master..c976329 · text

git.iko.soy/text

on c976329 help sign in

.hgtags+1

since splitsnapshots

unifiedsplit

@@ -64,3 +64,4 @@ 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

Data/Text.hs+13−1

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

60 , snoc

61 , append

62 , uncons

63+ , unsnoc

63 , head

64 , head

64 , last

65 , last

65 , tail

66 , tail

@@ -545,6 +546,17 @@ init (Text arr off len) | len <= 0 = emptyError "init"

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

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

546 #-}

547 #-}

547

548

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

550+-- 'Text', or 'Nothing' if empty.

551+unsnoc :: Text -> Maybe (Text, Char)

552+unsnoc (Text arr off len)

553+ | len <= 0 = Nothing

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

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

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

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

558+{-# INLINE [1] unsnoc #-}

559+

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

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

549-- fusion.

561-- fusion.

550null :: Text -> Bool

562null :: Text -> Bool

@@ -1099,8 +1111,8 @@ takeEnd n t@(Text arr off len)

1099iterNEnd :: Int -> Text -> Int

1111iterNEnd :: Int -> Text -> Int

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

1112iterNEnd n t@(Text _arr _off len) = loop (len-1) n

1101 where loop i !m

1113 where loop i !m

1114+ | m <= 0 = i+1

1102 | i <= 0 = 0

1115 | i <= 0 = 0

1103− | m <= 1 = i

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

1116 | otherwise = loop (i+d) (m-1)

1105 where d = reverseIter_ t i

1117 where d = reverseIter_ t i

1106

1118

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 < 0xDB00 || w > 0xD8FF = n

111+ m | w < 0xD800 || w > 0xDBFF = n

112 | otherwise = n+1

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

114

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

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

1{-# LANGUAGE Rank2Types #-}

2-- AUTOMATICALLY GENERATED - DO NOT EDIT

3-- Generated by scripts/CaseMapping.hs

4−-- CaseFolding-8.0.0.txt

4+-- CaseFolding-9.0.0.txt

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

5+-- Date: 2016-03-02, 18:54:54 GMT

6−-- SpecialCasing-8.0.0.txt

6+-- SpecialCasing-9.0.0.txt

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

7+-- Date: 2016-03-02, 18:55:13 GMT

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

10import Data.Char

@@ -371,6 +371,24 @@ 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')

374-- LATIN SMALL LETTER H WITH LINE BELOW

392-- LATIN SMALL LETTER H WITH LINE BELOW

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

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

376-- LATIN SMALL LETTER T WITH DIAERESIS

394-- LATIN SMALL LETTER T WITH DIAERESIS

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

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

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

546-- GREEK CAPITAL LETTER OMEGA WITH PROSGEGRAMMENI

564-- GREEK CAPITAL LETTER OMEGA WITH PROSGEGRAMMENI

547foldMapping '\x1ffc' s = Yield '\x03c9' (CC s '\x03b9' '\x0000')

565foldMapping '\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')

548-- LATIN CAPITAL LETTER J WITH CROSSED-TAIL

568-- LATIN CAPITAL LETTER J WITH CROSSED-TAIL

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

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

550-- LATIN CAPITAL LETTER CHI

570-- LATIN CAPITAL LETTER CHI

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

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

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

738-- ARMENIAN SMALL LIGATURE MEN XEH

758-- ARMENIAN SMALL LIGATURE MEN XEH

739foldMapping '\xfb17' s = Yield '\x0574' (CC s '\x056d' '\x0000')

759foldMapping '\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')

740-- OLD HUNGARIAN CAPITAL LETTER A

832-- OLD HUNGARIAN CAPITAL LETTER A

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

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

742-- OLD HUNGARIAN CAPITAL LETTER AA

834-- OLD HUNGARIAN CAPITAL LETTER AA

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

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

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

840-- OLD HUNGARIAN CAPITAL LETTER US

932-- OLD HUNGARIAN CAPITAL LETTER US

841foldMapping '\x10cb2' s = Yield '\x10cf2' (CC s '\x0000' '\x0000')

933foldMapping '\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')

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

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

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

@@ -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)

110+import Data.Char (isLetter, isSpace)

111import Data.Int (Int64)

112import Data.Text.Internal.Fusion.Types

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

@@ -462,15 +462,16 @@ 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− | letter' -> if letter

468+ | nonSpace -> if letter

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

469+ then lowerMapping c (nonSpace :*: 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)

472− where letter' = isLetter c

473+ letter' = isLetter c

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

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

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

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

475

476

476data Justify i s = Just1 !i !s

477data Justify i s = Just1 !i !s

Data/Text/Lazy.hs+15−6

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

68 , snoc

69 , append

70 , uncons

71+ , unsnoc

71 , head

72 , head

72 , last

73 , last

73 , tail

74 , tail

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

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

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

476-- * Basic functions

477-- * Basic functions

477

478

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

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

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

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

481cons :: Char -> Text -> Text

480cons :: Char -> Text -> Text

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

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

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

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

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

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

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

562 #-}

561 #-}

563

562

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+

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

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

565-- fusion.

573-- fusion.

566null :: Text -> Bool

574null :: Text -> Bool

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

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

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

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

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

1155 where takeChunk acc [] = acc

1163 where takeChunk acc [] = acc

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

1164+ takeChunk acc (t:ts)

1157− then (Chunk t' acc)

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

1166+ = chunk t' acc

1158− else takeChunk (Chunk t' acc) ts

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

1159 where t' = T.takeWhileEnd p t

1168 where t' = T.takeWhileEnd p t

1160{-# INLINE takeWhileEnd #-}

1169{-# INLINE takeWhileEnd #-}

1161

1170

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 decimal digit. Input is consumed

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

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

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

73−-- is case insensitive.

73+-- This function 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 decimal digit. Input is consumed

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

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

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

72−-- is case insensitive.

72+-- This function is case insensitive.

73--

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

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

cbits/cbits.c+5−10

@@ -207,10 +207,8 @@ _hs_text_decode_utf8_state(uint16_t *const dest, size_t *destoff,

207 uint32_t *codepoint0, uint32_t *state0)

208{

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

210− codepoint0, state0);

210+ codepoint0, state0);

211− if (*state0 == UTF8_REJECT)

211+ return *src;

212− ret -=1;

213− return ret;

214}

212}

215

213

216/*

214/*

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

222{

220{

223 uint32_t codepoint;

221 uint32_t codepoint;

224 uint32_t state = UTF8_ACCEPT;

222 uint32_t state = UTF8_ACCEPT;

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

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

226− &codepoint, &state);

224+ &codepoint, &state);

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

228− if (state != UTF8_ACCEPT)

229− ret -= 1;

230− return ret;

225+ return src;

231}

226}

232

227

233void

228void

changelog.md+27−1

@@ -1,10 +1,36 @@

1+1.2.3.0

3+* Bug fix: the lazy `takeWhileEnd` function violated the

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

6+* Spec compliance: `toCaseFold` now follows the Unicode 9.0 spec

7+ (updated from 8.0)

10+1.2.2.2

11+

12+* The `toTitle` function now correctly handles letters that

13+ immediately follow punctuation. Before, `"there's"` would turn into

14+ `"There'S"`. Now, it becomes `"There's"`.

15+

16+* The implementation of unstreaming is faster, resulting in operations

17+ such as `map` and `intersperse` speeding up by up to 30%, with

18+ smaller code generated.

19+

20+* The optimised length comparison function is now more likely to be

21+ used after some rewrite rule tweaking.

22+

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

24+

25+* Bug fix: a logic error in `takeWord16` is fixed.

26+

11.2.2.1

271.2.2.1

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

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

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

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

5 maintainers apologizes for the churn.

31 maintainers apologizes for the churn.

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

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

8 (updated from 7.0)

34 (updated from 7.0)

10* An STG lint error has been fixed

36* An STG lint error has been fixed

tests/Tests/Properties.hs+89−21

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

9 ) where

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

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

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

13import Data.Bits ((.&.))

