cache

type

cache result of f for cache key T

This is best explained by the following example:

ex_cache is
  # here we define the cache keys by which the results of the function will
  # be cached in the global cache
  cache_key(val sorted_array i32) is
  #             ^-------- the actual type we want to cache
  # ^------- a feature declaring a type which we use as a cache key
  #            wrapping the actual data we want to cache
  cache_kex(val i32) is
  # similarly, here, i32 => the actual type we want to cache and cache_kex
  # is the wrapping cache key
  # suppose now, that we have two functions fn and fm, which take no arguments but
  # which need a very long time for their calculations.
  fn Function cache_key := (() ->
    time.nano.sleep (time.durations.seconds 2)
    cache_key ([6,4,1,9].sort))
  fm Function cache_kex := (() ->
    time.nano.sleep (time.durations.seconds 3)
    cache_kex 42)
  # we define two more wrappers, which will actually call the functions if
  # their value is not in the cache, or return the stored value if already
  # computed
  from_cache => (cache cache_key fn).val
  #                              ^--- will only be computed the first time we run from_cache
  from_cachf => (cache cache_kex fm).val
  say from_cache # will take two seconds
  say from_cache # will take virtually no time
  say from_cachf # will take three seconds
  say from_cache # will take virtually no time
  say from_cachf # will take virtually no time
  say from_cachf # will take virtually no time

What are effects?

The cache implemented works on a global level.

Type Parameters

Any ¶

[src]

0.095dev (GIT hash 5a97c5766ea04e5412cbfaed4c53658737cd2e4b)