# LRU Cache from cache import AsyncLRU @AsyncLRU(maxsize=128) async def func(*args, **kwargs): """  maxsize : max number of results that are cached.  if max limit is reached the oldest result is deleted.  """ pass # TTL Cache from cache import AsyncTTL @AsyncTTL(time_to_live=60, maxsize=1024, skip_args=1) async def func(*args, **kwargs): """  time_to_live : max time for which a cached result is valid (in seconds)  maxsize : max number of results that are cached.  if max limit is reached the oldest result is deleted.  skip_args : Use `1` to skip first arg of func in determining cache key  """ pass # Supports primitive as well as non-primitive function parameter. # Currently TTL & LRU cache is supported.

class CustomDataClass: id: int value: int from cache import AsyncLRU @AsyncLRU(maxsize=128) async def func(model: "CustomDataClass"): ... # function logic ... # async-cache will work even if function parameters are: # 1. orm objects # 2. request object # 3. any other custom object type. # To refresh the function result use the `use_cache=False` param in the function invocation func(*args, use_cache=False, **kwargs)