a02c2da..master · text

git.iko.soy/text

39 files changed+251−1075

unifiedsplit

.gitignore+3−10 .hgtags−1 .travis.yml−111 Data/Text.hs+46−177 Data/Text/Array.hs+1−4 Data/Text/Encoding.hs+4−17 Data/Text/Encoding/Error.hs+1 Data/Text/Foreign.hs+2−5 Data/Text/IO.hs+1−4 Data/Text/Internal/Fusion/CaseMapping.hs+4−164 Data/Text/Internal/Fusion/Common.hs+37−44 Data/Text/Internal/Fusion/Size.hs+7−33 Data/Text/Internal/Fusion/Types.hs+4−4 Data/Text/Lazy.hs+20−54 Data/Text/Lazy/Builder.hs+1 Data/Text/Lazy/Builder/Int.hs+1−3 Data/Text/Lazy/Builder/RealFloat.hs−3 Data/Text/Lazy/Encoding.hs+3−13 Data/Text/Lazy/IO.hs+1 Data/Text/Lazy/Read.hs+4−3 Data/Text/Read.hs+4−3 Data/Text/Show.hs−2 Data/Text/Unsafe.hs+1−2 README.markdown+16−5 benchmarks/.gitignore+1−4 benchmarks/cabal.project−2 benchmarks/haskell/Benchmarks.hs−2 benchmarks/haskell/Benchmarks/Concat.hs−25 benchmarks/text-benchmarks.cabal+8−82 cabal.project−3 cbits/cbits.c+11−6 changelog.md+19−56 tests-and-benchmarks.markdown+8−13 tests/Makefile+1−6 tests/Tests/Properties.hs+21−94 tests/Tests/QuickCheckUtils.hs+1−1 tests/Tests/Regressions.hs−11 tests/text-tests.cabal+3−18 text.cabal+17−90

.gitignore+3−10

@@ -1,10 +1,3 @@

1−/dist/

2−/dist-boot/

3−/dist-install/

4−/dist-newstyle/

5−/cabal-dev/

6−/cabal.sandbox.config

7−/ghc.mk

8−/GNUmakefile

9−/.ghc.environment.*

10−/cabal.project.local

1+dist

2+cabal-dev

3+cabal.sandbox.config

.hgtags−1

@@ -64,4 +64,3 @@ f8a60279237c1c28e094868740d9188dd94fcc9e 1.2.0.0

6464631fd055e27fa10983e17e9ad36d1b285b85c661 1.2.1.3

65658221a68ed5ec37028357df9c6c25f52320e54ad9 1.2.2.0

6666ce1c5ac65eb9f5580114b4e380092b05a485e95d 1.2.2.1

67−d72c634a91ffe27ec5cd03921dd53efc8129d8f6 1.2.2.2

.travis.ymldeleted−111

@@ -1,111 +0,0 @@

1−# This Travis job script has been generated by a script via

2−#

3−# make_travis_yml_2.hs '-o' '.travis.yml' 'text.cabal'

4−#

5−# For more information, see https://github.com/hvr/multi-ghc-travis

6−#

7−language: c

8−sudo: false

9−

10−# Bryan threatens to turn off Travis-CI if he gets any email :-)

11−notifications:

12− email: false

13−

14−cache:

15− directories:

16− - $HOME/.cabal/packages

17− - $HOME/.cabal/store

18−

19−before_cache:

20− - rm -fv $HOME/.cabal/packages/hackage.haskell.org/build-reports.log

21− # remove files that are regenerated by 'cabal update'

22− - rm -fv $HOME/.cabal/packages/hackage.haskell.org/00-index.*

23− - rm -fv $HOME/.cabal/packages/hackage.haskell.org/*.json

24− - rm -fv $HOME/.cabal/packages/hackage.haskell.org/01-index.cache

25− - rm -fv $HOME/.cabal/packages/hackage.haskell.org/01-index.tar

26− - rm -fv $HOME/.cabal/packages/hackage.haskell.org/01-index.tar.idx

27−

28−matrix:

29− include:

30− - compiler: "ghc-7.0.4"

31− # env: TEST=--disable-tests BENCH=--disable-benchmarks

32− addons: {apt: {packages: [ghc-ppa-tools,cabal-install-head,ghc-7.0.4], sources: [hvr-ghc]}}

33− - compiler: "ghc-7.2.2"

34− # env: TEST=--disable-tests BENCH=--disable-benchmarks

35− addons: {apt: {packages: [ghc-ppa-tools,cabal-install-head,ghc-7.2.2], sources: [hvr-ghc]}}

36− - compiler: "ghc-7.4.2"

37− # env: TEST=--disable-tests BENCH=--disable-benchmarks

38− addons: {apt: {packages: [ghc-ppa-tools,cabal-install-head,ghc-7.4.2], sources: [hvr-ghc]}}

39− - compiler: "ghc-7.6.3"

40− # env: TEST=--disable-tests BENCH=--disable-benchmarks

41− addons: {apt: {packages: [ghc-ppa-tools,cabal-install-head,ghc-7.6.3], sources: [hvr-ghc]}}

42− - compiler: "ghc-7.8.4"

43− # env: TEST=--disable-tests BENCH=--disable-benchmarks

44− addons: {apt: {packages: [ghc-ppa-tools,cabal-install-head,ghc-7.8.4], sources: [hvr-ghc]}}

45− - compiler: "ghc-7.10.3"

46− # env: TEST=--disable-tests BENCH=--disable-benchmarks

47− addons: {apt: {packages: [ghc-ppa-tools,cabal-install-head,ghc-7.10.3], sources: [hvr-ghc]}}

48− - compiler: "ghc-8.0.2"

49− # env: TEST=--disable-tests BENCH=--disable-benchmarks

50− addons: {apt: {packages: [ghc-ppa-tools,cabal-install-head,ghc-8.0.2], sources: [hvr-ghc]}}

51− - compiler: "ghc-8.2.2"

52− # env: TEST=--disable-tests BENCH=--disable-benchmarks

53− addons: {apt: {packages: [ghc-ppa-tools,cabal-install-head,ghc-8.2.2], sources: [hvr-ghc]}}

54−

55−before_install:

56− - HC=${CC}

57− - HCPKG=${HC/ghc/ghc-pkg}

58− - unset CC

59− - PATH=/opt/ghc/bin:/opt/ghc-ppa-tools/bin:$PATH

60− - PKGNAME='text'

61−

62−install:

63− - cabal --version

64− - echo "$(${HC} --version) [$(${HC} --print-project-git-commit-id 2> /dev/null || echo '?')]"

65− - BENCH=${BENCH---enable-benchmarks}

66− - TEST=${TEST---enable-tests}

67− - HADDOCK=${HADDOCK-true}

68− - INSTALLED=${INSTALLED-true}

69− - travis_retry cabal update -v

70− - sed -i.bak 's/^jobs:/-- jobs:/' ${HOME}/.cabal/config

71− - rm -fv cabal.project.local

72− - "echo 'packages: .' > cabal.project"

73− - rm -f cabal.project.freeze

74− - cabal new-build -w ${HC} ${TEST} ${BENCH} --dep -j2 all

75− - cabal new-build -w ${HC} --disable-tests --disable-benchmarks --dep -j2 all

76−

77−# Here starts the actual work to be performed for the package under test;

78−# any command which exits with a non-zero exit code causes the build to fail.

79−script:

80− - if [ -f configure.ac ]; then autoreconf -i; fi

81− - rm -rf .ghc.environment.* dist/

82− - cabal sdist # test that a source-distribution can be generated

83− - cd dist/

84− - SRCTAR=(${PKGNAME}-*.tar.gz)

85− - SRC_BASENAME="${SRCTAR/%.tar.gz}"

86− - tar -xvf "./$SRC_BASENAME.tar.gz"

87− - cd "$SRC_BASENAME/"

88−## from here on, CWD is inside the extracted source-tarball

89− - rm -fv cabal.project.local

90− - "echo 'packages: .' > cabal.project"

91− # this builds all libraries and executables (without tests/benchmarks)

92− - rm -f cabal.project.freeze

93− - cabal new-build -w ${HC} --disable-tests --disable-benchmarks all

94− # this builds all libraries and executables (including tests/benchmarks)

95− # - rm -rf ./dist-newstyle

96−

97− # Build with installed constraints for packages in global-db

98− - if $INSTALLED; then

99− echo cabal new-build -w ${HC} --disable-tests --disable-benchmarks $(${HCPKG} list --global --simple-output --names-only | sed 's/$[a-zA-Z0-9-]\{1,\}$ */--constraint="\1 installed" /g') all | sh;

100− else echo "Not building with installed constraints"; fi

101−

102− # build & run tests, build benchmarks

103− - cabal new-build -w ${HC} ${TEST} ${BENCH} all

104− - if [ "x$TEST" = "x--enable-tests" ]; then cabal new-test -w ${HC} ${TEST} all; fi

105−

106− # haddock

107− - rm -rf ./dist-newstyle

108− - if $HADDOCK; then cabal new-haddock -w ${HC} --disable-tests --disable-benchmarks all; else echo "Skipping haddock generation";fi

109−

110−# REGENDATA ["-o",".travis.yml","text.cabal"]

111−# EOF

Data/Text.hs+46−177

@@ -15,6 +15,7 @@

1515--

1616-- License : BSD-style

1717-- Maintainer : bos@serpentine.com

18+-- Stability : experimental

1819-- Portability : GHC

1920--

2021-- A time and space-efficient implementation of Unicode text.

@@ -31,9 +32,8 @@

3132--

3233-- To use an extended and very rich family of functions for working

3334-- with Unicode text (including normalization, regular expressions,

34−-- non-standard encodings, text breaking, and locales), see the

35−-- <http://hackage.haskell.org/package/text-icu text-icu package >.

36−--

35+-- non-standard encodings, text breaking, and locales), see

36+-- <http://hackage.haskell.org/package/text-icu the text-icu package >.

3737

3838module Data.Text

3939 (

@@ -43,9 +43,6 @@ module Data.Text

4343 -- * Acceptable data

4444 -- $replacement

4545

46− -- * Definition of character

47− -- $character_definition

48−

4946 -- * Fusion

5047 -- $fusion

5148

@@ -63,7 +60,6 @@ module Data.Text

6360 , snoc

6461 , append

6562 , uncons

66− , unsnoc

6763 , head

6864 , last

6965 , tail

@@ -250,18 +246,6 @@ import qualified GHC.Exts as Exts

250246import Text.Printf (PrintfArg, formatArg, formatString)

251247#endif

252248

253−-- $character_definition

254−--

255−-- This package uses the term /character/ to denote Unicode /code points/.

256−--

257−-- Note that this is not the same thing as a grapheme (e.g. a

258−-- composition of code points that form one visual symbol). For

259−-- instance, consider the grapheme \"ä\". This symbol has two

260−-- Unicode representations: a single code-point representation

261−-- @U+00E4@ (the @LATIN SMALL LETTER A WITH DIAERESIS@ code point),

262−-- and a two code point representation @U+0061@ (the \"@A@\" code

263−-- point) and @U+0308@ (the @COMBINING DIAERESIS@ code point).

264−

265249-- $strict

266250--

267251-- This package provides both strict and lazy 'Text' types. The

@@ -350,13 +334,9 @@ instance Read Text where

350334 readsPrec p str = [(pack x,y) | (x,y) <- readsPrec p str]

351335

352336#if MIN_VERSION_base(4,9,0)

353−-- | Non-orphan 'Semigroup' instance only defined for

354−-- @base-4.9.0.0@ and later; orphan instances for older GHCs are

355−-- provided by

356−-- the [semigroups](http://hackage.haskell.org/package/semigroups)

357−-- package

358−--

359−-- @since 1.2.2.0

337+-- Semigroup orphan instances for older GHCs are provided by

338+-- 'semigroups` package

339+

360340instance Semigroup Text where

361341 (<>) = append

362342#endif

@@ -374,7 +354,6 @@ instance IsString Text where

374354 fromString = pack

375355

376356#if __GLASGOW_HASKELL__ >= 708

377−-- | @since 1.2.0.0

378357instance Exts.IsList Text where

379358 type Item Text = Char

380359 fromList = pack

@@ -385,7 +364,6 @@ instance Exts.IsList Text where

385364instance NFData Text where rnf !_ = ()

386365#endif

387366

388−-- | @since 1.2.1.0

389367instance Binary Text where

390368 put t = put (encodeUtf8 t)

391369 get = do

@@ -419,8 +397,6 @@ instance Data Text where

419397

420398#if MIN_VERSION_base(4,7,0)

421399-- | Only defined for @base-4.7.0.0@ and later

422−--

423−-- @since 1.2.2.0

424400instance PrintfArg Text where

425401 formatArg txt = formatString $ unpack txt

426402#endif

@@ -569,19 +545,6 @@ init (Text arr off len) | len <= 0 = emptyError "init"

569545 unstream (S.init (stream t)) = init t

570546 #-}

571547

572−-- | /O(1)/ Returns all but the last character and the last character of a

573−-- 'Text', or 'Nothing' if empty.

574−--

575−-- @since 1.2.3.0

576−unsnoc :: Text -> Maybe (Text, Char)

577−unsnoc (Text arr off len)

578− | len <= 0 = Nothing

579− | n < 0xDC00 || n > 0xDFFF = Just (text arr off (len-1), unsafeChr n)

580− | otherwise = Just (text arr off (len-2), U16.chr2 n0 n)

581− where n = A.unsafeIndex arr (off+len-1)

582− n0 = A.unsafeIndex arr (off+len-2)

583−{-# INLINE [1] unsnoc #-}

584−

585548-- | /O(1)/ Tests whether a 'Text' is empty or not. Subject to

586549-- fusion.

587550null :: Text -> Bool

@@ -661,15 +624,8 @@ compareLength t n = S.compareLengthI (stream t) n

661624-- -----------------------------------------------------------------------------

662625-- * Transformations

663626-- | /O(n)/ 'map' @f@ @t@ is the 'Text' obtained by applying @f@ to

664−-- each element of @t@.

665−--

666−-- Example:

667−--

668−-- >>> let message = pack "I am not angry. Not at all."

669−-- >>> T.map (\c -> if c == '.' then '!' else c) message

670−-- "I am not angry! Not at all!"

671−--

672−-- Subject to fusion. Performs replacement on invalid scalar values.

627+-- each element of @t@. Subject to fusion. Performs replacement on

628+-- invalid scalar values.

673629map :: (Char -> Char) -> Text -> Text

674630map f t = unstream (S.map (safe . f) (stream t))

675631{-# INLINE [1] map #-}

@@ -677,36 +633,18 @@ map f t = unstream (S.map (safe . f) (stream t))

677633-- | /O(n)/ The 'intercalate' function takes a 'Text' and a list of

678634-- 'Text's and concatenates the list after interspersing the first

679635-- argument between each element of the list.

680−--

681−-- Example:

682−--

683−-- >>> T.intercalate "NI!" ["We", "seek", "the", "Holy", "Grail"]

684−-- "WeNI!seekNI!theNI!HolyNI!Grail"

685636intercalate :: Text -> [Text] -> Text

686637intercalate t = concat . (F.intersperse t)

687638{-# INLINE intercalate #-}

688639

689640-- | /O(n)/ The 'intersperse' function takes a character and places it

690−-- between the characters of a 'Text'.

691−--

692−-- Example:

693−--

694−-- >>> T.intersperse '.' "SHIELD"

695−-- "S.H.I.E.L.D"

696−--

697−-- Subject to fusion. Performs replacement on invalid scalar values.

641+-- between the characters of a 'Text'. Subject to fusion. Performs

642+-- replacement on invalid scalar values.

698643intersperse :: Char -> Text -> Text

699644intersperse c t = unstream (S.intersperse (safe c) (stream t))

700645{-# INLINE intersperse #-}

701646

702−-- | /O(n)/ Reverse the characters of a string.

703−--

704−-- Example:

705−--

706−-- >>> T.reverse "desrever"

707−-- "reversed"

708−--

709−-- Subject to fusion.

647+-- | /O(n)/ Reverse the characters of a string. Subject to fusion.

710648reverse :: Text -> Text

711649reverse t = S.reverse (stream t)

712650{-# INLINE reverse #-}

@@ -725,14 +663,12 @@ reverse t = S.reverse (stream t)

725663-- @needle@ occurs in @replacement@, that occurrence will /not/ itself

726664-- be replaced recursively:

727665--

728−-- >>> replace "oo" "foo" "oo"

729−-- "foo"

666+-- > replace "oo" "foo" "oo" == "foo"

730667--

731668-- In cases where several instances of @needle@ overlap, only the

732669-- first one will be replaced:

733670--

734−-- >>> replace "ofo" "bar" "ofofo"

735−-- "barfo"

671+-- > replace "ofo" "bar" "ofofo" == "barfo"

736672--

737673-- In (unlikely) bad cases, this function's time complexity degrades

738674-- towards /O(n*m)/.

