2-- Module : Streamly.Internal.Data.Array
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.Stream.Chunked.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
46 , A.writeN -- drop new
48 , A.write -- full buffer
63 , producer -- experimental
68 , A.unsafeIndex -- XXX Rename to getIndexUnsafe??
70 , last -- XXX getIndexLast?
72 , getIndicesFromThenTo
73 -- , getIndicesFrom -- read from a given position to the end of file
74 -- , getIndicesUpto -- read from beginning up to the given position
76 -- , getIndicesFromRev -- read from a given position to the beginning of file
77 -- , getIndicesUptoRev -- read from end to the given position in file
95 , A.unsafeFreeze -- asImmutableUnsafe?
96 , A.unsafeThaw -- asMutableUnsafe?
105 -- * Streaming Operations
119#include "ArrayMacros.h"
121import Control.Exception (assert)
122import Control.Monad (when)
123import Control.Monad.IO.Class (MonadIO(..))
124import Data.Functor.Identity (Identity)
125import Data.Proxy (Proxy(..))
126import Data.Word (Word8)
127import Foreign.C.String (CString)
128import Foreign.Ptr (castPtr)
129import Foreign.Storable (Storable)
130import Streamly.Internal.Data.Unboxed
135import Prelude hiding (length, null, last, map, (!!), read, concat)
137import Streamly.Internal.Data.Array.Mut.Type (ArrayUnsafe(..))
138import Streamly.Internal.Data.Array.Type
139 (Array(..), length, asPtrUnsafe)
140import Streamly.Internal.Data.Fold.Type (Fold(..))
141import Streamly.Internal.Data.Producer.Type (Producer(..))
142import Streamly.Internal.Data.Stream (Stream)
143import Streamly.Internal.Data.Tuple.Strict (Tuple3Fused'(..))
144import Streamly.Internal.Data.Unfold.Type (Unfold(..))
145import Streamly.Internal.System.IO (unsafeInlineIO)
147import qualified Streamly.Internal.Data.Array.Mut.Type as MA
148import qualified Streamly.Internal.Data.Array.Mut as MA
149import qualified Streamly.Internal.Data.Array.Type as A
150import qualified Streamly.Internal.Data.Fold as FL
151import qualified Streamly.Internal.Data.Producer.Type as Producer
152import qualified Streamly.Internal.Data.Producer as Producer
153import qualified Streamly.Internal.Data.Ring.Unboxed as RB
154import qualified Streamly.Internal.Data.Stream.Common as P
155import qualified Streamly.Internal.Data.Stream.StreamD as D
156import qualified Streamly.Internal.Data.Stream.Type as Stream
157import qualified Streamly.Internal.Data.Unfold as Unfold
159-------------------------------------------------------------------------------
161-------------------------------------------------------------------------------
163-- | Create an 'Array' from the first N elements of a stream. The array is
164-- allocated to size N, if the stream terminates before N elements then the
165-- array may hold less than N elements.
168{-# INLINE fromStreamN #-}
169fromStreamN :: (MonadIO m, Unbox a) => Int -> Stream m a -> m (Array a)
171 when (n < 0) $ error "writeN: negative write count specified"
172 A.fromStreamDN n $ Stream.toStreamD m
174-- | Create an 'Array' from a stream. This is useful when we want to create a
175-- single array from a stream of unknown size. 'writeN' is at least twice
176-- as efficient when the size is already known.
178-- Note that if the input stream is too large memory allocation for the array
179-- may fail. When the stream size is not known, `arraysOf` followed by
180-- processing of indvidual arrays in the resulting stream should be preferred.
183{-# INLINE fromStream #-}
184fromStream :: (MonadIO m, Unbox a) => Stream m a -> m (Array a)
185fromStream m = P.fold A.write $ Stream.toStreamK m
186-- write m = A.fromStreamD $ D.fromStreamK m
188-------------------------------------------------------------------------------
190-------------------------------------------------------------------------------
192{-# INLINE_NORMAL producer #-}
193producer :: forall m a. (Monad m, Unbox a) => Producer m (Array a) a
195 Producer.translate A.unsafeThaw A.unsafeFreeze
196 $ MA.producerWith (return . unsafeInlineIO)
198-- | Unfold an array into a stream.
200{-# INLINE_NORMAL reader #-}
201reader :: forall m a. (Monad m, Unbox a) => Unfold m (Array a) a
202reader = Producer.simplify producer
204-- | Unfold an array into a stream, does not check the end of the array, the
205-- user is responsible for terminating the stream within the array bounds. For
206-- high performance application where the end condition can be determined by
207-- a terminating fold.
209-- Written in the hope that it may be faster than "read", however, in the case
210-- for which this was written, "read" proves to be faster even though the core
211-- generated with unsafeRead looks simpler.
