.

Thanks & Regards,

Abdul Aleem Mohammad
St Louis MO - USA

modified 8-Nov-11 20:03pm.

Abhinav S8-Nov-11 21:33

Abhinav S

8-Nov-11 21:33

indian143 wrote:
stretched to the max of the remaining browser size.

Using "*" in the layout width or height usually does this for you.

indian1439-Nov-11 7:16

9-Nov-11 7:16

Hi,

Eventhough, ContentControl is not stretching. It stretches when I put its HorizantalContentAlignment property set to stretch, but its content is displayed in the center. This is solving problem of one form but destroying the layout of all the other forms.
So I have to keep HorizantalContentAlignment=Left and HorizantalContentAlignment=Left and still I want the content control to be streched to the max width of the remaining in the browser.

I tried in the following way by using code behind, no way its listening to me. ContentControl is a stubborn control it looks Smile | :)

. It doesnt listen to what developers says and but developers have to listen to it Smile | :)

. Sorry for my words, but really I am frustrated Smile | :)

.

Remember here the actual width of the column in the grid is coming perfectly matching, and I am trying to set the width of the content control to that width. Still content control is not setting its width to the width to which I am setting to.

May be I am wrong, please advice me if you find any issues in my xaml or code behind. I requesting thanks a lot in advance.

Following is the xaml

<UserControl x:Class="Huron.UI.Healthcare.Configuration.Common.TaskWorkContainerView"
    xmlns="http://schemas.microsoft.com/winfx/2006/xaml/presentation" 
    xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml"
    xmlns:regions="clr-namespace:Microsoft.Practices.Prism.Regions;assembly=Microsoft.Practices.Prism"
    xmlns:local="clr-namespace:Huron.UI.Healthcare.Configuration.Common"
    DataContext="{Binding TaskWorkContainer,Source={StaticResource ConfigSupportViewModelLocator}}">

    <Grid ScrollViewer.HorizontalScrollBarVisibility="Auto" ShowGridLines="True" x:Name="parentGrid">
        <Grid.RowDefinitions>
            <RowDefinition Height="24" />
            <RowDefinition Height="*" />
            <RowDefinition Height="20" />
        </Grid.RowDefinitions>
        <Grid.ColumnDefinitions>
            <ColumnDefinition Width="10" />
            <ColumnDefinition x:Name="grid" MinWidth="744" Width="*" />
            <ColumnDefinition Width="10" />
        </Grid.ColumnDefinitions>

        <Canvas Grid.Row="0" Grid.Column="1" >

            <local:TaskWorkContainerHeaderView 
                Grid.Row="0" Grid.Column="1" 
                DataContext="{Binding Path=FormHeader}" 
                VerticalAlignment="Top" />

        </Canvas>

        <ContentControl Background="Purple" x:Name="parentContentControl"
                Margin="0,15,0,0"
                Grid.Row="1" Grid.Column="1"
				regions:RegionManager.RegionName="{Binding FormData.FormBodyRegionName}" 
				regions:RegionManager.RegionContext="{Binding FormData}" 
				VerticalContentAlignment="Stretch" 
				HorizontalContentAlignment="Left" 
                 HorizontalAlignment="Stretch"
                 SizeChanged="Content_SizeChanged"/>           
        

        <Canvas Grid.Row="2" Grid.Column="1" >

            <local:TaskWorkContainerFooterView 
                Grid.Row="2"  Grid.Column="1" 
                DataContext="{Binding Path=FormFooter}" 
                Visibility="{Binding Path=FooterVisibility}" 
                VerticalAlignment="Bottom" />

        </Canvas>
        
    </Grid>
</UserControl>

Following is the code behind

void Content_SizeChanged(object sender, SizeChangedEventArgs e)
{
    double minWidthOfParent;
    if ((parentGrid.ColumnDefinitions[1].MinWidth == null) || (parentGrid.ActualWidth==null))
    {
        minWidthOfParent = 744;
    }
    else
    {
        minWidthOfParent = parentGrid.ColumnDefinitions[1].ActualWidth;
    }
    parentContentControl.MinWidth = minWidthOfParent;
    parentContentControl.Width = minWidthOfParent;
}

Thanks,
Abdul.

