Forum OpenACS Improvement Proposals (TIPs): Re: TIP #57 (Proposed) Add caching db_* API (based on Greenpeace/Berkeley work) to 5.2

Jun, I think that sounds like a very error-prone way to deal with timed queries, which should depend on having date calculations as part of the query itself.

This is true whether or not one uses a caching db_* API, util_memoize, or ns_cache natively.

It would take a much better example than this for me to believe that per-query timeouts are more useful than dangerous.

If it is up to the develop to decide how long to cache when using util_memoize, I would vote (if I were a voter) for giving the developer the choice of when using db_* caching, perhaps with some guidance in the api-doc.

With or without caching, though, this is a great contribution. I'm hoping it gets approved!