Events

[kobezzza]

Really need something like:

var worker = operative(function (callback) {
    for (var i = 0; i < 1e9; i++) {
        if (i == 1e9 / 2)  {
             trigger('progress', '50%');
        }
    }

    callback('im done!');
});

worker.on('progress', function (val) { ... });
worker(function (val) { ... });

[padolsey]

I agree that there is a need for Operative to support the idea of progress. I have been playing with a few things, and tbh, I'm not convinced that adding an EventEmitter and a progress event is the best way to go. One simple alternative would be to allow the provided callback to be called multiple times. So you could potentially do:

var worker = operative(function(callback) {
    for (var i = 0; i < 1e9; i++) {
        if (i == 1e9 / 2)  {
             callback('progress', '50%');
        }
    }

    callback('complete');
});

worker(function(state, v) {
    if (state === 'progress') { ... }
    else if (state === 'complete') { ... }
});

The benefit with this is that:

• It's simple to implement (and lightweight)
• It doesn't impose a paradigm on the developer (instead you're free to manage the state-updates as you wish)

The only potential implementation problem with this is that, for the operative instance, it becomes unclear when it can delete the callback (freeing up memory) -- usually it'll delete it when the callback is first-called, but if the callback supports multiple calls it will just have to stick around on the operative instance until the operative itself is destroyed via operativeInstance.terminate().

Welcoming suggestions/thoughts...

[kobezzza]

Maybe create two different constructor?

For single usage

operative

For multiply usage (with "terminate" for destroy)

operative.connect

[chrisritter]

Any hope for an events api?

[evanworley]

@padolsey - I like the simplicity of the callback approach, and using a simple JSON object the developer can communicate data as structured as they like. I think it's totally acceptable to keep the callback around until terminate is called.

[chrisritter]

I like that approach

[chrisritter]

@padolsey -- any interest on changing the callback to continue until termination?

[padolsey]

Yes, that sounds reasonable. It may end up being an opt-in thing though. I'll play with it a bit and shall try to work something into the next release (which will hopefully include some more robust testing via BrowserStack).

[samzhao]

Is it possible to just bring the postMessage method into Operative?

There's also the Promise deferred.notify() method if we decide to use Promise.

[padolsey]

@samzhao AFAIK notify is not supported in ES6 Promises (nor its shims), so I don't think it'd be wise to support that (it'd mean operative would have a require a specific Promise implementation). Please correct me if I'm wrong here though.

Re: postMessage: You can still utilize postMessage on the underlying worker via operativeInstance.worker.postMessage(...). For operative's actual API, I think I prefer the callback approach for progress-events -- but I'm curious if there are merits to a generic per-operative postMessage as opposed to per-operative-method calls/callbacks?

[samzhao]

@padolsey, ah sorry I didn't know notify won't be supported in ES6. I'll try the postMessage.

And I personally think the generic operative postMessage is good enough. To distinguish between instances, we can just pass in a different message each time like so:

var a = Operative({
    var data = "foo";
    postMessage("progress-a", data);
});

var b = Operative({
    var data = "bar";
    postMessage("progress-b", data);
});

Operative.on("progress-a", callback);
Operative.on("progress-b", callback);

[padolsey]

As of 0.4.0-rc1 (currently available via npm) -- callbacks are no longer destroyed until the operative itself is terminated, so stuff like this will work fine:

var op = operative(function(callback) {
    for (var i = 0; i < 1e9; i++) {
        if (i == 1e9 / 2)  {
             callback('progress', '50%');
        }
    }

    callback('complete');
});

op(function(state, v) {
    if (state === 'progress') { ... }
    else if (state === 'complete') { ... }
});

@samzhao Re: postMessage -- I think that might add to the complexity of messaging using operative (especially for multi-method operatives). As it is, there is already callbacks and promises -- and operative takes care of passing things to the correct method within the worker context. A generic postMessage seems to produce little gain over changing the mechanics of callbacks (which is what I ended up doing). Happy to discuss further though -- as 0.4 is only in "release-candidate" at the moment.

[samzhao]

Thanks James! I think this is good :) And I agree that postMessage might not be much beneficial.

On Sat, Oct 18, 2014 at 1:03 PM, James Padolsey notifications@github.com wrote:

As of 0.4.0-rc1 (currently available via npm) -- callbacks are no longer destroyed until the operative itself is terminated, so stuff like this will work fine:

var op = operative(function(callback) { for (var i = 0; i < 1e9; i++) { if (i == 1e9 / 2) { callback('progress', '50%'); } }

callback('complete');

});

op(function(state, v) { if (state === 'progress') { ... } else if (state === 'complete') { ... } });

@samzhao https://github.com/samzhao Re: postMessage -- I think that might add to the complexity of messaging using operative (especially for multi-method operatives). As it is, there is already callbacks and promises -- and operative takes care of passing things to the correct method within the worker context. A generic postMessage seems to produce little gain over changing the mechanics of callbacks (which is what I ended up doing). Happy to discuss further though -- as 0.4 is only in "release-candidate" at the moment.

— Reply to this email directly or view it on GitHub https://github.com/padolsey/operative/issues/15#issuecomment-59627711.