chore: deprecate core module#311
Conversation
|
|
||
| This section details what behavior to expect when using Kyoto, and why such decisions were made. | ||
|
|
||
| ### Peer Selection |
There was a problem hiding this comment.
Is jumping from H1 to H3 a style choice?
| //! Usual sync on Testnet. | ||
|
|
||
| use kyoto::{chain::checkpoints::HeaderCheckpoint, core::builder::NodeBuilder}; | ||
| use kyoto::{builder::NodeBuilder, chain::checkpoints::HeaderCheckpoint}; |
There was a problem hiding this comment.
Not shown in the diff, but managed and tor examples will likely also need an upgrade. They require the additional features flag when building.
There was a problem hiding this comment.
Ah missed that ok, I am gonna try and add that feature flag matrix to CI in a separate PR too.
nyonson
force-pushed
the
break-up-core
branch
2 times, most recently
from
1088d1c to
3da5874
Compare
|
Ok everything iron'd out in 3da5874 except tor example, but that is broke on master right now too, will address separately. |
nyonson
force-pushed
the
break-up-core
branch
3 times, most recently
from
e526b87 to
1a6ffb4
Compare
| crate::client::{Client, Requester}, | ||
| crate::error::{ClientError, NodeError}, | ||
| crate::messages::{Event, Log, Progress, RejectPayload, SyncUpdate, Warning}, | ||
| crate::network::PeerTimeoutConfig, |
There was a problem hiding this comment.
This was previously used internally and did not hit the public API, but I think another method on the builder to allow users to use the PeerTimeoutConfig type would be a good follow up
|
We should probably remove the "Getting Started" section of the documentation in |
Remove the catch-all core module by promoting the implementation to the top level kyoto module. Further refactors can break this up into existing and new feature specific modules.
|
ACK 41a9ac6 |
2 participants
Remove the catch-all core module by promoting the implementation to the top level kyoto module. Some refinements were made by pushing things down into existing modules, like the peer_map to network. I still some smaller questions that can be dealt with later, like does broadcaster need to be its own module? Should that also live in network?
Related to #310