@@ -846,8 +782,6 @@ toUpper t = unstream (S.toUpper (stream t))

846782-- guides disagree on whether the book name \"The Hill of the Red

847783-- Fox\" is correctly title cased—but this function will

848784-- capitalize /every/ word.

849−--

850−-- @since 1.0.0.0

851785toTitle :: Text -> Text

852786toTitle t = unstream (S.toTitle (stream t))

853787{-# INLINE toTitle #-}

@@ -858,11 +792,8 @@ toTitle t = unstream (S.toTitle (stream t))

858792--

859793-- Examples:

860794--

861−-- >>> justifyLeft 7 'x' "foo"

862−-- "fooxxxx"

863−--

864−-- >>> justifyLeft 3 'x' "foobar"

865−-- "foobar"

795+-- > justifyLeft 7 'x' "foo" == "fooxxxx"

796+-- > justifyLeft 3 'x' "foobar" == "foobar"

866797justifyLeft :: Int -> Char -> Text -> Text

867798justifyLeft k c t

868799 | len >= k = t

@@ -883,11 +814,8 @@ justifyLeft k c t

883814--

884815-- Examples:

885816--

886−-- >>> justifyRight 7 'x' "bar"

887−-- "xxxxbar"

888−--

889−-- >>> justifyRight 3 'x' "foobar"

890−-- "foobar"

817+-- > justifyRight 7 'x' "bar" == "xxxxbar"

818+-- > justifyRight 3 'x' "foobar" == "foobar"

891819justifyRight :: Int -> Char -> Text -> Text

892820justifyRight k c t

893821 | len >= k = t

@@ -901,8 +829,7 @@ justifyRight k c t

901829--

902830-- Examples:

903831--

904−-- >>> center 8 'x' "HS"

905−-- "xxxHSxxx"

832+-- > center 8 'x' "HS" = "xxxHSxxx"

906833center :: Int -> Char -> Text -> Text

907834center k c t

908835 | len >= k = t

@@ -917,14 +844,6 @@ center k c t

917844-- of its 'Text' argument. Note that this function uses 'pack',

918845-- 'unpack', and the list version of transpose, and is thus not very

919846-- efficient.

920−--

921−-- Examples:

922−--

923−-- >>> transpose ["green","orange"]

924−-- ["go","rr","ea","en","ng","e"]

925−--

926−-- >>> transpose ["blue","red"]

927−-- ["br","le","ud","e"]

928847transpose :: [Text] -> [Text]

929848transpose ts = P.map pack (L.transpose (P.map unpack ts))

930849

@@ -1169,10 +1088,7 @@ iterN n t@(Text _arr _off len) = loop 0 0

11691088--

11701089-- Examples:

11711090--

1172−-- >>> takeEnd 3 "foobar"

1173−-- "bar"

1174−--

1175−-- @since 1.1.1.0

1091+-- > takeEnd 3 "foobar" == "bar"

11761092takeEnd :: Int -> Text -> Text

11771093takeEnd n t@(Text arr off len)

11781094 | n <= 0 = empty

@@ -1183,8 +1099,8 @@ takeEnd n t@(Text arr off len)

11831099iterNEnd :: Int -> Text -> Int

11841100iterNEnd n t@(Text _arr _off len) = loop (len-1) n

11851101 where loop i !m

1186− | m <= 0 = i+1

11871102 | i <= 0 = 0

1103+ | m <= 1 = i

11881104 | otherwise = loop (i+d) (m-1)

11891105 where d = reverseIter_ t i

11901106

@@ -1211,10 +1127,7 @@ drop n t@(Text arr off len)

12111127--

12121128-- Examples:

12131129--

1214−-- >>> dropEnd 3 "foobar"

1215−-- "foo"

1216−--

1217−-- @since 1.1.1.0

1130+-- > dropEnd 3 "foobar" == "foo"

12181131dropEnd :: Int -> Text -> Text

12191132dropEnd n t@(Text arr off len)

12201133 | n <= 0 = t

@@ -1244,10 +1157,7 @@ takeWhile p t@(Text arr off len) = loop 0

12441157-- satisfy @p@. Subject to fusion.

12451158-- Examples:

12461159--

1247−-- >>> takeWhileEnd (=='o') "foo"

1248−-- "oo"

1249−--

1250−-- @since 1.2.2.0

1160+-- > takeWhileEnd (=='o') "foo" == "oo"

12511161takeWhileEnd :: (Char -> Bool) -> Text -> Text

12521162takeWhileEnd p t@(Text arr off len) = loop (len-1) len

12531163 where loop !i !l | l <= 0 = t

@@ -1286,8 +1196,7 @@ dropWhile p t@(Text arr off len) = loop 0 0

12861196--

12871197-- Examples:

12881198--

1289−-- >>> dropWhileEnd (=='.') "foo..."

1290−-- "foo"

1199+-- > dropWhileEnd (=='.') "foo..." == "foo"

12911200dropWhileEnd :: (Char -> Bool) -> Text -> Text

12921201dropWhileEnd p t@(Text arr off len) = loop (len-1) len

12931202 where loop !i !l | l <= 0 = empty

@@ -1405,14 +1314,9 @@ tails t | null t = [empty]

14051314--

14061315-- Examples:

14071316--

1408−-- >>> splitOn "\r\n" "a\r\nb\r\nd\r\ne"

1409−-- ["a","b","d","e"]

1410−--

1411−-- >>> splitOn "aaa" "aaaXaaaXaaaXaaa"

1412−-- ["","X","X","X",""]

1413−--

1414−-- >>> splitOn "x" "x"

1415−-- ["",""]

1317+-- > splitOn "\r\n" "a\r\nb\r\nd\r\ne" == ["a","b","d","e"]

1318+-- > splitOn "aaa" "aaaXaaaXaaaXaaa" == ["","X","X","X",""]

1319+-- > splitOn "x" "x" == ["",""]

14161320--

14171321-- and

14181322--

@@ -1448,11 +1352,8 @@ splitOn pat@(Text _ _ l) src@(Text arr off len)

14481352-- resulting components do not contain the separators. Two adjacent

14491353-- separators result in an empty component in the output. eg.

14501354--

1451−-- >>> split (=='a') "aabbaca"

1452−-- ["","","bb","c",""]

1453−--

1454−-- >>> split (=='a') ""

1455−-- [""]

1355+-- > split (=='a') "aabbaca" == ["","","bb","c",""]

1356+-- > split (=='a') "" == [""]

14561357split :: (Char -> Bool) -> Text -> [Text]

14571358split _ t@(Text _off _arr 0) = [t]

14581359split p t = loop t

@@ -1465,11 +1366,8 @@ split p t = loop t

14651366-- element may be shorter than the other chunks, depending on the

14661367-- length of the input. Examples:

14671368--

1468−-- >>> chunksOf 3 "foobarbaz"

1469−-- ["foo","bar","baz"]

1470−--

1471−-- >>> chunksOf 4 "haskell.org"

1472−-- ["hask","ell.","org"]

1369+-- > chunksOf 3 "foobarbaz" == ["foo","bar","baz"]

1370+-- > chunksOf 4 "haskell.org" == ["hask","ell.","org"]

14731371chunksOf :: Int -> Text -> [Text]

14741372chunksOf k = go

14751373 where

@@ -1514,11 +1412,8 @@ filter p t = unstream (S.filter p (stream t))

15141412--

15151413-- Examples:

15161414--

1517−-- >>> breakOn "::" "a::b::c"

1518−-- ("a","::b::c")

1519−--

1520−-- >>> breakOn "/" "foobar"

1521−-- ("foobar","")

1415+-- > breakOn "::" "a::b::c" ==> ("a", "::b::c")

1416+-- > breakOn "/" "foobar" ==> ("foobar", "")

15221417--

15231418-- Laws:

15241419--

@@ -1546,8 +1441,7 @@ breakOn pat src@(Text arr off len)

15461441-- up to and including the last match of @needle@. The second is the

15471442-- remainder of @haystack@, following the match.

15481443--

1549−-- >>> breakOnEnd "::" "a::b::c"

1550−-- ("a::b::","c")

1444+-- > breakOnEnd "::" "a::b::c" ==> ("a::b::", "c")

15511445breakOnEnd :: Text -> Text -> (Text, Text)

15521446breakOnEnd pat src = (reverse b, reverse a)

15531447 where (a,b) = breakOn (reverse pat) (reverse src)

@@ -1562,11 +1456,10 @@ breakOnEnd pat src = (reverse b, reverse a)

15621456--

15631457-- Examples:

15641458--

1565−-- >>> breakOnAll "::" ""

1566−-- []

1567−--

1568−-- >>> breakOnAll "/" "a/b/c/"

1569−-- [("a","/b/c/"),("a/b","/c/"),("a/b/c","/")]

1459+-- > breakOnAll "::" ""

1460+-- > ==> []

1461+-- > breakOnAll "/" "a/b/c/"

1462+-- > ==> [("a", "/b/c/"), ("a/b", "/c/"), ("a/b/c", "/")]

15701463--

15711464-- In (unlikely) bad cases, this function's time complexity degrades

15721465-- towards /O(n*m)/.

@@ -1765,14 +1658,9 @@ isInfixOf needle haystack

17651658--

17661659-- Examples:

17671660--

1768−-- >>> stripPrefix "foo" "foobar"

1769−-- Just "bar"

1770−--

1771−-- >>> stripPrefix "" "baz"

1772−-- Just "baz"

1773−--

1774−-- >>> stripPrefix "foo" "quux"

1775−-- Nothing

1661+-- > stripPrefix "foo" "foobar" == Just "bar"

1662+-- > stripPrefix "" "baz" == Just "baz"

1663+-- > stripPrefix "foo" "quux" == Nothing

17761664--

17771665-- This is particularly useful with the @ViewPatterns@ extension to

17781666-- GHC, as follows:

@@ -1797,14 +1685,9 @@ stripPrefix p@(Text _arr _off plen) t@(Text arr off len)

17971685--

17981686-- Examples:

17991687--

1800−-- >>> commonPrefixes "foobar" "fooquux"

1801−-- Just ("foo","bar","quux")

1802−--

1803−-- >>> commonPrefixes "veeble" "fetzer"

1804−-- Nothing

1805−--

1806−-- >>> commonPrefixes "" "baz"

1807−-- Nothing

1688+-- > commonPrefixes "foobar" "fooquux" == Just ("foo","bar","quux")

1689+-- > commonPrefixes "veeble" "fetzer" == Nothing

1690+-- > commonPrefixes "" "baz" == Nothing

18081691commonPrefixes :: Text -> Text -> Maybe (Text,Text,Text)

18091692commonPrefixes t0@(Text arr0 off0 len0) t1@(Text arr1 off1 len1) = go 0 0

18101693 where

@@ -1821,14 +1704,9 @@ commonPrefixes t0@(Text arr0 off0 len0) t1@(Text arr1 off1 len1) = go 0 0

18211704--

18221705-- Examples:

18231706--

1824−-- >>> stripSuffix "bar" "foobar"

1825−-- Just "foo"

1826−--

1827−-- >>> stripSuffix "" "baz"

1828−-- Just "baz"

1829−--

1830−-- >>> stripSuffix "foo" "quux"

1831−-- Nothing

1707+-- > stripSuffix "bar" "foobar" == Just "foo"

1708+-- > stripSuffix "" "baz" == Just "baz"

1709+-- > stripSuffix "foo" "quux" == Nothing

18321710--

18331711-- This is particularly useful with the @ViewPatterns@ extension to

18341712-- GHC, as follows:

@@ -1875,12 +1753,3 @@ copy (Text arr off len) = Text (A.run go) 0 len

18751753 marr <- A.new len

18761754 A.copyI marr 0 arr off len

18771755 return marr

1878−

1879−

1880−-------------------------------------------------

1881−-- NOTE: the named chunk below used by doctest;

1882−-- verify the doctests via `doctest -fobject-code Data/Text.hs`

1883−

1884−-- $setup

1885−-- >>> :set -XOverloadedStrings

1886−-- >>> import qualified Data.Text as T

Data/Text/Array.hs+1−4

@@ -7,6 +7,7 @@

77--

88-- License : BSD-style

99-- Maintainer : bos@serpentine.com

10+-- Stability : experimental

1011-- Portability : portable

1112--

1213-- Packed, unboxed, heap-resident arrays. Suitable for performance

@@ -79,8 +80,6 @@ import GHC.Word (Word16(..))

7980import Prelude hiding (length, read)

8081

8182-- | Immutable array type.

82−--

83−-- The 'Array' constructor is exposed since @text-1.1.1.3@

8483data Array = Array {

8584 aBA :: ByteArray#

8685#if defined(ASSERTS)

@@ -89,8 +88,6 @@ data Array = Array {

8988 }

9089

9190-- | Mutable array type, for use in the ST monad.

92−--

93−-- The 'MArray' constructor is exposed since @text-1.1.1.3@

9491data MArray s = MArray {

9592 maBA :: MutableByteArray# s

9693#if defined(ASSERTS)

Data/Text/Encoding.hs+4−17

@@ -11,13 +11,14 @@

1111--

1212-- License : BSD-style

1313-- Maintainer : bos@serpentine.com

14+-- Stability : experimental

1415-- Portability : portable

1516--

1617-- Functions for converting 'Text' values to and from 'ByteString',

1718-- using several standard encodings.

1819--

1920-- To gain access to a much larger family of encodings, use the

20−-- <http://hackage.haskell.org/package/text-icu text-icu package>.

21+-- @text-icu@ package: <http://hackage.haskell.org/package/text-icu>

2122

2223module Data.Text.Encoding

2324 (

@@ -78,11 +79,7 @@ import Data.Text.Internal.Unsafe.Shift (shiftR)

7879import Data.Text.Show ()

7980import Data.Text.Unsafe (unsafeDupablePerformIO)

8081import Data.Word (Word8, Word32)

81−#if __GLASGOW_HASKELL__ >= 703

82−import Foreign.C.Types (CSize(CSize))

83−#else

84−import Foreign.C.Types (CSize)

85−#endif

82+import Foreign.C.Types (CSize(..))

8683import Foreign.ForeignPtr (withForeignPtr)

8784import Foreign.Marshal.Utils (with)

8885import Foreign.Ptr (Ptr, minusPtr, nullPtr, plusPtr)

@@ -212,8 +209,6 @@ decodeUtf8With onErr (PS fp off len) = runText $ \done -> do

212209-- or continuation where it is encountered.

213210

214211-- | A stream oriented decoding result.

215−--

216−-- @since 1.0.0.0

217212data Decoding = Some Text ByteString (ByteString -> Decoding)

218213

219214instance Show Decoding where

@@ -233,15 +228,11 @@ newtype DecoderState = DecoderState Word32 deriving (Eq, Show, Num, Storable)

233228-- thrown (either by this function or a continuation) that cannot be

234229-- caught in pure code. For more control over the handling of invalid

235230-- data, use 'streamDecodeUtf8With'.

236−--

237−-- @since 1.0.0.0

238231streamDecodeUtf8 :: ByteString -> Decoding

239232streamDecodeUtf8 = streamDecodeUtf8With strictDecode

240233

241234-- | Decode, in a stream oriented way, a 'ByteString' containing UTF-8

242235-- encoded text.

243−--

244−-- @since 1.0.0.0

245236streamDecodeUtf8With :: OnDecodeError -> ByteString -> Decoding

246237streamDecodeUtf8With onErr = decodeChunk B.empty 0 0

247238 where

@@ -317,8 +308,6 @@ decodeUtf8' = unsafeDupablePerformIO . try . evaluate . decodeUtf8With strictDec

317308{-# INLINE decodeUtf8' #-}

318309

319310-- | Encode text to a ByteString 'B.Builder' using UTF-8 encoding.

320−--

321−-- @since 1.1.0.0

322311encodeUtf8Builder :: Text -> B.Builder

323312encodeUtf8Builder = encodeUtf8BuilderEscaped (BP.liftFixedToBounded BP.word8)

324313

@@ -327,8 +316,6 @@ encodeUtf8Builder = encodeUtf8BuilderEscaped (BP.liftFixedToBounded BP.word8)

327316--

328317-- Use this function is to implement efficient encoders for text-based formats

329318-- like JSON or HTML.

