Performance of loading Xaml dynamically in Xamarin.Forms

As I mentioned in a previous post, you can quite easily load Xaml dynamically in your Xamarin.Forms App. But how about performance? How long does it take in comparison to “regular” loading of pre-compiled Xaml.

It’s anything but easy to really measure this. We could eventually use LoadFromXaml and measure the differences. I decided to create DataTemplates and use the CreateContent() method to create the actual Control.

<DataTemplate x:Key="Test1">
<Grid>
    <Grid.RowDefinitions>
        <RowDefinition Height="auto" />
        <RowDefinition Height="*" />

    </Grid.RowDefinitions>
    <Label Text="Hello World!" />
    <Button Grid.Row="1" Text="Press me" />
</Grid>
</DataTemplate>
<DataTemplate x:Key="Test2">
    <dynamiccontrol:XamlView>
        <dynamiccontrol:XamlView.Xaml>
            <![CDATA[
            <Grid>
                <Grid.RowDefinitions>
                    <RowDefinition Height="auto" />
                    <RowDefinition Height="*" />

                </Grid.RowDefinitions>
                <Label Text="Hello World!" />
                <Button Grid.Row="1" Text="Press me" />
            </Grid>
            ]]>
        </dynamiccontrol:XamlView.Xaml>
    </dynamiccontrol:XamlView>
           
</DataTemplate>

You can see that the Xaml is quite easy, but loading it dynamically takes over 20 times longer than compiled Xaml.

I tried to create 10,000 controls and measured the following values:
Compiled Xaml: 5.8s
Dynamic Xaml: 133.1s

I know the test is not very representative, but you should always keep in mind that there may be a performance problem when using this method. Especially when you want to create dynamic Layouts in ListViews.

You can slightly boost performance (5-10%), when you load the whole DataTemplate from Xaml, and then create the Control with it.

Dynamic DataTemplate: 124s

Never ever use Task.Result in Xamarin.Forms! Why?

I recently stumbled over a deadlock in our code. Someone called a method, that returned a Task, and instead of awaiting it, he called:

var res = Method().Result;

I reason for this code was just plain laziness, because using await requires an async-context, which probably isn’t that easy to achieve. But what’s the reason for that deadlock, and why is awaiting the result actually resolving it?

I don’t want to dick very deeply into the Task Parallel Library (TPL). But let’s just make clear, what a task is. A task is a “unit of execution”, it has no direct link to Threads (even though it’s in the same namespace). A Thread can run tasks, there can be one or more Threads in your program. Let’s assume that Threads run in parallel.

So now let’s make up some setting, we call a method (which returns and starts a Task, that’s what “async Task” does) and then wait for the result by calling “.Result”. That call will actually block the current thread and wait for the second task to finish.

When we look at the graphic above, we see that this call is working. We actually block Thread 1, but we have not created a deadlock. But we can never be sure, that the TaskScheduler (it is planning what Tasks runs on which Thread) really runs Task2 on Thread2. As soon as the TaskScheduler decides to run Task2 on Thread1, we have a classic deadlock:

But what is the difference to async/await pattern?

A couple of years ago, a friend described async-await pattern as syntactical sugar, it’s not a language feature, nor a real pattern. There is something true to that, but it makes programming Tasks so much easier. It actually splits your code into separate parts, and calls them in a sequence. So there will be no busy wait, no matter which Thread the Task we are waiting for is executed on.

Remove whitespaces of a string in C#

This seems to be really simple, so why am I writing this blog post?
Everyone I asked, had a different idea to achieve this, the first idea was to use Regular Expressions, which is really straight forward and simple, but there are several other methods with quite different performance.

Let’s have a look at all methods that came in mind:
1. String Replace
2. Regex
3. split directly with whitespace character
4. split & join each character
5. use a StringBuilder
(this list may not be complete, but I focused on these methods)

