diff --git a/Data/Text.hs b/Data/Text.hs --- a/Data/Text.hs +++ b/Data/Text.hs @@ -1,9 +1,9 @@ -{-# LANGUAGE BangPatterns, CPP, Rank2Types, UnboxedTuples #-} +{-# LANGUAGE BangPatterns, CPP, MagicHash, Rank2Types, UnboxedTuples #-} {-# OPTIONS_GHC -fno-warn-orphans #-} -- | -- Module : Data.Text --- Copyright : (c) 2009, 2010, 2011 Bryan O'Sullivan, +-- Copyright : (c) 2009, 2010, 2011, 2012 Bryan O'Sullivan, -- (c) 2009 Duncan Coutts, -- (c) 2008, 2009 Tom Harper -- @@ -222,6 +222,12 @@ import qualified Data.Text.Lazy as L import Data.Int (Int64) #endif +#if __GLASGOW_HASKELL__ >= 702 +import qualified GHC.CString as GHC +#else +import qualified GHC.Base as GHC +#endif +import GHC.Prim (Addr#) -- $strict -- @@ -371,6 +377,19 @@ unpack :: Text -> String unpack = S.unstreamList . stream {-# INLINE [1] unpack #-} + +-- | /O(n)/ Convert a literal string into a Text. +unpackCString# :: Addr# -> Text +unpackCString# addr# = unstream (S.streamCString# addr#) +{-# NOINLINE unpackCString# #-} + +{-# RULES "TEXT literal" forall a. + unstream (S.streamList (L.map safe (GHC.unpackCString# a))) + = unpackCString# a #-} + +{-# RULES "TEXT literal UTF8" forall a. + unstream (S.streamList (L.map safe (GHC.unpackCStringUtf8# a))) + = unpackCString# a #-} -- | /O(1)/ Convert a character into a Text. Subject to fusion. -- Performs replacement on invalid scalar values. diff --git a/Data/Text/Fusion/Common.hs b/Data/Text/Fusion/Common.hs --- a/Data/Text/Fusion/Common.hs +++ b/Data/Text/Fusion/Common.hs @@ -1,7 +1,7 @@ -{-# LANGUAGE BangPatterns, Rank2Types #-} +{-# LANGUAGE BangPatterns, MagicHash, Rank2Types #-} -- | -- Module : Data.Text.Fusion.Common --- Copyright : (c) Bryan O'Sullivan 2009 +-- Copyright : (c) Bryan O'Sullivan 2009, 2012 -- -- License : BSD-style -- Maintainer : bos@serpentine.com, rtomharper@googlemail.com, @@ -17,6 +17,7 @@ singleton , streamList , unstreamList + , streamCString# -- * Basic interface , cons @@ -104,10 +105,13 @@ (&&), fromIntegral, otherwise) import qualified Data.List as L import qualified Prelude as P +import Data.Bits (shiftL) import Data.Int (Int64) import Data.Text.Fusion.Internal import Data.Text.Fusion.CaseMapping (foldMapping, lowerMapping, upperMapping) import Data.Text.Fusion.Size +import GHC.Prim (Addr#, chr#, indexCharOffAddr#, ord#) +import GHC.Types (Char(..), Int(..)) singleton :: Char -> Stream Char singleton c = Stream next False 1 @@ -130,6 +134,35 @@ {-# INLINE [0] unstreamList #-} {-# RULES "STREAM streamList/unstreamList fusion" forall s. streamList (unstreamList s) = s #-} + +-- | Stream the UTF-8-like packed encoding used by GHC to represent +-- constant strings in generated code. +-- +-- This encoding uses the byte sequence "\xc0\x80" to represent NUL, +-- and the string is NUL-terminated. +streamCString# :: Addr# -> Stream Char +streamCString# addr = Stream step 0 unknownSize + where + step !i + | b == 0 = Done + | b <= 0x7f = Yield (C# b#) (i+1) + | b <= 0xdf = let !c = chr $ ((b-0xc0) `shiftL` 6) + next 1 + in