330−--

331−-- @since 1.1.0.0

332319{-# INLINE encodeUtf8BuilderEscaped #-}

333320-- TODO: Extend documentation with references to source code in @blaze-html@

334321-- or @aeson@ that uses this function.

@@ -386,7 +373,7 @@ encodeUtf8 :: Text -> ByteString

386373encodeUtf8 (Text arr off len)

387374 | len == 0 = B.empty

388375 | otherwise = unsafeDupablePerformIO $ do

389− fp <- mallocByteString (len*3) -- see https://github.com/haskell/text/issues/194 for why len*3 is enough

376+ fp <- mallocByteString (len*4)

390377 withForeignPtr fp $ \ptr ->

391378 with ptr $ \destPtr -> do

392379 c_encode_utf8 destPtr (A.aBA arr) (fromIntegral off) (fromIntegral len)

Data/Text/Encoding/Error.hs+1

@@ -10,6 +10,7 @@

1010--

1111-- License : BSD-style

1212-- Maintainer : bos@serpentine.com

13+-- Stability : experimental

1314-- Portability : GHC

1415--

1516-- Types and functions for dealing with encoding and decoding errors

Data/Text/Foreign.hs+2−5

@@ -5,6 +5,7 @@

55--

66-- License : BSD-style

77-- Maintainer : bos@serpentine.com

8+-- Stability : experimental

89-- Portability : GHC

910--

1011-- Support for using 'Text' data with native code via the Haskell

@@ -107,7 +108,7 @@ takeWord16 (I16 n) t@(Text arr off len)

107108 | n >= len || m >= len = t

108109 | otherwise = Text arr off m

109110 where

110− m | w < 0xD800 || w > 0xDBFF = n

111+ m | w < 0xDB00 || w > 0xD8FF = n

111112 | otherwise = n+1

112113 w = A.unsafeIndex arr (off+n-1)

113114

@@ -156,8 +157,6 @@ asForeignPtr t@(Text _arr _off len) = do

156157-- | /O(n)/ Decode a C string with explicit length, which is assumed

157158-- to have been encoded as UTF-8. If decoding fails, a

158159-- 'UnicodeException' is thrown.

159−--

160−-- @since 1.0.0.0

161160peekCStringLen :: CStringLen -> IO Text

162161peekCStringLen cs = do

163162 bs <- unsafePackCStringLen cs

@@ -170,7 +169,5 @@ peekCStringLen cs = do

170169-- The temporary storage is freed when the subcomputation terminates

171170-- (either normally or via an exception), so the pointer to the

172171-- temporary storage must /not/ be used after this function returns.

173−--

174−-- @since 1.0.0.0

175172withCStringLen :: Text -> (CStringLen -> IO a) -> IO a

176173withCStringLen t act = unsafeUseAsCStringLen (encodeUtf8 t) act

Data/Text/IO.hs+1−4

@@ -8,6 +8,7 @@

99-- License : BSD-style

1010-- Maintainer : bos@serpentine.com

11+-- Stability : experimental

1112-- Portability : GHC

1213--

1314-- Efficient locale-sensitive support for text I\/O.

@@ -335,7 +336,3 @@ putStrLn = hPutStrLn stdout

335336-- >

336337-- > putStr_Utf16LE :: Text -> IO ()

337338-- > putStr_Utf16LE t = B.putStr (encodeUtf16LE t)

338−--

339−-- On transcoding errors, an 'IOError' exception is thrown. You can

340−-- use the API in "Data.Text.Encoding" if you need more control over

341−-- error handling or transcoding.

Data/Text/Internal/Fusion/CaseMapping.hs+4−164

@@ -1,10 +1,10 @@

11{-# LANGUAGE Rank2Types #-}

22-- AUTOMATICALLY GENERATED - DO NOT EDIT

33-- Generated by scripts/CaseMapping.hs

4−-- CaseFolding-9.0.0.txt

5−-- Date: 2016-03-02, 18:54:54 GMT

6−-- SpecialCasing-9.0.0.txt

7−-- Date: 2016-03-02, 18:55:13 GMT

4+-- CaseFolding-8.0.0.txt

5+-- Date: 2015-01-13, 18:16:36 GMT [MD]

6+-- SpecialCasing-8.0.0.txt

7+-- Date: 2014-12-16, 23:08:04 GMT [MD]

99module Data.Text.Internal.Fusion.CaseMapping where

1010import Data.Char

@@ -371,24 +371,6 @@ foldMapping '\x13fb' s = Yield '\x13f3' (CC s '\x0000' '\x0000')

371371foldMapping '\x13fc' s = Yield '\x13f4' (CC s '\x0000' '\x0000')

372372-- CHEROKEE SMALL LETTER MV

373373foldMapping '\x13fd' s = Yield '\x13f5' (CC s '\x0000' '\x0000')

374−-- CYRILLIC SMALL LETTER ROUNDED VE

375−foldMapping '\x1c80' s = Yield '\x0432' (CC s '\x0000' '\x0000')

376−-- CYRILLIC SMALL LETTER LONG-LEGGED DE

377−foldMapping '\x1c81' s = Yield '\x0434' (CC s '\x0000' '\x0000')

378−-- CYRILLIC SMALL LETTER NARROW O

379−foldMapping '\x1c82' s = Yield '\x043e' (CC s '\x0000' '\x0000')

380−-- CYRILLIC SMALL LETTER WIDE ES

381−foldMapping '\x1c83' s = Yield '\x0441' (CC s '\x0000' '\x0000')

382−-- CYRILLIC SMALL LETTER TALL TE

383−foldMapping '\x1c84' s = Yield '\x0442' (CC s '\x0000' '\x0000')

384−-- CYRILLIC SMALL LETTER THREE-LEGGED TE

385−foldMapping '\x1c85' s = Yield '\x0442' (CC s '\x0000' '\x0000')

386−-- CYRILLIC SMALL LETTER TALL HARD SIGN

387−foldMapping '\x1c86' s = Yield '\x044a' (CC s '\x0000' '\x0000')

388−-- CYRILLIC SMALL LETTER TALL YAT

389−foldMapping '\x1c87' s = Yield '\x0463' (CC s '\x0000' '\x0000')

390−-- CYRILLIC SMALL LETTER UNBLENDED UK

391−foldMapping '\x1c88' s = Yield '\xa64b' (CC s '\x0000' '\x0000')

392374-- LATIN SMALL LETTER H WITH LINE BELOW

393375foldMapping '\x1e96' s = Yield '\x0068' (CC s '\x0331' '\x0000')

394376-- LATIN SMALL LETTER T WITH DIAERESIS

@@ -563,8 +545,6 @@ foldMapping '\x1ff6' s = Yield '\x03c9' (CC s '\x0342' '\x0000')

563545foldMapping '\x1ff7' s = Yield '\x03c9' (CC s '\x0342' '\x03b9')

564546-- GREEK CAPITAL LETTER OMEGA WITH PROSGEGRAMMENI

565547foldMapping '\x1ffc' s = Yield '\x03c9' (CC s '\x03b9' '\x0000')

566−-- LATIN CAPITAL LETTER SMALL CAPITAL I

567−foldMapping '\xa7ae' s = Yield '\x026a' (CC s '\x0000' '\x0000')

568548-- LATIN CAPITAL LETTER J WITH CROSSED-TAIL

569549foldMapping '\xa7b2' s = Yield '\x029d' (CC s '\x0000' '\x0000')

570550-- LATIN CAPITAL LETTER CHI

@@ -757,78 +737,6 @@ foldMapping '\xfb15' s = Yield '\x0574' (CC s '\x056b' '\x0000')

757737foldMapping '\xfb16' s = Yield '\x057e' (CC s '\x0576' '\x0000')

758738-- ARMENIAN SMALL LIGATURE MEN XEH

759739foldMapping '\xfb17' s = Yield '\x0574' (CC s '\x056d' '\x0000')

760−-- OSAGE CAPITAL LETTER A

761−foldMapping '\x104b0' s = Yield '\x104d8' (CC s '\x0000' '\x0000')

762−-- OSAGE CAPITAL LETTER AI

763−foldMapping '\x104b1' s = Yield '\x104d9' (CC s '\x0000' '\x0000')

764−-- OSAGE CAPITAL LETTER AIN

765−foldMapping '\x104b2' s = Yield '\x104da' (CC s '\x0000' '\x0000')

766−-- OSAGE CAPITAL LETTER AH

767−foldMapping '\x104b3' s = Yield '\x104db' (CC s '\x0000' '\x0000')

768−-- OSAGE CAPITAL LETTER BRA

769−foldMapping '\x104b4' s = Yield '\x104dc' (CC s '\x0000' '\x0000')

770−-- OSAGE CAPITAL LETTER CHA

771−foldMapping '\x104b5' s = Yield '\x104dd' (CC s '\x0000' '\x0000')

772−-- OSAGE CAPITAL LETTER EHCHA

773−foldMapping '\x104b6' s = Yield '\x104de' (CC s '\x0000' '\x0000')

774−-- OSAGE CAPITAL LETTER E

775−foldMapping '\x104b7' s = Yield '\x104df' (CC s '\x0000' '\x0000')

776−-- OSAGE CAPITAL LETTER EIN

777−foldMapping '\x104b8' s = Yield '\x104e0' (CC s '\x0000' '\x0000')

778−-- OSAGE CAPITAL LETTER HA

779−foldMapping '\x104b9' s = Yield '\x104e1' (CC s '\x0000' '\x0000')

780−-- OSAGE CAPITAL LETTER HYA

781−foldMapping '\x104ba' s = Yield '\x104e2' (CC s '\x0000' '\x0000')

782−-- OSAGE CAPITAL LETTER I

783−foldMapping '\x104bb' s = Yield '\x104e3' (CC s '\x0000' '\x0000')

784−-- OSAGE CAPITAL LETTER KA

785−foldMapping '\x104bc' s = Yield '\x104e4' (CC s '\x0000' '\x0000')

786−-- OSAGE CAPITAL LETTER EHKA

787−foldMapping '\x104bd' s = Yield '\x104e5' (CC s '\x0000' '\x0000')

788−-- OSAGE CAPITAL LETTER KYA

789−foldMapping '\x104be' s = Yield '\x104e6' (CC s '\x0000' '\x0000')

790−-- OSAGE CAPITAL LETTER LA

791−foldMapping '\x104bf' s = Yield '\x104e7' (CC s '\x0000' '\x0000')

792−-- OSAGE CAPITAL LETTER MA

793−foldMapping '\x104c0' s = Yield '\x104e8' (CC s '\x0000' '\x0000')

794−-- OSAGE CAPITAL LETTER NA

795−foldMapping '\x104c1' s = Yield '\x104e9' (CC s '\x0000' '\x0000')

796−-- OSAGE CAPITAL LETTER O

797−foldMapping '\x104c2' s = Yield '\x104ea' (CC s '\x0000' '\x0000')

798−-- OSAGE CAPITAL LETTER OIN

799−foldMapping '\x104c3' s = Yield '\x104eb' (CC s '\x0000' '\x0000')

800−-- OSAGE CAPITAL LETTER PA

801−foldMapping '\x104c4' s = Yield '\x104ec' (CC s '\x0000' '\x0000')

802−-- OSAGE CAPITAL LETTER EHPA

803−foldMapping '\x104c5' s = Yield '\x104ed' (CC s '\x0000' '\x0000')

804−-- OSAGE CAPITAL LETTER SA

805−foldMapping '\x104c6' s = Yield '\x104ee' (CC s '\x0000' '\x0000')

806−-- OSAGE CAPITAL LETTER SHA

807−foldMapping '\x104c7' s = Yield '\x104ef' (CC s '\x0000' '\x0000')

808−-- OSAGE CAPITAL LETTER TA

809−foldMapping '\x104c8' s = Yield '\x104f0' (CC s '\x0000' '\x0000')

810−-- OSAGE CAPITAL LETTER EHTA

811−foldMapping '\x104c9' s = Yield '\x104f1' (CC s '\x0000' '\x0000')

812−-- OSAGE CAPITAL LETTER TSA

813−foldMapping '\x104ca' s = Yield '\x104f2' (CC s '\x0000' '\x0000')

814−-- OSAGE CAPITAL LETTER EHTSA

815−foldMapping '\x104cb' s = Yield '\x104f3' (CC s '\x0000' '\x0000')

816−-- OSAGE CAPITAL LETTER TSHA

817−foldMapping '\x104cc' s = Yield '\x104f4' (CC s '\x0000' '\x0000')

818−-- OSAGE CAPITAL LETTER DHA

819−foldMapping '\x104cd' s = Yield '\x104f5' (CC s '\x0000' '\x0000')

820−-- OSAGE CAPITAL LETTER U

821−foldMapping '\x104ce' s = Yield '\x104f6' (CC s '\x0000' '\x0000')

822−-- OSAGE CAPITAL LETTER WA

823−foldMapping '\x104cf' s = Yield '\x104f7' (CC s '\x0000' '\x0000')

824−-- OSAGE CAPITAL LETTER KHA

825−foldMapping '\x104d0' s = Yield '\x104f8' (CC s '\x0000' '\x0000')

826−-- OSAGE CAPITAL LETTER GHA

827−foldMapping '\x104d1' s = Yield '\x104f9' (CC s '\x0000' '\x0000')

828−-- OSAGE CAPITAL LETTER ZA

829−foldMapping '\x104d2' s = Yield '\x104fa' (CC s '\x0000' '\x0000')

830−-- OSAGE CAPITAL LETTER ZHA

831−foldMapping '\x104d3' s = Yield '\x104fb' (CC s '\x0000' '\x0000')

832740-- OLD HUNGARIAN CAPITAL LETTER A

833741foldMapping '\x10c80' s = Yield '\x10cc0' (CC s '\x0000' '\x0000')

834742-- OLD HUNGARIAN CAPITAL LETTER AA

@@ -931,72 +839,4 @@ foldMapping '\x10cb0' s = Yield '\x10cf0' (CC s '\x0000' '\x0000')

931839foldMapping '\x10cb1' s = Yield '\x10cf1' (CC s '\x0000' '\x0000')

932840-- OLD HUNGARIAN CAPITAL LETTER US

933841foldMapping '\x10cb2' s = Yield '\x10cf2' (CC s '\x0000' '\x0000')

934−-- ADLAM CAPITAL LETTER ALIF

935−foldMapping '\x1e900' s = Yield '\x1e922' (CC s '\x0000' '\x0000')

936−-- ADLAM CAPITAL LETTER DAALI

937−foldMapping '\x1e901' s = Yield '\x1e923' (CC s '\x0000' '\x0000')

938−-- ADLAM CAPITAL LETTER LAAM

939−foldMapping '\x1e902' s = Yield '\x1e924' (CC s '\x0000' '\x0000')

940−-- ADLAM CAPITAL LETTER MIIM

941−foldMapping '\x1e903' s = Yield '\x1e925' (CC s '\x0000' '\x0000')

942−-- ADLAM CAPITAL LETTER BA

943−foldMapping '\x1e904' s = Yield '\x1e926' (CC s '\x0000' '\x0000')

944−-- ADLAM CAPITAL LETTER SINNYIIYHE

945−foldMapping '\x1e905' s = Yield '\x1e927' (CC s '\x0000' '\x0000')

946−-- ADLAM CAPITAL LETTER PE

947−foldMapping '\x1e906' s = Yield '\x1e928' (CC s '\x0000' '\x0000')

948−-- ADLAM CAPITAL LETTER BHE

949−foldMapping '\x1e907' s = Yield '\x1e929' (CC s '\x0000' '\x0000')

950−-- ADLAM CAPITAL LETTER RA

951−foldMapping '\x1e908' s = Yield '\x1e92a' (CC s '\x0000' '\x0000')

952−-- ADLAM CAPITAL LETTER E

953−foldMapping '\x1e909' s = Yield '\x1e92b' (CC s '\x0000' '\x0000')

954−-- ADLAM CAPITAL LETTER FA

955−foldMapping '\x1e90a' s = Yield '\x1e92c' (CC s '\x0000' '\x0000')

956−-- ADLAM CAPITAL LETTER I

957−foldMapping '\x1e90b' s = Yield '\x1e92d' (CC s '\x0000' '\x0000')

958−-- ADLAM CAPITAL LETTER O

959−foldMapping '\x1e90c' s = Yield '\x1e92e' (CC s '\x0000' '\x0000')

960−-- ADLAM CAPITAL LETTER DHA

961−foldMapping '\x1e90d' s = Yield '\x1e92f' (CC s '\x0000' '\x0000')

