úÎ.˜-Œportable experimental)Daniel Fischer <daniel.is.fischer@web.de>Unicode hyphen, U+2010 Low line or underscore, U+005F Double low line, U+2017 Wide low line, U+FF3F    portable experimental)Daniel Fischer <daniel.is.fischer@web.de><Transform identifiers in (haddock-produced) HTML files from ! camelCase to separated_words. 8The separation character is freely choosable, but it is ' recommended to take one of those in Transform.Separators. Since 5 underscore-separated identifiers are used in some  libraries, choosing lowLine may lead to confusion. 9The separation character is inserted between a lowercase ? character and an uppercase character immediately following. B If that uppercase character is followed by a lowercase letter, ) it is also transformed to lower case.  1 literals appearing in haddock comments are also & transformed. Deal with it or yell. :Transform identifiers in (non-literate) source files from ! camelCase to separated_words. 8The separation character is freely choosable, but it is ' recommended to take one of those in Transform.Separators. Since 5 underscore-separated identifiers are used in some  libraries, choosing lowLine may lead to confusion. 5 On the other hand, it is the only one which has a 4 fighting chance of producing valid Haskell code. 9The separation character is inserted between a lowercase ? character and an uppercase character immediately following. A If that uppercase charcter is followed by a lowercase letter, ) it is also transformed to lower case. 7Operators including two or more consecutive dashes are  handled correctly, i.e. |-- or --: are not treated # as the start of a line-comment. $Single quotes in identifiers, as in foldl' or f'2''d, are ? not considered to begin a character literal. An unfortunate > consequence of that and the simple algorithm is that in an  expression like   replicate 5'\\' :(with no space between number and character literal), the = closing quote is considered to begin a character literal. @Comments are not transformed, which may lead to inconsistencies = between code and comments. That may change in the future. portable experimental)Daniel Fischer <daniel.is.fischer@web.de>8Transform Haskell code written in separated_words style ' to the more common camelCase style. camelSource sep source removes all occurences of sep  in identifiers in source between two letters of which 3 the first is in lower case after processing and ( transforms the second to upper case.  Thus  camelSource '_' "a_b_c" == "aB_c" since after < processing the first underscore, the second is no longer # preceded by a lowercase letter. Comments (and  # literals) are not transformed, so , haddock comments may need manual fixing.       restyle-0.1.0Data.Transform.SeparatorsData.Transform.UnCamelData.Transform.CamelData.Transform.UtilshyphenlowLine doubleLowLine wideLowLine unCamelHTML unCamelSource camelSourceisOpCharisIdChar isIdPlain nestedComment lineCommentstringbaseGHC.BaseStringhumpstepskipTagcode removeSep