Async Class

[kdex]

Having the constructor return a `Promise` is generally considered bad 
practice, since it breaks `new X instanceof X`, which in general breaks 
inheritance.

So, for this idea to work, you really shouldn't mark `constructor` as `async` 
(since this might suggest it returning a `Promise`, which it mustn't); 
instead, you would have to make the constructor return `this`, then make the 
asynchronous initialization, and then execute the rest. In other words,

something like this:

```js
async class X {
	constructor() {
		await this.read();
		await this.write();
	}
	async read() { … }
	async write() { … }
}
console.log(new X());
```

would transpile like so:

```js
class X {
	constructor() {}
	async initialize() {
		await this.read();
		await this.write();
	}
	async read() { … }
	async write() { … }
}
(new X()).initialize().then(x => {
	console.log(x);
});
```

The next problem with this idea is that you can't statically decide what `new 
something` is about to construct. Since `something` might be an `async class`, 
the StatementList following the construction of `something` would always have 
to be wrapped in a callback.

Long story short, introducing this feature would slow down `new`/
`Reflect.construct`.

On Friday, February 16, 2018 5:12:10 AM CET Dimitrian Nine wrote:
> I cant find: was or not that proposal...
> 
> But can we have Async Class with async constructor and async super?
> 
> //AsyncClass
> async_class AsyncClass { // keyword async_class(example)
> constructor() {
> await something(); //now constructor async too
> }}
> 
> //AsyncObj in some async method
> let NewAsyncObj = await AsyncClass(); // create async obj
> 
> //Extends
> async_class ExtAsyncClass extends AsyncClass{
> constructor() {
> await super(); // now we can await super too
> await something();
> }}
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.mozilla.org/pipermail/es-discuss/attachments/20180218/bce50237/attachment.sig>