on master help sign in
Merge pull request #155 from RyanGlScott/master
6c4cc7fBryan O'Sullivan
Data/Text/Encoding.hs+7−18
@@ -55,13 +55,9 @@ module Data.Text.Encoding
55 , encodeUtf32LE
55 , encodeUtf32LE
56 , encodeUtf32BE
56 , encodeUtf32BE
57
57
58−#if MIN_VERSION_bytestring(0,10,4)
59 -- * Encoding Text using ByteString Builders
58 -- * Encoding Text using ByteString Builders
60− -- | /Note/ that these functions are only available if built against
61− -- @bytestring >= 0.10.4.0@.
62 , encodeUtf8Builder
59 , encodeUtf8Builder
63 , encodeUtf8BuilderEscaped
60 , encodeUtf8BuilderEscaped
64−#endif
65 ) where
61 ) where
66
62
67#if __GLASGOW_HASKELL__ >= 702
63#if __GLASGOW_HASKELL__ >= 702
@@ -70,24 +66,15 @@ import Control.Monad.ST.Unsafe (unsafeIOToST, unsafeSTToIO)
70import Control.Monad.ST (unsafeIOToST, unsafeSTToIO)
66import Control.Monad.ST (unsafeIOToST, unsafeSTToIO)
71#endif
67#endif
72
68
73−#if MIN_VERSION_bytestring(0,10,4)
74−import Data.Bits ((.&.))
75−import Data.Text.Internal.Unsafe.Char (ord)
76−import qualified Data.ByteString.Builder as B
77−import qualified Data.ByteString.Builder.Internal as B hiding (empty, append)
78−import qualified Data.ByteString.Builder.Prim as BP
79−import qualified Data.ByteString.Builder.Prim.Internal as BP
80−import qualified Data.Text.Internal.Encoding.Utf16 as U16
81−#endif
82−
83import Control.Exception (evaluate, try)
69import Control.Exception (evaluate, try)
84import Control.Monad.ST (runST)
70import Control.Monad.ST (runST)
71+import Data.Bits ((.&.))
85import Data.ByteString as B
72import Data.ByteString as B
86import Data.ByteString.Internal as B hiding (c2w)
73import Data.ByteString.Internal as B hiding (c2w)
87import Data.Text.Encoding.Error (OnDecodeError, UnicodeException, strictDecode)
74import Data.Text.Encoding.Error (OnDecodeError, UnicodeException, strictDecode)
88import Data.Text.Internal (Text(..), safe, text)
75import Data.Text.Internal (Text(..), safe, text)
89import Data.Text.Internal.Private (runText)
76import Data.Text.Internal.Private (runText)
90−import Data.Text.Internal.Unsafe.Char (unsafeWrite)
77+import Data.Text.Internal.Unsafe.Char (ord, unsafeWrite)
91import Data.Text.Internal.Unsafe.Shift (shiftR)
78import Data.Text.Internal.Unsafe.Shift (shiftR)
92import Data.Text.Show ()
79import Data.Text.Show ()
93import Data.Text.Unsafe (unsafeDupablePerformIO)
80import Data.Text.Unsafe (unsafeDupablePerformIO)
@@ -98,8 +85,13 @@ import Foreign.Marshal.Utils (with)
98import Foreign.Ptr (Ptr, minusPtr, nullPtr, plusPtr)
85import Foreign.Ptr (Ptr, minusPtr, nullPtr, plusPtr)
99import Foreign.Storable (Storable, peek, poke)
86import Foreign.Storable (Storable, peek, poke)
100import GHC.Base (ByteArray#, MutableByteArray#)
87import GHC.Base (ByteArray#, MutableByteArray#)
88+import qualified Data.ByteString.Builder as B
89+import qualified Data.ByteString.Builder.Internal as B hiding (empty, append)
90+import qualified Data.ByteString.Builder.Prim as BP
91+import qualified Data.ByteString.Builder.Prim.Internal as BP
101import qualified Data.Text.Array as A
92import qualified Data.Text.Array as A
102import qualified Data.Text.Internal.Encoding.Fusion as E
93import qualified Data.Text.Internal.Encoding.Fusion as E
94+import qualified Data.Text.Internal.Encoding.Utf16 as U16
103import qualified Data.Text.Internal.Fusion as F
95import qualified Data.Text.Internal.Fusion as F
104
96
105#include "text_cbits.h"
97#include "text_cbits.h"
@@ -315,8 +307,6 @@ decodeUtf8' :: ByteString -> Either UnicodeException Text
315decodeUtf8' = unsafeDupablePerformIO . try . evaluate . decodeUtf8With strictDecode
307decodeUtf8' = unsafeDupablePerformIO . try . evaluate . decodeUtf8With strictDecode
316{-# INLINE decodeUtf8' #-}
308{-# INLINE decodeUtf8' #-}
317
309
318−#if MIN_VERSION_bytestring(0,10,4)
319−
320-- | Encode text to a ByteString 'B.Builder' using UTF-8 encoding.
