A big change for #Platypush - and more are on their way before the next (very big) release.

The #YouTube integration has been completely rewritten to remove all the references to the YouTube API. I've tried my best to play fair, but the YouTube API has seen way too many breaking changes recently, as a result of Google's strategy against scrapers and 3rd-party clients. I just can't keep maintaining an integration with an API provided by a company with such a hostile stance against developers. I want to spend my time making new things work, not fixing stuff purposefully broken by someone else. Even just searching for videos now requires a registered and approved Google project, and the user to be logged in: this isn't exactly the kind of stuff that is easy for anybody to set up and run.

Also, scraping results from the Web interface is no longer possible unless the user has JS enabled - which means no more easy beautifulsoup scripts, one has to summon Selenium and its whole frontend suite to scrape stuff.

From now on, the YouTube integration will use #Piped as a backend instead. A simple public API, subscription to search results and feeds through simple RSS syndacation, and no more headaches with an enshittified tech-hostile giant. This is what the developer experience with YouTube used to be until a few years ago, and what it should have remained.

https://git.platypush.tech/platypush/platypush/commit/2b12984c81e83d54d1135300b0dc5031615fe6a3