How is this any different than existing solutions like aria2? I'd also like to see some benchmarks for the same download using the same chunk size settings with Rapel vs aria2.
I tried using existing solutions like aria2, but rapel is built to work on constrained machines with limited storage, like a raspberry pi running 24/7. You can use the post-part argument so that after downloading a chunk, you can do something with it (eg. upload it) and remove it afterward. Then, on another bigger machine, download the chunks and merge them