Thanks & Regards,

Abdul Aleem Mohammad
St Louis MO - USA

modified 9-Nov-11 13:26pm.

Mark Salsbery9-Nov-11 5:44

Mark Salsbery

9-Nov-11 5:44

I'm confused - how can an element be stretched the entire width and be centered at the same time?

I saw your code in your other post...Your grid cell had the correct "*" width setting. The cell was NOT in the left or right column, however, so you can't expect the contentcontrol to stretch to the left or right edges unless those outer columns have 0 width. I would remove all the alignment property settings and remove all margins from the controls. See how it behaves and work from there.

Mark Salsbery
Java | [Coffee]

indian1439-Nov-11 7:08

9-Nov-11 7:08

Hi,

Yes ContentControl is center aligned and then only it will stretch to max width. If we put it to the left aligned it will not stretch to the size of the browser, even by using the code behind. I tried by using the code behind as well. But it will stretch to the size of its contents. But they dont want the change to be in all the forms, they want change should be only in parent form.

Can I make its size be max size of the browser and still be left aligned. This Silverlight is freaking tough in sizing yaar.

Dear developers, can anybody please help me, I am in really need buddy. My sprint finishes by this weekend.

Thanks,
Abdul.

Thanks & Regards,

Abdul Aleem Mohammad
St Louis MO - USA

MVVM Recommendations

Tom Delany8-Nov-11 7:10

8-Nov-11 7:10

I'm in the middle of designing a WPF app that is going to have a view with a lot of information. It is a sort of Management Console for an app that we are developing.

I am envisioning some sort of tabbed interface to organize all of the various "things" that this UI has to deal with.

The model part is fairly straightforward; it is going to be accessing an SQL database with lots of different tables.

I am wondering if the "standard" way of doing things would be to create one large view model class that worked with the overall view, or if it is more appropriate to treat each tab like a separate view, with a separate view model for each? Or does that add needless complexity? Or does it even really matter? Confused | :confused:

Thanks for any advice on the matter. Smile | :)

WE ARE DYSLEXIC OF BORG. Refutance is systile. Your a$$ will be laminated.

There are 10 kinds of people in the world: People who know binary and people who don't.

SledgeHammer018-Nov-11 7:23

SledgeHammer01

8-Nov-11 7:23

Definitely each tab should be its own view and have its own VM. There is no view for managing the tabs per se. Its more of a control that data binds to a collection of views or VMs. Typically it'll bind to a collection of VMs and you'll use DataTemplates to map a View to the VMs and the "control" will auto create the view and wire up the data context, etc. Personally though, I find that backwards as I like to launch a view and have the xaml specify what the VM is through a ViewLocator.

Tom Delany8-Nov-11 7:52

8-Nov-11 7:52

Thanks for the feedback. I'm seriously considering using DevForce A/F to help me out.

http://drc.ideablade.com/xwiki/bin/view/Documentation/devforce-caliburn[^]

[Edit:] I'm thinking that if we ever decided we needed a web interface, I could leverage the Silverlight support in the above with minimal code changes.

WE ARE DYSLEXIC OF BORG. Refutance is systile. Your a$$ will be laminated.

There are 10 kinds of people in the world: People who know binary and people who don't.

Kevin Marois8-Nov-11 7:36

Kevin Marois

8-Nov-11 7:36

I'm doing the exact same thing right now. My app has tabs, and on each tab I add & remove user controls at runtime. So it's basically "nested' views. The MainWindowViewModel is responsible for creating the view and viewmodel and handling commands.

I created a TabInfo class. The MainWindowViewModel maintians a collection of TabInfo object. I then use that to manage the tabs.

