module Text.Highlighting.Kate.Syntax.Awk ( highlight, parseExpression, syntaxName, syntaxExtensions ) where
import Text.Highlighting.Kate.Definitions
import Text.Highlighting.Kate.Common
import qualified Text.Highlighting.Kate.Syntax.Alert
import Text.ParserCombinators.Parsec
import Control.Monad (when)
import Data.Map (fromList)
import Data.Maybe (fromMaybe, maybeToList)
import qualified Data.Set as Set
syntaxName :: String
syntaxName = "AWK"
syntaxExtensions :: String
syntaxExtensions = "*.awk"
highlight :: String -> Either String [SourceLine]
highlight input =
case runParser parseSource startingState "source" input of
Left err -> Left $ show err
Right result -> Right result
parseExpression :: GenParser Char SyntaxState LabeledSource
parseExpression = do
st <- getState
let oldLang = synStLanguage st
setState $ st { synStLanguage = "AWK" }
context <- currentContext <|> (pushContext "Base" >> currentContext)
result <- parseRules context
updateState $ \st -> st { synStLanguage = oldLang }
return result
parseSource = do
lineContents <- lookAhead wholeLine
updateState $ \st -> st { synStCurrentLine = lineContents }
result <- manyTill parseSourceLine eof
return $ map normalizeHighlighting result
startingState = SyntaxState {synStContexts = fromList [("AWK",["Base"])], synStLanguage = "AWK", synStCurrentLine = "", synStCharsParsedInLine = 0, synStPrevChar = '\n', synStCaseSensitive = True, synStKeywordCaseSensitive = True, synStCaptures = []}
parseSourceLine = manyTill parseExpressionInternal pEndLine
pEndLine = do
newline <|> (eof >> return '\n')
context <- currentContext
case context of
"Base" -> return ()
"String" -> return ()
"Comment" -> (popContext >> return ())
_ -> return ()
lineContents <- lookAhead wholeLine
updateState $ \st -> st { synStCurrentLine = lineContents, synStCharsParsedInLine = 0, synStPrevChar = '\n' }
withAttribute attr txt = do
when (null txt) $ fail "Parser matched no text"
let labs = attr : maybeToList (lookup attr styles)
st <- getState
let oldCharsParsed = synStCharsParsedInLine st
let prevchar = if null txt then '\n' else last txt
updateState $ \st -> st { synStCharsParsedInLine = oldCharsParsed + length txt, synStPrevChar = prevchar }
return (labs, txt)
styles = [("Keyword","kw"),("Builtin","dt"),("Function","fu"),("Decimal","dv"),("Float","fl"),("String","st"),("Comment","co"),("Pattern","st"),("Field","ot")]
parseExpressionInternal = do
context <- currentContext
parseRules context <|> (pDefault >>= withAttribute (fromMaybe "" $ lookup context defaultAttributes))
list_keywords = Set.fromList $ words $ "if else while do for in continue break print printf getline function return next exit"
list_builtins = Set.fromList $ words $ "ARGC ARGV CONVFMT ENVIRON FILENAME FNR FS NF NR OFMT OFS ORS RS RSTART RLENGTH SUBSEP"
list_functions = Set.fromList $ words $ "gsub gensub index length match split sprintf sub substr tolower toupper atan2 cos exp int log rand sin sqrt srand close fflush system"
regex_'5cb'28BEGIN'7cEND'29'5cb = compileRegex "\\b(BEGIN|END)\\b"
regex_'2f'28'5b'5e'5c'2f'5b'5d'7c'5c'5c'2e'7c'5c'5b'5c'5d'3f'28'5c'5b'5b'5e'5d'5d'2b'5c'5d'7c'2e'29'2b'5c'5d'29'2b'2f = compileRegex "/([^\\/[]|\\\\.|\\[\\]?(\\[[^]]+\\]|.)+\\])+/"
regex_'5c'24'5bA'2dZa'2dz0'2d9'5f'5d'2b = compileRegex "\\$[A-Za-z0-9_]+"
defaultAttributes = [("Base","Normal"),("String","String"),("Comment","Comment")]
parseRules "Base" =
do (attr, result) <- (((pRegExpr regex_'5cb'28BEGIN'7cEND'29'5cb >>= withAttribute "Pattern"))
<|>
((pRegExpr regex_'2f'28'5b'5e'5c'2f'5b'5d'7c'5c'5c'2e'7c'5c'5b'5c'5d'3f'28'5c'5b'5b'5e'5d'5d'2b'5c'5d'7c'2e'29'2b'5c'5d'29'2b'2f >>= withAttribute "Pattern"))
<|>
((pDetectChar False '{' >>= withAttribute "Keyword"))
<|>
((pDetectChar False '}' >>= withAttribute "Keyword"))
<|>
((pDetectChar False '#' >>= withAttribute "Comment") >>~ pushContext "Comment")
<|>
((pDetectChar False '"' >>= withAttribute "String") >>~ pushContext "String")
<|>
((pKeyword " \n\t.():!+,-<=>%&*/;?[]^{|}~\\" list_keywords >>= withAttribute "Keyword"))
<|>
((pKeyword " \n\t.():!+,-<=>%&*/;?[]^{|}~\\" list_builtins >>= withAttribute "Builtin"))
<|>
((pKeyword " \n\t.():!+,-<=>%&*/;?[]^{|}~\\" list_functions >>= withAttribute "Function"))
<|>
((pFloat >>= withAttribute "Float"))
<|>
((pInt >>= withAttribute "Decimal"))
<|>
((pRegExpr regex_'5c'24'5bA'2dZa'2dz0'2d9'5f'5d'2b >>= withAttribute "Field")))
return (attr, result)
parseRules "String" =
do (attr, result) <- (((pDetectChar False '"' >>= withAttribute "String") >>~ (popContext >> return ()))
<|>
((pHlCStringChar >>= withAttribute "String")))
return (attr, result)
parseRules "Comment" =
do (attr, result) <- ((Text.Highlighting.Kate.Syntax.Alert.parseExpression))
return (attr, result)
parseRules x = fail $ "Unknown context" ++ x