Merge pull request #904 from haskell/internal-modules

Add internal modules for bytestring and text compat

Author: phadej

aeson.cabal

@@ -84,7 +84,9 @@ library
 
   other-modules:
     Data.Aeson.Encoding.Builder
+    Data.Aeson.Internal.ByteString
     Data.Aeson.Internal.Functions
+    Data.Aeson.Internal.Text
     Data.Aeson.Parser.Time
     Data.Aeson.Parser.Unescape
     Data.Aeson.Types.Class
@@ -146,7 +148,7 @@ library
     cpp-options:    -DCFFI
     hs-source-dirs: src-ffi
     other-modules:  Data.Aeson.Parser.UnescapeFFI
-    build-depends:  text < 2.0
+    build-depends:  text <2.0
 
   if flag(ordered-keymap)
     cpp-options: -DUSE_ORDEREDMAP=1
@@ -157,7 +159,6 @@ test-suite aeson-tests
   hs-source-dirs:   tests
   main-is:          Tests.hs
   ghc-options:      -Wall -threaded -rtsopts
-
   other-modules:
     DataFamilies.Encoders
     DataFamilies.Instances

benchmarks/aeson-benchmarks.cabal

@@ -72,7 +72,9 @@ library
     Data.Aeson.Encoding.Builder
     Data.Aeson.Encoding.Internal
     Data.Aeson.Internal
+    Data.Aeson.Internal.ByteString
     Data.Aeson.Internal.Functions
+    Data.Aeson.Internal.Text
     Data.Aeson.Internal.Time
     Data.Aeson.Key
     Data.Aeson.KeyMap
@@ -122,6 +124,7 @@ executable aeson-benchmark-suite
     , time
     , unordered-containers
     , vector
+
   other-modules:
     AesonFoldable
     AesonMap
@@ -167,5 +170,4 @@ executable aeson-benchmark-suite
     other-modules: Compare.BufferBuilder
 
     -- this module won't
-    other-modules:
-      UnescapePureText1
+    other-modules: UnescapePureText1

src-pure/Data/Aeson/Parser/UnescapePure.hs

@@ -17,15 +17,17 @@ import           Foreign.ForeignPtr       (withForeignPtr)
 import           Foreign.Ptr              (Ptr, plusPtr)
 import           Foreign.Storable         (peek)
 
-import qualified Data.ByteString.Internal as BS
 import qualified Data.Primitive           as P
 import qualified Data.Text.Array          as T
 import qualified Data.Text.Internal       as T
 
+import Data.Aeson.Internal.ByteString
+
 #if !MIN_VERSION_text(2,0,0)
 import           Data.Word                (Word16)
 #endif
 
+
 unescapeText :: ByteString -> Either UnicodeException Text
 unescapeText = unsafeDupablePerformIO . try . unescapeTextIO
 
@@ -48,11 +50,10 @@ unescapeTextIO :: ByteString -> IO Text
 
 #if MIN_VERSION_text(2,0,0)
 
-unescapeTextIO bs = case bs of
-  BS.PS fptr off len -> withForeignPtr fptr $ \bsPtr -> do
-    let begin, end :: Ptr Word8
-        begin = plusPtr bsPtr off
-        end   = plusPtr begin len
+unescapeTextIO bs = withBS bs $ \fptr len ->
+  withForeignPtr fptr $ \begin -> do
+    let end :: Ptr Word8
+        end = plusPtr begin len
 
     arr <- P.newPrimArray len
 
@@ -397,11 +398,10 @@ unescapeTextIO bs = case bs of
     state_start (0 :: Int) begin
 #else
 
-unescapeTextIO bs = case bs of
-  BS.PS fptr off len -> withForeignPtr fptr $ \bsPtr -> do
-    let begin, end :: Ptr Word8
-        begin = plusPtr bsPtr off
-        end   = plusPtr begin len
+unescapeTextIO bs = withBS bs $ \fptr len ->
+  withForeignPtr fptr $ \begin -> do
+    let end :: Ptr Word8
+        end = plusPtr begin len
 
