{-# LANGUAGE BangPatterns,CPP #-}
{-# LANGUAGE Trustworthy #-}
{-# LANGUAGE ViewPatterns #-}

-- |
-- Module      : Data.Text.Lazy.Encoding
-- Copyright   : (c) 2009, 2010 Bryan O'Sullivan
--
-- License     : BSD-style
-- Maintainer  : bos@serpentine.com
-- Portability : portable
--
-- Functions for converting lazy 'Text' values to and from lazy
-- 'ByteString', using several standard encodings.
--
-- To gain access to a much larger family of encodings, use the
-- <http://hackage.haskell.org/package/text-icu text-icu package>.

module Data.Text.Lazy.Encoding
    (
    -- * Decoding ByteStrings to Text
    -- $strict

    -- ** Total Functions #total#
    -- $total
      decodeLatin1

    -- *** Catchable failure
    , decodeUtf8'

    -- *** Controllable error handling
    , decodeUtf8With
    , decodeUtf16LEWith
    , decodeUtf16BEWith
    , decodeUtf32LEWith
    , decodeUtf32BEWith

    -- ** Partial Functions
    -- $partial
    , decodeASCII
    , decodeUtf8
    , decodeUtf16LE
    , decodeUtf16BE
    , decodeUtf32LE
    , decodeUtf32BE

    -- * Encoding Text to ByteStrings
    , encodeUtf8
    , encodeUtf16LE
    , encodeUtf16BE
    , encodeUtf32LE
    , encodeUtf32BE

    -- * Encoding Text using ByteString Builders
    , encodeUtf8Builder
    , encodeUtf8BuilderEscaped
    ) where

import Control.Exception (evaluate, try)
import Data.Monoid (Monoid(..))
import Data.Text.Encoding.Error (OnDecodeError, UnicodeException, strictDecode)
import Data.Text.Internal.Lazy (Text(..), chunk, empty, foldrChunks)
import Data.Word (Word8)
import qualified Data.ByteString.Builder as B
import qualified Data.ByteString.Builder.Prim as BP
import qualified Data.ByteString.Lazy as B
import qualified Data.ByteString.Lazy.Internal as B
import qualified Data.Text.Encoding as TE
import qualified Data.Text.Internal.Encoding as TE
import qualified Data.Text.Internal.Lazy.Encoding.Fusion as E
import qualified Data.Text.Internal.Lazy.Fusion as F
import qualified Data.Text.Internal.StrictBuilder as SB
import Data.Text.Unsafe (unsafeDupablePerformIO)

-- $strict
--
-- All of the single-parameter functions for decoding bytestrings
-- encoded in one of the Unicode Transformation Formats (UTF) operate
-- in a /strict/ mode: each will throw an exception if given invalid
-- input.
--
-- Each function has a variant, whose name is suffixed with -'With',
-- that gives greater control over the handling of decoding errors.
-- For instance, 'decodeUtf8' will throw an exception, but
-- 'decodeUtf8With' allows the programmer to determine what to do on a
-- decoding error.

-- $total
--
-- These functions facilitate total decoding and should be preferred
-- over their partial counterparts.

-- $partial
--
-- These functions are partial and should only be used with great caution
-- (preferably not at all). See "Data.Text.Lazy.Encoding#g:total" for better
-- solutions.

-- | Decode a 'ByteString' containing 7-bit ASCII
-- encoded text.
decodeASCII :: B.ByteString -> Text
decodeASCII :: ByteString -> Text
decodeASCII = forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr (Text -> Text -> Text
chunk forall b c a. (b -> c) -> (a -> b) -> a -> c
. ByteString -> Text
TE.decodeASCII) Text
empty forall b c a. (b -> c) -> (a -> b) -> a -> c
. ByteString -> [ByteString]
B.toChunks

-- | Decode a 'ByteString' containing Latin-1 (aka ISO-8859-1) encoded text.
decodeLatin1 :: B.ByteString -> Text
decodeLatin1 :: ByteString -> Text
decodeLatin1 = forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr (Text -> Text -> Text
chunk forall b c a. (b -> c) -> (a -> b) -> a -> c
. ByteString -> Text
TE.decodeLatin1) Text
empty forall b c a. (b -> c) -> (a -> b) -> a -> c
. ByteString -> [ByteString]
B.toChunks

