c7b94a9 · text · git.iko.soy

git.iko.soy/text

Move breakSubstring into Compat module.

--HG-- extra : convert_revision : 4e3dadbee683f8925eed464ac8fee1dd1670cdf5

commitc7b94a9a8b32d8c782f794a9fbdcef7c526b3e3d authorBryan O'Sullivan <bos@serpentine.com> dateSun 5 Jul 2009, 06:15 (17 years ago) parentf584cbc

2 files changed+35−34

compare with master

unifiedsplit

Data/Text.hs−33 Data/Text/Compat.hs+35−1

Data/Text.hs−33

@@ -118,7 +118,6 @@ module Data.Text

118118 , splitTimes

119119 , splitChar

120120 , splitWith

121− , breakSubstring

122121

123122 -- ** Breaking into lines and words

124123 , lines

@@ -941,38 +940,6 @@ partition :: (Char -> Bool) -> Text -> (Text, Text)

941940partition p t = (filter p t, filter (not . p) t)

942941{-# INLINE partition #-}

943942

944−-- | /O(n)/ Break a string on a substring, returning a pair of the

945−-- part of the string prior to the match, and the rest of the string.

946−--

947−-- The following relationship holds:

948−--

949−-- > break (==c) l == breakSubstring (singleton c) l

950−--

951−-- For example, to tokenise a string, dropping delimiters:

952−--

953−-- > tokenise x y = h : if null t then [] else tokenise x (drop (length x) t)

954−-- > where (h,t) = breakSubstring x y

955−--

956−-- To skip to the first occurence of a string:

957−--

958−-- > snd (breakSubstring x y)

959−--

960−-- To take the parts of a string before a delimiter:

961−--

962−-- > fst (breakSubstring x y)

963−--

964−breakSubstring :: Text -- ^ String to search for

965− -> Text -- ^ String to search in

966− -> (Text,Text) -- ^ Head and tail of string broken at substring

967−

968−breakSubstring pat src = search 0 src

969− where

970− search !n !s

971− | null s = (src,empty) -- not found

972− | pat `isPrefixOf` s = (take n src,s)

973− | otherwise = search (n+1) (unsafeTail s)

974−{-# INLINE breakSubstring #-}

975−

976943-- | /O(n)/ 'filter', applied to a predicate and a 'Text',

977944-- returns a 'Text' containing those characters that satisfy the

978945-- predicate.

Data/Text/Compat.hs+35−1

@@ -1,3 +1,4 @@

1+{-# LANGUAGE BangPatterns #-}

12{-# OPTIONS_GHC -fno-warn-unused-imports #-}

23-- |

34-- Module : Data.Text.Compat

@@ -144,8 +145,41 @@ module Data.Text.Compat

144145

145146import Data.ByteString (ByteString)

146147import Data.Text hiding (split)

147−import Prelude (Char)

148+import Data.Text.Unsafe (unsafeTail)

149+import Prelude (Char, (+), otherwise)

148150

149151split :: Char -> Text -> [Text]

150152split = splitChar

151153{-# INLINE split #-}

154+

155+-- | /O(n)/ Break a string on a substring, returning a pair of the

156+-- part of the string prior to the match, and the rest of the string.

157+--

158+-- The following relationship holds:

159+--

160+-- > break (==c) l == breakSubstring (singleton c) l

161+--

162+-- For example, to tokenise a string, dropping delimiters:

163+--

164+-- > tokenise x y = h : if null t then [] else tokenise x (drop (length x) t)

165+-- > where (h,t) = breakSubstring x y

166+--

167+-- To skip to the first occurence of a string:

168+--

169+-- > snd (breakSubstring x y)

170+--

171+-- To take the parts of a string before a delimiter:

172+--

173+-- > fst (breakSubstring x y)

174+--

175+breakSubstring :: Text -- ^ String to search for

176+ -> Text -- ^ String to search in

177+ -> (Text,Text) -- ^ Head and tail of string broken at substring

178+

179+breakSubstring pat src = search 0 src

180+ where

181+ search !n !s

182+ | null s = (src,empty) -- not found

183+ | pat `isPrefixOf` s = (take n src,s)

184+ | otherwise = search (n+1) (unsafeTail s)

185+{-# INLINE breakSubstring #-}