310-- | Encode text to a ByteString 'B.Builder' using UTF-8 encoding.
321encodeUtf8Builder :: Text -> B.Builder
311encodeUtf8Builder :: Text -> B.Builder
322encodeUtf8Builder = encodeUtf8BuilderEscaped (BP.liftFixedToBounded BP.word8)
312encodeUtf8Builder = encodeUtf8BuilderEscaped (BP.liftFixedToBounded BP.word8)
@@ -377,7 +367,6 @@ encodeUtf8BuilderEscaped be =
377 outerLoop i (B.BufferRange op ope)
367 outerLoop i (B.BufferRange op ope)
378 where
368 where
379 poke8 j v = poke (op `plusPtr` j) (fromIntegral v :: Word8)
369 poke8 j v = poke (op `plusPtr` j) (fromIntegral v :: Word8)
380−#endif
381
370
382-- | Encode text using UTF-8 encoding.
371-- | Encode text using UTF-8 encoding.
383encodeUtf8 :: Text -> ByteString
372encodeUtf8 :: Text -> ByteString
Data/Text/Lazy/Encoding.hs+5−15
@@ -46,27 +46,23 @@ module Data.Text.Lazy.Encoding
46 , encodeUtf32LE
46 , encodeUtf32LE
47 , encodeUtf32BE
47 , encodeUtf32BE
48
48
49−#if MIN_VERSION_bytestring(0,10,4)
50 -- * Encoding Text using ByteString Builders
49 -- * Encoding Text using ByteString Builders
51 , encodeUtf8Builder
50 , encodeUtf8Builder
52 , encodeUtf8BuilderEscaped
51 , encodeUtf8BuilderEscaped
53−#endif
54 ) where
52 ) where
55
53
56import Control.Exception (evaluate, try)
54import Control.Exception (evaluate, try)
55+import Data.Monoid (Monoid(..))
57import Data.Text.Encoding.Error (OnDecodeError, UnicodeException, strictDecode)
56import Data.Text.Encoding.Error (OnDecodeError, UnicodeException, strictDecode)
58import Data.Text.Internal.Lazy (Text(..), chunk, empty, foldrChunks)
57import Data.Text.Internal.Lazy (Text(..), chunk, empty, foldrChunks)
59−import qualified Data.ByteString as S
60−import qualified Data.ByteString.Lazy as B
61−import qualified Data.ByteString.Lazy.Internal as B
62−import qualified Data.ByteString.Unsafe as B
63−#if MIN_VERSION_bytestring(0,10,4)
64import Data.Word (Word8)
58import Data.Word (Word8)
65−import Data.Monoid (Monoid(..))
