I'm confused about #2... Resuming downloads over HTTP has been solved since the late 90s / early 2000s. Apache added support for HTTP range requests in version 1.1 (1997) and it took a few years for everyone to upgrade their servers and browsers/downloaders to versions that support it.

AFAIK both Chrome and Firefox support it, but if your browser doesn't support it then GetRight still exists!

[–] LodeMike 1 points 3 months ago (1 children)

But what if your browser restarts? Wget? How do cookies complicate that?

[–] dan@upvote.au 1 points 3 months ago* (last edited 3 months ago) (2 children)

But what if your browser restarts?

It should be able to resume where it left off. The HTTP Range header lets a client specify which part of a file they want. For a partially downloaded file, it can skip the part that has already been downloaded. If you're downloading a 1000MB file and 700MB of it was already downloaded, you'd only need the last 300MB.

This is also how multithreaded downloaders work - each thread is requesting a different piece of the file.

I think wget may support resumable downloads but I'm not 100% sure.

[–] LodeMike 1 points 3 months ago

Yes but do they actually do that?

[–] nickwitha_k@lemmy.sdf.org 1 points 3 months ago

1 is easy. Just rsync over ssh.

[–] some_guy@lemmy.sdf.org 1 points 3 months ago

These are obviously problems that can’t be fixed. As you were.

Also, it’s normal that your fan is that loud. Quit yer bitchin’.

[–] purrtastic@lemmy.nz 1 points 3 months ago

Naming variables Cache invalidation

[–] nyan@lemmy.cafe 1 points 3 months ago

All but one of those were pretty much solved 20+ years ago. Problem is, they've since been unsolved.

load more comments