[support] how cache works: on the fly content creation (not node) and caching

Hi,

I've a lot of content that get created automatically on-the-fly from a DB.

Mapping this content to nodes is not practical (custom search, just a minority of the content get actually displayed etc...).

I need a means to check if the cache is still valid at *request* time. I can't know when the cache *will* expire at request time but I know if it is already expired.

Considering that the pk for page cache is the url and that a minority of the pages will actually be cached periodically invalidating all the urls checking if the underlying DB record is younger is highly inefficient.

Before just building my cache tables and build up my own cache system on top of cache API, is there an alternative approach?