-- | Decode a 'ByteString' containing UTF-8 encoded text.
decodeUtf8With :: OnDecodeError -> B.ByteString -> Text
decodeUtf8With :: OnDecodeError -> ByteString -> Text
decodeUtf8With OnDecodeError
onErr = Utf8State -> ByteString -> Text
loop Utf8State
TE.startUtf8State
  where
    chunkb :: StrictBuilder -> Text -> Text
chunkb StrictBuilder
builder Text
t | StrictBuilder -> Int
SB.sbLength StrictBuilder
builder forall a. Eq a => a -> a -> Bool
== Int
0 = Text
t
                    | Bool
otherwise = Text -> Text -> Text
Chunk (StrictBuilder -> Text
TE.strictBuilderToText StrictBuilder
builder) Text
t
    loop :: Utf8State -> ByteString -> Text
loop Utf8State
s (B.Chunk ByteString
b ByteString
bs) = case OnDecodeError
-> String
-> Utf8State
-> ByteString
-> (StrictBuilder, ByteString, Utf8State)
TE.decodeUtf8With2 OnDecodeError
onErr String
msg Utf8State
s ByteString
b of
      (StrictBuilder
builder, ByteString
_, Utf8State
s') -> StrictBuilder -> Text -> Text
chunkb StrictBuilder
builder (Utf8State -> ByteString -> Text
loop Utf8State
s' ByteString
bs)
    loop Utf8State
s ByteString
B.Empty = StrictBuilder -> Text -> Text
chunkb (OnDecodeError -> String -> Utf8State -> StrictBuilder
TE.skipIncomplete OnDecodeError
onErr String
msg Utf8State
s) Text
Empty
    msg :: String
msg = String
"Data.Text.Internal.Encoding: Invalid UTF-8 stream"

-- | Decode a 'ByteString' containing UTF-8 encoded text that is known
-- to be valid.
--
-- If the input contains any invalid UTF-8 data, an exception will be
-- thrown that cannot be caught in pure code.  For more control over
-- the handling of invalid data, use 'decodeUtf8'' or
-- 'decodeUtf8With'.
decodeUtf8 :: B.ByteString -> Text
decodeUtf8 :: ByteString -> Text
decodeUtf8 = OnDecodeError -> ByteString -> Text
decodeUtf8With OnDecodeError
strictDecode
{-# INLINE[0] decodeUtf8 #-}

-- | Decode a 'ByteString' containing UTF-8 encoded text..
--
-- If the input contains any invalid UTF-8 data, the relevant
-- exception will be returned, otherwise the decoded text.
--
-- /Note/: this function is /not/ lazy, as it must decode its entire
-- input before it can return a result.  If you need lazy (streaming)
-- decoding, use 'decodeUtf8With' in lenient mode.
decodeUtf8' :: B.ByteString -> Either UnicodeException Text
decodeUtf8' :: ByteString -> Either UnicodeException Text
decodeUtf8' ByteString
bs = forall a. IO a -> a
unsafeDupablePerformIO forall a b. (a -> b) -> a -> b
$ do
                   let t :: Text
t = ByteString -> Text
decodeUtf8 ByteString
bs
                   forall e a. Exception e => IO a -> IO (Either e a)
try (forall a. a -> IO a
evaluate (Text -> ()
rnf Text
t seq :: forall a b. a -> b -> b
`seq` Text
t))
  where
    rnf :: Text -> ()
rnf Text
Empty        = ()
    rnf (Chunk Text
_ Text
ts) = Text -> ()
rnf Text
ts
{-# INLINE decodeUtf8' #-}

