Asynchronous HTTP progress report & Pause/Resume downloads

[CaptainPRICE]

I would like to suggest to add a 'progress callback' to HTTP.

1. Expose a couple of variables via 'progress callback', such as total file size (in bytes), transferred size (downloaded/uploaded size so far, in bytes), transfer rate (download/upload speed, in bytes/sec). And perhaps also approximate time left (in seconds), so that we don't have to do calculation manually.
2. There's no way of stopping or even pausing/resuming a download/upload. Please consider implementing it.

I just tried to make HTTP downloader/uploader addon, but right now there's nothing more than 'completion callback', alone, is pretty limited functionality. I was aiming for something such as:

P.S. I know you may get the total file size by sending a HTTP HEAD request, but the thing is, a file might change between the two separate HTTP requests.

[CornerPin]

It is not possible to get the file size if server returns chunked data. Also, I think if you pause too much, the server will close the connection because of the time out.

[CaptainPRICE]

Also, I think if you pause too much, the server will close the connection because of the time out.

Yes. That's why there is "Resume capability" indication (see screenshot).

---

This makes me think, it would be best to have a class that would represent a HTTP file download/upload request, and a separate class to represent a response.

Please consider: API Draft.

I have spent some time to write that document. Thoughts? :)
P.S. Bear with me. English is not my native language.

[robotboy655]

I can see how some sort of ability to get download progress can be useful, but pausing simply cannot be added, we are using Steamworks HTTP ( https://partner.steamgames.com/doc/api/ISteamHTTP ) stuff and it doesn't have pausing.

[CaptainPRICE]

@robotboy655 I have updated API draft, please review the changes.
Is it okay? Is there anything else to drop or possibly add?