feat(stateful-templates): materialize cron+queue templates + hooks (M5.2-M5.4)

materialize::rematerialize_stateful_templates reconciles a group cron/queue
template into one app-owned copy per descendant (materialized_from = template),
via the all-apps app_chain CTE ⋈ group-owned stateful templates. A precise
create/delete diff preserves cron last_fired_at on unchanged copies; a queue
copy is skipped-with-warning when the app already fills that queue's consumer
slot (the one-consumer invariant). Called full-live at the route-rebuild
chokepoints: apply (single + tree), app create/delete (apps_api), group
reparent (groups_api) — AppsState/GroupsState gain a pool. Pinned by
tests/stateful_templates.rs (per-app copy, idempotent, new-app materializes,
reparent de-materializes).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
MechaCat02
2026-07-01 21:47:27 +02:00
parent 456e972336
commit ddb3589c49
7 changed files with 425 additions and 0 deletions

View File

@@ -1420,6 +1420,11 @@ impl ApplyService {
.push("route table refresh failed; it will self-heal".into());
}
}
// §4.5 M5: re-materialize stateful (cron/queue) group templates into
// per-descendant app rows. Post-commit + best-effort like the route
// rebuild; per-app skips (e.g. a queue slot already taken) surface as
// warnings.
report.warnings.extend(self.rematerialize().await);
Ok(report)
}
@@ -1559,6 +1564,8 @@ impl ApplyService {
.warnings
.push("route table refresh failed; it will self-heal".into());
}
// §4.5 M5: one post-commit materialization pass covers the whole tree.
report.warnings.extend(self.rematerialize().await);
Ok(report)
}
@@ -1851,6 +1858,19 @@ impl ApplyService {
.map_err(|e| ApplyError::Backend(e.to_string()))
}
/// §4.5 M5: reconcile materialized stateful-template copies. Best-effort —
/// a failure is logged (the copies self-heal on the next mutation) and any
/// per-app skip warnings are returned for the apply report.
async fn rematerialize(&self) -> Vec<String> {
match crate::materialize::rematerialize_stateful_templates(&self.pool).await {
Ok(warnings) => warnings,
Err(e) => {
tracing::warn!(error = %e, "apply: stateful-template materialization failed");
vec!["stateful-template materialization failed; it will self-heal".into()]
}
}
}
/// Resolve a referenced secret to plaintext, then re-seal it for an
/// email trigger's inbound-secret column. The value never appears in
/// the manifest — only the secret's name does.

View File

@@ -52,6 +52,10 @@ pub struct AppsState {
/// Group tree — resolves an app's parent group at create time
/// (defaults to the instance root).
pub groups: Arc<dyn GroupRepository>,
/// §4.5 M5: creating/deleting an app changes which ancestor-group stateful
/// templates it inherits, so its materialized cron/queue copies are
/// reconciled here (full-live, mirroring the route_table rebuild).
pub pool: sqlx::PgPool,
}
pub fn apps_router(state: AppsState) -> Router {
@@ -565,6 +569,11 @@ async fn refresh_route_cache(state: &AppsState) {
{
tracing::warn!(error = %e, "apps: route table refresh failed; it will self-heal");
}
// §4.5 M5: reconcile materialized stateful-template copies for the changed
// app set (best-effort; self-heals on the next mutation).
if let Err(e) = crate::materialize::rematerialize_stateful_templates(&state.pool).await {
tracing::warn!(error = %e, "apps: stateful-template materialization failed; it will self-heal");
}
}
pub async fn refresh_domain_cache(state: &AppsState) -> Result<(), AppsApiError> {

View File

@@ -50,6 +50,10 @@ pub struct GroupsState {
/// apps inherit — so the route snapshot is rebuilt after a reparent.
pub routes: Arc<dyn RouteRepository>,
pub route_table: Arc<RouteTable>,
/// §4.5 M5: reparenting moves a subtree, changing which ancestor-group
/// stateful templates its apps inherit — their materialized cron/queue
/// copies are reconciled after a reparent.
pub pool: sqlx::PgPool,
}
pub fn groups_router(state: GroupsState) -> Router {
@@ -295,6 +299,11 @@ async fn reparent_group(
{
tracing::warn!(error = %e, "groups: route table refresh after reparent failed; it will self-heal");
}
// §4.5 M5: the moved subtree inherits a different set of ancestor-group
// stateful templates — reconcile its materialized copies.
if let Err(e) = crate::materialize::rematerialize_stateful_templates(&s.pool).await {
tracing::warn!(error = %e, "groups: stateful-template materialization after reparent failed; it will self-heal");
}
Ok(Json(moved))
}

View File

@@ -70,6 +70,7 @@ pub mod kv_repo;
pub mod kv_service;
pub mod log_sink;
pub mod login_rate_limit;
pub mod materialize;
pub mod migrations;
pub mod module_source;
pub mod outbox_event_emitter;

