cached로 뷰 캐시하기

Some pages cost real work to build. The cached decorator stores the entire response, so repeat visitors get an instant copy.

You add @cache.cached() right above your view function. The first request runs it; later ones replay the stored result.

Stacking matters. Put @app.route on top and @cache.cached below it, so routing happens before caching kicks in.

You can override the global setting per view. The timeout argument says how many seconds this specific response stays fresh.

By default the cache key is the request path. So /report and /report/extra are stored as separate, independent entries.

Watch out: the default key skips the query string. So /list?page=2 and /list?page=3 would share one cached entry.

To split by query string, set query_string=True. Now each unique URL, parameters included, gets its own cached response.

Need full control? Pass a make_cache_key function that builds the key yourself, for example mixing in the current user.

Cache GET requests, never POST. Caching a write would replay an old body and skip the side effects users expect.

Return unless a condition holds and the view runs fresh every time. Handy to bypass caching for logged-in users.

The first request is a cache miss and does the work. Every later request within the timeout is a fast cache hit.

Your list page uses ?page= but every page shows the same cached content.

You learned to wrap views with @cache.cached, tune the timeout, key on the full URL, and skip caching for writes. 🚀