mirror of
https://github.com/facebook/Haxl.git
synced 2024-12-24 01:04:21 +03:00
70f5bad436
Summary: Here I try to populate the writes done as part of a Haxl computation in an IORef inside the Environment. `IVar` which is the synchornisation point, also acts as the point where we store intermediate writes for Haxl computations, so they can be memoized and reused whenever a memoized computation is done again. This is done inside `getIVarWithWrites` function. This works, because we create new IVars when running a memo computation or a data fetch, and it is only at these places where we need to create a new environment with empty writes to run the computation in. So I run every memoized computation in a new environment (with empty writes) and populate the writes in this new environment. At the end of the memoized computation, I look up these writes from the `IVar` and also add them to the original environment. This way ultimately all writes are correctly propagated upwards to the top level environment user passes to `runHaxl`. This logic lives inside `execMemoNow`. Reviewed By: simonmar Differential Revision: D14342181 fbshipit-source-id: a410dae1a477f27b480804b67b2212e7500997ab
163 lines
5.3 KiB
Haskell
163 lines
5.3 KiB
Haskell
-- Copyright (c) 2014-present, Facebook, Inc.
|
|
-- All rights reserved.
|
|
--
|
|
-- This source code is distributed under the terms of a BSD license,
|
|
-- found in the LICENSE file.
|
|
|
|
{-# LANGUAGE DeriveDataTypeable #-}
|
|
{-# LANGUAGE GADTs #-}
|
|
{-# LANGUAGE MultiParamTypeClasses #-}
|
|
{-# LANGUAGE OverloadedStrings #-}
|
|
{-# LANGUAGE StandaloneDeriving #-}
|
|
{-# LANGUAGE TypeFamilies #-}
|
|
{-# LANGUAGE GeneralizedNewtypeDeriving #-}
|
|
{-# LANGUAGE FlexibleInstances #-}
|
|
|
|
module ExampleDataSource (
|
|
-- * initialise the state
|
|
initGlobalState,
|
|
|
|
-- * requests for this data source
|
|
Id(..), ExampleReq(..),
|
|
countAardvarks,
|
|
listWombats,
|
|
) where
|
|
|
|
import Haxl.Prelude
|
|
import Prelude ()
|
|
|
|
import Haxl.Core
|
|
|
|
import Data.Typeable
|
|
import Data.Hashable
|
|
import Control.Concurrent
|
|
import System.IO
|
|
|
|
-- Here is an example minimal data source. Our data source will have
|
|
-- two requests:
|
|
--
|
|
-- countAardvarks :: String -> Haxl Int
|
|
-- listWombats :: Id -> Haxl [Id]
|
|
--
|
|
-- First, the data source defines a request type, with one constructor
|
|
-- for each request:
|
|
|
|
newtype Id = Id Int
|
|
deriving (Eq, Ord, Enum, Num, Integral, Real, Hashable, Typeable)
|
|
|
|
instance Show Id where
|
|
show (Id i) = show i
|
|
|
|
data ExampleReq a where
|
|
CountAardvarks :: String -> ExampleReq Int
|
|
ListWombats :: Id -> ExampleReq [Id]
|
|
deriving Typeable -- requests must be Typeable
|
|
|
|
-- The request type (ExampleReq) is parameterized by the result type of
|
|
-- each request. Each request might have a different result, so we use a
|
|
-- GADT - a data type in which each constructor may have different type
|
|
-- parameters. Here CountAardvarks is a request that takes a String
|
|
-- argument and its result is Int, whereas ListWombats takes an Id
|
|
-- argument and returns a [Id].
|
|
|
|
-- The request type needs instances for 'Eq1' and 'Hashable1'. These
|
|
-- are like 'Eq' and 'Hashable', but for types with one parameter
|
|
-- where the parameter is irrelevant for hashing and equality.
|
|
-- These two instances are used to support caching of requests.
|
|
|
|
-- We need Eq, but we have to derive it with a standalone declaration
|
|
-- like this, because plain deriving doesn't work with GADTs.
