Rewrote existing parser with a new logic to detect tweets

[Bhargav-Dave]

The earlier parser detected tweets by using RegEx queries and matching the Xpath to the tweet text elements with certain given Xpaths.

However, I discovered that every DIV on the twitter DOM that contains a tweet text has an attribute titled 'data-testid' whose value is always 'tweetText'

Hence, the new method uses the 'querySelectorAll()' method (ref: https://developer.mozilla.org/en-US/docs/Web/API/Document/querySelectorAll) of the DOM in order to select all such DIVs whose data-testid is set to tweetText. This gives an array of all DIVs in the current DOM that contain the spans which contains the tweetText.

This array is discovered in the file parser-v2.js and then sent to the file transform-v2.js and processed there where a for loop runs through all the DIVs and does the processing that was done in transform.js

[Bhargav-Dave]

Is related to: #179

[dennyabrain]

hey @Bhargav-Dave, there's two concerns with this approach :

1. data-testid is added to make testing easy for the twitter developer. its possible they will take it out someday
2. this works for the tweet text but can you extract tweet url, timestamp and author handle using this method?