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
94 , unsafeCast -- castUnsafe?
96 , unsafeAsCString -- asCStringUnsafe?
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)
123#if !(MIN_VERSION_base(4,11,0))
124import Data.Semigroup ((<>))
126import Data.Word (Word8)
127import Foreign.C.String (CString)
128import Foreign.Ptr (plusPtr, castPtr)
129import Foreign.Storable (Storable(..))
130import Prelude hiding (length, null, last, map, (!!), read, concat)
132import Streamly.Internal.Data.Array.Foreign.Mut.Type (ReadUState(..), touch)
133import Streamly.Internal.Data.Array.Foreign.Type
134 (Array(..), length, asPtrUnsafe)
135import Streamly.Internal.Data.Fold.Type (Fold(..))
136import Streamly.Internal.Data.Producer.Type (Producer(..))
137import Streamly.Internal.Data.Stream.Serial (SerialT(..))
138import Streamly.Internal.Data.Tuple.Strict (Tuple3Fused'(..))
139import Streamly.Internal.Data.Unfold.Type (Unfold(..))
140import Streamly.Internal.System.IO (unsafeInlineIO)
142import qualified Streamly.Internal.Data.Array.Foreign.Mut.Type as MA
143import qualified Streamly.Internal.Data.Array.Foreign.Mut as MA
144import qualified Streamly.Internal.Data.Array.Foreign.Type as A
145import qualified Streamly.Internal.Data.Fold as FL
146import qualified Streamly.Internal.Data.Producer as Producer
147import qualified Streamly.Internal.Data.Stream.IsStream.Type as IsStream
148import qualified Streamly.Internal.Data.Stream.Prelude as P
149import qualified Streamly.Internal.Data.Stream.StreamD as D
150import qualified Streamly.Internal.Data.Unfold as Unfold
151import qualified Streamly.Internal.Data.Ring.Foreign as RB
153-------------------------------------------------------------------------------
155-------------------------------------------------------------------------------
157-- | Create an 'Array' from the first N elements of a stream. The array is
158-- allocated to size N, if the stream terminates before N elements then the
159-- array may hold less than N elements.
162{-# INLINE fromStreamN #-}
163fromStreamN :: (MonadIO m, Storable a) => Int -> SerialT m a -> m (Array a)
164fromStreamN n (SerialT m) = do
165 when (n < 0) $ error "writeN: negative write count specified"
166 A.fromStreamDN n $ D.fromStreamK m
168-- | Create an 'Array' from a stream. This is useful when we want to create a
169-- single array from a stream of unknown size. 'writeN' is at least twice
170-- as efficient when the size is already known.
172-- Note that if the input stream is too large memory allocation for the array
173-- may fail. When the stream size is not known, `arraysOf` followed by
174-- processing of indvidual arrays in the resulting stream should be preferred.
177{-# INLINE fromStream #-}
178fromStream :: (MonadIO m, Storable a) => SerialT m a -> m (Array a)
179fromStream (SerialT m) = P.fold A.write m
180-- write m = A.fromStreamD $ D.fromStreamK m
182-------------------------------------------------------------------------------
184-------------------------------------------------------------------------------
186{-# INLINE_NORMAL producer #-}
187producer :: forall m a. (Monad m, Storable a) => Producer m (Array a) a
188producer = Producer step inject extract
191 {-# INLINE inject #-}
192 inject (Array contents start end) = return $ ReadUState contents end start
194 {-# INLINE_LATE step #-}
195 step (ReadUState contents end cur)
196 | assert (cur <= end) (cur == end) =
197 let x = unsafeInlineIO $ touch contents
198 in x `seq` return D.Stop
199 step (ReadUState contents end cur) = do
200 -- unsafeInlineIO allows us to run this in Identity monad for pure
201 -- toList/foldr case which makes them much faster due to not
202 -- accumulating the list and fusing better with the pure consumers.
204 -- This should be safe as the array contents are guaranteed to be
205 -- evaluated/written to before we peek at them.