215{-# INLINE_NORMAL readerUnsafe #-}
216readerUnsafe :: forall m a. (Monad m, Unbox a) => Unfold m (Array a) a
217readerUnsafe = Unfold step inject
220 inject (Array contents start end) =
221 return (ArrayUnsafe contents end start)
223 {-# INLINE_LATE step #-}
224 step (ArrayUnsafe contents end p) = do
225 -- unsafeInlineIO allows us to run this in Identity monad for pure
226 -- toList/foldr case which makes them much faster due to not
227 -- accumulating the list and fusing better with the pure consumers.
229 -- This should be safe as the array contents are guaranteed to be
230 -- evaluated/written to before we peek at them.
231 let !x = unsafeInlineIO $ peekWith contents p
232 let !p1 = INDEX_NEXT(p,a)
233 return $ D.Yield x (ArrayUnsafe contents end p1)
237-- >>> import qualified Streamly.Internal.Data.Array.Type as Array
238-- >>> null arr = Array.byteLength arr == 0
242null :: Array a -> Bool
243null arr = A.byteLength arr == 0
245-- | Like 'getIndex' but indexes the array in reverse from the end.
248{-# INLINE getIndexRev #-}
249getIndexRev :: forall a. Unbox a => Int -> Array a -> Maybe a
253 let elemPtr = RINDEX_OF(arrEnd arr, i, a)
254 if i >= 0 && elemPtr >= arrStart arr
255 then Just <$> peekWith (arrContents arr) elemPtr
260-- >>> import qualified Streamly.Internal.Data.Array as Array
261-- >>> last arr = Array.getIndexRev arr 0
265last :: Unbox a => Array a -> Maybe a
268-------------------------------------------------------------------------------
269-- Folds with Array as the container
270-------------------------------------------------------------------------------
272-- | @writeLastN n@ folds a maximum of @n@ elements from the end of the input
273-- stream to an 'Array'.
275{-# INLINE writeLastN #-}
277 (Storable a, Unbox a, MonadIO m) => Int -> Fold m a (Array a)
279 | n <= 0 = fmap (const mempty) FL.drain
280 | otherwise = A.unsafeFreeze <$> Fold step initial done
284 step (Tuple3Fused' rb rh i) a = do
285 rh1 <- liftIO $ RB.unsafeInsert rb rh a
286 return $ FL.Partial $ Tuple3Fused' rb rh1 (i + 1)
289 let f (a, b) = FL.Partial $ Tuple3Fused' a b (0 :: Int)
290 in fmap f $ liftIO $ RB.new n
292 done (Tuple3Fused' rb rh i) = do
293 arr <- liftIO $ MA.newPinned n
294 foldFunc i rh snoc' arr rb
296 -- XXX We should write a read unfold for ring.
297 snoc' b a = liftIO $ MA.snocUnsafe b a
300 | i < n = RB.unsafeFoldRingM
301 | otherwise = RB.unsafeFoldRingFullM
303-------------------------------------------------------------------------------
305-------------------------------------------------------------------------------
307-------------------------------------------------------------------------------
309-------------------------------------------------------------------------------
311-- | Given a sorted array, perform a binary search to find the given element.
312-- Returns the index of the element if found.
315{-# INLINE binarySearch #-}
316binarySearch :: a -> Array a -> Maybe Int
317binarySearch = undefined
319-- find/findIndex etc can potentially be implemented more efficiently on arrays
320-- compared to streams by using SIMD instructions.
321-- We can also return a bit array instead.
323-- | Perform a linear search to find all the indices where a given element is
324-- present in an array.
327findIndicesOf :: (a -> Bool) -> Unfold Identity (Array a) Int
328findIndicesOf = undefined
331findIndexOf :: (a -> Bool) -> Array a -> Maybe Int
332findIndexOf p = Unfold.fold Fold.one . Stream.unfold (findIndicesOf p)
334find :: (a -> Bool) -> Array a -> Bool
335find = Unfold.fold Fold.null . Stream.unfold (findIndicesOf p)
338-------------------------------------------------------------------------------
340-------------------------------------------------------------------------------
342-- XXX We can potentially use SIMD instructions on arrays to fold faster.
344-------------------------------------------------------------------------------
346-------------------------------------------------------------------------------
348-- | /O(1)/ Slice an array in constant time.
350-- Caution: The bounds of the slice are not checked.
355{-# INLINE getSliceUnsafe #-}
358 => Int -- ^ starting index
359 -> Int -- ^ length of the slice
362getSliceUnsafe index len (Array contents start e) =
363 let size = SIZE_OF(a)
364 start1 = start + (index * size)
365 end1 = start1 + (len * size)
366 in assert (end1 <= e) (Array contents start1 end1)
368-- | Split the array into a stream of slices using a predicate. The element
369-- matching the predicate is dropped.
