module Text.Highlighting.Kate.Syntax.Sgml
(highlight, parseExpression, syntaxName, syntaxExtensions)
where
import Text.Highlighting.Kate.Types
import Text.Highlighting.Kate.Common
import Text.ParserCombinators.Parsec hiding (State)
import Control.Monad.State
import Data.Char (isSpace)
syntaxName :: String
syntaxName = "SGML"
syntaxExtensions :: String
syntaxExtensions = "*.sgml"
highlight :: String -> [SourceLine]
highlight input = evalState (mapM parseSourceLine $ lines input) startingState
parseSourceLine :: String -> State SyntaxState SourceLine
parseSourceLine = mkParseSourceLine (parseExpression Nothing)
parseExpression :: Maybe (String,String)
-> KateParser Token
parseExpression mbcontext = do
(lang,cont) <- maybe currentContext return mbcontext
result <- parseRules (lang,cont)
optional $ do eof
updateState $ \st -> st{ synStPrevChar = '\n' }
pEndLine
return result
startingState = SyntaxState {synStContexts = [("SGML","Normal Text")], synStLineNumber = 0, synStPrevChar = '\n', synStPrevNonspace = False, synStContinuation = False, synStCaseSensitive = True, synStKeywordCaseSensitive = False, synStCaptures = []}
pEndLine = do
updateState $ \st -> st{ synStPrevNonspace = False }
context <- currentContext
contexts <- synStContexts `fmap` getState
st <- getState
if length contexts >= 2
then case context of
_ | synStContinuation st -> updateState $ \st -> st{ synStContinuation = False }
("SGML","Normal Text") -> return ()
("SGML","Attribute") -> return ()
("SGML","Value") -> return ()
("SGML","Value 2") -> return ()
("SGML","Comment") -> return ()
_ -> return ()
else 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)
regex_'3c'5cs'2a'5c'2f'3f'5cs'2a'5ba'2dzA'2dZ'5f'3a'5d'5ba'2dzA'2dZ0'2d9'2e'5f'3a'2d'5d'2a = compileRegex True "<\\s*\\/?\\s*[a-zA-Z_:][a-zA-Z0-9._:-]*"
regex_'5cs'2a'3d'5cs'2a = compileRegex True "\\s*=\\s*"
parseRules ("SGML","Normal Text") =
(((pString False "<!--" >>= withAttribute CommentTok) >>~ pushContext ("SGML","Comment"))
<|>
((pRegExpr regex_'3c'5cs'2a'5c'2f'3f'5cs'2a'5ba'2dzA'2dZ'5f'3a'5d'5ba'2dzA'2dZ0'2d9'2e'5f'3a'2d'5d'2a >>= withAttribute KeywordTok) >>~ pushContext ("SGML","Attribute"))
<|>
(currentContext >>= \x -> guard (x == ("SGML","Normal Text")) >> pDefault >>= withAttribute NormalTok))
parseRules ("SGML","Attribute") =
(((pDetect2Chars False '/' '>' >>= withAttribute KeywordTok) >>~ (popContext))
<|>
((pDetectChar False '>' >>= withAttribute KeywordTok) >>~ (popContext))
<|>
((pRegExpr regex_'5cs'2a'3d'5cs'2a >>= withAttribute NormalTok) >>~ pushContext ("SGML","Value"))
<|>
(currentContext >>= \x -> guard (x == ("SGML","Attribute")) >> pDefault >>= withAttribute OtherTok))
parseRules ("SGML","Value") =
(((pDetect2Chars False '/' '>' >>= withAttribute KeywordTok) >>~ (popContext >> popContext))
<|>
((pDetectChar False '>' >>= withAttribute KeywordTok) >>~ (popContext >> popContext))
<|>
((pDetectChar False '"' >>= withAttribute DataTypeTok) >>~ pushContext ("SGML","Value 2"))
<|>
(currentContext >>= \x -> guard (x == ("SGML","Value")) >> pDefault >>= withAttribute DataTypeTok))
parseRules ("SGML","Value 2") =
(((pDetectChar False '"' >>= withAttribute DataTypeTok) >>~ (popContext >> popContext))
<|>
(currentContext >>= \x -> guard (x == ("SGML","Value 2")) >> pDefault >>= withAttribute DataTypeTok))
parseRules ("SGML","Comment") =
(((pString False "-->" >>= withAttribute CommentTok) >>~ (popContext))
<|>
(currentContext >>= \x -> guard (x == ("SGML","Comment")) >> pDefault >>= withAttribute CommentTok))
parseRules x = parseRules ("SGML","Normal Text") <|> fail ("Unknown context" ++ show x)