| 1 |
{-# LANGUAGE FlexibleContexts #-} |
| 2 |
|
| 3 |
-- ------------------------------------------------------------ |
| 4 |
|
| 5 |
{- | |
| 6 |
Copyright : Copyright (C) 2014- Uwe Schmidt |
| 7 |
License : MIT |
| 8 |
|
| 9 |
Maintainer : Uwe Schmidt ([email protected]) |
| 10 |
Stability : stable |
| 11 |
Portability: portable |
| 12 |
|
| 13 |
csh style Glob Pattern Parser for Regular Expressions |
| 14 |
-} |
| 15 |
|
| 16 |
-- ------------------------------------------------------------ |
| 17 |
|
| 18 |
module Text.Regex.Glob.Generic.RegexParser |
| 19 |
( parseRegex |
| 20 |
, parseRegexNoCase |
| 21 |
) |
| 22 |
where |
| 23 |
|
| 24 |
import Data.Char (isLower, isUpper, |
| 25 |
toLower, toUpper) |
| 26 |
|
| 27 |
import Text.ParserCombinators.Parsec |
| 28 |
import Text.Regex.XMLSchema.Generic.Regex |
| 29 |
import Text.Regex.XMLSchema.Generic.StringLike |
| 30 |
|
| 31 |
-- ------------------------------------------------------------ |
| 32 |
|
| 33 |
-- | parse a glob pattern |
| 34 |
|
| 35 |
parseRegex :: StringLike s => s -> GenRegex s |
| 36 |
parseRegex |
| 37 |
= parseRegex' mkSymRng . toString |
| 38 |
|
| 39 |
parseRegexNoCase :: StringLike s => s -> GenRegex s |
| 40 |
parseRegexNoCase |
| 41 |
= parseRegex' mkNoCaseSymRng . toString |
| 42 |
|
| 43 |
parseRegex' :: StringLike s => (Char -> Char -> GenRegex s) -> String -> GenRegex s |
| 44 |
parseRegex' mkS |
| 45 |
= either (mkZero' . ("syntax error: " ++) . show) id |
| 46 |
. |
| 47 |
parse ( do |
| 48 |
r <- pattern mkS |
| 49 |
eof |
| 50 |
return r |
| 51 |
) "" |
| 52 |
|
| 53 |
-- ------------------------------------------------------------ |
| 54 |
|
| 55 |
pattern :: StringLike s => (Char -> Char -> GenRegex s) -> Parser (GenRegex s) |
| 56 |
pattern mkS |
| 57 |
= many part >>= return . mkSeqs |
| 58 |
where |
| 59 |
-- part :: Parser (GenRegex s) |
| 60 |
part |
| 61 |
= ( many1 (noneOf "\\?*[{") >>= return . mkWord' ) |
| 62 |
<|> |
| 63 |
( char '?' >> return mkDot ) |
| 64 |
<|> |
| 65 |
( char '*' >> return mkAll ) |
| 66 |
<|> |
| 67 |
( between (char '{') (char '}') wordList ) |
| 68 |
<|> |
| 69 |
( between (char '[') (char ']') charSet ) |
| 70 |
<|> |
| 71 |
( do c <- char '\\' >> anyChar |
| 72 |
return $ mkS c c |
| 73 |
) |
| 74 |
mkWord' |
| 75 |
= mkSeqs . map (\ c -> mkS c c) |
| 76 |
|
| 77 |
-- wordList :: Parser (GenRegex s) |
| 78 |
wordList |
| 79 |
= sepBy (many1 (noneOf ",}")) (char ',') |
| 80 |
>>= return . foldr mkAlt (mkZero' "") . map mkWord' |
| 81 |
|
| 82 |
-- charSet :: Parser (GenRegex s) |
| 83 |
charSet |
| 84 |
= ( do p1 <- charSet' anyChar |
| 85 |
ps <- many $ charSet' (noneOf "]") |
| 86 |
return $ foldr mkAlt (mkZero' "") (p1 : ps) |
| 87 |
) |
| 88 |
where |
| 89 |
charSet' cp |
| 90 |
= do c1 <- cp |
| 91 |
c2 <- rest c1 |
| 92 |
return $ mkS c1 c2 |
| 93 |
rest c1 |
| 94 |
= option c1 (char '-' >> anyChar) |
| 95 |
|
| 96 |
-- ------------------------------------------------------------ |
| 97 |
|
| 98 |
mkNoCaseSymRng :: StringLike s => Char -> Char -> GenRegex s |
| 99 |
mkNoCaseSymRng c1 c2 |
| 100 |
| isLower c1 |
| 101 |
&& |
| 102 |
isLower c2 |
| 103 |
= mkAlt (mkSymRng (toUpper c1) (toUpper c2)) (mkSymRng c1 c2) |
| 104 |
| isUpper c1 |
| 105 |
&& |
| 106 |
isUpper c2 |
| 107 |
= mkAlt (mkSymRng (toLower c1) (toLower c2)) (mkSymRng c1 c2) |
| 108 |
| otherwise |
| 109 |
= mkSymRng c1 c2 |
| 110 |
|
| 111 |
-- ------------------------------------------------------------ |