14−import Data.Char (chr, isDigit, isHexDigit, isLower, isSpace, isUpper, ord)

14+import Data.Char (chr, isDigit, isHexDigit, isLower, isSpace, isLetter, isUpper, ord)

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

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

17import Data.String (IsString(fromString))

@@ -23,6 +23,7 @@ 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)

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

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

27import Prelude hiding (replicate)

28import Prelude hiding (replicate)

28import Test.Framework (Test, testGroup)

29import Test.Framework (Test, testGroup)

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

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

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

31import Test.QuickCheck.Monadic

32import Test.QuickCheck.Monadic

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

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

34+import Test.QuickCheck.Unicode (char)

33import Tests.QuickCheckUtils

35import Tests.QuickCheckUtils

34import Tests.Utils

36import Tests.Utils

35import Text.Show.Functions ()

37import Text.Show.Functions ()

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

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

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

38import qualified Data.ByteString as B

40import qualified Data.ByteString as B

39import qualified Data.ByteString.Lazy as BL

41import qualified Data.ByteString.Lazy as BL

42+import qualified Data.Char as C

40import qualified Data.List as L

43import qualified Data.List as L

41import qualified Data.Text as T

44import qualified Data.Text as T

42import qualified Data.Text.Encoding as E

45import qualified Data.Text.Encoding as E

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

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

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

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

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

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

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

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

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

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

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

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

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

66−tl_unstreamChunks x = f 11 x == f 1000 x

69+tl_unstreamChunks x = f 11 x === f 1000 x

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

79−tl_ascii t = EL.decodeASCII (EL.encodeUtf8 a) == a

82+tl_ascii t = EL.decodeASCII (EL.encodeUtf8 a) === a

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

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

81−t_latin1 t = E.decodeLatin1 (encodeL1 a) == a

84+t_latin1 t = E.decodeLatin1 (encodeL1 a) === a

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

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

83−tl_latin1 t = EL.decodeLatin1 (encodeLazyL1 a) == a

86+tl_latin1 t = EL.decodeLatin1 (encodeLazyL1 a) === a

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

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

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

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

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

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

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

111 let b = E.encodeUtf8 t

114 let b = E.encodeUtf8 t

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

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

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

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

114− in (map snd . feedChunksOf 1 E.streamDecodeUtf8) b == ls

117+ in (map snd . feedChunksOf 1 E.streamDecodeUtf8) b === ls

115

118

116data Badness = Solo | Leading | Trailing

119data Badness = Solo | Leading | Trailing

117 deriving (Eq, Show)

120 deriving (Eq, Show)

@@ -177,6 +180,29 @@ genInvalidUTF8 = B.pack <$> oneof [

177 k <- choose (0,n)

180 k <- choose (0,n)

178 vectorOf k gen

181 vectorOf k gen

179

182

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+

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

206s_Eq s = (s==) `eq` ((S.streamList s==) . S.streamList)

181 where _types = s :: String

207 where _types = s :: String

182sf_Eq p s =

208sf_Eq p s =

@@ -201,8 +227,8 @@ t_mconcat = unsquare $

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

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

202tl_mconcat = unsquare $

228tl_mconcat = unsquare $

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

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

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

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

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

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

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

232t_IsString = fromString `eqP` (T.unpack . fromString)

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

233tl_IsString = fromString `eqP` (TL.unpack . fromString)

208

234

@@ -229,6 +255,13 @@ sf_uncons p = (uncons . L.filter p) `eqP`

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

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

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

256t_uncons = uncons `eqP` (fmap (second unpackS) . T.uncons)

231tl_uncons = uncons `eqP` (fmap (second unpackS) . TL.uncons)

257tl_uncons = uncons `eqP` (fmap (second unpackS) . TL.uncons)

258+

259+unsnoc xs@(_:_) = Just (init xs, last xs)

260+unsnoc [] = Nothing

261+

262+t_unsnoc = unsnoc `eqP` (fmap (first unpackS) . T.unsnoc)

263+tl_unsnoc = unsnoc `eqP` (fmap (first unpackS) . TL.unsnoc)

264+

232s_head = head `eqP` S.head

265s_head = head `eqP` S.head

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

266sf_head p = (head . L.filter p) `eqP` (S.head . S.filter p)

234t_head = head `eqP` T.head

267t_head = head `eqP` T.head

@@ -326,6 +359,19 @@ t_toUpper_upper t = p (T.toUpper t) >= p t

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

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

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

360tl_toUpper_upper t = p (TL.toUpper t) >= p t

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

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

362+t_toTitle_title t = all (<= 1) (caps w)

363+ where caps = fmap (T.length . T.filter isUpper) . T.words . T.toTitle

364+ -- TIL: there exist uppercase-only letters

365+ w = T.filter (\c -> if C.isUpper c then C.toLower c /= c else True) t

366+t_toTitle_1stNotLower = and . notLow . T.toTitle . T.filter stable

367+ where notLow = mapMaybe (fmap (not . isLower) . (T.find isLetter)) . T.words

368+ -- Surprise! The Spanish/Portuguese ordinal indicators changed

369+ -- from category Ll (letter, lowercase) to Lo (letter, other)

370+ -- in Unicode 7.0

371+ -- Oh, and there exist lowercase-only letters (see previous test)

372+ stable c = if isLower c

373+ then C.toUpper c /= c

374+ else c /= '\170' && c /= '\186'

329

375

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

376justifyLeft k c xs = xs ++ L.replicate (k - length xs) c

331justifyRight m n xs = L.replicate (m - length xs) n ++ xs

377justifyRight m n xs = L.replicate (m - length xs) n ++ xs

@@ -514,12 +560,20 @@ s_takeWhile p = L.takeWhile p `eqP` (unpackS . S.takeWhile p)

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

560s_takeWhile_s p = L.takeWhile p `eqP` (unpackS . S.unstream . S.takeWhile p)

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

561sf_takeWhile q p = (L.takeWhile p . L.filter q) `eqP`

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

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

563+noMatch = do

564+ c <- char

565+ d <- suchThat char (/= c)

566+ return (c,d)

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

567t_takeWhile p = L.takeWhile p `eqP` (unpackS . T.takeWhile p)

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

568tl_takeWhile p = L.takeWhile p `eqP` (unpackS . TL.takeWhile p)

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

569t_takeWhileEnd p = (L.reverse . L.takeWhile p . L.reverse) `eqP`

520 (unpackS . T.takeWhileEnd p)

570 (unpackS . T.takeWhileEnd p)

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

572+ T.takeWhileEnd (==d) (T.snoc t c)

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

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

522 (unpackS . TL.takeWhileEnd p)

574 (unpackS . TL.takeWhileEnd p)

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

576+ TL.takeWhileEnd (==d) (TL.snoc t c)

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

577s_dropWhile p = L.dropWhile p `eqP` (unpackS . S.dropWhile p)

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

578s_dropWhile_s p = L.dropWhile p `eqP` (unpackS . S.unstream . S.dropWhile p)

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

579sf_dropWhile q p = (L.dropWhile p . L.filter q) `eqP`

@@ -720,7 +774,7 @@ tl_indices (NotEmpty s) = lazyIndices s `eq` S.indices s

720 conc = T.concat . TL.toChunks

774 conc = T.concat . TL.toChunks

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

775t_indices_occurs = unsquare $ \(NotEmpty t) ts ->

722 let s = T.intercalate t ts

776 let s = T.intercalate t ts

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

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

724

778

725-- Bit shifts.

779-- Bit shifts.

726shiftL w = forAll (choose (0,width-1)) $ \k -> Bits.shiftL w k == U.shiftL w k

780shiftL w = forAll (choose (0,width-1)) $ \k -> Bits.shiftL w k == U.shiftL w k

@@ -812,18 +866,18 @@ tb_formatRealFloat_double (a::Double) = tb_formatRealFloat a

812-- Reading.

866-- Reading.

813

867

814t_decimal (n::Int) s =

868t_decimal (n::Int) s =

815− T.signed T.decimal (T.pack (show n) `T.append` t) == Right (n,t)

869+ T.signed T.decimal (T.pack (show n) `T.append` t) === Right (n,t)

816 where t = T.dropWhile isDigit s

870 where t = T.dropWhile isDigit s

817tl_decimal (n::Int) s =

871tl_decimal (n::Int) s =

818− TL.signed TL.decimal (TL.pack (show n) `TL.append` t) == Right (n,t)

872+ TL.signed TL.decimal (TL.pack (show n) `TL.append` t) === Right (n,t)

819 where t = TL.dropWhile isDigit s

873 where t = TL.dropWhile isDigit s

820t_hexadecimal m s ox =

874t_hexadecimal m s ox =

821− T.hexadecimal (T.concat [p, T.pack (showHex n ""), t]) == Right (n,t)

875+ T.hexadecimal (T.concat [p, T.pack (showHex n ""), t]) === Right (n,t)

822 where t = T.dropWhile isHexDigit s

876 where t = T.dropWhile isHexDigit s

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

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

824 n = getPositive m :: Int

878 n = getPositive m :: Int

825tl_hexadecimal m s ox =

879tl_hexadecimal m s ox =

826− TL.hexadecimal (TL.concat [p, TL.pack (showHex n ""), t]) == Right (n,t)

880+ TL.hexadecimal (TL.concat [p, TL.pack (showHex n ""), t]) === Right (n,t)

827 where t = TL.dropWhile isHexDigit s

881 where t = TL.dropWhile isHexDigit s

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

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

829 n = getPositive m :: Int

883 n = getPositive m :: Int

@@ -867,11 +921,11 @@ tl_write_read_line e m b t = write_read head TL.filter TL.hPutStrLn

867

921

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

922t_dropWord16 m t = dropWord16 m t `T.isSuffixOf` t

869t_takeWord16 m t = takeWord16 m t `T.isPrefixOf` t

923t_takeWord16 m t = takeWord16 m t `T.isPrefixOf` t

870−t_take_drop_16 m t = T.append (takeWord16 n t) (dropWord16 n t) == t

924+t_take_drop_16 m t = T.append (takeWord16 n t) (dropWord16 n t) === t

871 where n = small m

925 where n = small m

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

926t_use_from t = monadicIO $ assert . (==t) =<< run (useAsPtr t fromPtr)

873

927

874−t_copy t = T.copy t == t

928+t_copy t = T.copy t === t

875

929

876-- Regression tests.

930-- Regression tests.

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

931s_filter_eq s = S.filter p t == S.streamList (filter p s)

@@ -924,6 +978,14 @@ tests =

924 testGroup "errors" [

978 testGroup "errors" [

925 testProperty "t_utf8_err" t_utf8_err,

979 testProperty "t_utf8_err" t_utf8_err,

926 testProperty "t_utf8_err'" t_utf8_err'

980 testProperty "t_utf8_err'" t_utf8_err'

981+ ],

982+ testGroup "error recovery" [

983+ testProperty "t_decode_with_error2" t_decode_with_error2,

984+ testProperty "t_decode_with_error3" t_decode_with_error3,

985+ testProperty "t_decode_with_error4" t_decode_with_error4,

986+ testProperty "t_decode_with_error2'" t_decode_with_error2',

987+ testProperty "t_decode_with_error3'" t_decode_with_error3',

988+ testProperty "t_decode_with_error4'" t_decode_with_error4'

927 ]

989 ]

