c91deda..master · text

git.iko.soy/text

on master help sign in

.gitignore−4

since splitsnapshots

unifiedsplit

@@ -1,7 +1,3 @@

1dist

2−dist-boot

3−dist-install

4cabal-dev

2cabal-dev

5cabal.sandbox.config

3cabal.sandbox.config

6−ghc.mk

7−GNUmakefile

.hgtags−1

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

64631fd055e27fa10983e17e9ad36d1b285b85c661 1.2.1.3

658221a68ed5ec37028357df9c6c25f52320e54ad9 1.2.2.0

66ce1c5ac65eb9f5580114b4e380092b05a485e95d 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-8.2.1"

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

53− addons: {apt: {packages: [ghc-ppa-tools,cabal-install-head,ghc-8.2.1], 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+1−28

@@ -34,7 +34,6 @@

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

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

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

37−--

39module Data.Text

38module Data.Text

40 (

39 (

@@ -44,9 +43,6 @@ module Data.Text

44 -- * Acceptable data

43 -- * Acceptable data

45 -- $replacement

44 -- $replacement

47− -- * Definition of character

48− -- $character_definition

49−

50 -- * Fusion

46 -- * Fusion

51 -- $fusion

47 -- $fusion

@@ -64,7 +60,6 @@ module Data.Text

64 , snoc

60 , snoc

65 , append

61 , append

66 , uncons

62 , uncons

67− , unsnoc

68 , head

63 , head

69 , last

64 , last

70 , tail

65 , tail

@@ -251,17 +246,6 @@ import qualified GHC.Exts as Exts

251import Text.Printf (PrintfArg, formatArg, formatString)

246import Text.Printf (PrintfArg, formatArg, formatString)

252#endif

247#endif

253

248

254−-- $character-definition

255−--

256−-- This package uses the word /character/ to denote Unicode code points.

257−--

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

259−-- code points that form one visual symbol). For instance, consider the grapheme

260−-- @ä@. This symbol has two Unicode representations: a single code-point

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

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

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

264−

265-- $strict

249-- $strict

266--

250--

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

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

@@ -561,17 +545,6 @@ init (Text arr off len) | len <= 0 = emptyError "init"

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

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

562 #-}

546 #-}

563

547

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

565−-- 'Text', or 'Nothing' if empty.

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

567−unsnoc (Text arr off len)

568− | len <= 0 = Nothing

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

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

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

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

573−{-# INLINE [1] unsnoc #-}

574−

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

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

576-- fusion.

549-- fusion.

577null :: Text -> Bool

550null :: Text -> Bool

@@ -1126,8 +1099,8 @@ takeEnd n t@(Text arr off len)

1126iterNEnd :: Int -> Text -> Int

1099iterNEnd :: Int -> Text -> Int

1127iterNEnd n t@(Text _arr _off len) = loop (len-1) n

1100iterNEnd n t@(Text _arr _off len) = loop (len-1) n

1128 where loop i !m

1101 where loop i !m

1129− | m <= 0 = i+1

1130 | i <= 0 = 0

1102 | i <= 0 = 0

1103+ | m <= 1 = i

1131 | otherwise = loop (i+d) (m-1)

1104 | otherwise = loop (i+d) (m-1)

1132 where d = reverseIter_ t i

1105 where d = reverseIter_ t i

1133

1106

Data/Text/Encoding.hs+1−1

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

373encodeUtf8 (Text arr off len)

374 | len == 0 = B.empty

375 | otherwise = unsafeDupablePerformIO $ do

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

376+ fp <- mallocByteString (len*4)

377 withForeignPtr fp $ \ptr ->

378 with ptr $ \destPtr -> do

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

Data/Text/Foreign.hs+1−1

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

108 | n >= len || m >= len = t

109 | otherwise = Text arr off m

110 where

111− m | w < 0xD800 || w > 0xDBFF = n

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

112 | otherwise = n+1

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

114

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

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

1{-# LANGUAGE Rank2Types #-}

2-- AUTOMATICALLY GENERATED - DO NOT EDIT

3-- Generated by scripts/CaseMapping.hs

4−-- CaseFolding-9.0.0.txt

4+-- CaseFolding-8.0.0.txt

5−-- Date: 2016-03-02, 18:54:54 GMT

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

6−-- SpecialCasing-9.0.0.txt

6+-- SpecialCasing-8.0.0.txt

7−-- Date: 2016-03-02, 18:55:13 GMT

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

9module Data.Text.Internal.Fusion.CaseMapping where

10import Data.Char

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

371foldMapping '\x13fc' s = Yield '\x13f4' (CC s '\x0000' '\x0000')

372-- CHEROKEE SMALL LETTER MV

373foldMapping '\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')