-- | Encode text using UTF-8 encoding.
encodeUtf8 :: Text -> B.ByteString
encodeUtf8 :: Text -> ByteString
encodeUtf8 = forall a. (Text -> a -> a) -> a -> Text -> a
foldrChunks (ByteString -> ByteString -> ByteString
B.Chunk forall b c a. (b -> c) -> (a -> b) -> a -> c
. Text -> ByteString
TE.encodeUtf8) ByteString
B.Empty

-- | Encode text to a ByteString 'B.Builder' using UTF-8 encoding.
--
-- @since 1.1.0.0
encodeUtf8Builder :: Text -> B.Builder
encodeUtf8Builder :: Text -> Builder
encodeUtf8Builder =
    forall a. (Text -> a -> a) -> a -> Text -> a
foldrChunks (\Text
c Builder
b -> Text -> Builder
TE.encodeUtf8Builder Text
c forall a. Monoid a => a -> a -> a
`mappend` Builder
b) forall a. Monoid a => a
Data.Monoid.mempty

-- | Encode text using UTF-8 encoding and escape the ASCII characters using
-- a 'BP.BoundedPrim'.
--
-- Use this function is to implement efficient encoders for text-based formats
-- like JSON or HTML.
--
-- @since 1.1.0.0
{-# INLINE encodeUtf8BuilderEscaped #-}
encodeUtf8BuilderEscaped :: BP.BoundedPrim Word8 -> Text -> B.Builder
encodeUtf8BuilderEscaped :: BoundedPrim Word8 -> Text -> Builder
encodeUtf8BuilderEscaped BoundedPrim Word8
prim =
    forall a. (Text -> a -> a) -> a -> Text -> a
foldrChunks (\Text
c Builder
b -> BoundedPrim Word8 -> Text -> Builder
TE.encodeUtf8BuilderEscaped BoundedPrim Word8
prim Text
c forall a. Monoid a => a -> a -> a
`mappend` Builder
b) forall a. Monoid a => a
mempty

-- | Decode text from little endian UTF-16 encoding.
decodeUtf16LEWith :: OnDecodeError -> B.ByteString -> Text
decodeUtf16LEWith :: OnDecodeError -> ByteString -> Text
decodeUtf16LEWith OnDecodeError
onErr ByteString
bs = Stream Char -> Text
F.unstream (OnDecodeError -> ByteString -> Stream Char
E.streamUtf16LE OnDecodeError
onErr ByteString
bs)
{-# INLINE decodeUtf16LEWith #-}

-- | Decode text from little endian UTF-16 encoding.
--
-- If the input contains any invalid little endian UTF-16 data, an
-- exception will be thrown.  For more control over the handling of
-- invalid data, use 'decodeUtf16LEWith'.
decodeUtf16LE :: B.ByteString -> Text
decodeUtf16LE :: ByteString -> Text
decodeUtf16LE = OnDecodeError -> ByteString -> Text
decodeUtf16LEWith OnDecodeError
strictDecode
{-# INLINE decodeUtf16LE #-}

-- | Decode text from big endian UTF-16 encoding.
decodeUtf16BEWith :: OnDecodeError -> B.ByteString -> Text
decodeUtf16BEWith :: OnDecodeError -> ByteString -> Text
decodeUtf16BEWith OnDecodeError
onErr ByteString
bs = Stream Char -> Text
F.unstream (OnDecodeError -> ByteString -> Stream Char
E.streamUtf16BE OnDecodeError
onErr ByteString
bs)
{-# INLINE decodeUtf16BEWith #-}

-- | Decode text from big endian UTF-16 encoding.
--
-- If the input contains any invalid big endian UTF-16 data, an
-- exception will be thrown.  For more control over the handling of
-- invalid data, use 'decodeUtf16BEWith'.
decodeUtf16BE :: B.ByteString -> Text
decodeUtf16BE :: ByteString -> Text
decodeUtf16BE = OnDecodeError -> ByteString -> Text
decodeUtf16BEWith OnDecodeError
strictDecode
{-# INLINE decodeUtf16BE #-}

