X-Git-Url: https://git.martlubbers.net/?a=blobdiff_plain;f=lex.icl;h=aa34987969b5afcdc71c83451f91fe448ee9902f;hb=f5e125920d85bc53eda7b51f3f03e89fcaf3a0ce;hp=a1912bf959b332f43a0a731d5009886716bf4827;hpb=8de73bb36ae068011c2c1b1fc9b3c9091a897997;p=cc1516.git

diff --git a/lex.icl b/lex.icl
index a1912bf..aa34987 100644
--- a/lex.icl
+++ b/lex.icl
@@ -1,149 +1,118 @@
 implementation module lex
 
-import Data.Either
-import Data.List
-import StdString
-import System.CommandLine
-import StdFile
-import StdMisc
+import Control.Monad, Control.Applicative
+import Data.Either, Data.Func, Data.Void
+from StdFunc import o
 import StdBool
-from StdFunc import id, const
-import Data.Maybe
-import Control.Applicative
-import Control.Monad
-import Control.Monad.State
-import Data.Functor
-from Data.Func import $
-from Text import class Text, instance Text String
-import qualified Text
+import StdList
+import StdChar
+import StdString
+import StdTuple
 
-Start :: *World -> *World
-Start w
-# (args, w) = getCommandLine w // We lezen nu nog standaard van stdin
-# (out, w) = stdio w
-# (toparse, out) = readEntireFile out
-# out = out <<< print (lexer toparse)
-# (b, w) = fclose out w
-| not b = setReturnCode 1 w
-= w
-	where
-		readEntireFile :: *File -> *([Char], *File)
-		readEntireFile f
-		# (b, c, f) = freadc f
-		| not b = ([], f)
-		# (cs, f) = readEntireFile f
-		= ([c:cs], f)
+import yard
+import AST
+
+:: LexItem
+	= LexToken Int TokenValue
+	| LexSpace Int Int
+	| LexNL
+	| LexEOF
+	| LexItemError String
 
 lexer :: [Char] -> LexerOutput
-lexer [] = Right []
-lexer x = case lex x of
-	(Right t, rest) = lexer rest >>= \ts.Right [t:ts]
-	(Left e, _) = Left e
+lexer r = fst $ runParser (lexProgram 1 1) r
+
+lexProgram :: Int Int -> Parser Char [Token]
+lexProgram line column = lexToken >>= \t->case t of
+	LexEOF = pure []
+	LexNL = lexProgram (line+1) 1
+	(LexSpace l c) = lexProgram (line+l) (column+c)
+	(LexItemError e) = fail <?>
+		PositionalError line column ("LexerError: " +++ e)
+	(LexToken c t) = lexProgram line (column+c)
+		>>= \rest->pure [({line=line,col=column}, t):rest]
 
-print :: LexerOutput -> String
-print (Left l) = "Error: " +++ l
-print (Right x) = 'Text'.concat (print` 0 x)
+lexToken :: Parser Char LexItem
+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 <|>
+	//Character tokens
+	lexEscape <|> lexCharacter <|>
+	//Two char ops tokens
+	lexWord "::" DoubleColonToken <|> lexWord "!=" NotEqualToken <|>
+	lexWord "<=" LesserEqToken <|> lexWord ">=" GreaterEqToken <|>
+	lexWord "==" EqualsToken <|> lexWord "&&" AmpersandsToken <|>
+	lexWord "||" PipesToken <|> lexWord "[]" EmptyListToken <|>
+	lexWord "->" ArrowToken <|>
+	//One char ops tokens
+    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) <|>
+	//Whitespace
+	(satisfy isSpace >>| (pure $ LexSpace 0 1)) <|>
+	(eof >>| pure LexEOF)
 	where
-		print` :: Int [Token] -> [String]
-		print` _ [] = []
-		print` i [(IdentToken l):rest] = [toString l:print` i rest]
-		print` i [(NumberToken j):rest] = [toString j:print` i rest]
-		print` i [(CharToken c):rest] = ["'":toString c:"'":print` i rest]
-		print` i [VarToken:rest] = ["var ":print` i rest]
-		print` i [VoidToken:rest] = ["Void":print` i rest]
-		print` i [ReturnToken:rest] = ["return":print` i rest]
-		print` i [IfToken:rest] = ["if":print` i rest]
-		print` i [ElseToken:rest] = ["else":print` i rest]
-		print` i [WhileToken:rest] = ["while":print` i rest]
-		print` i [TrueToken:rest] = ["True":print` i rest]
-		print` i [FalseToken:rest] = ["False":print` i rest]
-		print` i [BraceOpenToken:rest] = ["(":print` i rest]
-		print` i [BraceCloseToken:rest] = [")":print` i rest]
-		print` i [CBraceOpenToken:rest] = ["{\n":tab (i+1):print` (i+1) rest]
-		print` i [CBraceCloseToken:rest] = ["}\n":tab i:print` i rest]
-		print` i [SquareOpenToken:rest] = ["[":print` i rest]
-		print` i [SquareCloseToken:rest] = ["]":print` i rest]
-		print` i [CommaToken:rest] = [", ":print` i rest]
-		print` i [ColonToken:rest] = [":":print` i rest]
-		print` i [SColonToken:CBraceCloseToken:rest] =
-			[";\n":tab (i-1):"}\n":print` (i-1) rest]
-		print` i [SColonToken:rest] = [";\n":tab i:print` i rest]
-		print` i [DotToken:rest] = [".":print` i rest]
-		print` i [PlusToken:rest] = [" + ":print` i rest]
-		print` i [DashToken:rest] = [" - ":print` i rest]
-		print` i [StarToken:rest] = [" * ":print` i rest]
-		print` i [SlashToken:rest] = [" / ":print` i rest]
-		print` i [PercentToken:rest] = [" % ":print` i rest]
-		print` i [AssignmentToken:rest] = [" = ":print` i rest]
-		print` i [LesserToken:rest] = [" < ":print` i rest]
-		print` i [BiggerToken:rest] = [" > ":print` i rest]
-		print` i [ExclamationToken:rest] = ["!":print` i rest]
-		print` i [DoubleColonToken:rest] = [" :: ":print` i rest]
-		print` i [LesserEqToken:rest] = [" <= ":print` i rest]
-		print` i [GreaterEqToken:rest] = [" >= ":print` i rest]
-		print` i [EqualsToken:rest] = [" == ":print` i rest]
-		print` i [AmpersandsToken:rest] = [" && ":print` i rest]
-		print` i [PipesToken:rest] = [" || ":print` i rest]
-		print` i [ArrowToken:rest] = [" -> ":print` i rest]
-		print` i [EndOfFileToken:rest] = ["\n":print` i rest]
+		lexWord :: String TokenValue -> Parser Char LexItem
+		lexWord s tv = list ls >>| pure (LexToken (length ls) tv)
+			where ls = fromString s
+
+		lexKw :: String TokenValue -> Parser Char LexItem
+		lexKw kw tv = lexWord kw tv <* check (not o isIdentChar)
+
+		lexComment :: Parser Char LexItem
+		lexComment = list (fromString "//") 
+			>>| top until (eof <|> (item '\n' >>| pure Void)) >>| 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 = some (satisfy isDigit) >>= \si->pure $
+			LexToken (length si) (NumberToken $ toInt $ toString si)
+
+		lexIdentifier :: Parser Char LexItem
+		lexIdentifier = some (satisfy isIdentChar)
+			>>= \si->pure $ LexToken (length si) (IdentToken $ toString si)
 
