camel-bridge 0.23.0

Bridge process lifecycle management for rust-camel (spawn, health, download)
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

use crate::process::BridgeError;
use tonic::transport::Channel;

/// Implemented by components that hold stateful resources in a bridge
/// (e.g. compiled XSDs, compiled XSLT stylesheets) and need to re-seed
/// those resources when the bridge process restarts.
///
/// The orchestration loop in each component calls `on_reconnect` after the
/// new bridge process announces its port and a mTLS channel is established.
///
/// # Contract
/// - MUST be non-blocking in the synchronous sense: implementors should spawn
///   a Tokio task for async work rather than blocking the reconnect loop.
/// - Returning `Err` is advisory: the reconnect loop logs the error but does
///   NOT abort — the bridge is considered live; individual resource re-seeds
///   may be retried lazily.
pub trait BridgeReconnectHandler: Send + Sync + std::fmt::Debug {
    /// Called after a new bridge process starts and its channel is connected.
    ///
    /// `channel` is the mTLS tonic channel to the new bridge.
    fn on_reconnect(&self, channel: &Channel) -> Result<(), BridgeError>;
}