206 let !x = unsafeInlineIO $ peek cur
207 return $ D.Yield x (ReadUState contents end (PTR_NEXT(cur,a)))
209 extract (ReadUState contents end cur) = return $ Array contents cur end
211-- | Unfold an array into a stream.
213-- /Since 0.7.0 (Streamly.Memory.Array)/
216{-# INLINE_NORMAL read #-}
217read :: forall m a. (Monad m, Storable a) => Unfold m (Array a) a
218read = Producer.simplify producer
220-- | Unfold an array into a stream, does not check the end of the array, the
221-- user is responsible for terminating the stream within the array bounds. For
222-- high performance application where the end condition can be determined by
223-- a terminating fold.
225-- Written in the hope that it may be faster than "read", however, in the case
226-- for which this was written, "read" proves to be faster even though the core
227-- generated with unsafeRead looks simpler.
231{-# INLINE_NORMAL unsafeRead #-}
232unsafeRead :: forall m a. (Monad m, Storable a) => Unfold m (Array a) a
233unsafeRead = Unfold step inject
236 inject (Array contents start end) =
237 return (ReadUState contents end start)
239 {-# INLINE_LATE step #-}
240 step (ReadUState contents end p) = do
241 -- unsafeInlineIO allows us to run this in Identity monad for pure
242 -- toList/foldr case which makes them much faster due to not
243 -- accumulating the list and fusing better with the pure consumers.
245 -- This should be safe as the array contents are guaranteed to be
246 -- evaluated/written to before we peek at them.
247 let !x = unsafeInlineIO $ do
251 let !p1 = PTR_NEXT(p,a)
252 return $ D.Yield x (ReadUState contents end p1)
256-- >>> import qualified Streamly.Internal.Data.Array.Foreign.Type as Array
257-- >>> null arr = Array.byteLength arr == 0
261null :: Array a -> Bool
262null arr = A.byteLength arr == 0
264-- | Like 'getIndex' but indexes the array in reverse from the end.
267{-# INLINE getIndexRev #-}
268getIndexRev :: forall a. Storable a => Int -> Array a -> Maybe a
273 let elemPtr = PTR_RINDEX(aEnd arr,i,a)
274 if i >= 0 && elemPtr >= ptr
275 then Just <$> peek elemPtr
280-- >>> import qualified Streamly.Internal.Data.Array.Foreign as Array
281-- >>> last arr = Array.getIndexRev arr 0
285last :: Storable a => Array a -> Maybe a
288-------------------------------------------------------------------------------
289-- Folds with Array as the container
290-------------------------------------------------------------------------------
292-- | @writeLastN n@ folds a maximum of @n@ elements from the end of the input
293-- stream to an 'Array'.
296{-# INLINE writeLastN #-}
297writeLastN :: (Storable a, MonadIO m) => Int -> Fold m a (Array a)
299 | n <= 0 = fmap (const mempty) FL.drain
300 | otherwise = A.unsafeFreeze <$> Fold step initial done
304 step (Tuple3Fused' rb rh i) a = do
305 rh1 <- liftIO $ RB.unsafeInsert rb rh a
306 return $ FL.Partial $ Tuple3Fused' rb rh1 (i + 1)
309 let f (a, b) = FL.Partial $ Tuple3Fused' a b (0 :: Int)
310 in fmap f $ liftIO $ RB.new n
312 done (Tuple3Fused' rb rh i) = do
313 arr <- liftIO $ MA.newArray n
314 foldFunc i rh snoc' arr rb
316 -- XXX We should write a read unfold for ring.
317 snoc' b a = liftIO $ MA.snocUnsafe b a
320 | i < n = RB.unsafeFoldRingM
321 | otherwise = RB.unsafeFoldRingFullM
323-------------------------------------------------------------------------------
325-------------------------------------------------------------------------------
327-------------------------------------------------------------------------------
329-------------------------------------------------------------------------------
331-- | Given a sorted array, perform a binary search to find the given element.
332-- Returns the index of the element if found.
