Find first duplicate in a string using RxJS

firstDuplicate.js

const str = `Mussum Ipsum, cacilds vidis litro abertis. Si u mundo tá muito paradis?
Toma um mé que o mundo vai girarzis! Viva Forevis aptent taciti sociosqu ad litora torquent.
Aenean aliquam molestie leo, vitae iaculis nisl.
Leite de capivaris, leite de mula manquis sem cabeça.
Posuere libero varius. Nullam a nisl ut ante blandit hendrerit. Aenean sit amet nisi.
Delegadis gente finis, bibendum egestas augue arcu ut est. Admodum accumsan disputationi eu sit.
Vide electram sadipscing et per. Sapien in monti palavris qui num significa nadis i pareci latim.
Interagi no mé, cursus quis, vehicula ac nisi. Manduma pindureta quium dia nois paga.
Mais vale um bebadis conhecidis, que um alcoolatra anonimis. Si num tem leite então bota uma pinga aí cumpadi!`;

const splitted = str.replace(/(\n|\t|\s{2,})/gi, ' ').split(' ');
const source$ = Rx.Observable.from(splitted);

source$
  /* 1. */ .scan((acc, item) => acc.push(item) && acc, [])
  /* 2. */ .takeWhile((data, item) => (new Set(data)).size === data.length)
  /* 3. */ .last()
  /* 4. */ .subscribe(x => console.log(x[x.length-1]));

// OUTPUT: mundo

/*
  1. accumulate each item into an array
  
  2. transform the current accumulated value into a set, which by default remove all duplicated items,
  and then compare its size to the actual data length. If they are different, it means that the first duplicated 
  item has been found.

  3. take the last accumulated array ...

  4. ... in which the last element is our first duplicate item, and print to the console.
 */