- PurgeSiteCacheAfterCommit queues the purge on a per-purger worker after
commit: one request in flight, saves committed meanwhile collapse into
one follow-up, so the admin request never waits for the site
- new cooler.SiteCachePurger.Wait drains the queue; Handle.Close waits for
it at shutdown (timeout plus one second)
- Purge stays the synchronous send; exported names and signatures unchanged
- the purge client's CheckRedirect returns http.ErrUseLastResponse: a 3xx
is logged as a failed purge with its status and the bearer token never
reaches the Location
- optional ReorderAfterSave(ctx, partition) runs after the new order is
written; an error rolls the reorder back through lifecycleFailure
- reorderPost runs in lagoon.Transaction so lagoon.AfterCommit work a hook
registers (a site-cache purge) runs only after the commit
- pact and cabana READMEs, admin hook table, tree reorder docs
- cooler.SiteCachePurger moved from the Journal plugin: loopback-only http,
32-character token, 2 s detached request, no token in logs or errors
- NewSiteCachePurgerFor names caller keys (for plugin alias keys)
- PurgeSiteCacheAfterCommit through lagoon.AfterCommit, one POST per save
- SiteCacheFromApp for hooks outside serve; Install publishes the purger
unless a plugin already did, and fails boot on an invalid config
- caching docs section, cooler README, configuration row
- Cache: Get, Set, Has, Forget, Remember (degrades on cache errors), Increment, Flush
with per-entry TTLs under <prefix>cache:
- memory driver on otter v2.3.0 bounded by cache.memory.max_entries; Redis driver
with SET PX, INCRBY and a Flush limited to SCAN MATCH <prefix>cache:* plus UNLINK
- Install publishes the Cache (sharing one Redis client with the rate limiter);
FromApp serves jobs and commands without a dial
- new docs/services/caching.md, configuration rows, cooler README