335{-# INLINE binarySearch #-}
336binarySearch :: a -> Array a -> Maybe Int
337binarySearch = undefined
339-- find/findIndex etc can potentially be implemented more efficiently on arrays
340-- compared to streams by using SIMD instructions.
341-- We can also return a bit array instead.
343-- | Perform a linear search to find all the indices where a given element is
344-- present in an array.
347findIndicesOf :: (a -> Bool) -> Unfold Identity (Array a) Int
348findIndicesOf = undefined
351findIndexOf :: (a -> Bool) -> Array a -> Maybe Int
352findIndexOf p = Unfold.fold Fold.head . Stream.unfold (findIndicesOf p)
354find :: (a -> Bool) -> Array a -> Bool
355find = Unfold.fold Fold.null . Stream.unfold (findIndicesOf p)
358-------------------------------------------------------------------------------
360-------------------------------------------------------------------------------
362-- XXX We can potentially use SIMD instructions on arrays to fold faster.
364-------------------------------------------------------------------------------
366-------------------------------------------------------------------------------
368-- | /O(1)/ Slice an array in constant time.
370-- Caution: The bounds of the slice are not checked.
375{-# INLINE getSliceUnsafe #-}
378 => Int -- ^ starting index
379 -> Int -- ^ length of the slice
382getSliceUnsafe index len (Array contents start e) =
383 let size = SIZE_OF(a)
384 fp1 = start `plusPtr` (index * size)
385 end = fp1 `plusPtr` (len * size)
386 in assert (end <= e) (Array contents fp1 end)
388-- | Split the array into a stream of slices using a predicate. The element
389-- matching the predicate is dropped.
392{-# INLINE splitOn #-}
393splitOn :: (Monad m, Storable a) =>
394 (a -> Bool) -> Array a -> SerialT m (Array a)
395splitOn predicate arr =
397 $ fmap (\(i, len) -> getSliceUnsafe i len arr)
398 $ D.sliceOnSuffix predicate (A.toStreamD arr)
400{-# INLINE genSlicesFromLen #-}
401genSlicesFromLen :: forall m a. (Monad m, Storable a)
402 => Int -- ^ from index
403 -> Int -- ^ length of the slice
404 -> Unfold m (Array a) (Int, Int)
405genSlicesFromLen from len =
406 Unfold.lmap A.unsafeThaw (MA.genSlicesFromLen from len)
408-- | Generate a stream of slices of specified length from an array, starting
409-- from the supplied array index. The last slice may be shorter than the
413{-# INLINE getSlicesFromLen #-}
414getSlicesFromLen :: forall m a. (Monad m, Storable a)
415 => Int -- ^ from index
416 -> Int -- ^ length of the slice
417 -> Unfold m (Array a) (Array a)
418getSlicesFromLen from len =
420 $ Unfold.lmap A.unsafeThaw (MA.getSlicesFromLen from len)
422-------------------------------------------------------------------------------
423-- Random reads and writes
424-------------------------------------------------------------------------------
426-- XXX Change this to a partial function instead of a Maybe type? And use
427-- MA.getIndex instead.
429-- XXX The signature should be "Int -> Array a -> Maybe a"
430-- XXX This is a released API so make this change in the next major release.
431-- | /O(1)/ Lookup the element at the given index. Index starts from 0.
434{-# INLINE getIndex #-}
435getIndex :: forall a. Storable a => Array a -> Int -> Maybe a
440 let elemPtr = PTR_INDEX(ptr,i,a)
441 if i >= 0 && PTR_VALID(elemPtr,aEnd arr,a)
442 then Just <$> peek elemPtr
445-- | Given a stream of array indices, read the elements on those indices from
446-- the supplied Array. An exception is thrown if an index is out of bounds.