392-- LATIN SMALL LETTER H WITH LINE BELOW

374-- LATIN SMALL LETTER H WITH LINE BELOW

393foldMapping '\x1e96' s = Yield '\x0068' (CC s '\x0331' '\x0000')

375foldMapping '\x1e96' s = Yield '\x0068' (CC s '\x0331' '\x0000')

394-- LATIN SMALL LETTER T WITH DIAERESIS

376-- LATIN SMALL LETTER T WITH DIAERESIS

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

563foldMapping '\x1ff7' s = Yield '\x03c9' (CC s '\x0342' '\x03b9')

545foldMapping '\x1ff7' s = Yield '\x03c9' (CC s '\x0342' '\x03b9')

564-- GREEK CAPITAL LETTER OMEGA WITH PROSGEGRAMMENI

546-- GREEK CAPITAL LETTER OMEGA WITH PROSGEGRAMMENI

565foldMapping '\x1ffc' s = Yield '\x03c9' (CC s '\x03b9' '\x0000')

547foldMapping '\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')

568-- LATIN CAPITAL LETTER J WITH CROSSED-TAIL

548-- LATIN CAPITAL LETTER J WITH CROSSED-TAIL

569foldMapping '\xa7b2' s = Yield '\x029d' (CC s '\x0000' '\x0000')

549foldMapping '\xa7b2' s = Yield '\x029d' (CC s '\x0000' '\x0000')

570-- LATIN CAPITAL LETTER CHI

550-- LATIN CAPITAL LETTER CHI

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

757foldMapping '\xfb16' s = Yield '\x057e' (CC s '\x0576' '\x0000')

737foldMapping '\xfb16' s = Yield '\x057e' (CC s '\x0576' '\x0000')

758-- ARMENIAN SMALL LIGATURE MEN XEH

738-- ARMENIAN SMALL LIGATURE MEN XEH

759foldMapping '\xfb17' s = Yield '\x0574' (CC s '\x056d' '\x0000')

739foldMapping '\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')

832-- OLD HUNGARIAN CAPITAL LETTER A

740-- OLD HUNGARIAN CAPITAL LETTER A

833foldMapping '\x10c80' s = Yield '\x10cc0' (CC s '\x0000' '\x0000')

741foldMapping '\x10c80' s = Yield '\x10cc0' (CC s '\x0000' '\x0000')

834-- OLD HUNGARIAN CAPITAL LETTER AA

742-- OLD HUNGARIAN CAPITAL LETTER AA

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

931foldMapping '\x10cb1' s = Yield '\x10cf1' (CC s '\x0000' '\x0000')

839foldMapping '\x10cb1' s = Yield '\x10cf1' (CC s '\x0000' '\x0000')

932-- OLD HUNGARIAN CAPITAL LETTER US

840-- OLD HUNGARIAN CAPITAL LETTER US

933foldMapping '\x10cb2' s = Yield '\x10cf2' (CC s '\x0000' '\x0000')

841foldMapping '\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')

1002foldMapping c s = Yield (toLower c) (CC s '\0' '\0')

842foldMapping c s = Yield (toLower c) (CC s '\0' '\0')

Data/Text/Internal/Fusion/Common.hs+9−10

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

107import qualified Data.List as L

108import qualified Prelude as P

109import Data.Bits (shiftL)

110−import Data.Char (isLetter, isSpace)

110+import Data.Char (isLetter)

111import Data.Int (Int64)

112import Data.Text.Internal.Fusion.Types

113import Data.Text.Internal.Fusion.CaseMapping (foldMapping, lowerMapping, titleMapping,

@@ -462,16 +462,15 @@ toTitle (Stream next0 s0 len) = Stream next (CC (False :*: s0) '\0' '\0') len

462 where

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

464 case next0 s of

465− Done -> Done

465+ Done -> Done

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

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

467 Yield c s'

468− | nonSpace -> if letter

468+ | letter' -> if letter

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

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

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

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

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

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

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

473− letter' = isLetter c

472+ where letter' = isLetter c

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

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

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

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

476

475

477data Justify i s = Just1 !i !s

476data Justify i s = Just1 !i !s

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

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

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

76-- small or too large.

77--

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

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

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

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

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

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

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

83data Stream a =

84 forall s. Stream

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

86 !s -- current state

87− !Size -- size hint in code units

87+ !Size -- size hint

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

90eq :: (Eq a) => Stream a -> Stream a -> Bool

Data/Text/Lazy.hs+6−15

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

68 , snoc

69 , append

70 , uncons

71− , unsnoc

72 , head

71 , head

73 , last

72 , last

74 , tail

73 , tail

@@ -476,7 +475,9 @@ fromStrict t = chunk t Empty

476-- -----------------------------------------------------------------------------

475-- -----------------------------------------------------------------------------

477-- * Basic functions

476-- * Basic functions

478

477

479−-- | /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.

