Cache Management

MiniFW can cache rendered page() and partial() HTML in memory. Set cache: true to cache indefinitely, or provide a millisecond time-to-live.

Cache Options

const permanent = page(renderCatalog, { cache: true });
const catalog = page(renderCatalog, { cache: { ttl: 60_000 } });
const live = page(renderCatalog, { cache: false });

Omit cache, or set it to false, to render each request. A TTL must be a positive number of milliseconds.

Cache Keys

MiniFW uses different keys for pages and partials. A key includes the route or partial name, URL pathname, query string, matched parameters, and HTMX state. Different route parameters, query values, and HTMX/full-page requests are cached separately.

Lifetime And Limits

The cache is process-local memory. It is cleared when the server restarts, is not shared between instances, and has no explicit maximum size or manual invalidation API. Expired entries are removed when their key is requested again.

Cache only output that is safe to reuse for the full cache key. Avoid caching per-user content unless the URL fully and safely distinguishes every response. For highly dynamic data, omit caching and render fresh HTML.

Cached responses are already minified. See Minification and mini() for the request pipeline.