URLs always prefixed by last word of extraction text

[midzdotdev]

The algorithm used to extract the URL is really poor. All URLs contain the last word of tweet text at the start of the url. Twitter has published its own module which contains all of their code for extracting stuff from text, including URLs called twitter-text, it's on GitHub.

[sebinsua]

Hello,

I'm happy to take a PR if you want to create one to use twitter-text.

There are quite a few issues with this package now. I have not been able to pay much attention to it, since I released it 2 years ago.

[midzdotdev]

I really like this module, you've done some great work. Its likely that I'll actually fork it or create my own.

May I ask how you found the URLs that are used in this module? Such as https://twitter.com/i/profiles/show/${username}/timeline.

Thanks :)