480cons :: Char -> Text -> Text

481cons :: Char -> Text -> Text

481cons c t = Chunk (T.singleton c) t

482cons c t = Chunk (T.singleton c) t

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

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

@@ -560,15 +561,6 @@ init Empty = emptyError "init"

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

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

561 #-}

562 #-}

562

563

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

564−-- empty.

565−--

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

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

568−unsnoc Empty = Nothing

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

570−{-# INLINE unsnoc #-}

571−

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

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

573-- fusion.

565-- fusion.

574null :: Text -> Bool

566null :: Text -> Bool

@@ -1161,10 +1153,9 @@ takeWhile p t0 = takeWhile' t0

1161takeWhileEnd :: (Char -> Bool) -> Text -> Text

1153takeWhileEnd :: (Char -> Bool) -> Text -> Text

1162takeWhileEnd p = takeChunk empty . L.reverse . toChunks

1154takeWhileEnd p = takeChunk empty . L.reverse . toChunks

1163 where takeChunk acc [] = acc

1155 where takeChunk acc [] = acc

1164− takeChunk acc (t:ts)

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

1165− | T.lengthWord16 t' < T.lengthWord16 t

1157+ then (Chunk t' acc)

1166− = chunk t' acc

1167− | otherwise = takeChunk (Chunk t' acc) ts

1158+ else takeChunk (Chunk t' acc) ts

1168 where t' = T.takeWhileEnd p t

1159 where t' = T.takeWhileEnd p t

1169{-# INLINE takeWhileEnd #-}

1160{-# INLINE takeWhileEnd #-}

1170

1161

Data/Text/Lazy/Builder/Int.hs−3

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

32import GHC.Num (quotRemInteger)

33import GHC.Types (Int(..))

34import Control.Monad.ST

35−#if MIN_VERSION_base(4,11,0)

36−import Prelude hiding ((<>))

37−#endif

39#ifdef __GLASGOW_HASKELL__

36#ifdef __GLASGOW_HASKELL__

40# if defined(INTEGER_GMP)

37# if defined(INTEGER_GMP)

Data/Text/Lazy/Builder/RealFloat.hs−3

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

24import Data.Text.Internal.Builder.RealFloat.Functions (roundTo)

25import Data.Text.Lazy.Builder

26import qualified Data.Text as T

27−#if MIN_VERSION_base(4,11,0)

28−import Prelude hiding ((<>))

29−#endif

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

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

32data FPFormat = Exponent

29data FPFormat = Exponent

Data/Text/Lazy/Read.hs+3−3

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

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

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

71−-- @\"0x\"@ followed by at least one hexadecimal digit. Input is

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

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

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

73−-- This function is case insensitive.

73+-- is case insensitive.

74--

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

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

Data/Text/Read.hs+3−3

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

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

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

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

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

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

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

72−-- This function is case insensitive.

72+-- is case insensitive.

73--

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

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

benchmarks/haskell/Benchmarks.hs−2

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

10import System.IO (IOMode (WriteMode), openFile, hSetEncoding, utf8)

12import qualified Benchmarks.Builder as Builder

13−import qualified Benchmarks.Concat as Concat

14import qualified Benchmarks.DecodeUtf8 as DecodeUtf8

13import qualified Benchmarks.DecodeUtf8 as DecodeUtf8

15import qualified Benchmarks.EncodeUtf8 as EncodeUtf8

14import qualified Benchmarks.EncodeUtf8 as EncodeUtf8

16import qualified Benchmarks.Equality as Equality

15import qualified Benchmarks.Equality as Equality

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

42 -- Traditional benchmarks

41 -- Traditional benchmarks

43 bs <- sequence

42 bs <- sequence

44 [ Builder.benchmark

43 [ Builder.benchmark

45− , Concat.benchmark

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

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

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

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

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

46 , 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−1

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

47 ghc-prim,

48 integer-gmp,

49 stringsearch,

50− transformers,

51 utf8-string,

50 utf8-string,

52 vector

51 vector

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,

206 const uint8_t *srcend,

207 uint32_t *codepoint0, uint32_t *state0)

208{

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

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

210−

210+ codepoint0, state0);

211+ if (*state0 == UTF8_REJECT)

212+ ret -=1;

211− return *src;

213+ return ret;

212}

214}

213

215

214/*

216/*

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

220{

222{

221 uint32_t codepoint;

223 uint32_t codepoint;

222 uint32_t state = UTF8_ACCEPT;

224 uint32_t state = UTF8_ACCEPT;

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

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

224− &codepoint, &state);

226+ &codepoint, &state);

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

228+ if (state != UTF8_ACCEPT)

229+ ret -= 1;

225− return src;

230+ return ret;

226}

231}

227

232

228void

233void

changelog.md+1−29

@@ -1,38 +1,10 @@

1−1.2.3.0

2−

3−* `unsnoc` implemented

4−

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

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

7−

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

9− (updated from 8.0)

10−

11−

12−1.2.2.2

13−

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

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

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

17−

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

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

20− smaller code generated.