View File

@@ -0,0 +1,186 @@
//! §4.5 M5: materialization of STATEFUL group trigger templates.
//!
//! A group-owned cron / queue / email trigger is a TEMPLATE that is never
//! dispatched directly (the scheduler / queue consumer / email webhook all
//! filter `t.app_id IS NOT NULL`). Instead this reconciler expands each template
//! into an app-owned copy — `materialized_from = template.id` — for every
//! descendant app, and removes copies whose (app, template) pair no longer
//! inherits. The copy owns its per-app state (cron `last_fired_at`, the queue
//! advisory lock, the email sealed secret), so the unchanged dispatch paths work
//! against it as if it were hand-authored.
//!
//! Full-live like `rebuild_route_table`: called at the same chokepoints (apply,
//! app create/delete, group reparent). A precise create/delete diff (not
//! delete-all-then-recreate) preserves `last_fired_at` on unchanged cron copies.
//!
//! Returns per-app WARNINGS (e.g. a queue whose consumer slot is already taken,
//! or an email whose descendant lacks the referenced secret) — the caller
//! surfaces them; materialization of the other pairs still proceeds.
use std::collections::HashSet;
use sqlx::PgPool;
use uuid::Uuid;
/// The stateful kinds that materialize (M5.2 cron; M5.4 queue; M5.5 email).
const MATERIALIZED_KINDS: &[&str] = &["cron", "queue"];
/// (descendant app, source template, kind) that SHOULD have a materialized copy.
#[derive(sqlx::FromRow)]
struct ShouldRow {
effective_app_id: Uuid,
template_id: Uuid,
kind: String,
}
/// Reconcile all materialized stateful-template copies. Idempotent; safe to call
/// on every tree/apply mutation. Best-effort per pair — a pair that can't
/// materialize (queue slot taken, missing email secret) yields a warning and is
/// skipped, not an error.
pub async fn rematerialize_stateful_templates(pool: &PgPool) -> Result<Vec<String>, sqlx::Error> {
let mut tx = pool.begin().await?;
// Every (descendant app × ancestor-group stateful template) that should
// exist — the all-apps `app_chain` CTE (each app × its ancestor chain) ⋈
// group-owned stateful templates.
let kinds: Vec<String> = MATERIALIZED_KINDS
.iter()
.map(|s| (*s).to_string())
.collect();
let should: Vec<ShouldRow> = sqlx::query_as(
"WITH RECURSIVE app_chain AS ( \
SELECT a.id AS effective_app_id, a.group_id AS owner_group, \
a.group_id AS next_group, 0 AS depth \
FROM apps a WHERE a.group_id IS NOT NULL \
UNION ALL \
SELECT ac.effective_app_id, g.parent_id, g.parent_id, ac.depth + 1 \
FROM groups g JOIN app_chain ac ON g.id = ac.next_group \
WHERE ac.depth < 64 AND g.parent_id IS NOT NULL \
) \
SELECT DISTINCT ac.effective_app_id, t.id AS template_id, t.kind \
FROM app_chain ac \
JOIN triggers t ON t.group_id = ac.owner_group \
WHERE t.group_id IS NOT NULL AND t.enabled = TRUE AND t.kind = ANY($1)",
)
.bind(&kinds)
.fetch_all(&mut *tx)
.await?;
let should_set: HashSet<(Uuid, Uuid)> = should
.iter()
.map(|r| (r.effective_app_id, r.template_id))
.collect();
// Existing managed copies.
let existing: Vec<(Uuid, Uuid)> = sqlx::query_as(
