have memoize recalculate in background before it's needed about webpy

[anandology]

What do you mean by just before? It is possible to trigger recomputation only when the function is called.

[aaronsw]

The first time you do the calculation, count how much time it takes (time_elapsed). Then spawn a thread, have it sleep(expires - time_elapsed) and then do the computation. Repeat.

[anandology]

So you want to start as many threads as there are items in the cache? Should it keep on recomputing after each timeout even though is not accessed?

It looks too complicated and inefficient to me.

[aaronsw]

I was only hoping to use it in cases where the function takes no arguments. The typical case I run into is that there's a database query which takes a little bit to do but the results are used on every page load. One wants to do the database query on a schedule, not on every page load. But maybe the solution is just to use background=True with a smaller expires time.

[aaronsw]

I convinced myself.