21−

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

23− used after some rewrite rule tweaking.

24−

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

26−

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

28−

291.2.2.1

11.2.2.1

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

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

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

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

33 maintainers apologizes for the churn.

5 maintainers apologizes for the churn.

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

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

36 (updated from 7.0)

8 (updated from 7.0)

38* An STG lint error has been fixed

10* An STG lint error has been fixed

tests/Tests/Properties.hs+21−94

@@ -9,9 +9,9 @@ module Tests.Properties

9 ) where

11import Control.Applicative ((<$>), (<*>))

12−import Control.Arrow ((***), first, second)

12+import Control.Arrow ((***), second)

13import 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)

15import Data.Int (Int8, Int16, Int32, Int64)

16import Data.Monoid (Monoid(..))

17import Data.String (IsString(fromString))

@@ -23,7 +23,6 @@ import Data.Text.Internal.Search (indices)

23import Data.Text.Lazy.Read as TL

24import Data.Text.Read as T

25import Data.Word (Word, Word8, Word16, Word32, Word64)

26−import Data.Maybe (mapMaybe)

27import Numeric (showEFloat, showFFloat, showGFloat, showHex)

26import Numeric (showEFloat, showFFloat, showGFloat, showHex)

28import Prelude hiding (replicate)

27import Prelude hiding (replicate)

29import Test.Framework (Test, testGroup)

28import Test.Framework (Test, testGroup)

@@ -31,7 +30,6 @@ import Test.Framework.Providers.QuickCheck2 (testProperty)

31import Test.QuickCheck hiding ((.&.))

30import Test.QuickCheck hiding ((.&.))

32import Test.QuickCheck.Monadic

31import Test.QuickCheck.Monadic

33import Test.QuickCheck.Property (Property(..))

32import Test.QuickCheck.Property (Property(..))

34−import Test.QuickCheck.Unicode (char)

35import Tests.QuickCheckUtils

33import Tests.QuickCheckUtils

36import Tests.Utils

34import Tests.Utils

37import Text.Show.Functions ()

35import Text.Show.Functions ()

@@ -39,7 +37,6 @@ import qualified Control.Exception as Exception

39import qualified Data.Bits as Bits (shiftL, shiftR)

37import qualified Data.Bits as Bits (shiftL, shiftR)

