module Text.Highlighting.Kate.Syntax.Bibtex
(highlight, parseExpression, syntaxName, syntaxExtensions)
where
import Text.Highlighting.Kate.Types
import Text.Highlighting.Kate.Common
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 = "BibTeX"
syntaxExtensions :: String
syntaxExtensions = "*.bib"
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 = "BibTeX" }
context <- currentContext <|> (pushContext "Normal" >> currentContext)
result <- parseRules context
optional $ eof >> pEndLine
updateState $ \st -> st { synStLanguage = oldLang }
return result
startingState = SyntaxState {synStContexts = fromList [("BibTeX",["Normal"])], synStLanguage = "BibTeX", synStLineNumber = 0, synStPrevChar = '\n', synStPrevNonspace = False, synStCaseSensitive = True, synStKeywordCaseSensitive = False, synStCaptures = []}
pEndLine = do
updateState $ \st -> st{ synStPrevNonspace = False }
context <- currentContext
case context of
"Normal" -> return ()
"PreambleCommand" -> return ()
"StringCommand" -> return ()
"Entry" -> return ()
"Field" -> return ()
"CurlyBracket" -> return ()
"QuotedText" -> return ()
_ -> 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_kw'5fentry = Set.fromList $ words $ "@article @book @booklet @conference @collection @electronic @inbook @incollection @inproceedings @manual @mastersthesis @misc @online @patent @periodical @proceedings @report @phdthesis @set @thesis @techreport @unpublished @www @person @company @place"
regex_'5ba'2dzA'2dZ0'2d9'5c'2d'5d'2b = compileRegex "[a-zA-Z0-9\\-]+"
regex_'5ba'2dzA'2dZ0'2d9'5f'40'5c'5c'2d'5c'5c'3a'5d'2b = compileRegex "[a-zA-Z0-9_@\\\\-\\\\:]+"
regex_'5ba'2dzA'2dZ0'2d9'5c'2d'5f'5c'2e'5d'2b = compileRegex "[a-zA-Z0-9\\-_\\.]+"
regex_'5b0'2d9'5d'2b = compileRegex "[0-9]+"
regex_'2e = compileRegex "."
regex_'5c'5c'28'5ba'2dzA'2dZ'40'5d'2b'7c'5b'5e_'5d'29 = compileRegex "\\\\([a-zA-Z@]+|[^ ])"
regex_'7d'24 = compileRegex "}$"
defaultAttributes = [("Normal",CommentTok),("PreambleCommand",NormalTok),("StringCommand",NormalTok),("Entry",NormalTok),("Field",NormalTok),("CurlyBracket",NormalTok),("QuotedText",StringTok)]
parseRules "Normal" =
(((pKeyword " \n\t.():!+,-<=>%&*/;?[]^{|}~" list_kw'5fentry >>= withAttribute KeywordTok) >>~ pushContext "Entry")
<|>
((pString False "@string" >>= withAttribute FunctionTok) >>~ pushContext "StringCommand")
<|>
((pString False "@preamble" >>= withAttribute FunctionTok) >>~ pushContext "PreambleCommand")
<|>
((pString False "@comment" >>= withAttribute CommentTok)))
parseRules "PreambleCommand" =
(((pDetectChar False '{' >>= withAttribute NormalTok) >>~ pushContext "CurlyBracket")
<|>
((popContext) >> currentContext >>= parseRules))
parseRules "StringCommand" =
(((pDetectChar False '{' >>= withAttribute NormalTok) >>~ pushContext "CurlyBracket")
<|>
((pRegExpr regex_'5ba'2dzA'2dZ0'2d9'5c'2d'5d'2b >>= withAttribute StringTok) >>~ pushContext "CurlyBracket")
<|>
((popContext) >> currentContext >>= parseRules))
parseRules "Entry" =
(((pDetectChar False '{' >>= withAttribute NormalTok))
<|>
((pRegExpr regex_'5ba'2dzA'2dZ0'2d9'5f'40'5c'5c'2d'5c'5c'3a'5d'2b >>= withAttribute OtherTok))
<|>
((pDetectChar False ',' >>= withAttribute NormalTok) >>~ pushContext "Field")
<|>
((pDetectChar False '}' >>= withAttribute NormalTok) >>~ (popContext)))
parseRules "Field" =
(((pFirstNonSpace >> pRegExpr regex_'5ba'2dzA'2dZ0'2d9'5c'2d'5f'5c'2e'5d'2b >>= withAttribute DataTypeTok))
<|>
((pDetectSpaces >>= withAttribute NormalTok))
<|>
((pDetectChar False '=' >>= withAttribute NormalTok))
<|>
((pDetectSpaces >>= withAttribute NormalTok))
<|>
((pDetectChar False '{' >>= withAttribute NormalTok) >>~ pushContext "CurlyBracket")
<|>
((lookAhead (pDetectChar False '}') >> (popContext) >> currentContext >>= parseRules))
<|>
((pDetectChar False '"' >>= withAttribute NormalTok) >>~ pushContext "QuotedText")
<|>
((pDetectChar False ',' >>= withAttribute NormalTok))
<|>
((pDetectChar False '#' >>= withAttribute NormalTok))
<|>
((pRegExpr regex_'5b0'2d9'5d'2b >>= withAttribute NormalTok))
<|>
((pRegExpr regex_'5ba'2dzA'2dZ0'2d9'5c'2d'5d'2b >>= withAttribute StringTok))
<|>
((pDetectSpaces >>= withAttribute NormalTok))
<|>
((pRegExpr regex_'2e >>= withAttribute AlertTok)))
parseRules "CurlyBracket" =
(((pDetectChar False '{' >>= withAttribute NormalTok) >>~ pushContext "CurlyBracket")
<|>
((pRegExpr regex_'5c'5c'28'5ba'2dzA'2dZ'40'5d'2b'7c'5b'5e_'5d'29 >>= withAttribute CharTok))
<|>
((pRegExpr regex_'7d'24 >>= withAttribute NormalTok) >>~ (popContext >> popContext))
<|>
((pDetectChar False '}' >>= withAttribute NormalTok) >>~ (popContext)))
parseRules "QuotedText" =
(((pDetectChar False '"' >>= withAttribute NormalTok) >>~ (popContext))
<|>
((pRegExpr regex_'5c'5c'28'5ba'2dzA'2dZ'40'5d'2b'7c'5b'5e_'5d'29 >>= withAttribute CharTok)))
parseRules "" = parseRules "Normal"
parseRules x = fail $ "Unknown context" ++ x