When consuming a component in another razor component, there is usually a need to call async methods triggered by an EventCallback property, onClick
for instance. Is there any particular way to call it to assure asynchronicity?
Let's assume the following example:
@* CASE 1*@
<MyButton OnClick=DoSomething>Click Me</MyButton>
@* CASE 2*@
<MyButton OnClick=@(async () => await DoSomething())>Click Me</MyButton>
@code {
private async Task DoSomething(){ ... }
}
After compiling, is there any distinction between using a delegate or calling the function directly? Should we prefer one or the other?
This question doesn't derive from an error or code running improperly but only from the fact that I get no feedback from Visual Studio and probably for a good reason but I would like to know if I'm writing improper code either way.
Choice Remark: Hard to pick a single answer for all provide valid points to the discussion, further reading is encouraged. It seems the confusion stems from a misunderstanding of the delegate's role in the call stack, in that way I think that Shaun's answer shows the most succinct and explicit example of that.
In this specific case, the first is better.
This code block is wrapping a Task within a Task. It wasteful on resources: each Task is a state machine object.
OnClick=@(async () => await DoSomething())>
However, what actually happens depends on MyButton
.
Here's a demo version. I've added two button event handlers, with comments on how each executes.
<button @attributes=this.AdditionalAttributes @onclick=this.OnButtonClickAsync></button>
@code {
[Parameter] public EventCallback<MouseEventArgs> OnClick { get; set; }
[Parameter] public RenderFragment? ChildContent { get; set; }
[Parameter(CaptureUnmatchedValues = true)] public Dictionary<string, object>? AdditionalAttributes { get; set; }
private async Task OnButtonClickAsync(MouseEventArgs e)
{
// you are awaiting the delegate in the parent component
// if it yields you'll await its completion
await this.OnClick.InvokeAsync(e);
// Code here will only execute after the parent delegate has completed, awaits and all
// any exceptions will bubble up to here
}
private void OnButtonClick(MouseEventArgs e)
{
// This is a Fire and Forget call
this.OnClick.InvokeAsync(e);
// code here will execute as soon as the parent delegate yields
// exceptions won't bubble up to here
}
}
More generally, it's a matter of personal preference.
I like clean markup, so I code like this.
<button disabled="@buttonCss"></button>
@code {
private bool _isDisabled;
private string buttonCss => _isDisabled ? "btn btn-danger" : "btn btn-success";
}
Others like succinct [inline] code.
<button disabled="@(_isDisabled ? "btn btn-danger": "btn btn-success")"></button>
@code {
private bool _isDisabled;
}