As others have noted, this doesn't seem to collect the full article text, just stuff that you would get from an RSS fee. From the title, I expected something more like newspaper3k[0]. I used that for an NLP class during undergrad to collect full-text news articles, in conjunction with Selenium (many mainstream sites don't work with just plain wget or requests).
Lately I've starting using EpubPress[1] to grab full-text articles and generate an ePub, which happens every night via cron. Then I can get a full digest on my iPad over sftp at my leisure. Sadly EpubPress is not very sophisticated, sites like Bloomberg or ArsTechnica return "are you a robot" challenges which it can't bypass.
I wish there was some kind of community driven library for retrieving full-text articles from common sites. In my vision of how that would work, users would contribute hand-crafted Selenium scripts to download and extract the article text, bypassing the bot-detection for each site. Then something like EpubPress would work a lot better.
The "modern web" just has too much junk to be interesting any more. Sometimes news sites publish articles I would like to read, but I'm not interested in dealing with 1000 different implementations of crappy mobile UIs, advertisements, animations, etc. I know reader view exists, but you still have to wait for the page to load, and it doesn't work very well for some sites. For the sites it doesn't break with, the experience with EpubPress is much better.
0 - https://github.com/codelucas/newspaper
1 - https://epub.press/