2-- Module : Streamly.Internal.Data.Array.Foreign
3-- Copyright : (c) 2019 Composewell Technologies
6-- Maintainer : streamly@composewell.com
7-- Stability : experimental
12-- * Arrays are finite and fixed in size
13-- * provide /O(1)/ access to elements
14-- * store only data and not functions
15-- * provide efficient IO interfacing
17-- 'Foldable' instance is not provided because the implementation would be much
18-- less efficient compared to folding via streams. 'Semigroup' and 'Monoid'
19-- instances should be used with care; concatenating arrays using binary
20-- operations can be highly inefficient. Instead, use
21-- 'Streamly.Internal.Data.Array.Stream.Foreign.toArray' to concatenate N
24-- Each array is one pointer visible to the GC. Too many small arrays (e.g.
25-- single byte) are only as good as holding those elements in a Haskell list.
26-- However, small arrays can be compacted into large ones to reduce the
27-- overhead. To hold 32GB memory in 32k sized buffers we need 1 million arrays
28-- if we use one array for each chunk. This is still significant to add
31module Streamly.Internal.Data.Array.Foreign
37 -- Pure, From Static Memory (Unsafe)
38 -- We can use fromPtrM#, fromCStringM# and fromAddrM# to create arrays from
39 -- a dynamic memory address which requires a finalizer.
53 , A.writeN -- drop new
55 , A.write -- full buffer
63 , unsafeRead -- XXX readUnsafe?
65 , producer -- experimental
70 , A.unsafeIndex -- XXX Rename to getIndexUnsafe??
72 , last -- XXX getIndexLast?
74 , getIndicesFromThenTo
75 -- , getIndicesFrom -- read from a given position to the end of file
76 -- , getIndicesUpto -- read from beginning up to the given position
78 -- , getIndicesFromRev -- read from a given position to the beginning of file
79 -- , getIndicesUptoRev -- read from end to the given position in file
97 , A.unsafeFreeze -- asImmutableUnsafe?
98 , A.unsafeThaw -- asMutableUnsafe?
107 -- * Streaming Operations
117#include "ArrayMacros.h"
119import Control.Exception (assert)
120import Control.Monad (when)
121import Control.Monad.IO.Class (MonadIO(..))
122import Data.Functor.Identity (Identity)
123import Data.Word (Word8)
124import Foreign.C.String (CString)
125import Foreign.Ptr (plusPtr, castPtr)
126import Foreign.Storable (Storable(..))
127import Prelude hiding (length, null, last, map, (!!), read, concat)
129import Streamly.Internal.Data.Array.Foreign.Mut.Type (ReadUState(..), touch)
130import Streamly.Internal.Data.Array.Foreign.Type
131 (Array(..), length, asPtrUnsafe)
132import Streamly.Internal.Data.Fold.Type (Fold(..))
133import Streamly.Internal.Data.Producer.Type (Producer(..))
134import Streamly.Internal.Data.Stream.Serial (SerialT(..))
135import Streamly.Internal.Data.Tuple.Strict (Tuple3Fused'(..))
136import Streamly.Internal.Data.Unfold.Type (Unfold(..))
137import Streamly.Internal.System.IO (unsafeInlineIO)
139import qualified Streamly.Internal.Data.Array.Foreign.Mut.Type as MA
140import qualified Streamly.Internal.Data.Array.Foreign.Mut as MA
141import qualified Streamly.Internal.Data.Array.Foreign.Type as A
142import qualified Streamly.Internal.Data.Fold as FL
143import qualified Streamly.Internal.Data.Producer.Type as Producer
144import qualified Streamly.Internal.Data.Producer as Producer
145import qualified Streamly.Internal.Data.Stream.Common as P
146import qualified Streamly.Internal.Data.Stream.StreamD as D
147import qualified Streamly.Internal.Data.Unfold as Unfold
148import qualified Streamly.Internal.Data.Ring.Foreign as RB
150-------------------------------------------------------------------------------
152-------------------------------------------------------------------------------
154-- | Create an 'Array' from the first N elements of a stream. The array is
155-- allocated to size N, if the stream terminates before N elements then the
156-- array may hold less than N elements.
159{-# INLINE fromStreamN #-}
160fromStreamN :: (MonadIO m, Storable a) => Int -> SerialT m a -> m (Array a)
161fromStreamN n (SerialT m) = do
162 when (n < 0) $ error "writeN: negative write count specified"
163 A.fromStreamDN n $ D.fromStreamK m
165-- | Create an 'Array' from a stream. This is useful when we want to create a
166-- single array from a stream of unknown size. 'writeN' is at least twice
167-- as efficient when the size is already known.
169-- Note that if the input stream is too large memory allocation for the array
170-- may fail. When the stream size is not known, `arraysOf` followed by
171-- processing of indvidual arrays in the resulting stream should be preferred.
174{-# INLINE fromStream #-}
175fromStream :: (MonadIO m, Storable a) => SerialT m a -> m (Array a)
176fromStream (SerialT m) = P.fold A.write m
177-- write m = A.fromStreamD $ D.fromStreamK m
179-------------------------------------------------------------------------------
181-------------------------------------------------------------------------------
183{-# INLINE_NORMAL producer #-}
184producer :: forall m a. (Monad m, Storable a) => Producer m (Array a) a
186 Producer.translate A.unsafeThaw A.unsafeFreeze
187 $ MA.producerWith (return . unsafeInlineIO)
189-- | Unfold an array into a stream.
191-- /Since 0.7.0 (Streamly.Memory.Array)/
194{-# INLINE_NORMAL read #-}
195read :: forall m a. (Monad m, Storable a) => Unfold m (Array a) a
196read = Producer.simplify producer
198-- | Unfold an array into a stream, does not check the end of the array, the
199-- user is responsible for terminating the stream within the array bounds. For
200-- high performance application where the end condition can be determined by
201-- a terminating fold.
203-- Written in the hope that it may be faster than "read", however, in the case
204-- for which this was written, "read" proves to be faster even though the core
205-- generated with unsafeRead looks simpler.
209{-# INLINE_NORMAL unsafeRead #-}
210unsafeRead :: forall m a. (Monad m, Storable a) => Unfold m (Array a) a
211unsafeRead = Unfold step inject
214 inject (Array contents start end) =
215 return (ReadUState contents end start)
217 {-# INLINE_LATE step #-}
218 step (ReadUState contents end p) = do
219 -- unsafeInlineIO allows us to run this in Identity monad for pure
220 -- toList/foldr case which makes them much faster due to not
221 -- accumulating the list and fusing better with the pure consumers.
223 -- This should be safe as the array contents are guaranteed to be
224 -- evaluated/written to before we peek at them.
225 let !x = unsafeInlineIO $ do
229 let !p1 = PTR_NEXT(p,a)
230 return $ D.Yield x (ReadUState contents end p1)
234-- >>> import qualified Streamly.Internal.Data.Array.Foreign.Type as Array
235-- >>> null arr = Array.byteLength arr == 0
239null :: Array a -> Bool
240null arr = A.byteLength arr == 0
242-- | Like 'getIndex' but indexes the array in reverse from the end.
245{-# INLINE getIndexRev #-}
246getIndexRev :: forall a. Storable a => Int -> Array a -> Maybe a
251 let elemPtr = PTR_RINDEX(aEnd arr,i,a)
252 if i >= 0 && elemPtr >= ptr
253 then Just <$> peek elemPtr
258-- >>> import qualified Streamly.Internal.Data.Array.Foreign as Array
259-- >>> last arr = Array.getIndexRev arr 0
263last :: Storable a => Array a -> Maybe a
266-------------------------------------------------------------------------------
267-- Folds with Array as the container
268-------------------------------------------------------------------------------
270-- | @writeLastN n@ folds a maximum of @n@ elements from the end of the input
271-- stream to an 'Array'.
274{-# INLINE writeLastN #-}
275writeLastN :: (Storable a, MonadIO m) => Int -> Fold m a (Array a)
277 | n <= 0 = fmap (const mempty) FL.drain
278 | otherwise = A.unsafeFreeze <$> Fold step initial done
282 step (Tuple3Fused' rb rh i) a = do
283 rh1 <- liftIO $ RB.unsafeInsert rb rh a
284 return $ FL.Partial $ Tuple3Fused' rb rh1 (i + 1)
287 let f (a, b) = FL.Partial $ Tuple3Fused' a b (0 :: Int)
288 in fmap f $ liftIO $ RB.new n
290 done (Tuple3Fused' rb rh i) = do
291 arr <- liftIO $ MA.newArray n
292 foldFunc i rh snoc' arr rb
294 -- XXX We should write a read unfold for ring.
295 snoc' b a = liftIO $ MA.snocUnsafe b a
298 | i < n = RB.unsafeFoldRingM
299 | otherwise = RB.unsafeFoldRingFullM
301-------------------------------------------------------------------------------
303-------------------------------------------------------------------------------
305-------------------------------------------------------------------------------
307-------------------------------------------------------------------------------
309-- | Given a sorted array, perform a binary search to find the given element.
310-- Returns the index of the element if found.
313{-# INLINE binarySearch #-}
314binarySearch :: a -> Array a -> Maybe Int
315binarySearch = undefined
317-- find/findIndex etc can potentially be implemented more efficiently on arrays
318-- compared to streams by using SIMD instructions.
319-- We can also return a bit array instead.
321-- | Perform a linear search to find all the indices where a given element is
322-- present in an array.
325findIndicesOf :: (a -> Bool) -> Unfold Identity (Array a) Int
326findIndicesOf = undefined
329findIndexOf :: (a -> Bool) -> Array a -> Maybe Int
330findIndexOf p = Unfold.fold Fold.head . Stream.unfold (findIndicesOf p)
332find :: (a -> Bool) -> Array a -> Bool
333find = Unfold.fold Fold.null . Stream.unfold (findIndicesOf p)
336-------------------------------------------------------------------------------
338-------------------------------------------------------------------------------
340-- XXX We can potentially use SIMD instructions on arrays to fold faster.
342-------------------------------------------------------------------------------
344-------------------------------------------------------------------------------
346-- | /O(1)/ Slice an array in constant time.
348-- Caution: The bounds of the slice are not checked.
353{-# INLINE getSliceUnsafe #-}
356 => Int -- ^ starting index
357 -> Int -- ^ length of the slice
360getSliceUnsafe index len (Array contents start e) =
361 let size = SIZE_OF(a)
362 fp1 = start `plusPtr` (index * size)
363 end = fp1 `plusPtr` (len * size)
364 in assert (end <= e) (Array contents fp1 end)
366-- | Split the array into a stream of slices using a predicate. The element
367-- matching the predicate is dropped.
370{-# INLINE splitOn #-}
371splitOn :: (Monad m, Storable a) =>
372 (a -> Bool) -> Array a -> SerialT m (Array a)
373splitOn predicate arr =
374 SerialT $ D.toStreamK
375 $ fmap (\(i, len) -> getSliceUnsafe i len arr)
376 $ D.sliceOnSuffix predicate (A.toStreamD arr)
378{-# INLINE genSlicesFromLen #-}
379genSlicesFromLen :: forall m a. (Monad m, Storable a)
380 => Int -- ^ from index
381 -> Int -- ^ length of the slice
382 -> Unfold m (Array a) (Int, Int)
383genSlicesFromLen from len =
384 Unfold.lmap A.unsafeThaw (MA.genSlicesFromLen from len)
386-- | Generate a stream of slices of specified length from an array, starting
387-- from the supplied array index. The last slice may be shorter than the
391{-# INLINE getSlicesFromLen #-}
392getSlicesFromLen :: forall m a. (Monad m, Storable a)
393 => Int -- ^ from index
394 -> Int -- ^ length of the slice
395 -> Unfold m (Array a) (Array a)
396getSlicesFromLen from len =
398 $ Unfold.lmap A.unsafeThaw (MA.getSlicesFromLen from len)
400-------------------------------------------------------------------------------
401-- Random reads and writes
402-------------------------------------------------------------------------------
404-- XXX Change this to a partial function instead of a Maybe type? And use
405-- MA.getIndex instead.
407-- | /O(1)/ Lookup the element at the given index. Index starts from 0.
410{-# INLINE getIndex #-}
411getIndex :: forall a. Storable a => Int -> Array a -> Maybe a
416 let elemPtr = PTR_INDEX(ptr,i,a)
417 if i >= 0 && PTR_VALID(elemPtr,aEnd arr,a)
418 then Just <$> peek elemPtr
421-- | Given a stream of array indices, read the elements on those indices from
422-- the supplied Array. An exception is thrown if an index is out of bounds.
