Skip to main content

PageCacheInvalidator

Trait PageCacheInvalidator 

Source
pub trait PageCacheInvalidator:
    Send
    + Sync
    + 'static {
    // Required method
    fn invalidate_url<'life0, 'life1, 'life2, 'async_trait>(
        &'life0 self,
        tenant_slug: &'life1 str,
        url_path: &'life2 str,
    ) -> Pin<Box<dyn Future<Output = ()> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait,
             'life1: 'async_trait,
             'life2: 'async_trait;

    // Provided methods
    fn invalidate_subtree<'life0, 'life1, 'life2, 'async_trait>(
        &'life0 self,
        tenant_slug: &'life1 str,
        url_path: &'life2 str,
    ) -> Pin<Box<dyn Future<Output = ()> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait,
             'life1: 'async_trait,
             'life2: 'async_trait { ... }
    fn invalidate_urls<'life0, 'life1, 'life2, 'async_trait>(
        &'life0 self,
        tenant_slug: &'life1 str,
        url_paths: &'life2 [String],
    ) -> Pin<Box<dyn Future<Output = ()> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait,
             'life1: 'async_trait,
             'life2: 'async_trait { ... }
}
Expand description

Hook called by admin handlers after a page-content mutation. Implementations should fan out to whatever cache backend the public router uses (in-memory CachePageLayer, Cloudflare, Varnish, …) and drop the keys identified by the URL list.

Failures are intentionally swallowed by the handler (cache invalidation is best-effort; a stale render is worse than a failed save). Implementations should log internally.

Required Methods§

Source

fn invalidate_url<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, tenant_slug: &'life1 str, url_path: &'life2 str, ) -> Pin<Box<dyn Future<Output = ()> + Send + 'async_trait>>
where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Drop the cache entry for the public render of url_path on the given tenant. Hosts pass the unprefixed page URL — the implementation appends the configured URL prefix + host(s) before computing the cache key.

Provided Methods§

Source

fn invalidate_subtree<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, tenant_slug: &'life1 str, url_path: &'life2 str, ) -> Pin<Box<dyn Future<Output = ()> + Send + 'async_trait>>
where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Drop every cache entry whose URL starts with url_path / (the page itself + every descendant). Called after a slug rename, a move, or a subtree delete.

Source

fn invalidate_urls<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, tenant_slug: &'life1 str, url_paths: &'life2 [String], ) -> Pin<Box<dyn Future<Output = ()> + Send + 'async_trait>>
where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Drop the cache entries for many page URLs at once. The default loops over Self::invalidate_url — preserving the per-URL behavior — but backends with a bulk-purge API (Cloudflare files, Azure CDN contentPaths) should override this to issue a single request instead of N. The purge fan-out in crate::task_queue::purge_urls calls this so a batch backend collapses a multi-URL save into one round-trip.

Dyn Compatibility§

This trait is dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety".

Implementors§