|
|
deriving instance Eq (ExampleReq a)
|
|
|
|
deriving instance Show (ExampleReq a)
|
|
|
|
instance ShowP ExampleReq where showp = show
|
|
|
|
instance Hashable (ExampleReq a) where
|
|
hashWithSalt s (CountAardvarks a) = hashWithSalt s (0::Int,a)
|
|
hashWithSalt s (ListWombats a) = hashWithSalt s (1::Int,a)
|
|
|
|
instance StateKey ExampleReq where
|
|
data State ExampleReq = ExampleState {
|
|
-- in here you can put any state that the
|
|
-- data source needs to maintain throughout the
|
|
-- run.
|
|
}
|
|
|
|
-- Next we need to define an instance of DataSourceName:
|
|
|
|
instance DataSourceName ExampleReq where
|
|
dataSourceName _ = "ExampleDataSource"
|
|
|
|
-- Next we need to define an instance of DataSource:
|
|
|
|
instance DataSource u ExampleReq where
|
|
-- I'll define exampleFetch below
|
|
fetch = exampleFetch
|
|
|
|
|
|
-- Every data source should define a function 'initGlobalState' that
|
|
-- initialises the state for that data source. The arguments to this
|
|
-- function might vary depending on the data source - we might need to
|
|
-- pass in resources from the environment, or parameters to set up the
|
|
-- data source.
|
|
initGlobalState :: IO (State ExampleReq)
|
|
initGlobalState = do
|
|
-- initialize the state here.
|
|
return ExampleState { }
|
|
|
|
|
|
-- The most important bit: fetching the data. The fetching function
|
|
-- takes a list of BlockedFetch, which is defined as
|
|
--
|
|
-- data BlockedFetch r
|
|
-- = forall a . BlockedFetch (r a) (ResultVar a)
|
|
--
|
|
-- That is, each BlockedFetch is a pair of
|
|
--
|
|
-- - the request to fetch (with result type a)
|
|
-- - a ResultVar to store either the result or an error
|
|
--
|
|
-- The job of fetch is to fetch the data and fill in all the ResultVars.
|
|
--
|
|
exampleFetch :: State ExampleReq -- current state
|
|
-> Flags -- tracing verbosity, etc.
|
|
-> u -- user environment
|
|
-> PerformFetch ExampleReq -- tells the framework how to fetch
|
|
|
|
exampleFetch _state _flags _user = SyncFetch $ mapM_ fetch1
|
|
|
|
-- There are two ways a data source can fetch data: synchronously or
|
|
-- asynchronously. See the type 'PerformFetch' in "Haxl.Core.Types" for
|
|
-- details.
|
|
|
|
fetch1 :: BlockedFetch ExampleReq -> IO ()
|
|
fetch1 (BlockedFetch (CountAardvarks "BANG") _) =
|
|
error "BANG" -- data sources should not throw exceptions, but in
|
|
-- the event that one does, the framework will
|
|
-- propagate the exception to the call site of
|
|
-- dataFetch.
|
|
fetch1 (BlockedFetch (CountAardvarks "BANG2") m) = do
|
|
putSuccess m 1
|
|
error "BANG2" -- the exception is propagated even if we have already
|
|
-- put the result with putSuccess
|
|
fetch1 (BlockedFetch (CountAardvarks "BANG3") _) = do
|
|
hPutStr stderr "BANG3"
|
|
killThread =<< myThreadId -- an asynchronous exception
|
|
fetch1 (BlockedFetch (CountAardvarks str) m) =
|
|
putSuccess m (length (filter (== 'a') str))
|
|
fetch1 (BlockedFetch (ListWombats a) r) =
|
|
if a > 999999
|
|
then putFailure r $ FetchError "too large"
|
|
else putSuccess r $ take (fromIntegral a) [1..]
|
|
|
|
|
|
-- Normally a data source will provide some convenient wrappers for
|
|
-- its requests:
|
|
|
|
countAardvarks :: String -> GenHaxl u w Int
|
|
countAardvarks str = dataFetch (CountAardvarks str)
|
|
|
|
listWombats :: Id -> GenHaxl u w [Id]
|
|
listWombats i = dataFetch (ListWombats i)
|