In a Blazor application, I have an <input type="text" />
and I'm trying to do something so that non-digit characters cannot be entered into it.
This is what I've so far, and it seems to be working pretty well:
<input type="text" @bind="InputContent" @bind:event="oninput" />
@code {
string inputContent;
string InputContent
{
get => inputContent;
set => inputContent = StripNonDigits(value);
}
string StripNonDigits(string str)
{
return new String(str.Where(c => char.IsDigit(c)).ToArray());
}
}
Now, I'm wondering:
@bind
?In JS, to do the very same thing, you would listen to the oninput
event and if the entered character is not a digit, you then do an event.preventDefault()
. You can't exactly do that in Blazor, as far as I know, is that right?
Here's why @enet's proposed solution, is in fact highly unreliable, flawed, inefficient, and naive. + Why his claims that my solution is problematic are actually false.
Update: He deleted his answer.
First, reasons why his solution is inelegant:
No comment:
No comment:
No comment:
No comment:
No comment:
Note: You could see all of these behaviors for yourself in this fiddle: https://blazorfiddle.com/s/xfik4hnq
RegEx is slow, inefficient, and often hard-to-read and parse. It should therefore be avoided wherever possible.
Seriously, no comment, pretty obvious.
Considering only the @code section, this solution, as presented by @enet, is about 23 lines of code, while mine, is about 12. (Both considering the curly braces on a line, etc.)
Note: It's also crucial to keep in mind that if this solution were to be even nearly as robust as the one I initially proposed (I'll explain that in just a bit), it would have to solve every single one of the problems that were just demonstrated, and in order to do so, it has to become much more complex than it already is, and it would require many more lines of code as well.
Here's why the solution that I initially proposed in my original question is the most optimal and elegant solution;
Solves ALL of the aforementioned problems, and it also does so in fewer lines of code + code that's easier-to-understand, shorter, nicer, and simpler.
Meanwhile, the only argument presented from the other side, in favor of the other solution, and against mine, was that my solution causes the disallowed characters to show up a fraction of a second, and then disappear. However, this was not observed at all in any of my tests.
Here's a GIF that demonstrates this:
Whenever I press a key on the keyboard, you can see on the console that a message like "Key pressed: a" is shown, but at the same time, you see that that character doesn't appear at all in the input: