GitLab

feat: reduce allocations

After profiling allocations on a bootstrap node, I've noticed that
peerInfoToPBPeer is a very heavy source of allocations. This change should
improve the situation significantly.

query.go: Remove shuffle comment

As far as I can tell the list of peers is not being shuffled in this function. In addition [`NearestPeers`](https://github.com/libp2p/go-libp2p-kbucket/blob/a8232c8ad2e784b3966d04f86a28f9223d1d11a8/table.go#L323) returns them in sorted order.

Hope I am not missing something.

fix: optimize isRelay

This function is called frequently, let's avoid allocating as much as possible.

feat: expose WANActive

fix: improve error handling in dual dht

1. FindPeer should succeed as long as we find the peer in one DHT.
2. If we get a kbucket error from one DHT, return the error from the other DHT.
3. Avoid returning the same error twice.

fix: dedup addresses

That way, external users can easily determine which DHT they should use. E.g.,
in go-ipfs, I'd like to call GetClosestPeers in the `ipfs dht` commands based on
the "active" DHT.

Fix bug in periodic peer pinging

Dual DHT scaffold

chore: linting fixes

fix: remove local provider check

fix: use the routing table filter

It doesn't actually work. Blocks are stored under /blocks and have been since
2015.

Furthermore, this is just wrong. We shouldn't go digging through the datastore
to see if we happen to have the block. If we want something like this, we should
include a blockstore in the provider manager.