-tab :: Int -> String
-tab 0 = ""
-tab i = "\t" +++ tab (i-1)
+		isIdentChar c = isAlphanum c || c == '_'
 
+		lexCharacter :: Parser Char LexItem
+		lexCharacter = item '\'' *> satisfy ((<>) '\'') <* item '\''
+			>>= \char->pure $ LexToken 3 (CharToken char)
 
-lex :: [Char] -> (Either String Token, [Char])
-lex [] = (Right EndOfFileToken, [])
-lex ['v':'a':'r':rest] = (Right VarToken, rest)
-lex ['V':'o':'i':'d':rest] = (Right VoidToken, rest)
-lex ['r':'e':'t':'u':'r':'n':rest] = (Right ReturnToken, rest)
-lex ['i':'f':rest] = (Right IfToken, rest)
-lex ['e':'l':'s':'e':rest] = (Right ElseToken, rest)
-lex ['w':'h':'i':'l':'e':rest] = (Right WhileToken, rest)
-lex ['T':'r':'u':'e':rest] = (Right TrueToken, rest)
-lex ['F':'a':'l':'s':'e':rest] = (Right FalseToken, rest)
-lex ['\'':'\\':x:'\'':rest] = case x of
-	'a' = (Right (CharToken (toChar 7)), rest) //Alarm
-	'b' = (Right (CharToken '\b'), rest) //Backspace
-	'f' = (Right (CharToken '\f'), rest) //Formfeed
-	'n' = (Right (CharToken '\n'), rest) //Newline
-	'r' = (Right (CharToken '\r'), rest) //Carriage Return
-	't' = (Right (CharToken '\t'), rest) //Horizontal tab
-	'v' = (Right (CharToken '\v'), rest) //Vertical tab
-	_ = (Left ("Unknown escape: \\" +++ toString x), rest)
-lex ['\'':x:'\'':rest] = (Right (CharToken x), rest)
-lex [':':':':rest] = (Right DoubleColonToken, rest)
-lex ['<':'=':rest] = (Right LesserEqToken, rest)
-lex ['>':'=':rest] = (Right GreaterEqToken, rest)
-lex ['=':'=':rest] = (Right EqualsToken, rest)
-lex ['&':'&':rest] = (Right AmpersandsToken, rest)
-lex ['|':'|':rest] = (Right PipesToken, rest)
-lex ['-':'>':rest] = (Right ArrowToken, rest)
-lex ['(':rest] = (Right BraceOpenToken, rest)
-lex [')':rest] = (Right BraceCloseToken, rest)
-lex ['{':rest] = (Right CBraceOpenToken, rest)
-lex ['}':rest] = (Right CBraceCloseToken, rest)
-lex ['[':rest] = (Right SquareOpenToken, rest)
-lex [']':rest] = (Right SquareCloseToken, rest)
-lex [',':rest] = (Right CommaToken, rest)
-lex [':':rest] = (Right ColonToken, rest)
-lex [';':rest] = (Right SColonToken, rest)
-lex ['.':rest] = (Right DotToken, rest)
-lex ['+':rest] = (Right PlusToken, rest)
-lex ['-':rest] = (Right DashToken, rest)
-lex ['*':rest] = (Right StarToken, rest)
-lex ['/':rest] = (Right SlashToken, rest)
-lex ['%':rest] = (Right PercentToken, rest)
-lex ['=':rest] = (Right AssignmentToken, rest)
-lex ['<':rest] = (Right LesserToken, rest)
-lex ['>':rest] = (Right BiggerToken, rest)
-lex ['!':rest] = (Right ExclamationToken, rest)
-lex [x:xs]
-| isSpace x = lex xs
-| isDigit x
-	# (is, rest) = span isDigit xs
-	= (Right (NumberToken [x:is]), rest)
-| isAlpha x
-	# (is, rest) = span (\c.isAlphanum c || c == '_') xs
-	= (Right (IdentToken [x:is]), rest)
-| otherwise = (Left ("Unexpected character: " +++ toString x), [])
+		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)