-- | Encode text using little endian UTF-16 encoding.
encodeUtf16LE :: Text -> B.ByteString
encodeUtf16LE :: Text -> ByteString
encodeUtf16LE Text
txt = [ByteString] -> ByteString
B.fromChunks (forall a. (Text -> a -> a) -> a -> Text -> a
foldrChunks ((:) forall b c a. (b -> c) -> (a -> b) -> a -> c
. Text -> ByteString
TE.encodeUtf16LE) [] Text
txt)
{-# INLINE encodeUtf16LE #-}

-- | Encode text using big endian UTF-16 encoding.
encodeUtf16BE :: Text -> B.ByteString
encodeUtf16BE :: Text -> ByteString
encodeUtf16BE Text
txt = [ByteString] -> ByteString
B.fromChunks (forall a. (Text -> a -> a) -> a -> Text -> a
foldrChunks ((:) forall b c a. (b -> c) -> (a -> b) -> a -> c
. Text -> ByteString
TE.encodeUtf16BE) [] Text
txt)
{-# INLINE encodeUtf16BE #-}

-- | Decode text from little endian UTF-32 encoding.
decodeUtf32LEWith :: OnDecodeError -> B.ByteString -> Text
decodeUtf32LEWith :: OnDecodeError -> ByteString -> Text
decodeUtf32LEWith OnDecodeError
onErr ByteString
bs = Stream Char -> Text
F.unstream (OnDecodeError -> ByteString -> Stream Char
E.streamUtf32LE OnDecodeError
onErr ByteString
bs)
{-# INLINE decodeUtf32LEWith #-}

-- | Decode text from little endian UTF-32 encoding.
--
-- If the input contains any invalid little endian UTF-32 data, an
-- exception will be thrown.  For more control over the handling of
-- invalid data, use 'decodeUtf32LEWith'.
decodeUtf32LE :: B.ByteString -> Text
decodeUtf32LE :: ByteString -> Text
decodeUtf32LE = OnDecodeError -> ByteString -> Text
decodeUtf32LEWith OnDecodeError
strictDecode
{-# INLINE decodeUtf32LE #-}

-- | Decode text from big endian UTF-32 encoding.
decodeUtf32BEWith :: OnDecodeError -> B.ByteString -> Text
decodeUtf32BEWith :: OnDecodeError -> ByteString -> Text
decodeUtf32BEWith OnDecodeError
onErr ByteString
bs = Stream Char -> Text
F.unstream (OnDecodeError -> ByteString -> Stream Char
E.streamUtf32BE OnDecodeError
onErr ByteString
bs)
{-# INLINE decodeUtf32BEWith #-}

-- | Decode text from big endian UTF-32 encoding.
--
-- If the input contains any invalid big endian UTF-32 data, an
-- exception will be thrown.  For more control over the handling of
-- invalid data, use 'decodeUtf32BEWith'.
decodeUtf32BE :: B.ByteString -> Text
decodeUtf32BE :: ByteString -> Text
decodeUtf32BE = OnDecodeError -> ByteString -> Text
decodeUtf32BEWith OnDecodeError
strictDecode
{-# INLINE decodeUtf32BE #-}

-- | Encode text using little endian UTF-32 encoding.
encodeUtf32LE :: Text -> B.ByteString
encodeUtf32LE :: Text -> ByteString
encodeUtf32LE Text
txt = [ByteString] -> ByteString
B.fromChunks (forall a. (Text -> a -> a) -> a -> Text -> a
foldrChunks ((:) forall b c a. (b -> c) -> (a -> b) -> a -> c
. Text -> ByteString
TE.encodeUtf32LE) [] Text
txt)
{-# INLINE encodeUtf32LE #-}

-- | Encode text using big endian UTF-32 encoding.
encodeUtf32BE :: Text -> B.ByteString
encodeUtf32BE :: Text -> ByteString
encodeUtf32BE Text
txt = [ByteString] -> ByteString
B.fromChunks (forall a. (Text -> a -> a) -> a -> Text -> a
foldrChunks ((:) forall b c a. (b -> c) -> (a -> b) -> a -> c
. Text -> ByteString
TE.encodeUtf32BE) [] Text
txt)
{-# INLINE encodeUtf32BE #-}