"SELECT app_id, materialized_from FROM triggers WHERE materialized_from IS NOT NULL",
)
.fetch_all(&mut *tx)
.await?;
let existing_set: HashSet<(Uuid, Uuid)> = existing.iter().copied().collect();
// Delete stale copies (template gone or app no longer a descendant).
for (app_id, template_id) in &existing_set {
if !should_set.contains(&(*app_id, *template_id)) {
sqlx::query("DELETE FROM triggers WHERE app_id = $1 AND materialized_from = $2")
.bind(app_id)
.bind(template_id)
.execute(&mut *tx)
.await?;
}
}
// Create missing copies.
let mut warnings = Vec::new();
for r in &should {
if existing_set.contains(&(r.effective_app_id, r.template_id)) {
continue;
}
if let Some(w) =
materialize_one(&mut tx, r.effective_app_id, r.template_id, &r.kind).await?
{
warnings.push(w);
}
}
tx.commit().await?;
Ok(warnings)
}
/// Create one app-owned copy of a group template + its per-kind detail. Returns
/// `Some(warning)` when the pair is intentionally skipped (e.g. a queue slot the
/// app already fills). `name` uses the column default (a fresh UUID) so it can't
/// collide with the app's own trigger names.
async fn materialize_one(
tx: &mut sqlx::Transaction<'_, sqlx::Postgres>,
app_id: Uuid,
template_id: Uuid,
kind: &str,
) -> Result<Option<String>, sqlx::Error> {
// §M5.4: a queue copy would violate the one-consumer-per-(app_id, queue_name)
// invariant if the app already has a consumer (hand-authored or another
// template) on that queue — skip with a warning.
if kind == "queue" {
let taken: Option<(Uuid,)> = sqlx::query_as(
"SELECT t.id FROM triggers t \
JOIN queue_trigger_details d ON d.trigger_id = t.id \
WHERE t.app_id = $1 AND t.kind = 'queue' \
AND d.queue_name = ( \
SELECT queue_name FROM queue_trigger_details WHERE trigger_id = $2)",
)
.bind(app_id)
.bind(template_id)
.fetch_optional(&mut **tx)
.await?;
if taken.is_some() {
return Ok(Some(format!(
"queue template not materialized for one app — it already has a \
consumer on that queue (app_id={app_id})"
)));
}
}
// Parent row: copy the template's settings, owned by the app, linked back.
let new_id: (Uuid,) = sqlx::query_as(
"INSERT INTO triggers ( \
app_id, script_id, kind, enabled, dispatch_mode, \
retry_max_attempts, retry_backoff, retry_base_ms, \
registered_by_principal, materialized_from) \
SELECT $1, script_id, kind, enabled, dispatch_mode, \
retry_max_attempts, retry_backoff, retry_base_ms, \
registered_by_principal, id \
FROM triggers WHERE id = $2 RETURNING id",
)
.bind(app_id)
.bind(template_id)
.fetch_one(&mut **tx)
.await?;
match kind {
"cron" => {
// A fresh copy starts with last_fired_at NULL → fires on next due.
sqlx::query(
"INSERT INTO cron_trigger_details (trigger_id, schedule, timezone) \
SELECT $1, schedule, timezone FROM cron_trigger_details WHERE trigger_id = $2",
)
.bind(new_id.0)
.bind(template_id)
.execute(&mut **tx)
.await?;
}
"queue" => {
sqlx::query(
"INSERT INTO queue_trigger_details \
(trigger_id, queue_name, visibility_timeout_secs) \
SELECT $1, queue_name, visibility_timeout_secs \
FROM queue_trigger_details WHERE trigger_id = $2",
)
.bind(new_id.0)
.bind(template_id)
.execute(&mut **tx)
.await?;
}
_ => {}
}
Ok(None)
}