40import qualified Data.ByteString as B

38import qualified Data.ByteString as B

41import qualified Data.ByteString.Lazy as BL

39import qualified Data.ByteString.Lazy as BL

42−import qualified Data.Char as C

43import qualified Data.List as L

40import qualified Data.List as L

44import qualified Data.Text as T

41import qualified Data.Text as T

45import qualified Data.Text.Encoding as E

42import qualified Data.Text.Encoding as E

@@ -63,10 +60,10 @@ t_pack_unpack = (T.unpack . T.pack) `eq` id

63tl_pack_unpack = (TL.unpack . TL.pack) `eq` id

60tl_pack_unpack = (TL.unpack . TL.pack) `eq` id

64t_stream_unstream = (S.unstream . S.stream) `eq` id

61t_stream_unstream = (S.unstream . S.stream) `eq` id

65tl_stream_unstream = (SL.unstream . SL.stream) `eq` id

62tl_stream_unstream = (SL.unstream . SL.stream) `eq` id

66−t_reverse_stream t = (S.reverse . S.reverseStream) t === t

63+t_reverse_stream t = (S.reverse . S.reverseStream) t == t

67−t_singleton c = [c] === (T.unpack . T.singleton) c

64+t_singleton c = [c] == (T.unpack . T.singleton) c

68−tl_singleton c = [c] === (TL.unpack . TL.singleton) c

65+tl_singleton c = [c] == (TL.unpack . TL.singleton) c

69−tl_unstreamChunks x = f 11 x === f 1000 x

66+tl_unstreamChunks x = f 11 x == f 1000 x

70 where f n = SL.unstreamChunks n . S.streamList

67 where f n = SL.unstreamChunks n . S.streamList

71tl_chunk_unchunk = (TL.fromChunks . TL.toChunks) `eq` id

68tl_chunk_unchunk = (TL.fromChunks . TL.toChunks) `eq` id

72tl_from_to_strict = (TL.fromStrict . TL.toStrict) `eq` id

69tl_from_to_strict = (TL.fromStrict . TL.toStrict) `eq` id

@@ -77,13 +74,13 @@ encodeL1 = B.pack . map (fromIntegral . fromEnum) . T.unpack

77encodeLazyL1 :: TL.Text -> BL.ByteString

74encodeLazyL1 :: TL.Text -> BL.ByteString

78encodeLazyL1 = BL.fromChunks . map encodeL1 . TL.toChunks

75encodeLazyL1 = BL.fromChunks . map encodeL1 . TL.toChunks

80−t_ascii t = E.decodeASCII (E.encodeUtf8 a) === a

77+t_ascii t = E.decodeASCII (E.encodeUtf8 a) == a

81 where a = T.map (\c -> chr (ord c `mod` 128)) t

78 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

83 where a = TL.map (\c -> chr (ord c `mod` 128)) t

80 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

85 where a = T.map (\c -> chr (ord c `mod` 256)) t

82 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

87 where a = TL.map (\c -> chr (ord c `mod` 256)) t

84 where a = TL.map (\c -> chr (ord c `mod` 256)) t

88t_utf8 = forAll genUnicode $ (E.decodeUtf8 . E.encodeUtf8) `eq` id

85t_utf8 = forAll genUnicode $ (E.decodeUtf8 . E.encodeUtf8) `eq` id

89t_utf8' = forAll genUnicode $ (E.decodeUtf8' . E.encodeUtf8) `eq` (id . Right)

86t_utf8' = forAll genUnicode $ (E.decodeUtf8' . E.encodeUtf8) `eq` (id . Right)

@@ -114,7 +111,7 @@ t_utf8_undecoded = forAll genUnicode $ \t ->

114 let b = E.encodeUtf8 t

111 let b = E.encodeUtf8 t

115 ls = concatMap (leftover . E.encodeUtf8 . T.singleton) . T.unpack $ t

112 ls = concatMap (leftover . E.encodeUtf8 . T.singleton) . T.unpack $ t

116 leftover = (++ [B.empty]) . init . tail . B.inits

113 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

118

115

119data Badness = Solo | Leading | Trailing

116data Badness = Solo | Leading | Trailing

120 deriving (Eq, Show)

117 deriving (Eq, Show)

@@ -180,33 +177,6 @@ genInvalidUTF8 = B.pack <$> oneof [

180 k <- choose (0,n)

177 k <- choose (0,n)

181 vectorOf k gen

178 vectorOf k gen

182

179

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−

210s_Eq s = (s==) `eq` ((S.streamList s==) . S.streamList)

180s_Eq s = (s==) `eq` ((S.streamList s==) . S.streamList)

211 where _types = s :: String

181 where _types = s :: String

212sf_Eq p s =

182sf_Eq p s =

@@ -231,8 +201,8 @@ t_mconcat = unsquare $

231 mconcat `eq` (unpackS . mconcat . L.map T.pack)

201 mconcat `eq` (unpackS . mconcat . L.map T.pack)

232tl_mconcat = unsquare $

202tl_mconcat = unsquare $

233 mconcat `eq` (unpackS . mconcat . L.map TL.pack)

203 mconcat `eq` (unpackS . mconcat . L.map TL.pack)

234−t_mempty = mempty === (unpackS (mempty :: T.Text))

204+t_mempty = mempty == (unpackS (mempty :: T.Text))

235−tl_mempty = mempty === (unpackS (mempty :: TL.Text))

205+tl_mempty = mempty == (unpackS (mempty :: TL.Text))

236t_IsString = fromString `eqP` (T.unpack . fromString)

206t_IsString = fromString `eqP` (T.unpack . fromString)

237tl_IsString = fromString `eqP` (TL.unpack . fromString)

207tl_IsString = fromString `eqP` (TL.unpack . fromString)

238

208

@@ -259,13 +229,6 @@ sf_uncons p = (uncons . L.filter p) `eqP`

259 (fmap (second unpackS) . S.uncons . S.filter p)

229 (fmap (second unpackS) . S.uncons . S.filter p)

260t_uncons = uncons `eqP` (fmap (second unpackS) . T.uncons)

230t_uncons = uncons `eqP` (fmap (second unpackS) . T.uncons)

261tl_uncons = uncons `eqP` (fmap (second unpackS) . TL.uncons)

231tl_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−

269s_head = head `eqP` S.head

232s_head = head `eqP` S.head

270sf_head p = (head . L.filter p) `eqP` (S.head . S.filter p)

233sf_head p = (head . L.filter p) `eqP` (S.head . S.filter p)

271t_head = head `eqP` T.head

234t_head = head `eqP` T.head

@@ -363,19 +326,6 @@ t_toUpper_upper t = p (T.toUpper t) >= p t

363 where p = T.length . T.filter isUpper

326 where p = T.length . T.filter isUpper

364tl_toUpper_upper t = p (TL.toUpper t) >= p t

327tl_toUpper_upper t = p (TL.toUpper t) >= p t

365 where p = TL.length . TL.filter isUpper

328 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'

