# kademlia_routing_table - Change Log ## [0.6.0] - Remove the xor_name dependency. Instead, `XorName` and other users need to implement the `Xorable` trait. - Rewrite the unit tests using `u8` instead of `XorName`. This should be both faster and more readable. - Use `u64` in the network tests instead of `XorName`. - Make the group size configurable: `bucket_size` specifies the maximum group size that close groups can be computed for, and `extra_entries` specifies how many additional entries are tolerated in each bucket. - Make `target_nodes` take a `route` argument that allows specifying any of `bucket_size` parallel routes to the destination. ## [0.5.2] - Reduce `PARALLELISM` to 4 again to reduce redundant traffic. - Swarm a message if the target is not in the routing table. - Fix several tests. ## [0.5.1] - Extend the API to facilitate finding unneeded connections and querying bucket sizes. ## [0.5.0] - Add functionality to detect unneeded routing table entries, so that unnecessarily large numbers of connections can be avoided. ## [0.4.1] - Don't block re-swarming messages and set `PARALLELISM` to 8. ## [0.4.0] - Made `closest_nodes_to` public. - Updated dependencies. ## [0.3.3] - Don't relay messages to the node they were received from. ## [0.3.2] - Allow get to return own node info. ## [0.3.1] - Add a `find` method. ## [0.3.0] - Make Routing generic again, without the previous restrictions, so it can store more peer information in addition to the `XorName` again. - Add lots of integration tests. ## [0.2.0] - Only manage `XorName`s, not connections or public IDs. - Keep contacts sorted in buckets to speed up several methods. ## [0.1.1] - Add the is_recipient method to move that logic out of the Routing crate. ## [0.1.0] - Major changes to the routing logic, so that some useful properties can be guaranteed. - Expand the documentation. ## [0.0.5] - Change the close group definition to ensure quorum can always be reached. - Relay messages to the nodes closest to the target instead of our close group. ## [0.0.4] - Add bucket index to NodeInfo\ - hash_node -> get (return option & NodeInfo) - reduced some if statements with if/else if blocks ## [0.0.3] - Remove unneeded library (clippy) - Fixed typo in parallelism methods ## [0.0.2] - Added functions to return constant values as usize - Moved constants to u8 - Added function to return dynamic quorum size - unpublished in crates.io ## [0.0.1] - Initial Implementation