962−-- ADLAM CAPITAL LETTER YHE

963−foldMapping '\x1e90e' s = Yield '\x1e930' (CC s '\x0000' '\x0000')

964−-- ADLAM CAPITAL LETTER WAW

965−foldMapping '\x1e90f' s = Yield '\x1e931' (CC s '\x0000' '\x0000')

966−-- ADLAM CAPITAL LETTER NUN

967−foldMapping '\x1e910' s = Yield '\x1e932' (CC s '\x0000' '\x0000')

968−-- ADLAM CAPITAL LETTER KAF

969−foldMapping '\x1e911' s = Yield '\x1e933' (CC s '\x0000' '\x0000')

970−-- ADLAM CAPITAL LETTER YA

971−foldMapping '\x1e912' s = Yield '\x1e934' (CC s '\x0000' '\x0000')

972−-- ADLAM CAPITAL LETTER U

973−foldMapping '\x1e913' s = Yield '\x1e935' (CC s '\x0000' '\x0000')

974−-- ADLAM CAPITAL LETTER JIIM

975−foldMapping '\x1e914' s = Yield '\x1e936' (CC s '\x0000' '\x0000')

976−-- ADLAM CAPITAL LETTER CHI

977−foldMapping '\x1e915' s = Yield '\x1e937' (CC s '\x0000' '\x0000')

978−-- ADLAM CAPITAL LETTER HA

979−foldMapping '\x1e916' s = Yield '\x1e938' (CC s '\x0000' '\x0000')

980−-- ADLAM CAPITAL LETTER QAAF

981−foldMapping '\x1e917' s = Yield '\x1e939' (CC s '\x0000' '\x0000')

982−-- ADLAM CAPITAL LETTER GA

983−foldMapping '\x1e918' s = Yield '\x1e93a' (CC s '\x0000' '\x0000')

984−-- ADLAM CAPITAL LETTER NYA

985−foldMapping '\x1e919' s = Yield '\x1e93b' (CC s '\x0000' '\x0000')

986−-- ADLAM CAPITAL LETTER TU

987−foldMapping '\x1e91a' s = Yield '\x1e93c' (CC s '\x0000' '\x0000')

988−-- ADLAM CAPITAL LETTER NHA

989−foldMapping '\x1e91b' s = Yield '\x1e93d' (CC s '\x0000' '\x0000')

990−-- ADLAM CAPITAL LETTER VA

991−foldMapping '\x1e91c' s = Yield '\x1e93e' (CC s '\x0000' '\x0000')

992−-- ADLAM CAPITAL LETTER KHA

993−foldMapping '\x1e91d' s = Yield '\x1e93f' (CC s '\x0000' '\x0000')

994−-- ADLAM CAPITAL LETTER GBE

995−foldMapping '\x1e91e' s = Yield '\x1e940' (CC s '\x0000' '\x0000')

996−-- ADLAM CAPITAL LETTER ZAL

997−foldMapping '\x1e91f' s = Yield '\x1e941' (CC s '\x0000' '\x0000')

998−-- ADLAM CAPITAL LETTER KPO

999−foldMapping '\x1e920' s = Yield '\x1e942' (CC s '\x0000' '\x0000')

1000−-- ADLAM CAPITAL LETTER SHA

1001−foldMapping '\x1e921' s = Yield '\x1e943' (CC s '\x0000' '\x0000')

1002842foldMapping c s = Yield (toLower c) (CC s '\0' '\0')

Data/Text/Internal/Fusion/Common.hs+37−44

@@ -1,4 +1,4 @@

1−{-# LANGUAGE PatternGuards, BangPatterns, MagicHash, Rank2Types #-}

1+{-# LANGUAGE BangPatterns, MagicHash, Rank2Types #-}

22-- |

33-- Module : Data.Text.Internal.Fusion.Common

44-- Copyright : (c) Bryan O'Sullivan 2009, 2012

@@ -107,7 +107,7 @@ import Prelude (Bool(..), Char, Eq(..), Int, Integral, Maybe(..),

107107import qualified Data.List as L

108108import qualified Prelude as P

109109import Data.Bits (shiftL)

110−import Data.Char (isLetter, isSpace)

110+import Data.Char (isLetter)

111111import Data.Int (Int64)

112112import Data.Text.Internal.Fusion.Types

113113import Data.Text.Internal.Fusion.CaseMapping (foldMapping, lowerMapping, titleMapping,

@@ -117,7 +117,7 @@ import GHC.Prim (Addr#, chr#, indexCharOffAddr#, ord#)

117117import GHC.Types (Char(..), Int(..))

118118

119119singleton :: Char -> Stream Char

120−singleton c = Stream next False (codePointsSize 1)

120+singleton c = Stream next False 1

121121 where next False = Yield c True

122122 next True = Done

123123{-# INLINE [0] singleton #-}

@@ -175,7 +175,7 @@ data C s = C0 !s

175175

176176-- | /O(n)/ Adds a character to the front of a Stream Char.

177177cons :: Char -> Stream Char -> Stream Char

178−cons !w (Stream next0 s0 len) = Stream next (C1 s0) (len + codePointsSize 1)

178+cons !w (Stream next0 s0 len) = Stream next (C1 s0) (len+1)

179179 where

180180 next (C1 s) = Yield w (C0 s)

181181 next (C0 s) = case next0 s of

@@ -189,7 +189,7 @@ data Snoc a = N

189189

190190-- | /O(n)/ Adds a character to the end of a stream.

191191snoc :: Stream Char -> Char -> Stream Char

192−snoc (Stream next0 xs0 len) w = Stream next (J xs0) (len + codePointsSize 1)

192+snoc (Stream next0 xs0 len) w = Stream next (J xs0) (len+1)

193193 where

194194 next (J xs) = case next0 xs of

195195 Done -> Yield w N

@@ -237,7 +237,7 @@ uncons :: Stream Char -> Maybe (Char, Stream Char)

237237uncons (Stream next s0 len) = loop_uncons s0

238238 where

239239 loop_uncons !s = case next s of

240− Yield x s1 -> Just (x, Stream next s1 (len - codePointsSize 1))

240+ Yield x s1 -> Just (x, Stream next s1 (len-1))

241241 Skip s' -> loop_uncons s'

242242 Done -> Nothing

243243{-# INLINE [0] uncons #-}

@@ -260,7 +260,7 @@ last (Stream next s0 _len) = loop0_last s0

260260-- | /O(1)/ Returns all characters after the head of a Stream Char, which must

261261-- be non-empty.

262262tail :: Stream Char -> Stream Char

263−tail (Stream next0 s0 len) = Stream next (C0 s0) (len - codePointsSize 1)

263+tail (Stream next0 s0 len) = Stream next (C0 s0) (len-1)

264264 where

265265 next (C0 s) = case next0 s of

266266 Done -> emptyError "tail"

@@ -278,7 +278,7 @@ data Init s = Init0 !s

278278-- | /O(1)/ Returns all but the last character of a Stream Char, which

279279-- must be non-empty.

280280init :: Stream Char -> Stream Char

281−init (Stream next0 s0 len) = Stream next (Init0 s0) (len - codePointsSize 1)

281+init (Stream next0 s0 len) = Stream next (Init0 s0) (len-1)

282282 where

283283 next (Init0 s) = case next0 s of

284284 Done -> emptyError "init"

@@ -318,14 +318,11 @@ lengthI (Stream next s0 _len) = loop_length 0 s0

318318-- greater than the number or if the stream can't possibly be as long

319319-- as the number supplied, and hence be more efficient.

320320compareLengthI :: Integral a => Stream Char -> a -> Ordering

321−compareLengthI (Stream next s0 len) n

322− -- Note that @len@ tracks code units whereas we want to compare the length

323− -- in code points. Specifically, a stream with hint @len@ may consist of

324− -- anywhere from @len/2@ to @len@ code points.

325− | Just r <- compareSize len n' = r

326− | otherwise = loop_cmp 0 s0

321+compareLengthI (Stream next s0 len) n =

322+ case compareSize len (fromIntegral n) of

323+ Just o -> o

324+ Nothing -> loop_cmp 0 s0

327325 where

328− n' = codePointsSize $ fromIntegral n

329326 loop_cmp !z s = case next s of

330327 Done -> compare z n

331328 Skip s' -> loop_cmp z s'

@@ -371,7 +368,7 @@ data I s = I1 !s

371368-- | /O(n)/ Take a character and place it between each of the

372369-- characters of a 'Stream Char'.

373370intersperse :: Char -> Stream Char -> Stream Char

374−intersperse c (Stream next0 s0 len) = Stream next (I1 s0) (len + unknownSize)

371+intersperse c (Stream next0 s0 len) = Stream next (I1 s0) len

375372 where

376373 next (I1 s) = case next0 s of

377374 Done -> Done

@@ -396,11 +393,9 @@ intersperse c (Stream next0 s0 len) = Stream next (I1 s0) (len + unknownSize)

396393-- functions may map one input character to two or three output

397394-- characters.

398395

399−-- | Map a 'Stream' through the given case-mapping function.

400396caseConvert :: (forall s. Char -> s -> Step (CC s) Char)

401397 -> Stream Char -> Stream Char

402−caseConvert remap (Stream next0 s0 len) =

403− Stream next (CC s0 '\0' '\0') (len `unionSize` 3*len)

398+caseConvert remap (Stream next0 s0 len) = Stream next (CC s0 '\0' '\0') len

404399 where

405400 next (CC s '\0' _) =

406401 case next0 s of

@@ -463,20 +458,19 @@ toLower = caseConvert lowerMapping

463458-- Fox\" is correctly title cased—but this function will

464459-- capitalize /every/ word.

465460toTitle :: Stream Char -> Stream Char

466−toTitle (Stream next0 s0 len) = Stream next (CC (False :*: s0) '\0' '\0') (len + unknownSize)

461+toTitle (Stream next0 s0 len) = Stream next (CC (False :*: s0) '\0' '\0') len

467462 where

468463 next (CC (letter :*: s) '\0' _) =

469464 case next0 s of

470− Done -> Done

471− Skip s' -> Skip (CC (letter :*: s') '\0' '\0')

465+ Done -> Done

466+ Skip s' -> Skip (CC (letter :*: s') '\0' '\0')

472467 Yield c s'

473− | nonSpace -> if letter

474− then lowerMapping c (nonSpace :*: s')

475− else titleMapping c (letter' :*: s')

476− | otherwise -> Yield c (CC (letter' :*: s') '\0' '\0')

477− where nonSpace = P.not (isSpace c)

478− letter' = isLetter c

479− next (CC s a b) = Yield a (CC s b '\0')

468+ | letter' -> if letter

469+ then lowerMapping c (letter' :*: s')

470+ else titleMapping c (letter' :*: s')

471+ | otherwise -> Yield c (CC (letter' :*: s') '\0' '\0')

472+ where letter' = isLetter c

473+ next (CC s a b) = Yield a (CC s b '\0')

480474{-# INLINE [0] toTitle #-}

481475

482476data Justify i s = Just1 !i !s

@@ -484,7 +478,7 @@ data Justify i s = Just1 !i !s

484478

485479justifyLeftI :: Integral a => a -> Char -> Stream Char -> Stream Char

486480justifyLeftI k c (Stream next0 s0 len) =

487− Stream next (Just1 0 s0) (larger (fromIntegral k * charSize c + len) len)

481+ Stream next (Just1 0 s0) (larger (fromIntegral k) len)

488482 where

489483 next (Just1 n s) =

490484 case next0 s of

@@ -704,7 +698,7 @@ replicateI n (Stream next0 s0 len) =

704698-- (a,b), in which case, a is the next Char in the string, and b is

705699-- the seed value for further production.

706700unfoldr :: (a -> Maybe (Char,a)) -> a -> Stream Char

707−unfoldr f s0 = Stream next s0 unknownSize

701+unfoldr f s0 = Stream next s0 1 -- HINT maybe too low

708702 where

709703 {-# INLINE next #-}

710704 next !s = case f s of

@@ -718,7 +712,7 @@ unfoldr f s0 = Stream next s0 unknownSize

718712-- 'unfoldr' when the length of the result is known.

719713unfoldrNI :: Integral a => a -> (b -> Maybe (Char,b)) -> b -> Stream Char

720714unfoldrNI n f s0 | n < 0 = empty

721− | otherwise = Stream next (0 :*: s0) (maxSize $ fromIntegral (n*2))

715+ | otherwise = Stream next (0 :*: s0) (fromIntegral (n*2)) -- HINT maybe too high

722716 where

723717 {-# INLINE next #-}

724718 next (z :*: s) = case f s of

@@ -730,12 +724,12 @@ unfoldrNI n f s0 | n < 0 = empty

730724-------------------------------------------------------------------------------

731725-- * Substreams

732726

733−-- | /O(n)/ @'take' n@, applied to a stream, returns the prefix of the

727+-- | /O(n)/ take n, applied to a stream, returns the prefix of the

734728-- stream of length @n@, or the stream itself if @n@ is greater than the

735729-- length of the stream.

736730take :: Integral a => a -> Stream Char -> Stream Char

737731take n0 (Stream next0 s0 len) =

738− Stream next (n0 :*: s0) (smaller len (codePointsSize $ fromIntegral n0))

732+ Stream next (n0 :*: s0) (smaller len (fromIntegral (max 0 n0)))

739733 where

740734 {-# INLINE next #-}

741735 next (n :*: s) | n <= 0 = Done

@@ -748,12 +742,12 @@ take n0 (Stream next0 s0 len) =

748742data Drop a s = NS !s

749743 | JS !a !s

750744

751−-- | /O(n)/ @'drop' n@, applied to a stream, returns the suffix of the

745+-- | /O(n)/ drop n, applied to a stream, returns the suffix of the

752746-- stream after the first @n@ characters, or the empty stream if @n@

753747-- is greater than the length of the stream.

754748drop :: Integral a => a -> Stream Char -> Stream Char

755749drop n0 (Stream next0 s0 len) =

756− Stream next (JS n0 s0) (len - codePointsSize (fromIntegral n0))

750+ Stream next (JS n0 s0) (len - fromIntegral (max 0 n0))

757751 where

758752 {-# INLINE next #-}

759753 next (JS n s)

@@ -768,10 +762,10 @@ drop n0 (Stream next0 s0 len) =

768762 Yield x s' -> Yield x (NS s')

769763{-# INLINE [0] drop #-}

770764

771−-- | 'takeWhile', applied to a predicate @p@ and a stream, returns the

772−-- longest prefix (possibly empty) of elements that satisfy @p@.

765+-- | takeWhile, applied to a predicate @p@ and a stream, returns the

766+-- longest prefix (possibly empty) of elements that satisfy p.

773767takeWhile :: (Char -> Bool) -> Stream Char -> Stream Char

774−takeWhile p (Stream next0 s0 len) = Stream next s0 (len - unknownSize)

768+takeWhile p (Stream next0 s0 len) = Stream next s0 len -- HINT maybe too high

775769 where

776770 {-# INLINE next #-}

777771 next !s = case next0 s of

@@ -781,9 +775,9 @@ takeWhile p (Stream next0 s0 len) = Stream next s0 (len - unknownSize)

781775 | otherwise -> Done

782776{-# INLINE [0] takeWhile #-}

783777

784−-- | @'dropWhile' p xs@ returns the suffix remaining after @'takeWhile' p xs@.

778+-- | dropWhile @p @xs returns the suffix remaining after takeWhile @p @xs.

785779dropWhile :: (Char -> Bool) -> Stream Char -> Stream Char

786−dropWhile p (Stream next0 s0 len) = Stream next (L s0) (len - unknownSize)

780+dropWhile p (Stream next0 s0 len) = Stream next (L s0) len -- HINT maybe too high

787781 where

788782 {-# INLINE next #-}

789783 next (L s) = case next0 s of

@@ -817,7 +811,7 @@ isPrefixOf (Stream next1 s1 _) (Stream next2 s2 _) = loop (next1 s1) (next2 s2)

817811-------------------------------------------------------------------------------

818812-- ** Searching by equality

819813

820−-- | /O(n)/ 'elem' is the stream membership predicate.

814+-- | /O(n)/ elem is the stream membership predicate.

821815elem :: Char -> Stream Char -> Bool

822816elem w (Stream next s0 _len) = loop_elem s0

823817 where

@@ -862,8 +856,7 @@ indexI (Stream next s0 _len) n0

862856-- returns a stream containing those characters that satisfy the

863857-- predicate.