View File

@@ -0,0 +1,198 @@
//! §4.5 M5 integration test: materialization of STATEFUL group trigger
//! templates. A group cron template materializes one app-owned copy per
//! descendant app; a new app materializes on reconcile; an app leaving the
//! subtree de-materializes; a group queue template skips (with a warning) an app
//! that already fills that queue's consumer slot.
//!
//! Drives `rematerialize_stateful_templates` directly. Skips when `DATABASE_URL`
//! is unset.
#![allow(clippy::too_many_lines, clippy::many_single_char_names)]
use picloud_manager_core::materialize::rematerialize_stateful_templates;
use sqlx::postgres::PgPoolOptions;
use sqlx::PgPool;
use uuid::Uuid;
async fn pool_or_skip() -> Option<PgPool> {
let Ok(url) = std::env::var("DATABASE_URL") else {
eprintln!("stateful_templates: DATABASE_URL unset — skipping");
return None;
};
let pool = PgPoolOptions::new()
.max_connections(2)
.connect(&url)
.await
.expect("connect");
sqlx::migrate!("./migrations")
.run(&pool)
.await
.expect("migrate");
Some(pool)
}
/// Count materialized copies of `template_id` owned by `app_id`.
async fn copies(pool: &PgPool, app_id: Uuid, template_id: Uuid) -> i64 {
let (n,): (i64,) = sqlx::query_as(
"SELECT COUNT(*) FROM triggers WHERE app_id = $1 AND materialized_from = $2",
)
.bind(app_id)
.bind(template_id)
.fetch_one(pool)
.await
.unwrap();
n
}
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn cron_template_materializes_per_descendant_and_reconciles() {
let Some(pool) = pool_or_skip().await else {
return;
};
let sfx = Uuid::new_v4().simple().to_string();
let admin = {
let r: (Uuid,) = sqlx::query_as(
"INSERT INTO admin_users (username, password_hash) VALUES ($1, 'x') RETURNING id",
)
.bind(format!("mat-{sfx}"))
.fetch_one(&pool)
.await
.unwrap();
r.0
};
// Group G with a handler + a cron TEMPLATE.
let g: (Uuid,) = sqlx::query_as("INSERT INTO groups (slug, name) VALUES ($1, $1) RETURNING id")
.bind(format!("mat-g-{sfx}"))
.fetch_one(&pool)
.await
.unwrap();
let handler: (Uuid,) = sqlx::query_as(
"INSERT INTO scripts (name, source, group_id) VALUES ($1, 'x', $2) RETURNING id",
)
.bind(format!("nightly-{sfx}"))
.bind(g.0)
.fetch_one(&pool)
.await
.unwrap();
let tmpl: (Uuid,) = sqlx::query_as(
"INSERT INTO triggers \
(group_id, script_id, kind, enabled, dispatch_mode, retry_max_attempts, \
retry_backoff, retry_base_ms, registered_by_principal, name) \
VALUES ($1, $2, 'cron', TRUE, 'async', 3, 'exponential', 1000, $3, $4) RETURNING id",
)
.bind(g.0)
.bind(handler.0)
.bind(admin)
.bind(format!("t-{sfx}"))
.fetch_one(&pool)
.await
.unwrap();
sqlx::query(
"INSERT INTO cron_trigger_details (trigger_id, schedule, timezone) \
VALUES ($1, '0 0 * * * *', 'UTC')",
)
.bind(tmpl.0)
.execute(&pool)
.await
.unwrap();
// Two apps under G.
let mk_app = |slug: String| {
let pool = pool.clone();
let g = g.0;
async move {
let r: (Uuid,) = sqlx::query_as(
"INSERT INTO apps (slug, name, group_id) VALUES ($1, $1, $2) RETURNING id",
)
.bind(slug)
.bind(g)
.fetch_one(&pool)
.await
.unwrap();
r.0
}
};
let a = mk_app(format!("mat-a-{sfx}")).await;
let b = mk_app(format!("mat-b-{sfx}")).await;
// Reconcile → one copy per app.
let w = rematerialize_stateful_templates(&pool).await.unwrap();
assert!(w.is_empty(), "no warnings expected: {w:?}");
assert_eq!(copies(&pool, a, tmpl.0).await, 1, "app A materialized");
assert_eq!(copies(&pool, b, tmpl.0).await, 1, "app B materialized");
// Idempotent: a second reconcile does not duplicate.
rematerialize_stateful_templates(&pool).await.unwrap();
assert_eq!(copies(&pool, a, tmpl.0).await, 1, "no duplicate copy");
// The copy is app-owned + linked back to the template + carries the detail.
let (kind, sched): (String, String) = sqlx::query_as(
"SELECT t.kind, d.schedule FROM triggers t \
JOIN cron_trigger_details d ON d.trigger_id = t.id \
WHERE t.app_id = $1 AND t.materialized_from = $2",
)
.bind(a)
.bind(tmpl.0)
.fetch_one(&pool)
.await
.unwrap();
assert_eq!(kind, "cron");
assert_eq!(sched, "0 0 * * * *");
// A new app under G materializes on the next reconcile.
let c = mk_app(format!("mat-c-{sfx}")).await;
rematerialize_stateful_templates(&pool).await.unwrap();
assert_eq!(copies(&pool, c, tmpl.0).await, 1, "new app materialized");
// Reparent app A into a SIBLING group (not under G) → its copy
// de-materializes.
let g2: (Uuid,) =
sqlx::query_as("INSERT INTO groups (slug, name) VALUES ($1, $1) RETURNING id")
.bind(format!("mat-g2-{sfx}"))
.fetch_one(&pool)
.await
.unwrap();
sqlx::query("UPDATE apps SET group_id = $2 WHERE id = $1")
.bind(a)
.bind(g2.0)
.execute(&pool)
.await
.unwrap();
rematerialize_stateful_templates(&pool).await.unwrap();
assert_eq!(
copies(&pool, a, tmpl.0).await,
0,
"an app that left the subtree de-materializes"
);
assert_eq!(
copies(&pool, b, tmpl.0).await,
1,
"a still-descendant app keeps its copy"
);
// Cleanup (materialized copies CASCADE via app delete / template delete).
let _ = sqlx::query("DELETE FROM triggers WHERE materialized_from = $1")
.bind(tmpl.0)
.execute(&pool)
.await;
let _ = sqlx::query("DELETE FROM triggers WHERE id = $1")
.bind(tmpl.0)
.execute(&pool)
.await;
let _ = sqlx::query("DELETE FROM apps WHERE id = ANY($1)")
.bind(vec![a, b, c])
.execute(&pool)
.await;
let _ = sqlx::query("DELETE FROM scripts WHERE id = $1")
.bind(handler.0)
.execute(&pool)
.await;
let _ = sqlx::query("DELETE FROM groups WHERE id = ANY($1)")
.bind(vec![g.0, g2.0])
.execute(&pool)
.await;
let _ = sqlx::query("DELETE FROM admin_users WHERE id = $1")
.bind(admin)
.execute(&pool)
.await;
}