X-Git-Url: https://git.martlubbers.net/?a=blobdiff_plain;f=src%2Flex.icl;h=9a414cbdcc953575434fb4e1b8985b20e5e788f3;hb=fa2b27e75142781978149dc3e906e8b5aae8e603;hp=fd4b564ede74a6a2984070a52e4a6dc82fefc97a;hpb=3d1c57710eb0b86f13df392f03131157aec22c21;p=cc1516.git

diff --git a/src/lex.icl b/src/lex.icl
index fd4b564..9a414cb 100644
--- a/src/lex.icl
+++ b/src/lex.icl
@@ -1,72 +1,120 @@
 implementation module lex
 
 import Control.Monad, Control.Applicative
-import Data.Either, Data.Func
+import Data.Either, Data.Func, Data.Maybe, Data.Functor
 from StdFunc import o
 import StdBool
 import StdList
 import StdChar
+import StdString
+
+from Text import class Text(textSize,concat), instance Text String
 
 import yard
 
 lexer :: [Char] -> LexerOutput
-lexer r = case runParser lexProgram r of
+lexer r = case runParser (lexProgram 1 1) r of
 	(Right p, _) = Right p
-	(Left e, _) = Left $ toString e
+	(Left e, _) = Left e
+
+lexProgram :: Int Int -> Parser Char [Token]
+lexProgram line column = lexToken >>= \t->case t of
+		LexEOF = pure []
+		(LexItemError e) = fail <?> 
+			PositionalError line column ("LexerError: " +++ e)
+		(LexToken c t) = lexProgram line (column+c)
+			>>= \rest->pure [{line=line, column=column, token=t}:rest]
+		LexNL = lexProgram (line+1) 1
+		(LexSpace l c) = lexProgram (line+l) (column+c)
 
-lexProgram :: Parser Char [Token]
-lexProgram = some lexToken <* many (satisfy isSpace) <* eof
-	>>= \ts->pure (map (\t->(0, 0, t)) ts)
+:: LexItem 
+	= LexToken Int TokenValue
+	| LexSpace Int Int
+	| LexNL 
+	| LexEOF 
+	| LexItemError String
 