928 ],

990 ],

929

991

@@ -967,6 +1029,8 @@ tests =

967 testProperty "sf_uncons" sf_uncons,

1029 testProperty "sf_uncons" sf_uncons,

968 testProperty "t_uncons" t_uncons,

1030 testProperty "t_uncons" t_uncons,

969 testProperty "tl_uncons" tl_uncons,

1031 testProperty "tl_uncons" tl_uncons,

1032+ testProperty "t_unsnoc" t_unsnoc,

1033+ testProperty "tl_unsnoc" tl_unsnoc,

970 testProperty "s_head" s_head,

1034 testProperty "s_head" s_head,

971 testProperty "sf_head" sf_head,

1035 testProperty "sf_head" sf_head,

972 testProperty "t_head" t_head,

1036 testProperty "t_head" t_head,

@@ -1030,7 +1094,9 @@ tests =

1030 testProperty "tl_toLower_lower" tl_toLower_lower,

1094 testProperty "tl_toLower_lower" tl_toLower_lower,

1031 testProperty "t_toUpper_length" t_toUpper_length,

1095 testProperty "t_toUpper_length" t_toUpper_length,

1032 testProperty "t_toUpper_upper" t_toUpper_upper,

1096 testProperty "t_toUpper_upper" t_toUpper_upper,

1033− testProperty "tl_toUpper_upper" tl_toUpper_upper

1097+ testProperty "tl_toUpper_upper" tl_toUpper_upper,

1098+ testProperty "t_toTitle_title" t_toTitle_title,

1099+ testProperty "t_toTitle_1stNotLower" t_toTitle_1stNotLower

1034 ],

1100 ],

1035

1101

1036 testGroup "justification" [

1102 testGroup "justification" [

@@ -1147,7 +1213,9 @@ tests =

1147 testProperty "t_takeWhile" t_takeWhile,

1213 testProperty "t_takeWhile" t_takeWhile,

1148 testProperty "tl_takeWhile" tl_takeWhile,

1214 testProperty "tl_takeWhile" tl_takeWhile,

1149 testProperty "t_takeWhileEnd" t_takeWhileEnd,

1215 testProperty "t_takeWhileEnd" t_takeWhileEnd,

1216+ testProperty "t_takeWhileEnd_null" t_takeWhileEnd_null,

1150 testProperty "tl_takeWhileEnd" tl_takeWhileEnd,

1217 testProperty "tl_takeWhileEnd" tl_takeWhileEnd,

1218+ testProperty "tl_takeWhileEnd_null" tl_takeWhileEnd_null,

1151 testProperty "sf_dropWhile" sf_dropWhile,

1219 testProperty "sf_dropWhile" sf_dropWhile,

1152 testProperty "s_dropWhile" s_dropWhile,

1220 testProperty "s_dropWhile" s_dropWhile,

1153 testProperty "s_dropWhile_s" s_dropWhile_s,

1221 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` arbitrary

109+ arbitrary = T.pack `fmap` string

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

111

112instance Arbitrary TL.Text where

tests/text-tests.cabal+18−3

@@ -20,6 +20,11 @@ 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+

23executable text-tests

28executable text-tests

24 main-is: Tests.hs

29 main-is: Tests.hs

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

48 HUnit >= 1.2,

53 HUnit >= 1.2,

49 QuickCheck >= 2.7,

54 QuickCheck >= 2.7,

50 base == 4.*,

55 base == 4.*,

51− bytestring,

52 deepseq,

56 deepseq,

53 directory,

57 directory,

54− quickcheck-unicode,

58+ quickcheck-unicode >= 1.0.1.0,

55 random,

59 random,

56 test-framework >= 0.4,

60 test-framework >= 0.4,

57 test-framework-hunit >= 0.2,

61 test-framework-hunit >= 0.2,

58 test-framework-quickcheck2 >= 0.2,

62 test-framework-quickcheck2 >= 0.2,

59 text-tests

63 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+

61executable text-tests-stdio

71executable text-tests-stdio

62 main-is: Tests/IO.hs

72 main-is: Tests/IO.hs

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

137 array,

147 array,

138 base == 4.*,

148 base == 4.*,

139 binary,

149 binary,

140− bytestring,

141 deepseq,

150 deepseq,

142 ghc-prim,

151 ghc-prim,

143 integer-gmp

152 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+2−2

@@ -1,5 +1,5 @@

1name: text

2−version: 1.2.2.1

2+version: 1.2.3.0

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

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

5synopsis: An efficient packed Unicode text type.

@@ -180,7 +180,7 @@ test-suite tests

180 deepseq,

181 directory,

182 ghc-prim,

183− quickcheck-unicode,

183+ quickcheck-unicode >= 1.0.1.0,

184 random,

185 test-framework >= 0.4,

186 test-framework-hunit >= 0.2,