// string replace
Stopwatch sw = Stopwatch.StartNew();
int n = 10000;
for (int i = 0; i < n; i++)
{
    var strRes = testString.Replace(" ", "");
    strRes = strRes.Replace("\t", "");
    strRes = strRes.Replace("\n", "");
    strRes = strRes.Replace("\r", "");
    //if (expected != strRes)
    //    throw new Exception();
}
sw.Stop();
Debug.WriteLine(sw.Elapsed);
// regex replace
sw.Restart();
for (int i = 0; i < n; i++)
{
    var strRes = Regex.Replace(testString, "\\s+", "");
    //if (expected != strRes)
    //    throw new Exception();
}
sw.Stop();
Debug.WriteLine(sw.Elapsed);
// split
sw.Restart();
for (int i = 0; i < n; i++)
{
    var parts = testString.Split(new[] { '\t', '\r', '\n', ' ' }, StringSplitOptions.RemoveEmptyEntries);
    //var parts = testString.Split(default(string[]), StringSplitOptions.RemoveEmptyEntries);
    var strRes = string.Join("", parts);
    //if (expected != strRes)
    //    throw new Exception();
}
sw.Stop();
Debug.WriteLine(sw.Elapsed);
// split 2
sw.Restart();
for (int i = 0; i < n; i++)
{
    var parts = testString.Split(default(string[]), StringSplitOptions.RemoveEmptyEntries);
    var strRes = string.Join("", parts);
    //if (expected != strRes)
    //    throw new Exception();
}
sw.Stop();
Debug.WriteLine(sw.Elapsed);
// StringBuilder
sw.Restart();
for (int i = 0; i < n; i++)
{
    StringBuilder resSb = new StringBuilder();
    foreach (var chr in testString)
    {
        if (chr != '\t' && chr != '\r' && chr != '\n' && chr != ' ')
        {
            resSb.Append(chr);
        }
    }
    var strRes = resSb.ToString();
    //if (expected != strRes)
    //    throw new Exception();
}
sw.Stop();
Debug.WriteLine(sw.Elapsed);

The Results

1. String Replace: 41ms
2. Regex: 113ms
3. split directly with whitespace character: 27ms
4. split & join each character: 27ms
5. use a StringBuilder: 9ms

Dynamically create controls in Xamarin Forms

This is a very broad topic. You can actually always dynamically create controls in code behind, by just adding them to the UI during runtime. This is very easy, but it’s not generic at all, you need to code everything you would actually do in Xaml.

Another solution (which I actually prefer) is to use a ListLayout (maybe a Bindable StackLayout) and use a TemplateSelector to switch between pre defined DataTemplates based on the ViewModel, that is being used. But this will actually not give you more flexibility, but it’s definitely a better solution than the first shot.

We actually want something really flexible, we try to achieve instanciating an unknown Xaml from whatever source we have (Text, Internet, UserInput,.. ), and even provide binding to further use the UserInput in ViewModel. So there are 2 parts to this:
1. load xaml during runtime,
2. somehow achieve binding.

When you have a look in your obj-folder of a compiled Xamarin app, you will find a file called something like “MainPage.xaml.g.cs”

[global::Xamarin.Forms.Xaml.XamlFilePathAttribute("MainPage.xaml")]
    public partial class MainPage : global::Xamarin.Forms.ContentPage {
        
        [global::System.CodeDom.Compiler.GeneratedCodeAttribute("Xamarin.Forms.Build.Tasks.XamlG", "2.0.0.0")]
        private void InitializeComponent() {
            global::Xamarin.Forms.Xaml.Extensions.LoadFromXaml(this, typeof(MainPage));
        }
    }

There is another overload of the global extension function Xamarin.Forms.Xaml.Extensions.LoadFromXaml(..) which takes a string representation of your xaml to create a control.

We can use this, to create a ContentControl with a BindableProperty, which creates the given Xaml at runtime.

public class XamlView : ContentView
{
       
    public static readonly BindableProperty XamlProperty =
                            BindableProperty.Create(nameof(Xaml), typeof(string), typeof(XamlView), propertyChanged: OnXamlChanged);

    private static void OnXamlChanged(BindableObject bindable, object oldValue, object newValue)
    {
        try
        {
            var xamlView = (XamlView)bindable;
            ContentView view = new ContentView();
            // we add the default xaml namespace to our surrounding ContentView,
            // so it doesn't need to be defined in xaml
            view = view.LoadFromXaml("" +
                (xamlView.Xaml ?? string.Empty) +
                "");
            xamlView.Content = view;
        }
        catch (Exception ex)
        {
            // we should actually handle that exception, maybe put it on the screen as label
        }
    }

    public string Xaml
    {
        get => (string)GetValue(XamlProperty);
        set => SetValue(XamlProperty, value);
    }
}

usage:

<Grid BackgroundColor="Red">
    <Grid.RowDefinitions>
        <RowDefinition Height="*" />
        <RowDefinition Height="*" />
    </Grid.RowDefinitions>
    <Entry x:Name="XamlEditor" Text="{Binding ContentXaml}" />
    <dynamiccontrol:XamlView
        Grid.Row="1"
        Xaml="{Binding ContentXaml}" />
</Grid>

You could also bind the XamlView.Xaml directly to an Entry.Text using ReferenceBinding, but I prefer using a ViewModel instead.