379

329

380justifyLeft k c xs = xs ++ L.replicate (k - length xs) c

330justifyLeft k c xs = xs ++ L.replicate (k - length xs) c

381justifyRight m n xs = L.replicate (m - length xs) n ++ xs

331justifyRight 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)

564s_takeWhile_s p = L.takeWhile p `eqP` (unpackS . S.unstream . S.takeWhile p)

514s_takeWhile_s p = L.takeWhile p `eqP` (unpackS . S.unstream . S.takeWhile p)

565sf_takeWhile q p = (L.takeWhile p . L.filter q) `eqP`

515sf_takeWhile q p = (L.takeWhile p . L.filter q) `eqP`

566 (unpackS . S.takeWhile p . S.filter q)

516 (unpackS . S.takeWhile p . S.filter q)

567−noMatch = do

568− c <- char

569− d <- suchThat char (/= c)

570− return (c,d)

571t_takeWhile p = L.takeWhile p `eqP` (unpackS . T.takeWhile p)

517t_takeWhile p = L.takeWhile p `eqP` (unpackS . T.takeWhile p)

572tl_takeWhile p = L.takeWhile p `eqP` (unpackS . TL.takeWhile p)

518tl_takeWhile p = L.takeWhile p `eqP` (unpackS . TL.takeWhile p)

573t_takeWhileEnd p = (L.reverse . L.takeWhile p . L.reverse) `eqP`

519t_takeWhileEnd p = (L.reverse . L.takeWhile p . L.reverse) `eqP`

574 (unpackS . T.takeWhileEnd p)

520 (unpackS . T.takeWhileEnd p)

575−t_takeWhileEnd_null t = forAll noMatch $ \(c,d) -> T.null $

576− T.takeWhileEnd (==d) (T.snoc t c)

577tl_takeWhileEnd p = (L.reverse . L.takeWhile p . L.reverse) `eqP`

521tl_takeWhileEnd p = (L.reverse . L.takeWhile p . L.reverse) `eqP`

578 (unpackS . TL.takeWhileEnd p)

522 (unpackS . TL.takeWhileEnd p)

579−tl_takeWhileEnd_null t = forAll noMatch $ \(c,d) -> TL.null $

580− TL.takeWhileEnd (==d) (TL.snoc t c)

581s_dropWhile p = L.dropWhile p `eqP` (unpackS . S.dropWhile p)

523s_dropWhile p = L.dropWhile p `eqP` (unpackS . S.dropWhile p)

582s_dropWhile_s p = L.dropWhile p `eqP` (unpackS . S.unstream . S.dropWhile p)

524s_dropWhile_s p = L.dropWhile p `eqP` (unpackS . S.unstream . S.dropWhile p)

583sf_dropWhile q p = (L.dropWhile p . L.filter q) `eqP`

525sf_dropWhile q p = (L.dropWhile p . L.filter q) `eqP`

@@ -778,7 +720,7 @@ tl_indices (NotEmpty s) = lazyIndices s `eq` S.indices s

778 conc = T.concat . TL.toChunks

720 conc = T.concat . TL.toChunks

779t_indices_occurs = unsquare $ \(NotEmpty t) ts ->

721t_indices_occurs = unsquare $ \(NotEmpty t) ts ->

780 let s = T.intercalate t ts

722 let s = T.intercalate t ts

781− in Slow.indices t s === indices t s

723+ in Slow.indices t s == indices t s

782

724

783-- Bit shifts.

725-- Bit shifts.

784shiftL w = forAll (choose (0,width-1)) $ \k -> Bits.shiftL w k == U.shiftL w k

726shiftL 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

870-- Reading.

812-- Reading.

871

813

872t_decimal (n::Int) s =

814t_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)

874 where t = T.dropWhile isDigit s

816 where t = T.dropWhile isDigit s

875tl_decimal (n::Int) s =

817tl_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)

877 where t = TL.dropWhile isDigit s

819 where t = TL.dropWhile isDigit s

878t_hexadecimal m s ox =

820t_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)

880 where t = T.dropWhile isHexDigit s

822 where t = T.dropWhile isHexDigit s

881 p = if ox then "0x" else ""

823 p = if ox then "0x" else ""

882 n = getPositive m :: Int

824 n = getPositive m :: Int

883tl_hexadecimal m s ox =

825tl_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)

885 where t = TL.dropWhile isHexDigit s

827 where t = TL.dropWhile isHexDigit s

886 p = if ox then "0x" else ""

828 p = if ox then "0x" else ""

887 n = getPositive m :: Int

829 n = getPositive m :: Int

@@ -925,11 +867,11 @@ tl_write_read_line e m b t = write_read head TL.filter TL.hPutStrLn

925

867

926t_dropWord16 m t = dropWord16 m t `T.isSuffixOf` t

868t_dropWord16 m t = dropWord16 m t `T.isSuffixOf` t

927t_takeWord16 m t = takeWord16 m t `T.isPrefixOf` t

869t_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

929 where n = small m

871 where n = small m

930t_use_from t = monadicIO $ assert . (==t) =<< run (useAsPtr t fromPtr)

872t_use_from t = monadicIO $ assert . (==t) =<< run (useAsPtr t fromPtr)

931

873

932−t_copy t = T.copy t === t

874+t_copy t = T.copy t == t

933

875

934-- Regression tests.