     arr <- P.newPrimArray len
 

src/Data/Aeson/Internal/ByteString.hs

@@ -0,0 +1,58 @@
+{-# LANGUAGE BangPatterns #-}
+{-# LANGUAGE CPP #-}
+{-# LANGUAGE MagicHash #-}
+module Data.Aeson.Internal.ByteString (
+    mkBS, 
+    withBS,
+) where
+
+import Data.ByteString.Internal (ByteString (..))
+import Data.Word (Word8)
+import Foreign.ForeignPtr (ForeignPtr)
+
+#if !MIN_VERSION_bytestring(0,11,0)
+#if MIN_VERSION_base(4,10,0)
+import GHC.ForeignPtr (plusForeignPtr)
+#else
+import GHC.ForeignPtr (ForeignPtr(ForeignPtr))
+import GHC.Types (Int (..))
+import GHC.Prim (plusAddr#)
+#endif
+#endif
+
+mkBS :: ForeignPtr Word8 -> Int -> ByteString
+#if MIN_VERSION_bytestring(0,11,0)
+mkBS dfp n = BS dfp n
+#else
+mkBS dfp n = PS dfp 0 n
+#endif
+{-# INLINE mkBS #-}
+
+withBS :: ByteString -> (ForeignPtr Word8 -> Int -> r) -> r
+#if MIN_VERSION_bytestring(0,11,0)
+withBS (BS !sfp !slen)       kont = kont sfp slen
+#else
+withBS (PS !sfp !soff !slen) kont = kont (plusForeignPtr sfp soff) slen
+#endif
+{-# INLINE withBS #-}
+
+#if !MIN_VERSION_bytestring(0,11,0)
+#if !MIN_VERSION_base(4,10,0)
+-- |Advances the given address by the given offset in bytes.
+--
+-- The new 'ForeignPtr' shares the finalizer of the original,
+-- equivalent from a finalization standpoint to just creating another
+-- reference to the original. That is, the finalizer will not be
+-- called before the new 'ForeignPtr' is unreachable, nor will it be
+-- called an additional time due to this call, and the finalizer will
+-- be called with the same address that it would have had this call
+-- not happened, *not* the new address.
+plusForeignPtr :: ForeignPtr a -> Int -> ForeignPtr b
+plusForeignPtr (ForeignPtr addr guts) (I# offset) = ForeignPtr (plusAddr# addr offset) guts
+{-# INLINE [0] plusForeignPtr #-}
+{-# RULES
+"ByteString plusForeignPtr/0" forall fp .
+   plusForeignPtr fp 0 = fp
+ #-}
+#endif
+#endif

src/Data/Aeson/Internal/Text.hs

@@ -0,0 +1,20 @@
+{-# LANGUAGE CPP #-}
+module Data.Aeson.Internal.Text (
+    unsafeDecodeASCII,
+) where
+
+import           Data.ByteString    (ByteString)
+import           Data.Text          (Text)
+
+import qualified Data.Text.Encoding as TE
+
+-- | The input is assumed to contain only 7bit ASCII characters (i.e. @< 0x80@).
+--   We use TE.decodeLatin1 here because TE.decodeASCII is currently (text-1.2.4.0)
+--   deprecated and equal to TE.decodeUtf8, which is slower than TE.decodeLatin1.
+unsafeDecodeASCII :: ByteString -> Text
+
+#if MIN_VERSION_text(2,0,0)
+unsafeDecodeASCII = TE.decodeASCII
+#else
+unsafeDecodeASCII = TE.decodeLatin1
+#endif

src/Data/Aeson/Parser/Internal.hs

@@ -59,7 +59,6 @@ import Data.Function (fix)
 import Data.Functor.Compat (($>))
 import Data.Scientific (Scientific)
 import Data.Text (Text)
-import qualified Data.Text.Encoding as TE
 import Data.Vector (Vector)
 import qualified Data.Vector as Vector (empty, fromList, fromListN, reverse)
 import qualified Data.Attoparsec.ByteString as A
@@ -72,6 +71,7 @@ import qualified Data.ByteString.Lazy.Char8 as C
 import qualified Data.ByteString.Builder as B
 import qualified Data.Scientific as Sci
 import Data.Aeson.Parser.Unescape (unescapeText)
+import Data.Aeson.Internal.Text
 
 -- $setup
 -- >>> :set -XOverloadedStrings
@@ -342,17 +342,6 @@ jstring_ = do
     Just w | w < 0x20 -> fail "unescaped control character"
     _                 -> jstringSlow s
 
-#if MIN_VERSION_text(2,0,0)
-unsafeDecodeASCII :: B.ByteString -> Text
-unsafeDecodeASCII = TE.decodeASCII
-#else
--- | The input is assumed to contain only 7bit ASCII characters (i.e. @< 0x80@).
---   We use TE.decodeLatin1 here because TE.decodeASCII is currently (text-1.2.4.0)
---   deprecated and equal to TE.decodeUtf8, which is slower than TE.decodeLatin1.
-unsafeDecodeASCII :: B.ByteString -> Text
-unsafeDecodeASCII = TE.decodeLatin1
-#endif
-
 jstringSlow :: B.ByteString -> Parser Text
 {-# INLINE jstringSlow #-}
 jstringSlow s' = {-# SCC "jstringSlow" #-} do