372{-# INLINE splitOn #-}
373splitOn :: (Monad m, Unbox a) =>
374 (a -> Bool) -> Array a -> Stream m (Array a)
375splitOn predicate arr =
377 $ fmap (\(i, len) -> getSliceUnsafe i len arr)
378 $ D.sliceOnSuffix predicate (A.toStreamD arr)
380{-# INLINE genSlicesFromLen #-}
381genSlicesFromLen :: forall m a. (Monad m, Unbox a)
382 => Int -- ^ from index
383 -> Int -- ^ length of the slice
384 -> Unfold m (Array a) (Int, Int)
385genSlicesFromLen from len =
386 Unfold.lmap A.unsafeThaw (MA.genSlicesFromLen from len)
388-- | Generate a stream of slices of specified length from an array, starting
389-- from the supplied array index. The last slice may be shorter than the
393{-# INLINE getSlicesFromLen #-}
394getSlicesFromLen :: forall m a. (Monad m, Unbox a)
395 => Int -- ^ from index
396 -> Int -- ^ length of the slice
397 -> Unfold m (Array a) (Array a)
398getSlicesFromLen from len =
400 $ Unfold.lmap A.unsafeThaw (MA.getSlicesFromLen from len)
402-------------------------------------------------------------------------------
403-- Random reads and writes
404-------------------------------------------------------------------------------
406-- XXX Change this to a partial function instead of a Maybe type? And use
407-- MA.getIndex instead.
409-- | /O(1)/ Lookup the element at the given index. Index starts from 0.
411{-# INLINE getIndex #-}
412getIndex :: forall a. Unbox a => Int -> Array a -> Maybe a
416 let elemPtr = INDEX_OF(arrStart arr, i, a)
417 if i >= 0 && INDEX_VALID(elemPtr, arrEnd arr, a)
418 then Just <$> peekWith (arrContents arr) 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, Unbox a) => Stream m Int -> Unfold m (Array a) a
441 let unf = MA.getIndicesD (return . unsafeInlineIO) $ D.fromStreamK $ Stream.toStreamK m
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
474{-# INLINE runPipe #-}
475runPipe :: (MonadIO m, Unbox a, Unbox b)
476 => Pipe m a b -> Array a -> m (Array b)
477runPipe f arr = P.runPipe (toArrayMinChunk (length arr)) $ f (A.read arr)
480-- XXX For transformations that cannot change the number of elements e.g. "map"
481-- we can use a predetermined array length.
483-- | Transform an array into another array using a stream transformation
487{-# INLINE streamTransform #-}
488streamTransform :: forall m a b. (MonadIO m, Unbox a, Unbox b)
489 => (Stream m a -> Stream m b) -> Array a -> m (Array b)
490streamTransform f arr =
491 P.fold (A.writeWith (length arr)) $ Stream.toStreamK $ f (A.read arr)
493-------------------------------------------------------------------------------
495-------------------------------------------------------------------------------
497-- | Cast an array having elements of type @a@ into an array having elements of
498-- type @b@. The array size must be a multiple of the size of type @b@
499-- otherwise accessing the last element of the array may result into a crash or
509castUnsafe (Array contents start end) =
510 Array contents start end
512-- | Cast an @Array a@ into an @Array Word8@.
515asBytes :: Array a -> Array Word8
518-- | Cast an array having elements of type @a@ into an array having elements of
519-- type @b@. The length of the array should be a multiple of the size of the
520-- target element otherwise 'Nothing' is returned.
523cast :: forall a b. (Unbox b) => Array a -> Maybe (Array b)
525 let len = A.byteLength arr
526 r = len `mod` SIZE_OF(b)
529 else Just $ castUnsafe arr
531-- | Convert an array of any type into a null terminated CString Ptr.
535-- /O(n) Time: (creates a copy of the array)/
539asCStringUnsafe :: Array a -> (CString -> IO b) -> IO b
540asCStringUnsafe arr act = do
541 -- XXX Ensure a pinned allocation here.
542 let arr1 = asBytes arr <> A.fromList [0]
543 asPtrUnsafe arr1 $ \ptr -> act (castPtr ptr)
545-------------------------------------------------------------------------------
547-------------------------------------------------------------------------------
549-- XXX We can directly use toStreamD and D.fold here.
551-- | Fold an array using a 'Fold'.
555fold :: forall m a b. (Monad m, Unbox a) => Fold m a b -> Array a -> m b
556fold f arr = P.fold f (Stream.toStreamK (A.read arr))
558-- | Fold an array using a stream fold operation.
561{-# INLINE streamFold #-}
562streamFold :: (Monad m, Unbox a) => (Stream m a -> m b) -> Array a -> m b
563streamFold f arr = f (A.read 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