876-- Regression tests.

935s_filter_eq s = S.filter p t == S.streamList (filter p s)

877s_filter_eq s = S.filter p t == S.streamList (filter p s)

@@ -982,15 +924,6 @@ tests =

982 testGroup "errors" [

924 testGroup "errors" [

983 testProperty "t_utf8_err" t_utf8_err,

925 testProperty "t_utf8_err" t_utf8_err,

984 testProperty "t_utf8_err'" t_utf8_err'

926 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

994 ]

927 ]

995 ],

928 ],

996

929

@@ -1034,8 +967,6 @@ tests =

1034 testProperty "sf_uncons" sf_uncons,

967 testProperty "sf_uncons" sf_uncons,

1035 testProperty "t_uncons" t_uncons,

968 testProperty "t_uncons" t_uncons,

1036 testProperty "tl_uncons" tl_uncons,

969 testProperty "tl_uncons" tl_uncons,

1037− testProperty "t_unsnoc" t_unsnoc,

1038− testProperty "tl_unsnoc" tl_unsnoc,

1039 testProperty "s_head" s_head,

970 testProperty "s_head" s_head,

1040 testProperty "sf_head" sf_head,

971 testProperty "sf_head" sf_head,

1041 testProperty "t_head" t_head,

972 testProperty "t_head" t_head,

@@ -1099,9 +1030,7 @@ tests =

1099 testProperty "tl_toLower_lower" tl_toLower_lower,

1030 testProperty "tl_toLower_lower" tl_toLower_lower,

1100 testProperty "t_toUpper_length" t_toUpper_length,

1031 testProperty "t_toUpper_length" t_toUpper_length,

1101 testProperty "t_toUpper_upper" t_toUpper_upper,

1032 testProperty "t_toUpper_upper" t_toUpper_upper,

1102− testProperty "tl_toUpper_upper" tl_toUpper_upper,

1033+ testProperty "tl_toUpper_upper" tl_toUpper_upper

1103− testProperty "t_toTitle_title" t_toTitle_title,

1104− testProperty "t_toTitle_1stNotLower" t_toTitle_1stNotLower

1105 ],

1034 ],

1106

1035

1107 testGroup "justification" [

1036 testGroup "justification" [

@@ -1218,9 +1147,7 @@ tests =

1218 testProperty "t_takeWhile" t_takeWhile,

1147 testProperty "t_takeWhile" t_takeWhile,

1219 testProperty "tl_takeWhile" tl_takeWhile,

1148 testProperty "tl_takeWhile" tl_takeWhile,

1220 testProperty "t_takeWhileEnd" t_takeWhileEnd,

1149 testProperty "t_takeWhileEnd" t_takeWhileEnd,

1221− testProperty "t_takeWhileEnd_null" t_takeWhileEnd_null,

1222 testProperty "tl_takeWhileEnd" tl_takeWhileEnd,

1150 testProperty "tl_takeWhileEnd" tl_takeWhileEnd,

1223− testProperty "tl_takeWhileEnd_null" tl_takeWhileEnd_null,

1224 testProperty "sf_dropWhile" sf_dropWhile,

1151 testProperty "sf_dropWhile" sf_dropWhile,

1225 testProperty "s_dropWhile" s_dropWhile,

1152 testProperty "s_dropWhile" s_dropWhile,

1226 testProperty "s_dropWhile_s" s_dropWhile_s,

1153 testProperty "s_dropWhile_s" s_dropWhile_s,

tests/Tests/QuickCheckUtils.hs+1−1

@@ -106,7 +106,7 @@ smallArbitrary = sized $ \n -> resize (smallish n) arbitrary

106 where smallish = round . (sqrt :: Double -> Double) . fromIntegral . abs

107

108instance Arbitrary T.Text where

109− arbitrary = T.pack `fmap` string

109+ arbitrary = T.pack `fmap` arbitrary

110 shrink = map T.pack . shrink . T.unpack

111

112instance Arbitrary TL.Text where

tests/Tests/Regressions.hs−11

@@ -72,16 +72,6 @@ mapAccumL_resize = do

72 assertEqual "mapAccumL should correctly size buffers for two-word results"

73 (count * 2) (T.lengthWord16 (snd val))

