nim-libp2p
nim-libp2p copied to clipboard
vacp2p
libp2p implementation in Nim
This is to add more delay related metrics to evaluate GossipSub improvement proposals. See https://github.com/libp2p/specs/pull/413 for related discussions
Mostly relevant in test, but also nice to have in other scenarios Also, various speedups in the gossipsub CI
removing some of the repeated metrics code for per-topic metrics in PubSub and co.
Alternative to #646 This PR adds a generic builder system: ```nim import builder3 proc setup(i: int, x: float, res: var string) {.setupproc.} = res = $(i.float + x) proc setup(i:...
Epic issue to follow the various improvements for gossipsub - [x] Don't relay message to the peer who sent it to us (#625) - [x] Validate pubsub subscriptions (#627) -...
Our CI fails than more than it succeeds. Starting to get frustrating. Current issues: - go-libp2p-daemon bug: https://github.com/libp2p/interop/issues/48 - windows hangs sometimes on websocket tests? Caused by a recent dependency...
Issue to track what's needed to support the WebRTC transport (aka libp2p-webrtc-direct) ## Transport protocols The WebRTC stack: | RTCPeerConnection | DataChannel | |-------------------------------|-------------------| | SRTP | SCTP | |...
Fixes: #667. Add switch.listenError callback that is called when a transport fails to start. This is a pessimistic approach meaning that it will fail on the first address that fails...
background: https://github.com/status-im/nim-libp2p/pull/668, https://github.com/status-im/nim-libp2p/pull/662 Right now, the switch is setup like so (pseudocode): ```nim let switch = createSwitch(transports, addresses, etc) await switch.start() # rest of the owl await switch.stop() ``` This...
In `gossipsub` and `floodsub` `rpcHandler` procs, descore peers for which `ValidationResult.Reject` is returned. See https://github.com/status-im/nim-libp2p/pull/688#discussion_r794334077 for more info.