peer-to-peer

Hi. I've been experiencing good reliability syncing Maildirs with Syncthing for a few weeks already - however, from time to time an occasional empty mail may now show up in my IMAP folders/e-mail client. I suspect that this is due to the partial transfers and/or temp files being read by dovecot (perhaps locking them somehow could help with this?). I was able to ignore dovecot's index files in Sync

Hello!

When text line in list is too long and part of it is moved on the next line, list number is placed on the last line. When I use <br>, list number is placed on the first line of multiline text.

Demo is below:

Demo text:

1. [ ] task 1
2. [ ] task 2 task 2 task 2

Here is what I've noticed:

• ipfs.block.get
  • docs say it returns a Block, but it returns a Uint8Array
  • docs are missing the preload option
• ipfs.block.put
  • docs say it can also take a Block as a parameter, but it can only be a Uint8Array
  • docs say it can take a cid as an option, but the implementation ignores such a parameter
  • docs say it returns a Block, b

All of the CI builds are flavors of linux. It'd be great to add some Windows / Mac ones.

Anybody know github actions and want to help out?

Background

If my node is asked to forward N sats to peer X via channel C, lnd is free to pick any other channel it might have to peer X. As far as I understand the code (switch.go:handlePacketForward), all channels to X (not just C) are checked for eligability, and a random eligable channel is then used for the actual forward.

As part of this eligability check (implemented in link.go:Che

Since the URLs are just a gate pass for the video call, there might be a chance of getting intruders into the call with a simple brute force.

Users feel more secure if we have some passcode or approval from the creator to join into call.

This was cut from the v0.2.0 release for present time constraints however making ipfs-http more accessible as "the" ipfs binary would make the project more accessible to anyone just wanting to try it out. At the moment I see the following obstacles:

• http/README.md has no tutorial on how to try out ipfs-http
• ipfs-http supports only `i