864858filter :: (Char -> Bool) -> Stream Char -> Stream Char

865−filter p (Stream next0 s0 len) =

866− Stream next s0 (len - unknownSize) -- HINT maybe too high

859+filter p (Stream next0 s0 len) = Stream next s0 len -- HINT maybe too high

867860 where

868861 next !s = case next0 s of

869862 Done -> Done

Data/Text/Internal/Fusion/Size.hs+7−33

@@ -1,4 +1,4 @@

1−{-# LANGUAGE CPP, PatternGuards #-}

1+{-# LANGUAGE CPP #-}

22{-# OPTIONS_GHC -fno-warn-missing-methods #-}

33-- |

44-- Module : Data.Text.Internal.Fusion.Internal

@@ -19,16 +19,11 @@

1919module Data.Text.Internal.Fusion.Size

2020 (

2121 Size

22− -- * Sizes

22+ , exactly

2323 , exactSize

2424 , maxSize

2525 , betweenSize

2626 , unknownSize

27− , unionSize

28− , charSize

29− , codePointsSize

30− -- * Querying sizes

31− , exactly

3227 , smaller

3328 , larger

3429 , upperBound

@@ -37,13 +32,11 @@ module Data.Text.Internal.Fusion.Size

3732 , isEmpty

3833 ) where

3934

40−import Data.Char (ord)

4135import Data.Text.Internal (mul)

4236#if defined(ASSERTS)

4337import Control.Exception (assert)

4438#endif

4539

46−-- | A size in UTF-16 code units.

4740data Size = Between {-# UNPACK #-} !Int {-# UNPACK #-} !Int -- ^ Lower and upper bounds on size.

4841 | Unknown -- ^ Unknown size.

4942 deriving (Eq, Show)

@@ -53,17 +46,6 @@ exactly (Between na nb) | na == nb = Just na

5346exactly _ = Nothing

5447{-# INLINE exactly #-}

5548

56−-- | The 'Size' of the given code point.

57−charSize :: Char -> Size

58−charSize c

59− | ord c < 0x10000 = exactSize 1

60− | otherwise = exactSize 2

61−

62−-- | The 'Size' of @n@ code points.

63−codePointsSize :: Int -> Size

64−codePointsSize n = Between n (2*n)

65−{-# INLINE codePointsSize #-}

66−

6749exactSize :: Int -> Size

6850exactSize n =

6951#if defined(ASSERTS)

@@ -89,10 +71,6 @@ betweenSize m n =

8971 Between m n

9072{-# INLINE betweenSize #-}

9173

92−unionSize :: Size -> Size -> Size

93−unionSize (Between a b) (Between c d) = Between (min a c) (max b d)

94−unionSize _ _ = Unknown

95−

9674unknownSize :: Size

9775unknownSize = Unknown

9876{-# INLINE unknownSize #-}

@@ -162,15 +140,11 @@ lowerBound _ (Between n _) = n

162140lowerBound k _ = k

163141{-# INLINE lowerBound #-}

164142

165−-- | Determine the ordering relationship between two 'Size's, or 'Nothing' in

166−-- the indeterminate case.

167−compareSize :: Size -> Size -> Maybe Ordering

168−compareSize (Between ma mb) (Between na nb)

169− | mb < na = Just LT

170− | ma > nb = Just GT

171− | ma == mb

172− , ma == na

173− , ma == nb = Just EQ

143+compareSize :: Size -> Int -> Maybe Ordering

144+compareSize (Between ma mb) n

145+ | mb < n = Just LT

146+ | ma > n = Just GT

147+ | ma == n && mb == n = Just EQ

174148compareSize _ _ = Nothing

175149

176150

Data/Text/Internal/Fusion/Types.hs+4−4

@@ -75,16 +75,16 @@ instance (Ord a) => Ord (Stream a) where

7575-- unstreaming functions must be able to cope with the hint being too

7676-- small or too large.

7777--

78−-- The size hint tries to track the UTF-16 code units in a stream,

79−-- but often counts the number of code points instead. It can easily

78+-- The size hint tries to track the UTF-16 code points in a stream,

79+-- but often counts the number of characters instead. It can easily

8080-- undercount if, for instance, a transformed stream contains astral

81−-- plane code points (those above 0x10000).

81+-- plane characters (those above 0x10000).

8282

8383data Stream a =

8484 forall s. Stream

8585 (s -> Step s a) -- stepper function

8686 !s -- current state

87− !Size -- size hint in code units

87+ !Size -- size hint

8888

8989-- | /O(n)/ Determines if two streams are equal.

9090eq :: (Eq a) => Stream a -> Stream a -> Bool

Data/Text/Lazy.hs+20−54

@@ -13,6 +13,7 @@

1313--

1414-- License : BSD-style

1515-- Maintainer : bos@serpentine.com

16+-- Stability : experimental

1617-- Portability : GHC

1718--

1819-- A time and space-efficient implementation of Unicode text using

@@ -67,7 +68,6 @@ module Data.Text.Lazy

6768 , snoc

6869 , append

6970 , uncons

70− , unsnoc

7171 , head

7272 , last

7373 , tail

@@ -273,16 +273,14 @@ import Text.Printf (PrintfArg, formatArg, formatString)

273273-- $replacement

274274--

275275-- A 'Text' value is a sequence of Unicode scalar values, as defined

276−-- in

277−-- <http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf#page=35 §3.9, definition D76 of the Unicode 5.2 standard >.

278−-- As such, a 'Text' cannot contain values in the range U+D800 to

279−-- U+DFFF inclusive. Haskell implementations admit all Unicode code

280−-- points

281−-- (<http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf#page=13 §3.4, definition D10 >)

282−-- as 'Char' values, including code points from this invalid range.

283−-- This means that there are some 'Char' values that are not valid

284−-- Unicode scalar values, and the functions in this module must handle

285−-- those cases.

276+-- in §3.9, definition D76 of the Unicode 5.2 standard:

277+-- <http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf#page=35>. As

278+-- such, a 'Text' cannot contain values in the range U+D800 to U+DFFF

279+-- inclusive. Haskell implementations admit all Unicode code points

280+-- (§3.4, definition D10) as 'Char' values, including code points

281+-- from this invalid range. This means that there are some 'Char'

282+-- values that are not valid Unicode scalar values, and the functions

283+-- in this module must handle those cases.

286284--

287285-- Within this module, many functions construct a 'Text' from one or

288286-- more 'Char' values. Those functions will substitute 'Char' values

@@ -296,8 +294,8 @@ import Text.Printf (PrintfArg, formatArg, formatString)

296294-- range U+D800 through U+DFFF are used by UTF-16 to denote surrogate

297295-- code points, and so cannot be represented. The functions replace

298296-- invalid scalar values, instead of dropping them, as a security

299−-- measure. For details, see

300−-- <http://unicode.org/reports/tr36/#Deletion_of_Noncharacters Unicode Technical Report 36, §3.5 >.)

297+-- measure. For details, see Unicode Technical Report 36, §3.5:

298+-- <http://unicode.org/reports/tr36#Deletion_of_Noncharacters>)

301299

302300equal :: Text -> Text -> Bool

303301equal Empty Empty = True

@@ -344,13 +342,9 @@ instance Read Text where

344342 readsPrec p str = [(pack x,y) | (x,y) <- readsPrec p str]

345343

346344#if MIN_VERSION_base(4,9,0)

347−-- | Non-orphan 'Semigroup' instance only defined for

348−-- @base-4.9.0.0@ and later; orphan instances for older GHCs are

349−-- provided by

350−-- the [semigroups](http://hackage.haskell.org/package/semigroups)

351−-- package

352−--

353−-- @since 1.2.2.0

345+-- Semigroup orphan instances for older GHCs are provided by

346+-- 'semigroups` package

347+

354348instance Semigroup Text where

355349 (<>) = append

356350#endif

@@ -368,7 +362,6 @@ instance IsString Text where

368362 fromString = pack

369363

370364#if __GLASGOW_HASKELL__ >= 708

371−-- | @since 1.2.0.0

372365instance Exts.IsList Text where

373366 type Item Text = Char

374367 fromList = pack

@@ -381,7 +374,6 @@ instance NFData Text where

381374 rnf (Chunk _ ts) = rnf ts

382375#endif

383376

384−-- | @since 1.2.1.0

385377instance Binary Text where

386378 put t = put (encodeUtf8 t)

387379 get = do

@@ -405,8 +397,6 @@ instance Data Text where

405397

406398#if MIN_VERSION_base(4,7,0)

407399-- | Only defined for @base-4.7.0.0@ and later

408−--

409−-- @since 1.2.2.0

410400instance PrintfArg Text where

411401 formatArg txt = formatString $ unpack txt

412402#endif

@@ -485,7 +475,9 @@ fromStrict t = chunk t Empty

485475-- -----------------------------------------------------------------------------

486476-- * Basic functions

487477

488−-- | /O(1)/ Adds a character to the front of a 'Text'. Subject to fusion.

478+-- | /O(n)/ Adds a character to the front of a 'Text'. This function

479+-- is more costly than its 'List' counterpart because it requires

480+-- copying a new array. Subject to fusion.

489481cons :: Char -> Text -> Text

490482cons c t = Chunk (T.singleton c) t

491483{-# INLINE [1] cons #-}

@@ -569,17 +561,6 @@ init Empty = emptyError "init"

569561 unstream (S.init (stream t)) = init t

570562 #-}

571563

572−-- | /O(n\/c)/ Returns the 'init' and 'last' of a 'Text', or 'Nothing' if

573−-- empty.

574−--

575−-- * It is no faster than using 'init' and 'last'.

576−--

577−-- @since 1.2.3.0

578−unsnoc :: Text -> Maybe (Text, Char)

579−unsnoc Empty = Nothing

580−unsnoc ts@(Chunk _ _) = Just (init ts, last ts)

581−{-# INLINE unsnoc #-}

582−

583564-- | /O(1)/ Tests whether a 'Text' is empty or not. Subject to

584565-- fusion.

585566null :: Text -> Bool

@@ -839,8 +820,6 @@ toUpper t = unstream (S.toUpper (stream t))

839820-- guides disagree on whether the book name \"The Hill of the Red

840821-- Fox\" is correctly title cased—but this function will

841822-- capitalize /every/ word.

842−--

843−-- @since 1.0.0.0

844823toTitle :: Text -> Text

845824toTitle t = unstream (S.toTitle (stream t))

846825{-# INLINE toTitle #-}

@@ -993,8 +972,6 @@ mapAccumR f = go

993972

994973-- | @'repeat' x@ is an infinite 'Text', with @x@ the value of every

995974-- element.

996−--

997−-- @since 1.2.0.5

998975repeat :: Char -> Text

999976repeat c = let t = Chunk (T.replicate smallChunkSize (T.singleton c)) t

1000977 in t

@@ -1012,8 +989,6 @@ replicate n t

1012989

1013990-- | 'cycle' ties a finite, non-empty 'Text' into a circular one, or

1014991-- equivalently, the infinite repetition of the original 'Text'.

1015−--

1016−-- @since 1.2.0.5

1017992cycle :: Text -> Text

1018993cycle Empty = emptyError "cycle"

1019994cycle t = let t' = foldrChunks Chunk t' t

@@ -1023,8 +998,6 @@ cycle t = let t' = foldrChunks Chunk t' t

1023998-- of @f@ to @x@:

1024999--

10251000-- > iterate f x == [x, f x, f (f x), ...]

1026−--

1027−-- @since 1.2.0.5

10281001iterate :: (Char -> Char) -> Char -> Text

10291002iterate f c = let t c' = Chunk (T.singleton c') (t (f c'))

10301003 in t c

@@ -1088,8 +1061,6 @@ take i t0 = take' i t0

10881061-- Examples:

10891062--

10901063-- > takeEnd 3 "foobar" == "bar"

1091−--

1092−-- @since 1.1.1.0

10931064takeEnd :: Int64 -> Text -> Text

10941065takeEnd n t0

10951066 | n <= 0 = empty

@@ -1128,8 +1099,6 @@ drop i t0

11281099-- Examples:

11291100--

11301101-- > dropEnd 3 "foobar" == "foo"

1131−--

1132−-- @since 1.1.1.0

11331102dropEnd :: Int64 -> Text -> Text

11341103dropEnd n t0

11351104 | n <= 0 = t0

@@ -1181,15 +1150,12 @@ takeWhile p t0 = takeWhile' t0

11811150-- Examples:

11821151--

11831152-- > takeWhileEnd (=='o') "foo" == "oo"

1184−--

1185−-- @since 1.2.2.0

11861153takeWhileEnd :: (Char -> Bool) -> Text -> Text

11871154takeWhileEnd p = takeChunk empty . L.reverse . toChunks

11881155 where takeChunk acc [] = acc

1189− takeChunk acc (t:ts)

1190− | T.lengthWord16 t' < T.lengthWord16 t

1191− = chunk t' acc

1192− | otherwise = takeChunk (Chunk t' acc) ts

1156+ takeChunk acc (t:ts) = if T.length t' < T.length t

1157+ then (Chunk t' acc)

1158+ else takeChunk (Chunk t' acc) ts

11931159 where t' = T.takeWhileEnd p t

11941160{-# INLINE takeWhileEnd #-}

11951161

Data/Text/Lazy/Builder.hs+1

@@ -11,6 +11,7 @@

1111-- License : BSD-style (see LICENSE)

1212--

1313-- Maintainer : Johan Tibell <johan.tibell@gmail.com>

14+-- Stability : experimental

1415-- Portability : portable to Hugs and GHC

1516--

1617-- Efficient construction of lazy @Text@ values. The principal

Data/Text/Lazy/Builder/Int.hs+1−3

@@ -9,6 +9,7 @@

1010-- License: BSD-style

1111-- Maintainer: Bryan O'Sullivan <bos@serpentine.com>

12+-- Stability: experimental

1213-- Portability: portable

1314--

1415-- Efficiently write an integral value to a 'Builder'.

@@ -31,9 +32,6 @@ import GHC.Base (quotInt, remInt)

3132import GHC.Num (quotRemInteger)

3233import GHC.Types (Int(..))

3334import Control.Monad.ST

34−#if MIN_VERSION_base(4,11,0)

35−import Prelude hiding ((<>))

36−#endif

3735

3836#ifdef __GLASGOW_HASKELL__

3937# if defined(INTEGER_GMP)

Data/Text/Lazy/Builder/RealFloat.hs−3

@@ -24,9 +24,6 @@ import Data.Text.Lazy.Builder.Int (decimal)

2424import Data.Text.Internal.Builder.RealFloat.Functions (roundTo)

2525import Data.Text.Lazy.Builder

2626import qualified Data.Text as T

27−#if MIN_VERSION_base(4,11,0)

28−import Prelude hiding ((<>))

29−#endif

3027

3128-- | Control the rendering of floating point numbers.

3229data FPFormat = Exponent

Data/Text/Lazy/Encoding.hs+3−13

@@ -8,13 +8,14 @@

88--

99-- License : BSD-style

1010-- Maintainer : bos@serpentine.com

11+-- Stability : experimental

1112-- Portability : portable

1213--

1314-- Functions for converting lazy 'Text' values to and from lazy

1415-- 'ByteString', using several standard encodings.

1516--

16−-- To gain access to a much larger family of encodings, use the

17−-- <http://hackage.haskell.org/package/text-icu text-icu package>.

17+-- To gain access to a much larger variety of encodings, use the

18+-- @text-icu@ package: <http://hackage.haskell.org/package/text-icu>

1819

1920module Data.Text.Lazy.Encoding

2021 (

@@ -140,7 +141,6 @@ decodeUtf8' bs = unsafeDupablePerformIO $ do

140141 rnf (Chunk _ ts) = rnf ts

141142{-# INLINE decodeUtf8' #-}

142143

143−-- | Encode text using UTF-8 encoding.

144144encodeUtf8 :: Text -> B.ByteString

145145encodeUtf8 Empty = B.empty

146146encodeUtf8 lt@(Chunk t _) =

@@ -154,20 +154,10 @@ encodeUtf8 lt@(Chunk t _) =

154154 firstChunkSize = min B.smallChunkSize (4 * (T.length t + 1))

155155 strategy = B.safeStrategy firstChunkSize B.defaultChunkSize

156156

157−-- | Encode text to a ByteString 'B.Builder' using UTF-8 encoding.

158−--

159−-- @since 1.1.0.0

160157encodeUtf8Builder :: Text -> B.Builder

161158encodeUtf8Builder =

162159 foldrChunks (\c b -> TE.encodeUtf8Builder c `mappend` b) Data.Monoid.mempty

163160

164−-- | Encode text using UTF-8 encoding and escape the ASCII characters using

165−-- a 'BP.BoundedPrim'.

166−--

167−-- Use this function is to implement efficient encoders for text-based formats

168−-- like JSON or HTML.

