-
-
Notifications
You must be signed in to change notification settings - Fork 3.1k
DHT Refactoring #160
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
We have a good number of methods that make a single one hop request to a single given peer, i have most of these methods suffixed with 'Single' but it would be nice to come up with a name that more clearly describes their purpose and refactor them to have a similar interface. and potentially put them all in their own file. These methods are:
|
The clustering idea needs to be abstracted out more, and some pleasant interface should be made. Maybe having a channel to read peers from would be the easiest interface to satisfy, then the object can automatically move to more distant peers as it runs out of closer ones. |
@jbenet what do you think of having a peer type for the dht that contains context about routing levels? Something like a LeveledPeer, that (for coral type routing scenarios) contains information about which level of cluster its in |
All this sgtm. there's also a dht interface lying around somewhere in an issue (that both standard kademlia, coral, etc can implement for the general "routing-over-dht" |
find self in DHT when bootstrapping
This ensures cid-codec introduced in #8568 gets correctly passed (+ we maintain backward-compatibility with CIDv0) This commit was moved from ipfs/go-ipfs-http-client@9c9f43f
As @jbenet is quick to mention, the DHT could use a bit more abstraction, this issue will be a place to discuss those changes.
The text was updated successfully, but these errors were encountered: