[RxJS] Hot Observable, by .share()

.share() is an alias for .publish().refCount().

So if the source is not yet completed, no matter how many subscribers subscribe to the source, they share the same source. 

const clock$ = Rx.Observable.interval(500).share().take(6);

const randomNum$ = clock$
  .map(i => Math.random() * 100).share();

const smallNum$ = randomNum$
  .filter(x => x <= 50)
  .toArray();

const largeNum$ = randomNum$
  .filter(x => x > 50)
  .toArray();

randomNum$.subscribe(x => console.log('random: ' + x));
smallNum$.subscribe(x => console.log('small:', x));
largeNum$.subscribe(x => console.log('large:', x));

/*

Console Run  Clear
"random: 49.87840398986816"
"random: 75.01024609865293"
"random: 32.59613439667008"
"random: 63.4234109489461"
"random: 35.58020574147034"
"random: 74.94599860014348"
"small:"
[49.87840398986816, 32.59613439667008, 35.58020574147034]
"large:"
[75.01024609865293, 63.4234109489461, 74.94599860014348]
*/

It is important to know share the same source is only before the source stream completed, if it is already completed, then the new subscribers will trigger another source running.

For example, in the code, we change largeNum$ happens after 4s of first subscriber.

setTimeout(() => largeNum$.subscribe(x => console.log('large:', x)), 4000)

Because source will complete after 3s, therefor it will trigger a new source:

/*
"random: 74.91154828671043"
"random: 10.964684522348733"
"random: 29.076967396825903"
"random: 20.070493440627235"
"random: 22.44421045844409"
"random: 14.233614544120798"
"small:"
[10.964684522348733, 29.076967396825903, 20.070493440627235, 22.44421045844409, 14.233614544120798]
"large:"
[93.04544926644354, 65.4090612653734, 67.15475480984114]
*/

As we can see, in the large array, all the numbers are not from random we log out before.