Everything makes sense in someone's mind

Tom Delany8-Nov-11 7:53

8-Nov-11 7:53

Cool. Thanks for the feedback! Smile | :)

WE ARE DYSLEXIC OF BORG. Refutance is systile. Your a$$ will be laminated.

There are 10 kinds of people in the world: People who know binary and people who don't.

Abhinav S8-Nov-11 21:38

Abhinav S

8-Nov-11 21:38

Each tab should be a separate view / view model and you should be plugging in /removing views on demand.

There is no concept of creating one large view model that works for all views.

Tom Delany9-Nov-11 3:57

9-Nov-11 3:57

Thanks. That seems to be the general consensus. Sorry if it was a stupid question.

WE ARE DYSLEXIC OF BORG. Refutance is systile. Your a$$ will be laminated.

There are 10 kinds of people in the world: People who know binary and people who don't.

Wayne Gaylard9-Nov-11 4:23

Wayne Gaylard

9-Nov-11 4:23

Tom Delany wrote:
Sorry if it was a stupid question

There is no such thing as a stupid question Smile | :)

Only stupid answers. If you are unsure, it is always better to ask.

When I was a coder, we worked on algorithms. Today, we memorize APIs for countless libraries — those libraries have the algorithms - Eric Allman

Tom Delany9-Nov-11 4:51

9-Nov-11 4:51

Thanks

WE ARE DYSLEXIC OF BORG. Refutance is systile. Your a$$ will be laminated.

There are 10 kinds of people in the world: People who know binary and people who don't.

Abhinav S9-Nov-11 6:41

Abhinav S

9-Nov-11 6:41

Tom Delany wrote:
Sorry if it was a stupid question.

Nope. It was not.

_Maxxx_14-Nov-11 17:55

_Maxxx_

14-Nov-11 17:55

IMHO ...
If each tab in your tabbed interface is, essentially, stand alone, and could be presented in a different way (say in a tree view, or pop-up form etc.) then you need to model that view using a view model. the fact that a particular View attached to that view model happens to sit on a tab control somewhere is 'nothing to do with' the ViewModel.

OTOH if the separation into tabs is more just a gui convenience (that is, all the tabs are necessary and inter-related) then the ViewModel should model all of the controls on all of the tabs - and the View just 'happens' to split them into tabs. Some other implementation may decide to present all of the data in one large, scrolling form, for example, still bound to the single view model.

this isn't a criticism, as I do it myself, but I believe one of the aims of using ViewModels is that you try to design the ViewModel first rather than thinking 'I want a tabbed interface' you think 'I want to perform these functions." The ViewModel contains the data to be presented/updated, and offers the functionality - then you design the view to be able to provide the interface to the user.

So in the case of the 'Management Console' style, i would think you would have several ViewModels, one for each piece of functionality, and a ManagementConsoleViewmodel that has, as properties, each of the view models it requires to perform its functionality.

The View that binds to the 'big' view model may place the smaller views on tabs, in trees, or on the sides of an animated 3D spinning cube - by binding to its view model it has access to all it needs regardless, and each of the smaller viewmodels can also be viewed in whatever way you want - and indeed can also be used independently if required.

MVVM# - See how I did MVVM my way
___________________________________________
Man, you're a god. - walterhevedeich 26/05/2011

.\\axxx
(That's an 'M')

Tom Delany20-Nov-11 11:27

20-Nov-11 11:27

Sorry for the slow reply. My father passed away on the date that you sent this to me and I had to leave town suddenly.

Thank you for the advice. I appreciate it. Smile | :)

I do not in fact take your comments as criticism. In actuality, I have been trying to code up the view models first, but in parallel I have also been trying to get some idea what the view will end up looking like. As you say, it may indeed end up looking completely different that a tabbed interface. If I construct the view models properly (and that's a big if), I should be able to swap in any sort of views that I desire with minimal fuss.

WE ARE DYSLEXIC OF BORG. Refutance is systile. Your a$$ will be laminated.

There are 10 kinds of people in the world: People who know binary and people who don't.