May 16, 2020

I started work on a new Selenium based Python script project as I find it pretty frustrating how YouTube has elminated the ability to export your Subscriptions as an OPML file anymore.

I believe they are doing this purely for the fact that there’s YouTube Premium now, and they want to squeeze as much ad revenue as possible by forcing people onto their site to view content.

However, the lack of being able to export your Subscription feed into an OPML XML file doesn’t help anyone at all who accesses their site via Screen Readers who happen to use (ex: blind people who can’t see video content but want to keep up with the audio versions of their Subscription content on YouTube purely on text based titles for videos).

This project aims to solve this issue by logging into a user’s YouTube account via Selenium with Python bindings + the Firefox web driver so that the user can rip the list of the YouTube channels into a list to then later provide to a RSS reader (ex: Newsboat, etc): https://github.com/SBanya/YoutubeSubscriptionList/

I’m not finished with it by any means yet, but am getting close but I’m running into a few issues with the Selenium driver itself as I had to obtain the latest Gecko driver here, and had to un-tar it to a specific location on my harddrive: https://github.com/mozilla/geckodriver/releases

I also learned how to use w3m more efficiently, especially with a lot of the options enabled within the ‘o’ menu to quickly refer to each line with ‘g’ + line number.

I also learned how to use w3mman which is a way to put a man page within w3m which is useful since most man pages have external links at the bottom for documentation references.

Using w3m alone is pretty sweet.

However, I would love to have my entire Tmux session to be based from Newsboat with w3m and other CLI utilities, so its been fun incorporating these tools into a terminal only workflow.