448-- This is the most general operation. We can implement other operations in
453-- let u = lmap (\arr -> (0, length arr - 1)) Unfold.enumerateFromTo
454-- in Unfold.lmap f (getIndices arr)
457-- let i = length arr - 1
458-- in Unfold.lmap f (getIndicesFromThenTo i (i - 1) 0)
462{-# INLINE getIndices #-}
463getIndices :: (Monad m, Storable a) => SerialT m Int -> Unfold m (Array a) a
464getIndices (SerialT stream) =
465 let unf = MA.getIndicesD (return . unsafeInlineIO) $ D.fromStreamK stream
466 in Unfold.lmap A.unsafeThaw unf
468-- | Unfolds @(from, then, to, array)@ generating a finite stream whose first
469-- element is the array value from the index @from@ and the successive elements
470-- are from the indices in increments of @then@ up to @to@. Index enumeration
471-- can occur downwards or upwards depending on whether @then@ comes before or
475-- getIndicesFromThenTo =
476-- let f (from, next, to, arr) =
477-- (Stream.enumerateFromThenTo from next to, arr)
478-- in Unfold.lmap f getIndices
482{-# INLINE getIndicesFromThenTo #-}
483getIndicesFromThenTo :: Unfold m (Int, Int, Int, Array a) a
484getIndicesFromThenTo = undefined
486-------------------------------------------------------------------------------
487-- Transform via stream operations
488-------------------------------------------------------------------------------
490-- for non-length changing operations we can use the original length for
491-- allocation. If we can predict the length then we can use the prediction for
492-- new allocation. Otherwise we can use a hint and adjust dynamically.
495-- | Transform an array into another array using a pipe transformation
499{-# INLINE runPipe #-}
500runPipe :: (MonadIO m, Storable a, Storable b)
501 => Pipe m a b -> Array a -> m (Array b)
502runPipe f arr = P.runPipe (toArrayMinChunk (length arr)) $ f (A.read arr)
505-- XXX For transformations that cannot change the number of elements e.g. "map"
506-- we can use a predetermined array length.
508-- | Transform an array into another array using a stream transformation
512{-# INLINE streamTransform #-}
513streamTransform :: forall m a b. (MonadIO m, Storable a, Storable b)
514 => (SerialT m a -> SerialT m b) -> Array a -> m (Array b)
515streamTransform f arr =
516 P.fold (A.writeWith (length arr)) $ getSerialT $ f (A.toStream arr)
518-------------------------------------------------------------------------------
520-------------------------------------------------------------------------------
522-- | Cast an array having elements of type @a@ into an array having elements of
523-- type @b@. The array size must be a multiple of the size of type @b@
524-- otherwise accessing the last element of the array may result into a crash or
534unsafeCast (Array contents start end) =
535 Array contents (castPtr start) (castPtr end)
537-- | Cast an @Array a@ into an @Array Word8@.
541asBytes :: Array a -> Array Word8
544-- | Cast an array having elements of type @a@ into an array having elements of
545-- type @b@. The length of the array should be a multiple of the size of the
546-- target element otherwise 'Nothing' is returned.
550cast :: forall a b. (Storable b) => Array a -> Maybe (Array b)
552 let len = A.byteLength arr
553 r = len `mod` SIZE_OF(b)
556 else Just $ unsafeCast arr
558-- | Convert an array of any type into a null terminated CString Ptr.
562-- /O(n) Time: (creates a copy of the array)/
566unsafeAsCString :: Array a -> (CString -> IO b) -> IO b
567unsafeAsCString arr act = do
568 let arr1 = asBytes arr <> A.fromList [0]
569 asPtrUnsafe arr1 $ \ptr -> act (castPtr ptr)
571-------------------------------------------------------------------------------
573-------------------------------------------------------------------------------
575-- | Fold an array using a 'Fold'.
579fold :: forall m a b. (MonadIO m, Storable a) => Fold m a b -> Array a -> m b
580fold f arr = P.fold f (getSerialT (A.toStream arr))
582-- | Fold an array using a stream fold operation.
585{-# INLINE streamFold #-}
586streamFold :: (MonadIO m, Storable a) => (SerialT m a -> m b) -> Array a -> m b
587streamFold 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