Struct SsrCache 

pub struct SsrCache { /* private fields */ }

Multi-tier SSR cache

Architecture

1. Hot cache (L1/L2): Thread-local, 8 entries per thread
2. Cold cache (RAM): Shared DashMap with LRU eviction

Entries found in cold cache are automatically promoted to hot cache.

Implementations

impl SsrCache

pub fn new(max_cold_entries: usize) -> Self

Create a new SSR cache

Arguments

• max_cold_entries - Maximum entries in the cold cache

pub fn with_ttl(max_cold_entries: usize, ttl_secs: u64) -> Self

Create a new SSR cache with TTL

Arguments

• max_cold_entries - Maximum entries in the cold cache
• ttl_secs - Time-to-live in seconds (0 = no expiration)

pub fn try_get(&self, url: &str) -> Option<Arc<str>>

Try to get cached HTML

Checks hot cache first, then cold cache. Cold hits are promoted to hot cache.

pub fn insert(&self, url: &str, html: Arc<str>)

Insert HTML into cache

pub fn invalidate(&self, url: &str)

Invalidate a single cached URL

Removes from cold cache and bumps generation to clear hot caches. Other hot-cached entries will be re-promoted from cold on next access.

pub fn invalidate_prefix(&self, prefix: &str) -> usize

Invalidate all cached URLs that start with the given prefix

Example: cache.invalidate_prefix("/products") removes /products, /products/123, /products/foo/bar, etc.

Also bumps the generation counter to clear all hot caches, ensuring stale entries don’t survive in thread-local caches.

pub fn clear(&self)

Clear the cache, including hot caches and metrics

pub fn size(&self) -> usize

Get current cold cache size

pub fn metrics(&self) -> CacheMetrics

Get cache metrics