aboutsummaryrefslogtreecommitdiffhomepage
path: root/crates/shirabe/tests/plugin/subscriber_test.rs
AgeCommit message (Collapse)Author
2026-08-04test(plugin): tolerate the _composer_tmp rename in the removal testnsfisis
The PHP worker is shared across the test binary, so whichever subscriber test runs first owns the bare class name and every later install registers the plugin under a _composer_tmpN rename (as upstream does). The exact-name lookup made the removal test depend on lock acquisition order. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04feat(plugin): remove subscribed listeners when a plugin is removednsfisis
Wire removePlugin to EventDispatcher::removeListener now that subscriber listeners carry the plugin's P-table handle: PHP's $candidate[0] === $listener identity check maps to phandle equality on Callable::PhpMethod. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04feat(plugin): dispatch plugin event subscribers through the RPC workernsfisis
Wires the addPlugin subscriber branch end to end: EventSubscriberInterface and Capable become fallible and dyn-compatible (their sole implementor is the PHP plugin proxy, which answers getSubscribedEvents over RPC), listeners register as Callable::PhpMethod and are invoked with a per-call event handle, and the R table now drops entries when a child-side stub destructs. The R table keeps its IndexMap with monotonically increasing handles, so released handles are never reused and no generation counter is needed. Upstream has no subscriber-plugin test, so the path is covered by a Shirabe-owned fixture exercising all three getSubscribedEvents shapes. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>