And this is how it looks like:

In my next post, I will talk about binding the Data from the XamlView to a ViewModel.

Copy an entire synced Realm

In our company, we use realm.io as mobile Database with the ROS Cloud to sync data between mobile devices and the server. Main advantage is, that we can build Apps in a fraction of time, without juggling with synchronization.

One of our main challenges was the disability to copy data from one realm to another. But there is still a good solution for this, even though you have to do a lot of coding for that.

Motivation

There are a lot of reasons, why copying realms is a really neat feature. First of all would be versioning your data. But it’s also very neat, when you have several instances (for development, testing etc.) to copy the whole data when you go productive or need some test data in development.

Challenge 1: generating the Schema

When we copy one realm to another, we actually don’t know the used types. We could include a library which contains the types, but then our “RealmCopy” will only work for a specific realm, which is really baloney. That’s as if you develop a copy-program, that can only copy text-files.

So we need to read the source realm, get the schema and create the same schema in the destination. In .Net we can open a realm dynamically, which is pretty cool, because this gives us a chance to get the information we need. Only problem – we can not create or change a schema on a dynamically opened realm.

Solution? We can create dynamic types, and open the destination realm, this will automatically create a schema on the destination realm. (It’s really a little bit awkward, but that’s currently the only chance in .Net – with Javascript, we can mutate the realm directly, but that’s probably due to the nature of that language)

Open a realm as dynamic:

var username = "";
var password = "";
var realmUrl = "";
var realmName = "TestRealm";
var user = await User.LoginAsync(Credentials.UsernamePassword(username, password, false), new Uri(realmUrl));

var configuration = new FullSyncConfiguration(new Uri(realmName, UriKind.Relative), user)
{
IsDynamic = true
};
var srcRealm = await Realm.GetInstanceAsync(configuration);

now we can go through the scheme and create types:

var types = new List();
foreach (var scheme in schemasToProcess)
{
// create an assembly for our type
var assemblyName = new AssemblyName($"SomeAssemblyName{name}");
var assemblyBuilder = AssemblyBuilder.DefineDynamicAssembly(assemblyName, AssemblyBuilderAccess.Run);
// create an module for the type
var moduleBuilder = assemblyBuilder.DefineDynamicModule("MainModule");
// get the typebuilder as
var typeBuilder = moduleBuilder.DefineType(name,
TypeAttributes.Public |
TypeAttributes.Class
, parent);
// add properties to type
foreach (var prop in schema)
{
var attributes = new List();
Type propType;
// woven property is needed, so Realm will add it to the destination schema
attributes.Add(typeof(WovenPropertyAttribute));
if (prop.IsPrimaryKey)
{
// add primary key attribute
attributes.Add(typeof(PrimaryKeyAttribute));
}
else if (prop.IsIndexed)
{
// add indexed attribute
attributes.Add(typeof(IndexedAttribute));
}

// get the property type
propType = prop.GetPropertyType(knownTypes);

// set required attribute, when its not nullable
if ((prop.Type & PropertyType.Nullable) != PropertyType.Nullable)
{
attributes.Add(typeof(RequiredAttribute));
}

// create the property
MyTypeBuilder.CreateProperty(typeBuilder, prop.Name, propType, attributes.ToArray());
}
types.Add(typeBuilder.CreateType());
}

In one of my previous blog posts I mentioned how to create types using the TypeBuilder, so I won’t code it out here.
The function GetPropertyType is an extension, it’s just a mapping of the Realm-PropertyType to a CLR type.
Now we can open the destination Realm:

var configuration = new FullSyncConfiguration(new Uri(realmName, UriKind.Relative), user);
configuration.ObjectClasses = types;
var destRealm = await Realm.GetInstanceAsync(configuration);

Challenge 2: copying the realm

This doesn’t actually seem to be a challenge. We go through all objects in source realm and copy it to a new RealmObject, which is created in destination. You can actually use our generated types to open the source realm and use straight reflection for mapping all properties. But be careful, if you did something wrong, you would probably mutate the source realm (what you might not want). You can stick to the dynamic Realm instead and use CallSiteCache to read the properties and reflection to set it to the destination object. (more infos)

Challenge 3: cascaded objects

An object may not only reference atomic types, but also have other RealmObjects as properties. It can even contain cycles. This is really tricky, I solved it by initially creating a TypeBuilder for each SchemaItem, then we can use this TypeBuilder as a PropertyType, without building the actual type. Creating objects can be done recursively, but keep track of the objects, you already created, so you can avoid cycles.

If you have any questions, feel free to ask. ..have fun coding.