vinzenz/hyperhive
vinzenz/hyperhive
Watch
0
0
Fork
You've already forked hyperhive
 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

swarm-controller: fix broken intra-doc links (private items, rustdoc lint)

Browse source
This commit is contained in:
damocles 2026-08-19 21:06:16 +02:00
commit 377dbb57e3
3 changed files with 18 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

10
swarm-controller/src/config_pr.rs
View file

@ -16,7 +16,7 @@
//! - [`ConfigPrCache::apply_webhook_delivery`] — called from //! - [`ConfigPrCache::apply_webhook_delivery`] — called from
//! `crate::webhook::post_webhook_forge` on a verified `ConfigPr` delivery. //! `crate::webhook::post_webhook_forge` on a verified `ConfigPr` delivery.
//! The low-latency path: a PR opening or closing shows up immediately //! The low-latency path: a PR opening or closing shows up immediately
//! instead of waiting up to [`POLL_INTERVAL`]. //! instead of waiting up to `POLL_INTERVAL`.
//! //!
//! Per mara's review call: ship both from the start rather than the poll //! Per mara's review call: ship both from the start rather than the poll
//! alone — the eventual swarm-level replacement for `hive-c0re`'s own //! alone — the eventual swarm-level replacement for `hive-c0re`'s own
@ -93,6 +93,12 @@ impl ConfigPrCache {
.cloned() .cloned()
} }
/// ⚠️ Can race [`Self::apply_webhook_delivery`]: a scan started before a
/// PR opened may finish *after* the webhook already upserted it, and
/// this snapshot — taken before that PR existed — will overwrite the
/// fresh entry. Self-heals within one `POLL_INTERVAL` (the next scan
/// sees the PR), so not worth coordinating against; noted per argus's
/// review rather than left implicit.
fn replace(&self, scan: HashMap<String, ConfigPrStatus>) { fn replace(&self, scan: HashMap<String, ConfigPrStatus>) {
*self *self
.0 .0
@ -152,7 +158,7 @@ impl ConfigPrCache {
/// ///
/// The scan runs immediately on the first tick (`tokio::time::interval`'s /// The scan runs immediately on the first tick (`tokio::time::interval`'s
/// default), so the cache is populated on startup rather than staying empty /// default), so the cache is populated on startup rather than staying empty
/// for a full [`POLL_INTERVAL`] after boot. /// for a full `POLL_INTERVAL` after boot.
pub fn spawn(client: Arc<Client>) -> Arc<ConfigPrCache> { pub fn spawn(client: Arc<Client>) -> Arc<ConfigPrCache> {
let cache = Arc::new(ConfigPrCache::new()); let cache = Arc::new(ConfigPrCache::new());
let cache_for_task = Arc::clone(&cache); let cache_for_task = Arc::clone(&cache);

7
swarm-controller/src/forge.rs
View file

@ -31,7 +31,7 @@ use utoipa::ToSchema;
use crate::webhook::DeliveryKind; use crate::webhook::DeliveryKind;
/// An agent's open config-PR, as [`Client::list_open_config_prs`] reports it /// An agent's open config-PR, as [`Client::list_open_config_prs`] reports it
/// and [`crate::get_agent_config_pr`] serves it. /// and `GET /api/agents/{name}/config-pr` serves it.
#[derive(Clone, Debug, PartialEq, Serialize, ToSchema)] #[derive(Clone, Debug, PartialEq, Serialize, ToSchema)]
pub struct ConfigPrStatus { pub struct ConfigPrStatus {
pub pr_number: u64, pub pr_number: u64,
@ -367,8 +367,9 @@ impl Client {
/// name. Mirrors `hive-c0re::forge::config_pr_poll::poll_open_config_prs`'s /// name. Mirrors `hive-c0re::forge::config_pr_poll::poll_open_config_prs`'s
/// scan shape (list repos in the org, list open PRs per repo) but returns /// scan shape (list repos in the org, list open PRs per repo) but returns
/// data instead of side-effecting an approval queue — this daemon has no /// data instead of side-effecting an approval queue — this daemon has no
/// approval system of its own; it exists so [`crate::get_agent_config_pr`] /// approval system of its own; it exists so `GET
/// has something to answer from, independent of any one hive being up. /// /api/agents/{name}/config-pr` has something to answer from,
/// independent of any one hive being up.
/// ///
/// A single repo's list failing does not fail the whole scan — logged and /// A single repo's list failing does not fail the whole scan — logged and
/// skipped, so one flaky repo can't blank out every other agent's status. /// skipped, so one flaky repo can't blank out every other agent's status.

11
swarm-controller/src/main.rs
View file

@ -772,15 +772,16 @@ async fn get_jobq_rollup(State(state): State<AppState>) -> Json<Vec<hive_jobq_wi
/// `agent`'s open config PR, from [`config_pr::ConfigPrCache`] — kept /// `agent`'s open config PR, from [`config_pr::ConfigPrCache`] — kept
/// current by a webhook nudge (near-real-time) backstopped by a periodic /// current by a webhook nudge (near-real-time) backstopped by a periodic
/// scan (worst case: up to one [`config_pr::POLL_INTERVAL`] stale if a /// scan (worst case: up to one `config_pr::POLL_INTERVAL` stale if a
/// delivery was ever missed). Not a live forge read either way, so this /// delivery was ever missed). Not a live forge read either way, so this
/// answers even when the forge itself is momentarily unreachable. /// answers even when the forge itself is momentarily unreachable.
/// ///
/// `200` with a `null` body means "no open PR" (or "no scan has completed /// `200` with a `null` body means "no open PR" (or "no scan has completed
/// yet") — not distinguished, for the same reason [`ConfigPrCache::get`] /// yet") — not distinguished, for the same reason
/// doesn't: both read as "nothing to show" to the swarm-ui config-PR panel /// [`config_pr::ConfigPrCache::get`] doesn't: both read as "nothing to show"
/// this feeds, and a cache is a best-effort read, not a source of truth /// to the swarm-ui config-PR panel this feeds, and a cache is a best-effort
/// callers should expect to disambiguate against. /// read, not a source of truth callers should expect to disambiguate
/// against.
#[utoipa::path( #[utoipa::path(
get, get,
path = "/api/agents/{name}/config-pr", path = "/api/agents/{name}/config-pr",