59+import qualified Data.ByteString as S
66import qualified Data.ByteString.Builder as B
60import qualified Data.ByteString.Builder as B
67import qualified Data.ByteString.Builder.Extra as B (safeStrategy, toLazyByteStringWith)
61import qualified Data.ByteString.Builder.Extra as B (safeStrategy, toLazyByteStringWith)
68import qualified Data.ByteString.Builder.Prim as BP
62import qualified Data.ByteString.Builder.Prim as BP
69−#endif
63+import qualified Data.ByteString.Lazy as B
64+import qualified Data.ByteString.Lazy.Internal as B
65+import qualified Data.ByteString.Unsafe as B
70import qualified Data.Text as T
66import qualified Data.Text as T
71import qualified Data.Text.Encoding as TE
67import qualified Data.Text.Encoding as TE
72import qualified Data.Text.Internal.Lazy.Encoding.Fusion as E
68import qualified Data.Text.Internal.Lazy.Encoding.Fusion as E
@@ -146,7 +142,6 @@ decodeUtf8' bs = unsafeDupablePerformIO $ do
146{-# INLINE decodeUtf8' #-}
142{-# INLINE decodeUtf8' #-}
147
143
148encodeUtf8 :: Text -> B.ByteString
144encodeUtf8 :: Text -> B.ByteString
149−#if MIN_VERSION_bytestring(0,10,4)
150encodeUtf8 Empty = B.empty
145encodeUtf8 Empty = B.empty
151encodeUtf8 lt@(Chunk t _) =
146encodeUtf8 lt@(Chunk t _) =
152 B.toLazyByteStringWith strategy B.empty $ encodeUtf8Builder lt
147 B.toLazyByteStringWith strategy B.empty $ encodeUtf8Builder lt
@@ -168,11 +163,6 @@ encodeUtf8BuilderEscaped :: BP.BoundedPrim Word8 -> Text -> B.Builder
168encodeUtf8BuilderEscaped prim =
163encodeUtf8BuilderEscaped prim =
169 foldrChunks (\c b -> TE.encodeUtf8BuilderEscaped prim c `mappend` b) mempty
164 foldrChunks (\c b -> TE.encodeUtf8BuilderEscaped prim c `mappend` b) mempty
170
165
171−#else
172−encodeUtf8 (Chunk c cs) = B.Chunk (TE.encodeUtf8 c) (encodeUtf8 cs)
173−encodeUtf8 Empty = B.Empty
174−#endif
175−
176-- | Decode text from little endian UTF-16 encoding.
166-- | Decode text from little endian UTF-16 encoding.
177decodeUtf16LEWith :: OnDecodeError -> B.ByteString -> Text
167decodeUtf16LEWith :: OnDecodeError -> B.ByteString -> Text
178decodeUtf16LEWith onErr bs = F.unstream (E.streamUtf16LE onErr bs)
168decodeUtf16LEWith onErr bs = F.unstream (E.streamUtf16LE onErr bs)
benchmarks/text-benchmarks.cabal+11−1
@@ -15,6 +15,11 @@ build-type: Simple
15
15
16cabal-version: >=1.2
16cabal-version: >=1.2
17
17
18+flag bytestring-builder
19+ description: Depend on the bytestring-builder package for backwards compatibility.
