Couldn’t stop myself from going back to @happybaristas today. 🤷♂️
Was working on that web extension and got the weirdest problems ever with an existing, otherwise fully functional component that I wrote.
I really prefer Rust. There I at least know what I screwed up even if I fight with the compiler all the time. I’m going to write some tool to create extended #twtxt timelines. It’s an interesting project because it looks super simple from the outside but has lots of little nifty complexities hidden in the details.
The main goal is to fetch some text files and split the lines up into the timestamp and text itself. Then bring everything together, sorted by timestamp. But it could be hundreds of files with megabytes of text. All of them need to be fetched one by one because they come from different origins and many of them send a last-updated header.
Lots of possibilities for optimization and screw-ups 😂