169−--

170−-- @since 1.1.0.0

171161{-# INLINE encodeUtf8BuilderEscaped #-}

172162encodeUtf8BuilderEscaped :: BP.BoundedPrim Word8 -> Text -> B.Builder

173163encodeUtf8BuilderEscaped prim =

Data/Text/Lazy/IO.hs+1

@@ -8,6 +8,7 @@

99-- License : BSD-style

1010-- Maintainer : bos@serpentine.com

11+-- Stability : experimental

1112-- Portability : GHC

1213--

1314-- Efficient locale-sensitive support for lazy text I\/O.

Data/Text/Lazy/Read.hs+4−3

@@ -11,6 +11,7 @@

1111--

1212-- License : BSD-style

1313-- Maintainer : bos@serpentine.com

14+-- Stability : experimental

1415-- Portability : GHC

1516--

1617-- Functions used frequently when reading textual data.

@@ -67,9 +68,9 @@ decimal txt

6768 go n d = (n * 10 + fromIntegral (digitToInt d))

6869

6970-- | Read a hexadecimal integer, consisting of an optional leading

70−-- @\"0x\"@ followed by at least one hexadecimal digit. Input is

71−-- consumed until a non-hex-digit or end of string is reached.

72−-- This function is case insensitive.

71+-- @\"0x\"@ followed by at least one decimal digit. Input is consumed

72+-- until a non-hex-digit or end of string is reached. This function

73+-- is case insensitive.

7374--

7475-- This function does not handle leading sign characters. If you need

7576-- to handle signed input, use @'signed' 'hexadecimal'@.

Data/Text/Read.hs+4−3

@@ -9,6 +9,7 @@

99--

1010-- License : BSD-style

1111-- Maintainer : bos@serpentine.com

12+-- Stability : experimental

1213-- Portability : GHC

1314--

1415-- Functions used frequently when reading textual data.

@@ -66,9 +67,9 @@ decimal txt

6667 go n d = (n * 10 + fromIntegral (digitToInt d))

6768

6869-- | Read a hexadecimal integer, consisting of an optional leading

69−-- @\"0x\"@ followed by at least one hexadecimal digit. Input is

70−-- consumed until a non-hex-digit or end of string is reached.

71−-- This function is case insensitive.

70+-- @\"0x\"@ followed by at least one decimal digit. Input is consumed

71+-- until a non-hex-digit or end of string is reached. This function

72+-- is case insensitive.

7273--

7374-- This function does not handle leading sign characters. If you need

7475-- to handle signed input, use @'signed' 'hexadecimal'@.

Data/Text/Show.hs−2

@@ -46,8 +46,6 @@ unpack = S.unstreamList . stream

4646-- fusion.

4747--

4848-- This is exposed solely for people writing GHC rewrite rules.

49−--

50−-- @since 1.2.1.1

5149unpackCString# :: Addr# -> Text

5250unpackCString# addr# = unstream (S.streamCString# addr#)

5351{-# NOINLINE unpackCString# #-}

Data/Text/Unsafe.hs+1−2

@@ -4,6 +4,7 @@

55-- License : BSD-style

66-- Maintainer : bos@serpentine.com

7+-- Stability : experimental

78-- Portability : portable

89--

910-- A module containing unsafe 'Text' operations, for very very careful

@@ -97,8 +98,6 @@ reverseIter (Text arr off _len) i

9798-- | /O(1)/ Iterate one step backwards through a UTF-16 array,

9899-- returning the delta to add (i.e. a negative number) to give the

99100-- next offset to iterate at.

100−--

101−-- @since 1.1.1.0

102101reverseIter_ :: Text -> Int -> Int

103102reverseIter_ (Text arr off _len) i

104103 | m < 0xDC00 || m > 0xDFFF = -1

README.markdown+16−5

@@ -1,18 +1,29 @@

1−# `text`: Fast, packed Unicode strings, using stream fusion

1+# Text: Fast, packed Unicode strings, using stream fusion

33This package provides the Data.Text library, a library for the space-

44and time-efficient manipulation of Unicode text in Haskell.

6−**Please refer to the [package description on Hackage](https://hackage.haskell.org/package/text#description) for more information.**

7+# Normalization, conversion, and collation, oh my!

9+This library intentionally provides a simple API based on the

10+Haskell prelude's list manipulation functions. For more complicated

11+real-world tasks, such as Unicode normalization, conversion to and

12+from a larger variety of encodings, and collation, use the [text-icu

13+package](http://hackage.haskell.org/cgi-bin/hackage-scripts/package/text-icu).

14+

15+That library uses the well-respected and liberally licensed ICU

16+library to provide these facilities.

17+

718

819# Get involved!

920

1021Please report bugs via the

11−[github issue tracker](https://github.com/haskell/text/issues).

22+[github issue tracker](https://github.com/bos/text/issues).

1223

13−Master [git repository](https://github.com/haskell/text):

24+Master [git repository](https://github.com/bos/text):

1425

15−* `git clone git://github.com/haskell/text.git`

26+* `git clone git://github.com/bos/text.git`

1627

1728There's also a [Mercurial mirror](https://bitbucket.org/bos/text):

1829

benchmarks/.gitignore+1−4

@@ -1,4 +1 @@

1−/dist/

2−/dist-newstyle/

3−/.ghc.environment.*

4−/cabal.project.local

1+dist

benchmarks/cabal.projectdeleted−2

@@ -1,2 +0,0 @@

1−-- NB: we use a separate project

2−packages: .

benchmarks/haskell/Benchmarks.hs−2

@@ -10,7 +10,6 @@ import System.FilePath ((</>))

1010import System.IO (IOMode (WriteMode), openFile, hSetEncoding, utf8)

1111

1212import qualified Benchmarks.Builder as Builder

13−import qualified Benchmarks.Concat as Concat

1413import qualified Benchmarks.DecodeUtf8 as DecodeUtf8

1514import qualified Benchmarks.EncodeUtf8 as EncodeUtf8

1615import qualified Benchmarks.Equality as Equality

@@ -42,7 +41,6 @@ benchmarks = do

4241 -- Traditional benchmarks

4342 bs <- sequence

4443 [ Builder.benchmark

45− , Concat.benchmark

4644 , DecodeUtf8.benchmark "html" (tf "libya-chinese.html")

4745 , DecodeUtf8.benchmark "xml" (tf "yiwiki.xml")

4846 , DecodeUtf8.benchmark "ascii" (tf "ascii.txt")

benchmarks/haskell/Benchmarks/Concat.hsdeleted−25

@@ -1,25 +0,0 @@

1−{-# LANGUAGE OverloadedStrings #-}

2−

3−module Benchmarks.Concat (benchmark) where

4−

5−import Control.Monad.Trans.Writer

6−import Criterion (Benchmark, bgroup, bench, whnf)

7−import Data.Text as T

8−

9−benchmark :: IO Benchmark

10−benchmark = return $ bgroup "Concat"

11− [ bench "append" $ whnf (append4 "Text 1" "Text 2" "Text 3") "Text 4"

12− , bench "concat" $ whnf (concat4 "Text 1" "Text 2" "Text 3") "Text 4"

13− , bench "write" $ whnf (write4 "Text 1" "Text 2" "Text 3") "Text 4"

14− ]

15−

16−append4, concat4, write4 :: Text -> Text -> Text -> Text -> Text

17−

18−{-# NOINLINE append4 #-}

19−append4 x1 x2 x3 x4 = x1 `append` x2 `append` x3 `append` x4

20−

21−{-# NOINLINE concat4 #-}

22−concat4 x1 x2 x3 x4 = T.concat [x1, x2, x3, x4]

23−

24−{-# NOINLINE write4 #-}

25−write4 x1 x2 x3 x4 = execWriter $ tell x1 >> tell x2 >> tell x3 >> tell x4

benchmarks/text-benchmarks.cabal+8−82

@@ -12,7 +12,8 @@ author: Jasper Van der Jeugt <jaspervdj@gmail.com>,

1212maintainer: jaspervdj@gmail.com

1313category: Text

1414build-type: Simple

15−cabal-version: >=1.8

15+

16+cabal-version: >=1.2

1617

1718flag bytestring-builder

1819 description: Depend on the bytestring-builder package for backwards compatibility.

@@ -25,12 +26,16 @@ flag llvm

2526 manual: True

2627

2728executable text-benchmarks

29+ hs-source-dirs: haskell ..

30+ c-sources: ../cbits/cbits.c

31+ cbits/time_iconv.c

32+ include-dirs: ../include

33+ main-is: Benchmarks.hs

2834 ghc-options: -Wall -O2 -rtsopts

2935 if flag(llvm)

3036 ghc-options: -fllvm

3137 cpp-options: -DHAVE_DEEPSEQ -DINTEGER_GMP

32− build-depends: array,

33− base == 4.*,

38+ build-depends: base == 4.*,

3439 binary,

3540 blaze-builder,

3641 bytestring-lexing >= 0.5.0,

@@ -42,7 +47,6 @@ executable text-benchmarks

4247 ghc-prim,

4348 integer-gmp,

4449 stringsearch,

45− transformers,

4650 utf8-string,

4751 vector

4852

@@ -52,87 +56,9 @@ executable text-benchmarks

5256 else

5357 build-depends: bytestring >= 0.10.4

5458

55− -- modules for benchmark proper

56− c-sources: cbits/time_iconv.c

57− hs-source-dirs: haskell

58− main-is: Benchmarks.hs

59− other-modules:

60− Benchmarks.Builder

61− Benchmarks.Concat

62− Benchmarks.DecodeUtf8

63− Benchmarks.EncodeUtf8

64− Benchmarks.Equality

65− Benchmarks.FileRead

66− Benchmarks.FoldLines

67− Benchmarks.Mul

68− Benchmarks.Programs.BigTable

69− Benchmarks.Programs.Cut

70− Benchmarks.Programs.Fold

71− Benchmarks.Programs.Sort

72− Benchmarks.Programs.StripTags

73− Benchmarks.Programs.Throughput

74− Benchmarks.Pure

75− Benchmarks.ReadNumbers

76− Benchmarks.Replace

77− Benchmarks.Search

78− Benchmarks.Stream

79− Benchmarks.WordFrequencies

80−

81− -- Source code for IUT (implementation under test)

82− -- "borrowed" from parent folder

83− include-dirs: ../include

84− c-sources: ../cbits/cbits.c

85− hs-source-dirs: ..

86− other-modules:

87− Data.Text

88− Data.Text.Array

89− Data.Text.Encoding

90− Data.Text.Encoding.Error

91− Data.Text.Foreign

92− Data.Text.IO

93− Data.Text.Internal

94− Data.Text.Internal.Builder

95− Data.Text.Internal.Builder.Functions

96− Data.Text.Internal.Builder.Int.Digits

97− Data.Text.Internal.Builder.RealFloat.Functions

98− Data.Text.Internal.Encoding.Fusion

99− Data.Text.Internal.Encoding.Fusion.Common

100− Data.Text.Internal.Encoding.Utf16

101− Data.Text.Internal.Encoding.Utf32

102− Data.Text.Internal.Encoding.Utf8

103− Data.Text.Internal.Functions

104− Data.Text.Internal.Fusion

105− Data.Text.Internal.Fusion.CaseMapping

106− Data.Text.Internal.Fusion.Common

107− Data.Text.Internal.Fusion.Size

108− Data.Text.Internal.Fusion.Types

109− Data.Text.Internal.IO

110− Data.Text.Internal.Lazy

111− Data.Text.Internal.Lazy.Encoding.Fusion

112− Data.Text.Internal.Lazy.Fusion

113− Data.Text.Internal.Lazy.Search

114− Data.Text.Internal.Private

115− Data.Text.Internal.Read

116− Data.Text.Internal.Search

117− Data.Text.Internal.Unsafe

118− Data.Text.Internal.Unsafe.Char

119− Data.Text.Internal.Unsafe.Shift

120− Data.Text.Lazy

121− Data.Text.Lazy.Builder

122− Data.Text.Lazy.Builder.Int

123− Data.Text.Lazy.Builder.RealFloat

124− Data.Text.Lazy.Encoding

125− Data.Text.Lazy.IO

126− Data.Text.Lazy.Internal

127− Data.Text.Lazy.Read

128− Data.Text.Read

129− Data.Text.Unsafe

130− Data.Text.Show

131−

13259executable text-multilang

13360 hs-source-dirs: haskell

13461 main-is: Multilang.hs

135− other-modules: Timer

13662 ghc-options: -Wall -O2

13763 build-depends: base == 4.*,

13864 bytestring,

cabal.projectdeleted−3

@@ -1,3 +0,0 @@

1−-- See http://cabal.readthedocs.io/en/latest/nix-local-build-overview.html

2−packages: .

3−tests: True

cbits/cbits.c+11−6

@@ -206,9 +206,11 @@ _hs_text_decode_utf8_state(uint16_t *const dest, size_t *destoff,

206206 const uint8_t *srcend,

207207 uint32_t *codepoint0, uint32_t *state0)

208208{

209− _hs_text_decode_utf8_int(dest, destoff, src, srcend, codepoint0, state0);

210−

211− return *src;

209+ uint8_t const *ret = _hs_text_decode_utf8_int(dest, destoff, src, srcend,

210+ codepoint0, state0);

211+ if (*state0 == UTF8_REJECT)

212+ ret -=1;

213+ return ret;

212214}

213215

214216/*

@@ -220,9 +222,12 @@ _hs_text_decode_utf8(uint16_t *const dest, size_t *destoff,

220222{

221223 uint32_t codepoint;

222224 uint32_t state = UTF8_ACCEPT;

223− _hs_text_decode_utf8_int(dest, destoff, &src, srcend,

224− &codepoint, &state);

225− return src;

225+ uint8_t const *ret = _hs_text_decode_utf8_int(dest, destoff, &src, srcend,

226+ &codepoint, &state);

227+ /* Back up if we have an incomplete or invalid encoding */

228+ if (state != UTF8_ACCEPT)

229+ ret -= 1;

230+ return ret;

226231}

227232

228233void

changelog.md+19−56

@@ -1,52 +1,15 @@

1−### 1.2.3.0

2−

3−* Spec compliance: `toCaseFold` now follows the Unicode 9.0 spec

4− (updated from 8.0).

5−

6−* Bug fix: the lazy `takeWhileEnd` function violated the

7− [lazy text invariant](https://github.com/bos/text/blob/1.2.3.0/Data/Text/Internal/Lazy.hs#L51)

8− (gh-184).

9−

10−* Bug fix: Fixed usage of size hints causing incorrect behavior (gh-197).

11−

12−* New function: `unsnoc` (gh-173).

13−

14−* Reduce memory overhead in `encodeUTF8` (gh-194).

15−

16−* Improve UTF-8 decoder error-recovery (gh-182).

17−

18−* Minor documentation improvements (`@since` annotations, more

19− examples, clarifications).

20−

21−#### 1.2.2.2

22−

23−* The `toTitle` function now correctly handles letters that

24− immediately follow punctuation. Before, `"there's"` would turn into

25− `"There'S"`. Now, it becomes `"There's"`.

26−

27−* The implementation of unstreaming is faster, resulting in operations

28− such as `map` and `intersperse` speeding up by up to 30%, with

29− smaller code generated.

30−

31−* The optimised length comparison function is now more likely to be

32− used after some rewrite rule tweaking.

33−

34−* Bug fix: an off-by-one bug in `takeEnd` is fixed.

35−

36−* Bug fix: a logic error in `takeWord16` is fixed.

37−

38−#### 1.2.2.1

1+1.2.2.1

392

403* The switch to `integer-pure` in 1.2.2.0 was apparently mistaken.

414 The build flag has been renamed accordingly. Your army of diligent

425 maintainers apologizes for the churn.

436

44−* Spec compliance: `toCaseFold` now follows the Unicode 8.0 spec

7+* Spec compliance: toCaseFold now follows the Unicode 8.0 spec

458 (updated from 7.0)

469

4710* An STG lint error has been fixed

4811

49−### 1.2.2.0

12+1.2.2.0

5013

5114* The `integer-simple` package, upon which this package optionally

5215 depended, has been replaced with `integer-pure`. The build flag has

@@ -62,7 +25,7 @@

6225 * if `base` >= 4.7: `PrintfArg`

6326 * if `base` >= 4.9: `Semigroup`

6427

65−#### 1.2.1.3

28+1.2.1.3

6629

6730* Bug fix: As it turns out, moving the literal rewrite rules to simplifier

6831 phase 2 does not prevent competition with the `unpack` rule, which is

@@ -70,7 +33,7 @@

7033 test environment mistake. Moving literal rules back to phase 1 finally

7134 fixes GHC Trac #10528 correctly.