20+ default: False
21+ manual: False
22+
18flag llvm
23flag llvm
19 description: use LLVM
24 description: use LLVM
20 default: False
25 default: False
@@ -33,7 +38,6 @@ executable text-benchmarks
33 build-depends: base == 4.*,
38 build-depends: base == 4.*,
34 binary,
39 binary,
35 blaze-builder,
40 blaze-builder,
36− bytestring,
37 bytestring-lexing >= 0.5.0,
41 bytestring-lexing >= 0.5.0,
38 containers,
42 containers,
39 criterion >= 0.10.0.0,
43 criterion >= 0.10.0.0,
@@ -46,6 +50,12 @@ executable text-benchmarks
46 utf8-string,
50 utf8-string,
47 vector
51 vector
48
52
53+ if flag(bytestring-builder)
54+ build-depends: bytestring >= 0.9 && < 0.10.4,
55+ bytestring-builder >= 0.10.4
56+ else
57+ build-depends: bytestring >= 0.10.4
58+
49executable text-multilang
59executable text-multilang
50 hs-source-dirs: haskell
60 hs-source-dirs: haskell
51 main-is: Multilang.hs
61 main-is: Multilang.hs
tests/literal-rule-test.sh+1−1
@@ -6,7 +6,7 @@ function check_firings() {
6 rule=$1
6 rule=$1
7 expected=$2
7 expected=$2
8 build="ghc -O -ddump-rule-firings LiteralRuleTest.hs"
8 build="ghc -O -ddump-rule-firings LiteralRuleTest.hs"
9− build="$build -i.. -I../include -DMIN_VERSION_bytestring(a,b,c)=1"
9+ build="$build -i.. -I../include"
10 touch LiteralRuleTest.hs
10 touch LiteralRuleTest.hs
11 echo -n "Want to see $expected firings of rule $rule... " >&2
11 echo -n "Want to see $expected firings of rule $rule... " >&2
12 firings=$($build 2>&1 | grep "Rule fired: $rule\$" | wc -l)
12 firings=$($build 2>&1 | grep "Rule fired: $rule\$" | wc -l)
text.cabal+15−4
@@ -64,6 +64,11 @@ extra-source-files:
64 tests/scripts/*.sh
64 tests/scripts/*.sh
65 tests/text-tests.cabal
65 tests/text-tests.cabal
66
66
67+flag bytestring-builder
68+ description: Depend on the bytestring-builder package for backwards compatibility.
69+ default: False
70+ manual: False
71+
67flag developer
72flag developer
68 description: operate in developer mode
73 description: operate in developer mode
69 default: False
74 default: False
@@ -133,10 +138,11 @@ library
133 deepseq >= 1.1.0.0,
138 deepseq >= 1.1.0.0,
134 ghc-prim >= 0.2
139 ghc-prim >= 0.2
135
140
136− if impl(ghc >= 7.7)
141+ if flag(bytestring-builder)
137− build-depends: bytestring >= 0.10.4.0
142+ build-depends: bytestring >= 0.9 && < 0.10.4,
143+ bytestring-builder >= 0.10.4
138 else
144 else
139− build-depends: bytestring >= 0.9
145+ build-depends: bytestring >= 0.10.4
140
146
141 cpp-options: -DHAVE_DEEPSEQ
147 cpp-options: -DHAVE_DEEPSEQ
142 ghc-options: -Wall -fwarn-tabs -funbox-strict-fields -O2
148 ghc-options: -Wall -fwarn-tabs -funbox-strict-fields -O2
@@ -171,7 +177,6 @@ test-suite tests
171 array,
177 array,
172 base,
178 base,
173 binary,
179 binary,
174− bytestring,
175 deepseq,
180 deepseq,
176 directory,
181 directory,
177 ghc-prim,
182 ghc-prim,
@@ -181,6 +186,12 @@ test-suite tests
181 test-framework-hunit >= 0.2,
186 test-framework-hunit >= 0.2,
182 test-framework-quickcheck2 >= 0.2
187 test-framework-quickcheck2 >= 0.2
183
188
189+ if flag(bytestring-builder)
190+ build-depends: bytestring >= 0.9 && < 0.10.4,
191+ bytestring-builder >= 0.10.4
192+ else
193+ build-depends: bytestring >= 0.10.4
194+
184 if flag(integer-simple)
195 if flag(integer-simple)
185 cpp-options: -DINTEGER_SIMPLE
196 cpp-options: -DINTEGER_SIMPLE
186 build-depends: integer-simple >= 0.1 && < 0.5
197 build-depends: integer-simple >= 0.1 && < 0.5