Fossil itself is a C binary, not a database. Maybe they meant that Fossil’s source code is hosted in Fossil, or that Fossil repositories are SQLite files? I don’t exactly know either.
I'm not sure if you're being facetious and making fun of my saying stream a timestamp instead of stream to a particular time in the video, but if so I guess https://news.ycombinator.com/item?id=46364765 suggested a way.
I just expected you would stream to point X in the stream that would be to the timestamp set to start and then to point y which would be the end. Obviously it would have to be able to figure out how the streamed file would map to time, which I don't know how to do which is why I said I would like a tool that did it other than announcing I made a tool that did it myself.
Of course obviously some tools like yt-dlp etc. have this capability with the --download-sections property but I want something for torrents.
I vibe coded a little tool [0] that can stream range requests from torrents.
It's a little buggy and super rough around the edges, but it's definitely possible because your torrent client can prioritize piece requests and http standards support http range requests, just requesting parts of a document. I lightly tested it with VLC and seeking the playback to the middle of a video
You negotiate the header to find the video length, to then issue http get requests with the offset to the timestamp. Sometimes there’s an API that cuts with ffmpeg and returns the buffer. Sometimes you just need to fetch the raw bytes between offset+0 and offset+n.
I can confirm this, from my experience. Many organizations have foregone ms in lieu of redhat and oracle, and redhat is slowly injecting openshift where they can fit it.
You can see why people like the docker experience, you can manage to do all that in a single interface, instead of one off scripts touching a ton of little things
reply