7235

73−#### 1.2.1.2

36+1.2.1.2

7437

7538* Bug fix: Run literal rewrite rules in simplifier phase 2.

7639 The behavior of the simplifier changed in GHC 7.10.2,

@@ -78,21 +41,21 @@

7841 and long compilation times. See

7942 [GHC Trac #10528](https://ghc.haskell.org/trac/ghc/ticket/10528).

8043

81−#### 1.2.1.1

44+1.2.1.1

8245

8346* Expose unpackCString#, which you should never use.

8447

85−### 1.2.1.0

48+1.2.1.0

8649

8750* Added Binary instances for both Text types. (If you have previously

8851 been using the text-binary package to get a Binary instance, it is

8952 now obsolete.)

9053

91−#### 1.2.0.6

54+1.2.0.6

9255

9356* Fixed a space leak in UTF-8 decoding

9457

95−#### 1.2.0.5

58+1.2.0.5

9659

9760* Feature parity: repeat, cycle, iterate are now implemented for lazy

9861 Text, and the Data instance is more complete

@@ -112,26 +75,26 @@

11275* Spec compliance: toCaseFold now follows the Unicode 7.0 spec

11376 (updated from 6.3)

11477

115−#### 1.2.0.4

78+1.2.0.4

11679

11780* Fixed an incompatibility with base < 4.5

11881

119−#### 1.2.0.3

82+1.2.0.3

12083

12184* Update formatRealFloat to correspond to the definition in versions

12285 of base newer than 4.5 (https://github.com/bos/text/issues/105)

12386

124−#### 1.2.0.2

87+1.2.0.2

12588

12689* Bumped lower bound on deepseq to 1.4 for compatibility with the

12790 upcoming GHC 7.10

12891

129−#### 1.2.0.1

92+1.2.0.1

13093

13194* Fixed a buffer overflow in rendering of large Integers

13295 (https://github.com/bos/text/issues/99)

13396

134−## 1.2.0.0

97+1.2.0.0

13598

13699* Fixed an integer overflow in the replace function

137100 (https://github.com/bos/text/issues/81)

@@ -144,7 +107,7 @@

144107

145108* Added an instance of IsList for GHC 7.8 and above

146109

147−### 1.1.1.0

110+1.1.1.0

148111

149112* The Data.Data instance now allows gunfold to work, via a virtual

150113 pack constructor

@@ -154,12 +117,12 @@

154117* Comparing the length of a Text against a number can now

155118 short-circuit in more cases

156119

157−#### 1.1.0.1

120+1.1.0.1

158121

159122* streamDecodeUtf8: fixed gh-70, did not return all unconsumed bytes

160123 in single-byte chunks

161124

162−## 1.1.0.0

125+1.1.0.0

163126

164127* encodeUtf8: Performance is improved by up to 4x.

165128

@@ -178,12 +141,12 @@

178141 use at your own risk - there are no API stability guarantees for

179142 internal modules!

180143

181−#### 1.0.0.1

144+1.0.0.1

182145

183146* decodeUtf8: Fixed a regression that caused us to incorrectly

184147 identify truncated UTF-8 as valid (gh-61)

185148

186−# 1.0.0.0

149+1.0.0.0

187150

188151* Added support for Unicode 6.3.0 to case conversion functions

189152

tests-and-benchmarks.markdown+8−13

@@ -13,13 +13,10 @@ you can clone from one of the following locations:

1313* Git mirror repository:

1414 [github.com/bos/text-test-data](https://github.com/bos/text-test-data)

1515

16−You can clone either repository into the `tests` subdirectory using

17−

18− cd tests/

19− make text-test-data # to clone from mercurial, OR

20− make VCS=git text-test-data # to clone from git

21−

22−Many tests and benchmarks will fail if the test files are missing.

16+You should clone that repository into the `tests` subdirectory (your

17+clone must be named `text-test-data` locally), then run `make -C

18+tests/text-test-data` to uncompress the test files. Many tests and

19+benchmarks will fail if the test files are missing.

2320

2421Functional tests

2522----------------

@@ -57,12 +54,10 @@ To compile the benchmarks, navigate to the `benchmarks` subdirectory and run

5754

5855 ./dist/build/text-benchmarks/text-benchmarks

5956

60−Or if you have a recent enough `cabal`, you can build and run the

61−benchmarks via

57+However, since there's quite a lot of benchmarks, you usually don't want to

58+run them all. Instead, use the `-l` flag to get a list of benchmarks:

6259

63− cabal new-run exe:text-benchmarks -- --help

60+ ./dist/build/text-benchmarks/text-benchmarks

6461

65−However, since there's quite a lot of benchmarks, you usually don't want to

66−run them all. Instead, use the `-l` flag to get a list of benchmarks

67−and run the ones you want to inspect. If you want to configure the benchmarks

62+And run the ones you want to inspect. If you want to configure the benchmarks

6863further, the exact parameters can be changed in `Benchmarks.hs`.

tests/Makefile+1−6

@@ -1,4 +1,3 @@

1−VCS = hg

21count = 1000

43all: coverage literal-rule-test

@@ -13,11 +12,7 @@ build: text-test-data

1312 cabal build

1413

1514text-test-data:

16−ifeq ($(VCS),git)

17− git clone https://github.com/bos/text-test-data.git

18−else

1915 hg clone https://bitbucket.org/bos/text-test-data

20−endif

2116 $(MAKE) -C text-test-data

2217

2318coverage/text-tests.tix:

@@ -42,4 +37,4 @@ coverage/hpc_index.html: coverage/coverage.tix

4237clean:

4338 rm -rf dist coverage .hpc

4439

45−.PHONY: all build clean coverage literal-rule-test

40+.PHONY: build coverage all literal-rule-test

tests/Tests/Properties.hs+21−94

@@ -9,9 +9,9 @@ module Tests.Properties

99 ) where

1010

1111import Control.Applicative ((<$>), (<*>))

12−import Control.Arrow ((***), first, second)

12+import Control.Arrow ((***), second)

1313import Data.Bits ((.&.))

14−import Data.Char (chr, isDigit, isHexDigit, isLower, isSpace, isLetter, isUpper, ord)

14+import Data.Char (chr, isDigit, isHexDigit, isLower, isSpace, isUpper, ord)

1515import Data.Int (Int8, Int16, Int32, Int64)

1616import Data.Monoid (Monoid(..))

1717import Data.String (IsString(fromString))

@@ -23,7 +23,6 @@ import Data.Text.Internal.Search (indices)

2323import Data.Text.Lazy.Read as TL

2424import Data.Text.Read as T

2525import Data.Word (Word, Word8, Word16, Word32, Word64)

26−import Data.Maybe (mapMaybe)

2726import Numeric (showEFloat, showFFloat, showGFloat, showHex)

2827import Prelude hiding (replicate)

2928import Test.Framework (Test, testGroup)

@@ -31,7 +30,6 @@ import Test.Framework.Providers.QuickCheck2 (testProperty)

3130import Test.QuickCheck hiding ((.&.))

3231import Test.QuickCheck.Monadic

3332import Test.QuickCheck.Property (Property(..))

34−import Test.QuickCheck.Unicode (char)

3533import Tests.QuickCheckUtils

3634import Tests.Utils

3735import Text.Show.Functions ()

@@ -39,7 +37,6 @@ import qualified Control.Exception as Exception

3937import qualified Data.Bits as Bits (shiftL, shiftR)

4038import qualified Data.ByteString as B

4139import qualified Data.ByteString.Lazy as BL

42−import qualified Data.Char as C

4340import qualified Data.List as L

4441import qualified Data.Text as T

4542import qualified Data.Text.Encoding as E

@@ -63,10 +60,10 @@ t_pack_unpack = (T.unpack . T.pack) `eq` id

6360tl_pack_unpack = (TL.unpack . TL.pack) `eq` id

6461t_stream_unstream = (S.unstream . S.stream) `eq` id

6562tl_stream_unstream = (SL.unstream . SL.stream) `eq` id

66−t_reverse_stream t = (S.reverse . S.reverseStream) t === t

67−t_singleton c = [c] === (T.unpack . T.singleton) c

68−tl_singleton c = [c] === (TL.unpack . TL.singleton) c

69−tl_unstreamChunks x = f 11 x === f 1000 x

63+t_reverse_stream t = (S.reverse . S.reverseStream) t == t

64+t_singleton c = [c] == (T.unpack . T.singleton) c

65+tl_singleton c = [c] == (TL.unpack . TL.singleton) c

66+tl_unstreamChunks x = f 11 x == f 1000 x

7067 where f n = SL.unstreamChunks n . S.streamList

7168tl_chunk_unchunk = (TL.fromChunks . TL.toChunks) `eq` id

7269tl_from_to_strict = (TL.fromStrict . TL.toStrict) `eq` id

@@ -77,13 +74,13 @@ encodeL1 = B.pack . map (fromIntegral . fromEnum) . T.unpack

7774encodeLazyL1 :: TL.Text -> BL.ByteString

7875encodeLazyL1 = BL.fromChunks . map encodeL1 . TL.toChunks

7976

80−t_ascii t = E.decodeASCII (E.encodeUtf8 a) === a

77+t_ascii t = E.decodeASCII (E.encodeUtf8 a) == a

8178 where a = T.map (\c -> chr (ord c `mod` 128)) t

82−tl_ascii t = EL.decodeASCII (EL.encodeUtf8 a) === a

79+tl_ascii t = EL.decodeASCII (EL.encodeUtf8 a) == a

8380 where a = TL.map (\c -> chr (ord c `mod` 128)) t

84−t_latin1 t = E.decodeLatin1 (encodeL1 a) === a

81+t_latin1 t = E.decodeLatin1 (encodeL1 a) == a

8582 where a = T.map (\c -> chr (ord c `mod` 256)) t

86−tl_latin1 t = EL.decodeLatin1 (encodeLazyL1 a) === a

83+tl_latin1 t = EL.decodeLatin1 (encodeLazyL1 a) == a

8784 where a = TL.map (\c -> chr (ord c `mod` 256)) t

8885t_utf8 = forAll genUnicode $ (E.decodeUtf8 . E.encodeUtf8) `eq` id

8986t_utf8' = forAll genUnicode $ (E.decodeUtf8' . E.encodeUtf8) `eq` (id . Right)

@@ -114,7 +111,7 @@ t_utf8_undecoded = forAll genUnicode $ \t ->

114111 let b = E.encodeUtf8 t

115112 ls = concatMap (leftover . E.encodeUtf8 . T.singleton) . T.unpack $ t

116113 leftover = (++ [B.empty]) . init . tail . B.inits

117− in (map snd . feedChunksOf 1 E.streamDecodeUtf8) b === ls

114+ in (map snd . feedChunksOf 1 E.streamDecodeUtf8) b == ls

118115

119116data Badness = Solo | Leading | Trailing

120117 deriving (Eq, Show)

@@ -180,33 +177,6 @@ genInvalidUTF8 = B.pack <$> oneof [

180177 k <- choose (0,n)

181178 vectorOf k gen

182179

183−-- See http://unicode.org/faq/utf_bom.html#gen8

184−-- A sequence such as <110xxxxx2 0xxxxxxx2> is illegal ...

185−-- When faced with this illegal byte sequence ... a UTF-8 conformant process

186−-- must treat the first byte 110xxxxx2 as an illegal termination error

187−-- (e.g. filter it out or replace by 0xFFFD) ...

188−-- ... and continue processing at the second byte 0xxxxxxx2

189−t_decode_with_error2 =

190− E.decodeUtf8With (\_ _ -> Just 'x') (B.pack [0xC2, 97]) === "xa"

191−t_decode_with_error3 =

192− E.decodeUtf8With (\_ _ -> Just 'x') (B.pack [0xE0, 97, 97]) === "xaa"

193−t_decode_with_error4 =

194− E.decodeUtf8With (\_ _ -> Just 'x') (B.pack [0xF0, 97, 97, 97]) === "xaaa"

195−

196−t_decode_with_error2' =

197− case E.streamDecodeUtf8With (\_ _ -> Just 'x') (B.pack [0xC2, 97]) of

198− E.Some x _ _ -> x === "xa"

199−t_decode_with_error3' =

200− case E.streamDecodeUtf8With (\_ _ -> Just 'x') (B.pack [0xC2, 97, 97]) of

201− E.Some x _ _ -> x === "xaa"

202−t_decode_with_error4' =

203− case E.streamDecodeUtf8With (\_ _ -> Just 'x') (B.pack [0xC2, 97, 97, 97]) of

204− E.Some x _ _ -> x === "xaaa"

205−

206−t_infix_concat bs1 text bs2 rep =

207− text `T.isInfixOf`

208− E.decodeUtf8With (\_ _ -> rep) (B.concat [bs1, E.encodeUtf8 text, bs2])

209−

210180s_Eq s = (s==) `eq` ((S.streamList s==) . S.streamList)

211181 where _types = s :: String

212182sf_Eq p s =

@@ -231,8 +201,8 @@ t_mconcat = unsquare $

231201 mconcat `eq` (unpackS . mconcat . L.map T.pack)

232202tl_mconcat = unsquare $

233203 mconcat `eq` (unpackS . mconcat . L.map TL.pack)

234−t_mempty = mempty === (unpackS (mempty :: T.Text))

235−tl_mempty = mempty === (unpackS (mempty :: TL.Text))

204+t_mempty = mempty == (unpackS (mempty :: T.Text))

205+tl_mempty = mempty == (unpackS (mempty :: TL.Text))

236206t_IsString = fromString `eqP` (T.unpack . fromString)

237207tl_IsString = fromString `eqP` (TL.unpack . fromString)

238208

@@ -259,13 +229,6 @@ sf_uncons p = (uncons . L.filter p) `eqP`

259229 (fmap (second unpackS) . S.uncons . S.filter p)

260230t_uncons = uncons `eqP` (fmap (second unpackS) . T.uncons)

261231tl_uncons = uncons `eqP` (fmap (second unpackS) . TL.uncons)

262−

263−unsnoc xs@(_:_) = Just (init xs, last xs)

264−unsnoc [] = Nothing

265−

266−t_unsnoc = unsnoc `eqP` (fmap (first unpackS) . T.unsnoc)

267−tl_unsnoc = unsnoc `eqP` (fmap (first unpackS) . TL.unsnoc)

268−

269232s_head = head `eqP` S.head

270233sf_head p = (head . L.filter p) `eqP` (S.head . S.filter p)

271234t_head = head `eqP` T.head

@@ -363,19 +326,6 @@ t_toUpper_upper t = p (T.toUpper t) >= p t

363326 where p = T.length . T.filter isUpper

364327tl_toUpper_upper t = p (TL.toUpper t) >= p t

365328 where p = TL.length . TL.filter isUpper

366−t_toTitle_title t = all (<= 1) (caps w)

367− where caps = fmap (T.length . T.filter isUpper) . T.words . T.toTitle

368− -- TIL: there exist uppercase-only letters

369− w = T.filter (\c -> if C.isUpper c then C.toLower c /= c else True) t

370−t_toTitle_1stNotLower = and . notLow . T.toTitle . T.filter stable

371− where notLow = mapMaybe (fmap (not . isLower) . (T.find isLetter)) . T.words

372− -- Surprise! The Spanish/Portuguese ordinal indicators changed

373− -- from category Ll (letter, lowercase) to Lo (letter, other)

374− -- in Unicode 7.0

375− -- Oh, and there exist lowercase-only letters (see previous test)

376− stable c = if isLower c

377− then C.toUpper c /= c

378− else c /= '\170' && c /= '\186'

379329

380330justifyLeft k c xs = xs ++ L.replicate (k - length xs) c

381331justifyRight m n xs = L.replicate (m - length xs) n ++ xs

@@ -564,20 +514,12 @@ s_takeWhile p = L.takeWhile p `eqP` (unpackS . S.takeWhile p)

564514s_takeWhile_s p = L.takeWhile p `eqP` (unpackS . S.unstream . S.takeWhile p)

565515sf_takeWhile q p = (L.takeWhile p . L.filter q) `eqP`

566516 (unpackS . S.takeWhile p . S.filter q)

567−noMatch = do

568− c <- char

569− d <- suchThat char (/= c)

570− return (c,d)

571517t_takeWhile p = L.takeWhile p `eqP` (unpackS . T.takeWhile p)

572518tl_takeWhile p = L.takeWhile p `eqP` (unpackS . TL.takeWhile p)

573519t_takeWhileEnd p = (L.reverse . L.takeWhile p . L.reverse) `eqP`

574520 (unpackS . T.takeWhileEnd p)

575−t_takeWhileEnd_null t = forAll noMatch $ \(c,d) -> T.null $

576− T.takeWhileEnd (==d) (T.snoc t c)

577521tl_takeWhileEnd p = (L.reverse . L.takeWhile p . L.reverse) `eqP`

578522 (unpackS . TL.takeWhileEnd p)

579−tl_takeWhileEnd_null t = forAll noMatch $ \(c,d) -> TL.null $

580− TL.takeWhileEnd (==d) (TL.snoc t c)

581523s_dropWhile p = L.dropWhile p `eqP` (unpackS . S.dropWhile p)

582524s_dropWhile_s p = L.dropWhile p `eqP` (unpackS . S.unstream . S.dropWhile p)

583525sf_dropWhile q p = (L.dropWhile p . L.filter q) `eqP`

@@ -778,7 +720,7 @@ tl_indices (NotEmpty s) = lazyIndices s `eq` S.indices s

778720 conc = T.concat . TL.toChunks

779721t_indices_occurs = unsquare $ \(NotEmpty t) ts ->

780722 let s = T.intercalate t ts

781− in Slow.indices t s === indices t s

723+ in Slow.indices t s == indices t s

782724

783725-- Bit shifts.