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−

85tests :: F.Test

75tests :: F.Test

86tests = F.testGroup "Regressions"

76tests = F.testGroup "Regressions"

87 [ F.testCase "hGetContents_crash" hGetContents_crash

77 [ F.testCase "hGetContents_crash" hGetContents_crash

@@ -89,5 +79,4 @@ tests = F.testGroup "Regressions"

89 , F.testCase "mapAccumL_resize" mapAccumL_resize

79 , F.testCase "mapAccumL_resize" mapAccumL_resize

90 , F.testCase "replicate_crash" replicate_crash

80 , F.testCase "replicate_crash" replicate_crash

91 , F.testCase "utf8_decode_unsafe" utf8_decode_unsafe

81 , F.testCase "utf8_decode_unsafe" utf8_decode_unsafe

92− , F.testCase "t197" t197

93 ]

82 ]

tests/text-tests.cabal+3−18

@@ -20,11 +20,6 @@ flag hpc

20 default: False

21 manual: True

23−flag bytestring-builder

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

25− default: False

26− manual: False

27−

28executable text-tests

23executable text-tests

29 main-is: Tests.hs

24 main-is: Tests.hs

@@ -53,21 +48,16 @@ executable text-tests

53 HUnit >= 1.2,

48 HUnit >= 1.2,

54 QuickCheck >= 2.7,

49 QuickCheck >= 2.7,

55 base == 4.*,

50 base == 4.*,

51+ bytestring,

56 deepseq,

52 deepseq,

57 directory,

53 directory,

58− quickcheck-unicode >= 1.0.1.0,

54+ quickcheck-unicode,

59 random,

55 random,

60 test-framework >= 0.4,

56 test-framework >= 0.4,

61 test-framework-hunit >= 0.2,

57 test-framework-hunit >= 0.2,

62 test-framework-quickcheck2 >= 0.2,

58 test-framework-quickcheck2 >= 0.2,

63 text-tests

59 text-tests

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−

71executable text-tests-stdio

61executable text-tests-stdio

72 main-is: Tests/IO.hs

62 main-is: Tests/IO.hs

@@ -147,12 +137,7 @@ library

147 array,

137 array,

148 base == 4.*,

138 base == 4.*,

149 binary,

139 binary,

140+ bytestring,

150 deepseq,

141 deepseq,

151 ghc-prim,

142 ghc-prim,

152 integer-gmp

143 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+7−9

@@ -1,7 +1,7 @@

1name: text

2−version: 1.2.3.0

2+version: 1.2.2.1

3−homepage: https://github.com/haskell/text

3+homepage: https://github.com/bos/text

4−bug-reports: https://github.com/haskell/text/issues

4+bug-reports: https://github.com/bos/text/issues

5synopsis: An efficient packed Unicode text type.

6description:

7 .

39category: Data, Text

40build-type: Simple

41cabal-version: >= 1.8

42−tested-with: GHC==8.2.1, GHC==8.0.2, GHC==7.10.3, GHC==7.8.4,

43− GHC==7.6.3, GHC==7.4.2, GHC==7.2.2, GHC==7.0.4

44extra-source-files:

42extra-source-files:

45 -- scripts/CaseFolding.txt

43 -- scripts/CaseFolding.txt

46 -- scripts/SpecialCasing.txt

44 -- scripts/SpecialCasing.txt

@@ -168,21 +166,21 @@ test-suite tests

168 include-dirs: include

166 include-dirs: include

169

167

170 ghc-options:

168 ghc-options:

171− -Wall -threaded -O1 -rtsopts

169+ -Wall -threaded -O0 -rtsopts

172

170

173 cpp-options:

171 cpp-options:

174 -DASSERTS -DHAVE_DEEPSEQ -DTEST_SUITE

172 -DASSERTS -DHAVE_DEEPSEQ -DTEST_SUITE

175

173

176 build-depends:

174 build-depends:

177 HUnit >= 1.2,

175 HUnit >= 1.2,

178− QuickCheck >= 2.7 && < 2.10,

176+ QuickCheck >= 2.7,

179 array,

177 array,

180 base,

178 base,

181 binary,

179 binary,

182 deepseq,

180 deepseq,

183 directory,

181 directory,

184 ghc-prim,

182 ghc-prim,

185− quickcheck-unicode >= 1.0.1.0,

183+ quickcheck-unicode,

186 random,

184 random,

187 test-framework >= 0.4,

185 test-framework >= 0.4,

188 test-framework-hunit >= 0.2,

186 test-framework-hunit >= 0.2,

@@ -203,7 +201,7 @@ test-suite tests

203

201

204source-repository head

202source-repository head

205 type: git

203 type: git

206− location: https://github.com/haskell/text

204+ location: https://github.com/bos/text

207

205

208source-repository head

206source-repository head

209 type: mercurial

207 type: mercurial