module Text.Highlighting.Kate.Syntax.Awk
(highlight, parseExpression, syntaxName, syntaxExtensions)
where
import Text.Highlighting.Kate.Types
import Text.Highlighting.Kate.Common
import qualified Text.Highlighting.Kate.Syntax.Alert
import Text.ParserCombinators.Parsec hiding (State)
import Data.Map (fromList)
import Control.Monad.State
import Data.Char (isSpace)
import Data.Maybe (fromMaybe)
import qualified Data.Set as Set
syntaxName :: String
syntaxName = "AWK"
syntaxExtensions :: String
syntaxExtensions = "*.awk"
highlight :: String -> [SourceLine]
highlight input = evalState (mapM parseSourceLine $ lines input) startingState
parseSourceLine :: String -> State SyntaxState SourceLine
parseSourceLine = mkParseSourceLine parseExpressionInternal pEndLine
parseExpression :: KateParser Token
parseExpression = do
st <- getState
let oldLang = synStLanguage st
setState $ st { synStLanguage = "AWK" }
context <- currentContext <|> (pushContext "Base" >> currentContext)
result <- parseRules context
optional $ eof >> pEndLine
updateState $ \st -> st { synStLanguage = oldLang }
return result
startingState = SyntaxState {synStContexts = fromList [("AWK",["Base"])], synStLanguage = "AWK", synStLineNumber = 0, synStPrevChar = '\n', synStPrevNonspace = False, synStCaseSensitive = True, synStKeywordCaseSensitive = True, synStCaptures = []}
pEndLine = do
updateState $ \st -> st{ synStPrevNonspace = False }
context <- currentContext
case context of
"Base" -> return ()
"String" -> return ()
"Comment" -> (popContext) >> pEndLine
_ -> return ()
withAttribute attr txt = do
when (null txt) $ fail "Parser matched no text"
updateState $ \st -> st { synStPrevChar = last txt
, synStPrevNonspace = synStPrevNonspace st || not (all isSpace txt) }
return (attr, txt)
parseExpressionInternal = do
context <- currentContext
parseRules context <|> (pDefault >>= withAttribute (fromMaybe NormalTok $ 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_'5c'24'5bA'2dZa'2dz0'2d9'5f'5d'2b = compileRegex "\\$[A-Za-z0-9_]+"
defaultAttributes = [("Base",NormalTok),("String",StringTok),("Comment",CommentTok)]
parseRules "Base" =
(((pRegExpr regex_'5cb'28BEGIN'7cEND'29'5cb >>= withAttribute StringTok))
<|>
((pDetectChar False '{' >>= withAttribute KeywordTok))
<|>
((pDetectChar False '}' >>= withAttribute KeywordTok))
<|>
((pDetectChar False '#' >>= withAttribute CommentTok) >>~ pushContext "Comment")
<|>
((pDetectChar False '"' >>= withAttribute StringTok) >>~ pushContext "String")
<|>
((pKeyword " \n\t.():!+,-<=>%&*/;?[]^{|}~\\" list_keywords >>= withAttribute KeywordTok))
<|>
((pKeyword " \n\t.():!+,-<=>%&*/;?[]^{|}~\\" list_builtins >>= withAttribute DataTypeTok))
<|>
((pKeyword " \n\t.():!+,-<=>%&*/;?[]^{|}~\\" list_functions >>= withAttribute FunctionTok))
<|>
((pFloat >>= withAttribute FloatTok))
<|>
((pInt >>= withAttribute DecValTok))
<|>
((pRegExpr regex_'5c'24'5bA'2dZa'2dz0'2d9'5f'5d'2b >>= withAttribute OtherTok)))
parseRules "String" =
(((pDetectChar False '"' >>= withAttribute StringTok) >>~ (popContext))
<|>
((pHlCStringChar >>= withAttribute StringTok)))
parseRules "Comment" =
((Text.Highlighting.Kate.Syntax.Alert.parseExpression >>= ((withAttribute CommentTok) . snd)))
parseRules "" = parseRules "Base"
parseRules x = fail $ "Unknown context" ++ x