Yield c (i+2) + | b <= 0xef = let !c = chr $ ((b-0xe0) `shiftL` 12) + + (next 1 `shiftL` 6) + + next 2 + in Yield c (i+3) + | otherwise = let !c = chr $ ((b-0xf0) `shiftL` 18) + + (next 1 `shiftL` 12) + + (next 2 `shiftL` 6) + + next 3 + in Yield c (i+4) + where b = I# (ord# b#) + next n = I# (ord# (at# (i+n))) - 0x80 + !b# = at# i + at# (I# i#) = indexCharOffAddr# addr i# + chr (I# i#) = C# (chr# i#) +{-# INLINE [0] streamCString# #-} -- ---------------------------------------------------------------------------- -- * Basic stream functions diff --git a/Data/Text/Lazy.hs b/Data/Text/Lazy.hs --- a/Data/Text/Lazy.hs +++ b/Data/Text/Lazy.hs @@ -1,8 +1,8 @@ {-# OPTIONS_GHC -fno-warn-orphans #-} -{-# LANGUAGE BangPatterns, CPP #-} +{-# LANGUAGE BangPatterns, MagicHash, CPP #-} -- | -- Module : Data.Text.Lazy --- Copyright : (c) 2009, 2010 Bryan O'Sullivan +-- Copyright : (c) 2009, 2010, 2012 Bryan O'Sullivan -- -- License : BSD-style -- Maintainer : bos@serpentine.com, rtomharper@googlemail.com, @@ -217,6 +217,12 @@ import Data.Text.Internal (firstf, safe, textP) import qualified Data.Text.Util as U import Data.Text.Lazy.Search (indices) +#if __GLASGOW_HASKELL__ >= 702 +import qualified GHC.CString as GHC +#else +import qualified GHC.Base as GHC +#endif +import GHC.Prim (Addr#) -- $fusion -- @@ -350,6 +356,19 @@ unpack :: Text -> String unpack t = S.unstreamList (stream t) {-# INLINE [1] unpack #-} + +-- | /O(n)/ Convert a literal string into a Text. +unpackCString# :: Addr# -> Text +unpackCString# addr# = unstream (S.streamCString# addr#) +{-# NOINLINE unpackCString# #-} + +{-# RULES "TEXT literal" forall a. + unstream (S.streamList (L.map safe (GHC.unpackCString# a))) + = unpackCString# a #-} + +{-# RULES "TEXT literal UTF8" forall a. + unstream (S.streamList (L.map safe (GHC.unpackCStringUtf8# a))) + = unpackCString# a #-} -- | /O(1)/ Convert a character into a Text. Subject to fusion. -- Performs replacement on invalid scalar values. diff --git a/text.cabal b/text.cabal --- a/text.cabal +++ b/text.cabal @@ -1,5 +1,5 @@ name: text -version: 0.11.1.13 +version: 0.11.2.0 homepage: https://github.com/bos/text bug-reports: https://github.com/bos/text/issues synopsis: An efficient packed Unicode text type. @@ -31,6 +31,15 @@ the @text-icu@ package: . + —— RELEASE NOTES —— + . + Changes in 0.11.2.0: + . + * String literals are now converted directly from the format in + which GHC stores them into 'Text', without an intermediate + transformation through 'String', and without inlining of + conversion code at each site where a string literal is declared. + . license: BSD3 license-file: LICENSE author: Bryan O'Sullivan @@ -152,11 +161,13 @@ base >= 4 && < 5, bytestring >= 0.9 && < 0.10, deepseq >= 1.1, + ghc-prim, directory >= 1.0 && < 1.2, + ghc-prim, random >= 1.0 && < 1.1, QuickCheck >= 2.4 && < 2.5, HUnit >= 1.2 && < 1.3, - test-framework >= 0.4 && < 0.5, + test-framework >= 0.4 && < 0.7, test-framework-quickcheck2 >= 0.2 && < 0.3, test-framework-hunit >= 0.2 && < 0.3