-lexToken :: Parser Char TokenValue
+lexToken :: Parser Char LexItem
 lexToken = 
-    //Comments
-    (list (fromString "//") >>| until top (item '\n') >>| lexToken) <|>
-    (list (fromString "/*") >>| until top (list (fromString "*/")) >>| lexToken) <|>
+	//Comments
+	lexBlockComment <|> lexComment <|>
 	//Keyword tokens
-	(lexKw "var" VarToken) <|>
-	(lexKw "Void" VoidToken) <|>
-	(lexKw "return" ReturnToken) <|>
-	(lexKw "if" IfToken) <|>
-	(lexKw "else" ElseToken) <|>
-	(lexKw "while" WhileToken) <|>
-	(lexKw "True" TrueToken) <|>
-	(lexKw "False" FalseToken) <|>
-	(lexKw "Int" IntTypeToken) <|>
-	(lexKw "Bool" BoolTypeToken) <|>
-	(lexKw "Char" CharTypeToken) <|>
-	//Escape chars tokens
-	(liftM CharToken $ item '\'' *> item '\\' *> lexEscape <* item '\'') <|>
-	//Normal chars tokens
-	(liftM CharToken $ item '\'' *> satisfy ((<>) '\'') <* item '\'') <|>
+	lexKw "var" VarToken <|> lexKw "Void" VoidToken <|>
+	lexKw "return" ReturnToken <|> lexKw "if" IfToken <|>
+	lexKw "else" ElseToken <|> lexKw "while" WhileToken <|>
+	lexKw "True" TrueToken <|> lexKw "False" FalseToken <|>
+	lexKw "Int" IntTypeToken <|> lexKw "Bool" BoolTypeToken <|>
+	lexKw "Char" CharTypeToken <|>
+	//Character tokens
+	lexEscape <|> lexCharacter <|>
 	//Two char ops tokens
-	(lexOp "::" DoubleColonToken) <|> (lexOp "!=" NotEqualToken) <|>
-	(lexOp "<=" LesserEqToken) <|> (lexOp ">=" GreaterEqToken) <|>
-	(lexOp "==" EqualsToken) <|> (lexOp "&&" AmpersandsToken) <|>
-	(lexOp "||" PipesToken) <|> (lexOp "[]" EmptyListToken) <|>
-	(lexOp "->" ArrowToken) <|> 
+	lexWord "::" DoubleColonToken <|> lexWord "!=" NotEqualToken <|>
+	lexWord "<=" LesserEqToken <|> lexWord ">=" GreaterEqToken <|>
+	lexWord "==" EqualsToken <|> lexWord "&&" AmpersandsToken <|>
+	lexWord "||" PipesToken <|> lexWord "[]" EmptyListToken <|>
+	lexWord "->" ArrowToken <|> 
 	//One char ops tokens
-    (lexOp "(" BraceOpenToken) <|>
-	(lexOp ")" BraceCloseToken) <|> (lexOp "{" CBraceOpenToken) <|>
-	(lexOp "}" CBraceCloseToken) <|> (lexOp "[" SquareOpenToken) <|>
-	(lexOp "]" SquareCloseToken) <|> (lexOp "," CommaToken) <|>
-	(lexOp ":" ColonToken) <|> (lexOp ";" SColonToken) <|>
-	(lexOp "." DotToken) <|> (lexOp "+" PlusToken) <|>
-	(lexOp "*" StarToken) <|> (lexOp "/" SlashToken) <|>
-	(lexOp "%" PercentToken) <|> (lexOp "=" AssignmentToken) <|>
-	(lexOp "<" LesserToken) <|> (lexOp ">" BiggerToken) <|>
-	(lexOp "!" ExclamationToken) <|> (lexOp "-" DashToken) <|>
-	//Number tokens
-	(liftM (NumberToken o toInt o toString) $ some $ satisfy isDigit) <|>
-	//Ident tokens
-	(liftM IdentToken $ some $ satisfy isIdentChar) <|>
-	(satisfy isSpace >>| lexToken)
-//	(eof >>| pure EndOfFileToken) 
+    lexWord "(" BraceOpenToken <|>
+	lexWord ")" BraceCloseToken <|> lexWord "{" CBraceOpenToken <|>
+	lexWord "}" CBraceCloseToken <|> lexWord "[" SquareOpenToken <|>
+	lexWord "]" SquareCloseToken <|> lexWord "," CommaToken <|>
+	lexWord ":" ColonToken <|> lexWord ";" SColonToken <|>
+	lexWord "." DotToken <|> lexWord "+" PlusToken <|>
+	lexWord "*" StarToken <|> lexWord "/" SlashToken <|>
+	lexWord "%" PercentToken <|> lexWord "=" AssignmentToken <|>
+	lexWord "<" LesserToken <|> lexWord ">" BiggerToken <|>
+	lexWord "!" ExclamationToken <|> lexWord "-" DashToken <|>
+	//Number and identifier tokens
+	lexNumber <|> lexIdentifier <|>
+	(item '\n' >>| pure LexNL) <|>
+	(satisfy isSpace >>| (pure $ LexSpace 0 1)) <|>
+	(eof >>| pure LexEOF)
 	where
+		lexWord :: String TokenValue -> Parser Char LexItem
+		lexWord s tv = list (fromString s) >>| pure (LexToken (textSize s) tv)
+
+		lexKw :: String TokenValue -> Parser Char LexItem
+		lexKw kw tv = lexWord kw tv <* check (not o isIdentChar)
+
+		lexUntilNL = top until (eof <|> (item '\n' >>| pure Void))
+
+		lexComment :: Parser Char LexItem
+    	lexComment = list (fromString "//") >>| lexUntilNL 
+			>>= \chars->pure LexNL
+
+		lexBlockComment :: Parser Char LexItem
+		lexBlockComment = list (fromString "/*") 
+			>>| (top until (list (fromString "*/")))
+			>>= \chars->pure $ widthHeight chars 0 0
+			where
+				widthHeight :: [Char] Int Int -> LexItem
+				widthHeight [] l c = LexSpace l c
+				widthHeight ['\n':xs] l _ = widthHeight xs (l+1) 0
+				widthHeight [x:xs] l c = widthHeight xs l (c+1)
+
+		lexNumber :: Parser Char LexItem
+		lexNumber = toString <$> some (satisfy isDigit) 
+			>>= \si->pure $ LexToken (textSize si) (NumberToken $ toInt si)
+
+		lexIdentifier :: Parser Char LexItem
+		lexIdentifier = toString <$> some (satisfy isIdentChar)
+			>>= \si->pure $ LexToken (textSize si) (IdentToken si)
+
 		isIdentChar c = isAlphanum c || c == '_'
-		lexOp s tv = list (fromString s) >>| pure tv
-		lexKw kw tv = lexOp kw tv <* check (not o isIdentChar) >>| pure tv
-		lexEscape = (
-			lexOp "a" (toChar 7) <|> lexOp "b" '\b' <|> lexOp "f" '\f' <|>
-			lexOp "n" '\n' <|> lexOp "r" '\t' <|> lexOp "v" '\v' <|>
-			lexOp "'" '\'') <?> ("Unknown escape", 0)
+
+		lexCharacter :: Parser Char LexItem
+		lexCharacter = item '\'' *> satisfy ((<>) '\'') <* item '\''
+			>>= \char->pure $ LexToken 3 (CharToken char)
+
+		lexEscape :: Parser Char LexItem
+		lexEscape = item '\'' *> item '\\' *> top <* item '\''
+			>>= \char->pure case char of
+				'a' = LexToken 4 (CharToken $ toChar 7)
+				'b' = LexToken 4 (CharToken '\b')
+				'b' = LexToken 4 (CharToken '\b')
+				'f' = LexToken 4 (CharToken '\f')
+				'n' = LexToken 4 (CharToken '\n')
+				'r' = LexToken 4 (CharToken '\t')
+				'v' = LexToken 4 (CharToken '\v')
+				'\'' =LexToken 4 (CharToken '\'')
+				c = (LexItemError $ "Unknown escape: " +++ toString c)