module Text.Highlighting.Kate.Syntax.Dtd ( 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 = "DTD"
syntaxExtensions :: String
syntaxExtensions = "*.dtd"
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 = "DTD" }
context <- currentContext <|> (pushContext "Normal" >> 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 [("DTD",["Normal"])], synStLanguage = "DTD", 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
"Normal" -> return ()
"Comment" -> return ()
"PI" -> return ()
"Declaration" -> return ()
"String" -> return ()
"InlineComment" -> (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 = [("Comment","co"),("Processing Instruction","kw"),("Declaration","dt"),("Name","fu"),("Delimiter","dv"),("Symbol","fl"),("Keyword","kw"),("String","st"),("Entity","dv"),("Local","dv")]
parseExpressionInternal = do
context <- currentContext
parseRules context <|> (pDefault >>= withAttribute (fromMaybe "" $ lookup context defaultAttributes))
list_Category = Set.fromList $ words $ "EMPTY ANY CDATA ID IDREF IDREFS NMTOKEN NMTOKENS ENTITY ENTITIES NOTATION PUBLIC SYSTEM NDATA"
list_Keywords = Set.fromList $ words $ "#PCDATA #REQUIRED #IMPLIED #FIXED"
regex_'28'2d'7cO'29'5cs'28'2d'7cO'29 = compileRegex "(-|O)\\s(-|O)"
regex_'28'25'7c'26'29'28'23'5b0'2d9'5d'2b'7c'23'5bxX'5d'5b0'2d9A'2dFa'2df'5d'2b'7c'5b'5c'2d'5cw'5cd'5c'2e'3a'5f'5d'2b'29'3b = compileRegex "(%|&)(#[0-9]+|#[xX][0-9A-Fa-f]+|[\\-\\w\\d\\.:_]+);"
regex_'25'5cs = compileRegex "%\\s"
regex_'5cb'5b'5c'2d'5cw'5cd'5c'2e'3a'5f'5d'2b'5cb = compileRegex "\\b[\\-\\w\\d\\.:_]+\\b"
regex_'25'5b'5c'2d'5cw'5cd'5c'2e'3a'5f'5d'2b'3b = compileRegex "%[\\-\\w\\d\\.:_]+;"
defaultAttributes = [("Normal","Normal"),("Comment","Comment"),("PI","Normal"),("Declaration","Normal"),("String","String"),("InlineComment","Comment")]
parseRules "Normal" =
do (attr, result) <- (((pDetectSpaces >>= withAttribute "Normal"))
<|>
((pString False "<!--" >>= withAttribute "Comment") >>~ pushContext "Comment")
<|>
((pString False "<?xml" >>= withAttribute "Processing Instruction") >>~ pushContext "PI")
<|>
((pString False "<!ELEMENT" >>= withAttribute "Declaration") >>~ pushContext "Declaration")
<|>
((pString False "<!ATTLIST" >>= withAttribute "Declaration") >>~ pushContext "Declaration")
<|>
((pString False "<!NOTATION" >>= withAttribute "Declaration") >>~ pushContext "Declaration")
<|>
((pString False "<!ENTITY" >>= withAttribute "Declaration") >>~ pushContext "Declaration")
<|>
((pDetectIdentifier >>= withAttribute "Normal")))
return (attr, result)
parseRules "Comment" =
do (attr, result) <- (((pDetectSpaces >>= withAttribute "Comment"))
<|>
((pString False "-->" >>= withAttribute "Comment") >>~ (popContext >> return ()))
<|>
((Text.Highlighting.Kate.Syntax.Alert.parseExpression >>= ((withAttribute "") . snd)))
<|>
((pDetectIdentifier >>= withAttribute "Comment")))
return (attr, result)
parseRules "PI" =
do (attr, result) <- ((pDetect2Chars False '?' '>' >>= withAttribute "Processing Instruction") >>~ (popContext >> return ()))
return (attr, result)
parseRules "Declaration" =
do (attr, result) <- (((pString False "<!--" >>= withAttribute "Comment") >>~ pushContext "Comment")
<|>
((pDetect2Chars False '-' '-' >>= withAttribute "Comment") >>~ pushContext "InlineComment")
<|>
((pDetectChar False '>' >>= withAttribute "Declaration") >>~ (popContext >> return ()))
<|>
((pDetectChar False '"' >>= withAttribute "String") >>~ pushContext "String")
<|>
((pRegExpr regex_'28'2d'7cO'29'5cs'28'2d'7cO'29 >>= withAttribute "Declaration"))
<|>
((pAnyChar "(|)," >>= withAttribute "Delimiter"))
<|>
((pRegExpr regex_'28'25'7c'26'29'28'23'5b0'2d9'5d'2b'7c'23'5bxX'5d'5b0'2d9A'2dFa'2df'5d'2b'7c'5b'5c'2d'5cw'5cd'5c'2e'3a'5f'5d'2b'29'3b >>= withAttribute "Entity"))
<|>
((pAnyChar "?*+-&" >>= withAttribute "Symbol"))
<|>
((pRegExpr regex_'25'5cs >>= withAttribute "Local"))
<|>
((pKeyword " \n\t.():!+,-<=>%&*/;?[]^{|}~\\" list_Category >>= withAttribute "Keyword"))
<|>
((pKeyword " \n\t.():!+,-<=>%&*/;?[]^{|}~\\" list_Keywords >>= withAttribute "Keyword"))
<|>
((pRegExpr regex_'5cb'5b'5c'2d'5cw'5cd'5c'2e'3a'5f'5d'2b'5cb >>= withAttribute "Name")))
return (attr, result)
parseRules "String" =
do (attr, result) <- (((pDetectSpaces >>= withAttribute "String"))
<|>
((pDetectChar False '"' >>= withAttribute "String") >>~ (popContext >> return ()))
<|>
((pRegExpr regex_'25'5b'5c'2d'5cw'5cd'5c'2e'3a'5f'5d'2b'3b >>= withAttribute "Entity")))
return (attr, result)
parseRules "InlineComment" =
do (attr, result) <- (((pDetectSpaces >>= withAttribute "Comment"))
<|>
((pDetect2Chars False '-' '-' >>= withAttribute "Comment") >>~ (popContext >> return ()))
<|>
((Text.Highlighting.Kate.Syntax.Alert.parseExpression >>= ((withAttribute "") . snd)))
<|>
((pDetectIdentifier >>= withAttribute "Comment")))
return (attr, result)
parseRules x = fail $ "Unknown context" ++ x