424-- This is the most general operation. We can implement other operations in
429-- let u = lmap (\arr -> (0, length arr - 1)) Unfold.enumerateFromTo
430-- in Unfold.lmap f (getIndices arr)
433-- let i = length arr - 1
434-- in Unfold.lmap f (getIndicesFromThenTo i (i - 1) 0)
438{-# INLINE getIndices #-}
439getIndices :: (Monad m, Storable a) => SerialT m Int -> Unfold m (Array a) a
440getIndices (SerialT stream) =
441 let unf = MA.getIndicesD (return . unsafeInlineIO) $ D.fromStreamK stream
442 in Unfold.lmap A.unsafeThaw unf
444-- | Unfolds @(from, then, to, array)@ generating a finite stream whose first
445-- element is the array value from the index @from@ and the successive elements
446-- are from the indices in increments of @then@ up to @to@. Index enumeration
447-- can occur downwards or upwards depending on whether @then@ comes before or
451-- getIndicesFromThenTo =
452-- let f (from, next, to, arr) =
453-- (Stream.enumerateFromThenTo from next to, arr)
454-- in Unfold.lmap f getIndices
458{-# INLINE getIndicesFromThenTo #-}
459getIndicesFromThenTo :: Unfold m (Int, Int, Int, Array a) a
460getIndicesFromThenTo = undefined
462-------------------------------------------------------------------------------
463-- Transform via stream operations
464-------------------------------------------------------------------------------
466-- for non-length changing operations we can use the original length for
467-- allocation. If we can predict the length then we can use the prediction for
468-- new allocation. Otherwise we can use a hint and adjust dynamically.
471-- | Transform an array into another array using a pipe transformation
475{-# INLINE runPipe #-}
476runPipe :: (MonadIO m, Storable a, Storable b)
477 => Pipe m a b -> Array a -> m (Array b)
478runPipe f arr = P.runPipe (toArrayMinChunk (length arr)) $ f (A.read arr)
481-- XXX For transformations that cannot change the number of elements e.g. "map"
482-- we can use a predetermined array length.
484-- | Transform an array into another array using a stream transformation
488{-# INLINE streamTransform #-}
489streamTransform :: forall m a b. (MonadIO m, Storable a, Storable b)
490 => (SerialT m a -> SerialT m b) -> Array a -> m (Array b)
491streamTransform f arr =
492 P.fold (A.writeWith (length arr)) $ getSerialT $ f (A.toStream arr)
494-------------------------------------------------------------------------------
496-------------------------------------------------------------------------------
498-- | Cast an array having elements of type @a@ into an array having elements of
499-- type @b@. The array size must be a multiple of the size of type @b@
500-- otherwise accessing the last element of the array may result into a crash or
510castUnsafe (Array contents start end) =
511 Array contents (castPtr start) (castPtr end)
513-- | Cast an @Array a@ into an @Array Word8@.
517asBytes :: Array a -> Array Word8
520-- | Cast an array having elements of type @a@ into an array having elements of
521-- type @b@. The length of the array should be a multiple of the size of the
522-- target element otherwise 'Nothing' is returned.
526cast :: forall a b. (Storable b) => Array a -> Maybe (Array b)
528 let len = A.byteLength arr
529 r = len `mod` SIZE_OF(b)
532 else Just $ castUnsafe arr
534-- | Convert an array of any type into a null terminated CString Ptr.
538-- /O(n) Time: (creates a copy of the array)/
542asCStringUnsafe :: Array a -> (CString -> IO b) -> IO b
543asCStringUnsafe arr act = do
544 let arr1 = asBytes arr <> A.fromList [0]
545 asPtrUnsafe arr1 $ \ptr -> act (castPtr ptr)
547-------------------------------------------------------------------------------
549-------------------------------------------------------------------------------
551-- | Fold an array using a 'Fold'.
555fold :: forall m a b. (Monad m, Storable a) => Fold m a b -> Array a -> m b
556fold f arr = P.fold f (getSerialT (A.toStream arr))
558-- | Fold an array using a stream fold operation.
561{-# INLINE streamFold #-}
562streamFold :: (Monad m, Storable a) => (SerialT m a -> m b) -> Array a -> m b
563streamFold f arr = f (A.toStream arr)
keysesc to close
- j k
- move down, up
- enter
- open
- esc
- back, or up a level
- 1 2 3 4
- readme, files, commits, refs
- /
- filter
- y
- copy the clone command
- c
- compare the chosen commit with another (or shift-click both)
- ?
- show or hide this list