X-Git-Url: https://git.martlubbers.net/?a=blobdiff_plain;f=lex.icl;h=de200df2f4dcd62930ea1e74bf20f97fb7b9872a;hb=75ac4a13efdc20b444a5eb6cb2d16425c82d333e;hp=a1912bf959b332f43a0a731d5009886716bf4827;hpb=8de73bb36ae068011c2c1b1fc9b3c9091a897997;p=cc1516.git diff --git a/lex.icl b/lex.icl index a1912bf..de200df 100644 --- a/lex.icl +++ b/lex.icl @@ -1,149 +1,80 @@ implementation module lex -import Data.Either -import Data.List +import Control.Monad +import Data.Either, Data.Maybe, Data.Map, Data.Func import StdString -import System.CommandLine -import StdFile -import StdMisc +import StdTuple 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 -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) +SingleCharTokens :: Map Char TokenValue +SingleCharTokens = fromList [ + ('(', BraceOpenToken), (')', BraceCloseToken), ('{', CBraceOpenToken), + ('}', CBraceCloseToken), ('[', SquareOpenToken), (']', SquareCloseToken), + (',', CommaToken), (':', ColonToken), (';', SColonToken), + ('.', DotToken), ('+', PlusToken), ('*', StarToken), ('/', SlashToken), + ('%', PercentToken), ('=', AssignmentToken), ('<', LesserToken), + ('>', BiggerToken), ('!', ExclamationToken)] -lexer :: [Char] -> LexerOutput -lexer [] = Right [] -lexer x = case lex x of - (Right t, rest) = lexer rest >>= \ts.Right [t:ts] - (Left e, _) = Left e +EscapeMap :: Map Char Char +EscapeMap = fromList [('a', toChar 7), ('b', '\b'), ('f', '\f'), ('n', '\n'), + ('r', '\r'), ('t', '\t'), ('v', '\v')] -print :: LexerOutput -> String -print (Left l) = "Error: " +++ l -print (Right x) = 'Text'.concat (print` 0 x) +lexer :: [Char] -> LexerOutput +lexer x = lexer` x 1 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] + lexer` :: [Char] Int -> LexerOutput + lexer` [] i = Right [] + lexer` x i = case lex x i of + (Right t, rest, i) = lexer` rest i >>= \ts.Right [t:ts] + (Left e, _, _) = Left e -tab :: Int -> String -tab 0 = "" -tab i = "\t" +++ tab (i-1) +ret:: TokenValue [Char] Int -> (Either String Token, [Char], Int) +ret tv rest i = (Right (i, tv), rest, i) +err :: String Int -> (Either String Token, [Char], Int) +err s i = (Left ("Line " +++ toString i +++ ": " +++ s), [], i) -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), []) +lex :: [Char] -> (Int -> (Either String Token, [Char], Int)) +lex [] = ret EndOfFileToken [] +lex ['/':'/':xs] = lex (dropWhile ((<>) '\n') xs) +lex ['/':'*':xs] = \i.let (c, r) = span ((<>) ('*','/')) $ zip2 [' ':xs] xs in + lex (tl $ map snd $ tl $ r) (i + length [x\\(x, _)<-c | x == '\n']) +lex ['v':'a':'r':xs] = ret VarToken xs +lex ['V':'o':'i':'d':xs] = ret VoidToken xs +lex ['r':'e':'t':'u':'r':'n':xs] = ret ReturnToken xs +lex ['i':'f':xs] = ret IfToken xs +lex ['e':'l':'s':'e':xs] = ret ElseToken xs +lex ['w':'h':'i':'l':'e':xs] = ret WhileToken xs +lex ['T':'r':'u':'e':xs] = ret TrueToken xs +lex ['F':'a':'l':'s':'e':xs] = ret FalseToken xs +lex ['I':'n':'t':xs] = ret IntTypeToken xs +lex ['C':'h':'a':'r':xs] = ret CharTypeToken xs +lex ['B':'o':'o':'l':xs] = ret BoolTypeToken xs +lex [':':':':xs] = ret DoubleColonToken xs +lex ['!':'=':xs] = ret NotEqualToken xs +lex ['<':'=':xs] = ret LesserEqToken xs +lex ['>':'=':xs] = ret GreaterEqToken xs +lex ['=':'=':xs] = ret EqualsToken xs +lex ['&':'&':xs] = ret AmpersandsToken xs +lex ['|':'|':xs] = ret PipesToken xs +lex ['-':'>':xs] = ret ArrowToken xs +lex ['\'':x:'\'':xs] = ret (CharToken x) xs +lex ['\'':'\\':x:'\'':xs] = case get x EscapeMap of + Just t = ret (CharToken t) xs + _ = err ("Unknown escape: \\" +++ toString x) +lex ['-':t=:[x:xs]] +| isDigit x = let (n, r) = span isDigit t in + ret (NumberToken (toInt $ toString ['-':n])) r +| otherwise = ret DashToken xs +lex t=:[x:xs] = case get x SingleCharTokens of + (Just tok) = ret tok xs + Nothing + | x == '\n' = \i.lex xs (i+1) + | isSpace x = lex xs + | isDigit x = let (v, r) = span isDigit t in + ret (NumberToken (toInt $ toString v)) r + | isAlpha x = let (v, r) = span isIdent t in ret (IdentToken v) r + with isIdent c = isAlphanum c || c == '_' + | otherwise = err ("Unexpected character: " +++ toString x)