CODEWITHSUNDEEP
rxjs
Mark van Straten
Mark van Straten

Reputation: 9425

Difference in signature of disposable returned in Rx.Observable.create

I am trying to figure out the 'correct' way of implementing the dispose handler on my custom-created Observable using Rx.Observable.create

named dispose function returned:

Rx.Observable.create(obs => {
  //emit values
  obs.onComplete();

  return function dispose() {
    console.log('diposed');
  };
});

anonymous function returned:

Rx.Observable.create(obs => {
  //emit values
  obs.onComplete();

  return () => {
    console.log('disposed');
  };
});

explicit Disposable.create returned:

Rx.Observable.create(obs => {
  //emit values
  obs.onComplete();

  return Rx.Disposable.create(() => {
    console.log('disposed');
  });
});

They all seem to work exactly alike but the documentation is not really clear about what is the preferred way to go.

Upvotes: 1

Views: 138

Answers (1)

paulpdaniels
paulpdaniels

Reputation: 18663

It will come down to really what your preference is on what you want to do. Though I think the three uses are actually, undefined, function, Disposable. The choice between named function and anonymous lambda is more of a JavaScript distinction rather than an Rx one so I encourage you to read up on the differences (semantic this, callstack readability, etc...).

From a usability standpoint it doesn't really matter because the framework will wrap false-y values and functions in either a Disposable.empty or Disposable.create respectively.

One advantage of the (arrow) function approach is that if you at some point upgrade to RxJS5 where Disposable -> Subscription you can avoid having to make a whole bunch of refactorings where ever you explicitly called that type (or do some sort of global aliasing), since the new version does the same sort of conversion magic.

The choice is up to you.

Upvotes: 2

Related Questions