784726shiftL w = forAll (choose (0,width-1)) $ \k -> Bits.shiftL w k == U.shiftL w k

@@ -870,18 +812,18 @@ tb_formatRealFloat_double (a::Double) = tb_formatRealFloat a

870812-- Reading.

871813

872814t_decimal (n::Int) s =

873− T.signed T.decimal (T.pack (show n) `T.append` t) === Right (n,t)

815+ T.signed T.decimal (T.pack (show n) `T.append` t) == Right (n,t)

874816 where t = T.dropWhile isDigit s

875817tl_decimal (n::Int) s =

876− TL.signed TL.decimal (TL.pack (show n) `TL.append` t) === Right (n,t)

818+ TL.signed TL.decimal (TL.pack (show n) `TL.append` t) == Right (n,t)

877819 where t = TL.dropWhile isDigit s

878820t_hexadecimal m s ox =

879− T.hexadecimal (T.concat [p, T.pack (showHex n ""), t]) === Right (n,t)

821+ T.hexadecimal (T.concat [p, T.pack (showHex n ""), t]) == Right (n,t)

880822 where t = T.dropWhile isHexDigit s

881823 p = if ox then "0x" else ""

882824 n = getPositive m :: Int

883825tl_hexadecimal m s ox =

884− TL.hexadecimal (TL.concat [p, TL.pack (showHex n ""), t]) === Right (n,t)

826+ TL.hexadecimal (TL.concat [p, TL.pack (showHex n ""), t]) == Right (n,t)

885827 where t = TL.dropWhile isHexDigit s

886828 p = if ox then "0x" else ""

887829 n = getPositive m :: Int

@@ -925,11 +867,11 @@ tl_write_read_line e m b t = write_read head TL.filter TL.hPutStrLn

925867

926868t_dropWord16 m t = dropWord16 m t `T.isSuffixOf` t

927869t_takeWord16 m t = takeWord16 m t `T.isPrefixOf` t

928−t_take_drop_16 m t = T.append (takeWord16 n t) (dropWord16 n t) === t

870+t_take_drop_16 m t = T.append (takeWord16 n t) (dropWord16 n t) == t

929871 where n = small m

930872t_use_from t = monadicIO $ assert . (==t) =<< run (useAsPtr t fromPtr)

931873

932−t_copy t = T.copy t === t

874+t_copy t = T.copy t == t

933875

934876-- Regression tests.

935877s_filter_eq s = S.filter p t == S.streamList (filter p s)

@@ -982,15 +924,6 @@ tests =

982924 testGroup "errors" [

983925 testProperty "t_utf8_err" t_utf8_err,

984926 testProperty "t_utf8_err'" t_utf8_err'

985− ],

986− testGroup "error recovery" [

987− testProperty "t_decode_with_error2" t_decode_with_error2,

988− testProperty "t_decode_with_error3" t_decode_with_error3,

989− testProperty "t_decode_with_error4" t_decode_with_error4,

990− testProperty "t_decode_with_error2'" t_decode_with_error2',

991− testProperty "t_decode_with_error3'" t_decode_with_error3',

992− testProperty "t_decode_with_error4'" t_decode_with_error4',

993− testProperty "t_infix_concat" t_infix_concat

994927 ]

995928 ],

996929

@@ -1034,8 +967,6 @@ tests =

1034967 testProperty "sf_uncons" sf_uncons,

1035968 testProperty "t_uncons" t_uncons,

1036969 testProperty "tl_uncons" tl_uncons,

1037− testProperty "t_unsnoc" t_unsnoc,

1038− testProperty "tl_unsnoc" tl_unsnoc,

1039970 testProperty "s_head" s_head,

1040971 testProperty "sf_head" sf_head,

1041972 testProperty "t_head" t_head,

@@ -1099,9 +1030,7 @@ tests =

10991030 testProperty "tl_toLower_lower" tl_toLower_lower,

11001031 testProperty "t_toUpper_length" t_toUpper_length,

11011032 testProperty "t_toUpper_upper" t_toUpper_upper,

1102− testProperty "tl_toUpper_upper" tl_toUpper_upper,

1103− testProperty "t_toTitle_title" t_toTitle_title,

1104− testProperty "t_toTitle_1stNotLower" t_toTitle_1stNotLower

1033+ testProperty "tl_toUpper_upper" tl_toUpper_upper

11051034 ],

11061035

11071036 testGroup "justification" [

@@ -1218,9 +1147,7 @@ tests =

12181147 testProperty "t_takeWhile" t_takeWhile,

12191148 testProperty "tl_takeWhile" tl_takeWhile,

12201149 testProperty "t_takeWhileEnd" t_takeWhileEnd,

1221− testProperty "t_takeWhileEnd_null" t_takeWhileEnd_null,

12221150 testProperty "tl_takeWhileEnd" tl_takeWhileEnd,

1223− testProperty "tl_takeWhileEnd_null" tl_takeWhileEnd_null,

12241151 testProperty "sf_dropWhile" sf_dropWhile,

12251152 testProperty "s_dropWhile" s_dropWhile,

12261153 testProperty "s_dropWhile_s" s_dropWhile_s,

tests/Tests/QuickCheckUtils.hs+1−1

@@ -106,7 +106,7 @@ smallArbitrary = sized $ \n -> resize (smallish n) arbitrary

106106 where smallish = round . (sqrt :: Double -> Double) . fromIntegral . abs

107107

108108instance Arbitrary T.Text where

109− arbitrary = T.pack `fmap` string

109+ arbitrary = T.pack `fmap` arbitrary

110110 shrink = map T.pack . shrink . T.unpack

111111

112112instance Arbitrary TL.Text where

tests/Tests/Regressions.hs−11

@@ -72,16 +72,6 @@ mapAccumL_resize = do

7272 assertEqual "mapAccumL should correctly size buffers for two-word results"

7373 (count * 2) (T.lengthWord16 (snd val))

7474

75−-- See GitHub #197

76−t197 :: IO ()

77−t197 =

78− assertBool "length (filter (==',') \"0,00\") should be 1" (currencyParser "0,00")

79− where

80− currencyParser x = cond == 1

81− where

82− cond = length fltr

83− fltr = filter (== ',') x

84−

8575tests :: F.Test

8676tests = F.testGroup "Regressions"

8777 [ F.testCase "hGetContents_crash" hGetContents_crash

@@ -89,5 +79,4 @@ tests = F.testGroup "Regressions"

8979 , F.testCase "mapAccumL_resize" mapAccumL_resize

9080 , F.testCase "replicate_crash" replicate_crash

9181 , F.testCase "utf8_decode_unsafe" utf8_decode_unsafe

92− , F.testCase "t197" t197

9382 ]

tests/text-tests.cabal+3−18

@@ -20,11 +20,6 @@ flag hpc

2020 default: False

2121 manual: True

2222

23−flag bytestring-builder

24− description: Depend on the bytestring-builder package for backwards compatibility.

25− default: False

26− manual: False

27−

2823executable text-tests

2924 main-is: Tests.hs

3025

@@ -53,21 +48,16 @@ executable text-tests

5348 HUnit >= 1.2,

5449 QuickCheck >= 2.7,

5550 base == 4.*,

51+ bytestring,

5652 deepseq,

5753 directory,

58− quickcheck-unicode >= 1.0.1.0,

54+ quickcheck-unicode,

5955 random,

6056 test-framework >= 0.4,

6157 test-framework-hunit >= 0.2,

6258 test-framework-quickcheck2 >= 0.2,

6359 text-tests

6460

65− if flag(bytestring-builder)

66− build-depends: bytestring >= 0.9 && < 0.10.4,

67− bytestring-builder >= 0.10.4

68− else

69− build-depends: bytestring >= 0.10.4

70−

7161executable text-tests-stdio

7262 main-is: Tests/IO.hs

7363

@@ -147,12 +137,7 @@ library

147137 array,

148138 base == 4.*,

149139 binary,

140+ bytestring,

150141 deepseq,

151142 ghc-prim,

152143 integer-gmp

153−

154− if flag(bytestring-builder)

155− build-depends: bytestring >= 0.9 && < 0.10.4,

156− bytestring-builder >= 0.10.4

157− else

158− build-depends: bytestring >= 0.10.4

text.cabal+17−90

@@ -1,7 +1,7 @@

11name: text

2−version: 1.2.3.0

3−homepage: https://github.com/haskell/text

4−bug-reports: https://github.com/haskell/text/issues

2+version: 1.2.2.1

3+homepage: https://github.com/bos/text

4+bug-reports: https://github.com/bos/text/issues

55synopsis: An efficient packed Unicode text type.

66description:

77 .

@@ -14,13 +14,11 @@ description:

1414 in terms of large data quantities and high speed.

1515 .

1616 The 'Text' type provides character-encoding, type-safe case

17− conversion via whole-string case conversion functions (see "Data.Text").

18− It also provides a range of functions for converting 'Text' values to

19− and from 'ByteStrings', using several standard encodings

20− (see "Data.Text.Encoding").

17+ conversion via whole-string case conversion functions. It also

18+ provides a range of functions for converting 'Text' values to and from

19+ 'ByteStrings', using several standard encodings.

2120 .

22− Efficient locale-sensitive support for text IO is also supported

23− (see "Data.Text.IO").

21+ Efficient locale-sensitive support for text IO is also supported.

2422 .

2523 These modules are intended to be imported qualified, to avoid name

2624 clashes with Prelude functions, e.g.

@@ -30,21 +28,17 @@ description:

3028 To use an extended and very rich family of functions for working

3129 with Unicode text (including normalization, regular expressions,

3230 non-standard encodings, text breaking, and locales), see

33− the [text-icu package](https://hackage.haskell.org/package/text-icu)

34− based on the well-respected and liberally

35− licensed [ICU library](http://site.icu-project.org/).

31+ the @text-icu@ package:

32+ <http://hackage.haskell.org/package/text-icu>

3633

3734license: BSD2

3835license-file: LICENSE

3936author: Bryan O'Sullivan <bos@serpentine.com>

4037maintainer: Bryan O'Sullivan <bos@serpentine.com>

4138copyright: 2009-2011 Bryan O'Sullivan, 2008-2009 Tom Harper

42−bug-reports: https://github.com/haskell/text/issues

4339category: Data, Text

4440build-type: Simple

4541cabal-version: >= 1.8

46−tested-with: GHC==8.2.2, GHC==8.0.2, GHC==7.10.3, GHC==7.8.4,

47− GHC==7.6.3, GHC==7.4.2, GHC==7.2.2, GHC==7.0.4

4842extra-source-files:

4943 -- scripts/CaseFolding.txt

5044 -- scripts/SpecialCasing.txt

@@ -71,9 +65,7 @@ extra-source-files:

7165 tests/text-tests.cabal

7266

7367flag bytestring-builder

74− description:

75− Depend on the [bytestring-builder](https://hackage.haskell.org/package/bytestring-builder)

76− package for backwards compatibility.

68+ description: Depend on the bytestring-builder package for backwards compatibility.

7769 default: False

7870 manual: False

7971

@@ -83,9 +75,7 @@ flag developer

8375 manual: True

8476

8577flag integer-simple

86− description:

87− Use the [simple integer library](http://hackage.haskell.org/package/integer-simple)

88− instead of [integer-gmp](http://hackage.haskell.org/package/integer-gmp)

78+ description: Use the simple integer library instead of GMP

8979 default: False

9080 manual: False

9181

@@ -170,90 +160,27 @@ library

170160

171161test-suite tests

172162 type: exitcode-stdio-1.0

163+ hs-source-dirs: tests .

164+ main-is: Tests.hs

173165 c-sources: cbits/cbits.c

174166 include-dirs: include

175167

176168 ghc-options:

177− -Wall -threaded -rtsopts

169+ -Wall -threaded -O0 -rtsopts

178170

179171 cpp-options:

180172 -DASSERTS -DHAVE_DEEPSEQ -DTEST_SUITE

181173

182− -- modules specific to test-suite

183− hs-source-dirs: tests

184− main-is: Tests.hs

185− other-modules:

186− Tests.Properties

187− Tests.Properties.Mul

188− Tests.QuickCheckUtils

189− Tests.Regressions

190− Tests.SlowFunctions

191− Tests.Utils

192−

193− -- Same as in `library` stanza; this is needed by cabal for accurate

194− -- file-monitoring as well as to avoid `-Wmissing-home-modules`

195− -- warnings We can't use an inter-package library dependency because

196− -- of different `ghc-options`/`cpp-options` (as a side-benefitt,

197− -- this enables per-component build parallelism in `cabal

198− -- new-build`!); We could, however, use cabal-version:2.2's `common`

199− -- blocks at some point in the future to reduce the duplication.

200− hs-source-dirs: .

201− other-modules:

202− Data.Text

203− Data.Text.Array

204− Data.Text.Encoding

205− Data.Text.Encoding.Error

206− Data.Text.Foreign

207− Data.Text.IO

208− Data.Text.Internal

209− Data.Text.Internal.Builder

210− Data.Text.Internal.Builder.Functions

211− Data.Text.Internal.Builder.Int.Digits

212− Data.Text.Internal.Builder.RealFloat.Functions

213− Data.Text.Internal.Encoding.Fusion

214− Data.Text.Internal.Encoding.Fusion.Common

215− Data.Text.Internal.Encoding.Utf16

216− Data.Text.Internal.Encoding.Utf32

217− Data.Text.Internal.Encoding.Utf8

218− Data.Text.Internal.Functions

219− Data.Text.Internal.Fusion

220− Data.Text.Internal.Fusion.CaseMapping

221− Data.Text.Internal.Fusion.Common

222− Data.Text.Internal.Fusion.Size

223− Data.Text.Internal.Fusion.Types

224− Data.Text.Internal.IO

225− Data.Text.Internal.Lazy

226− Data.Text.Internal.Lazy.Encoding.Fusion

227− Data.Text.Internal.Lazy.Fusion

228− Data.Text.Internal.Lazy.Search

229− Data.Text.Internal.Private

230− Data.Text.Internal.Read

231− Data.Text.Internal.Search

232− Data.Text.Internal.Unsafe

233− Data.Text.Internal.Unsafe.Char

234− Data.Text.Internal.Unsafe.Shift

235− Data.Text.Lazy

236− Data.Text.Lazy.Builder

237− Data.Text.Lazy.Builder.Int

238− Data.Text.Lazy.Builder.RealFloat

239− Data.Text.Lazy.Encoding

240− Data.Text.Lazy.IO

241− Data.Text.Lazy.Internal

242− Data.Text.Lazy.Read

243− Data.Text.Read

244− Data.Text.Unsafe

245− Data.Text.Show

246−

247174 build-depends:

248175 HUnit >= 1.2,

249− QuickCheck >= 2.7 && < 2.10,

176+ QuickCheck >= 2.7,

250177 array,

251178 base,

252179 binary,

253180 deepseq,

254181 directory,

255182 ghc-prim,

256− quickcheck-unicode >= 1.0.1.0,

183+ quickcheck-unicode,

257184 random,

258185 test-framework >= 0.4,

259186 test-framework-hunit >= 0.2,

@@ -274,7 +201,7 @@ test-suite tests

274201

275202source-repository head

276203 type: git

277− location: https://github.com/haskell/text

204+ location: https://github.com/bos/text

278205

279206source-repository head

280207 type: mercurial