implementation module lex
-from Text import class Text, instance Text String
-import Control.Monad
-import Data.Either
-import Data.Functor
-import Data.List
-import StdString
+import Control.Monad, Control.Applicative
+import Data.Either, Data.Func, Data.Void
+from StdFunc import o
import StdBool
-import StdFile
-import System.CommandLine
-import qualified Text
+import StdList
+import StdChar
+import StdString
+import StdTuple
-Start :: *World -> *World
-Start w
-# (out, w) = stdio w
-# (toparse, out) = readEntireFile out
-# (_, w) = fclose (out <<< "\n" <<< toString (lexer toparse)) w
-= w
+import yard
+import AST
-readEntireFile :: *File -> *([Char], *File)
-readEntireFile f
-# (b, c, f) = freadc f
-| not b = ([], f)
-# (cs, f) = readEntireFile f
-= ([c:cs], f)
+:: 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]
+
+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
+ 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)
-untilNext :: [Char] -> [Char]
-untilNext [] = []
-untilNext ['*':'/':rest] = rest
-untilNext [x:rest] = untilNext rest
+ lexNumber :: Parser Char LexItem
+ lexNumber = some (satisfy isDigit) >>= \si->pure $
+ LexToken (length si) (NumberToken $ toInt $ toString si)
-lex :: [Char] -> (Either String Token, [Char])
-lex [] = (Right EndOfFileToken, [])
-//Comments
-lex ['/':'/':rest] = lex (tl (dropWhile ((<>) '\n') rest))
-lex ['/':'*':rest] = lex (untilNext rest)
-//Keyword tokens
-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 ['I':'n':'t':rest] = (Right IntTypeToken, rest)
-lex ['C':'h':'a':'r':rest] = (Right CharTypeToken, rest)
-lex ['B':'o':'o':'l':rest] = (Right BoolTypeToken, rest)
-//Two character tokens
-lex [':':':':rest] = (Right DoubleColonToken, rest)
-lex ['!':'=':rest] = (Right NotEqualToken, 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)
-//One character tokens
-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)
-//Value tokens
-lex ['\'':x:'\'':rest] = (Right (CharToken x), 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), [])
-lex [x:xs]
-| isSpace x = lex xs
-| isDigit x = let (is, rest) = span isDigit xs in
- (Right (NumberToken [x:is]), rest)
-| isAlpha x = let (is, rest) = span (\c.isAlphanum c || c == '_') xs in
- (Right (IdentToken [x:is]), rest)
-| otherwise = (Left ("Unexpected character: " +++ toString x), [])
+ lexIdentifier :: Parser Char LexItem
+ lexIdentifier = some (satisfy isIdentChar)
+ >>= \si->pure $ LexToken (length si) (IdentToken $ toString si)
-instance toString LexerOutput where
- toString (Left l) = "Error: " +++ l
- toString (Right x) = 'Text'.concat (print 0 x)
+ isIdentChar c = isAlphanum c || c == '_'
-tab :: Int -> [String]
-tab i = replicate i "\t"
+ lexCharacter :: Parser Char LexItem
+ lexCharacter = item '\'' *> satisfy ((<>) '\'') <* item '\''
+ >>= \char->pure $ LexToken 3 (CharToken char)
-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 [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 [VoidToken:rest] = ["Void":print i rest]
-print i [IntTypeToken:rest] = ["Int":print i rest]
-print i [CharTypeToken:rest] = ["Char":print i rest]
-print i [BoolTypeToken:rest] = ["Bool":print i rest]
-print i [DoubleColonToken:rest] = [" :: ":print i rest]
-print i [NotEqualToken: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 [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] = case rest of
- [CBraceCloseToken:_] = ["}\n":tab (i-2)] ++ print (i-1) rest
- _ = ["}\n":tab (i-1)] ++ print (i-1) 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:rest] = case rest of
- [CBraceCloseToken:_] = [";\n":tab (i-1)] ++ print i 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 [EndOfFileToken:rest] = ["\n":print i rest]
+ 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)