tip-lib-0.1: src/Tip/Parser/LexTIP.x
-- -*- haskell -*-
-- This Alex file was machine-generated by the BNF converter
{
{-# OPTIONS -fno-warn-incomplete-patterns #-}
{-# OPTIONS_GHC -w #-}
module Tip.Parser.LexTIP where
import qualified Data.Bits
import Data.Word (Word8)
}
$l = [a-zA-Z\192 - \255] # [\215 \247] -- isolatin1 letter FIXME
$c = [A-Z\192-\221] # [\215] -- capital isolatin1 letter FIXME
$s = [a-z\222-\255] # [\247] -- small isolatin1 letter FIXME
$d = [0-9] -- digit
$i = [$l $d _ '] -- identifier character
$u = [\0-\255] -- universal: any character
@rsyms = -- symbols and non-identifier-like reserved words
\( | "check" \- "sat" | \) | "declare" \- "datatypes" | "declare" \- "sort" | "declare" \- "fun" | "define" \- "funs" \- "rec" | "assert" \- "not" | \= \> | \@ | \= | \+ | \- | \* | \> | \> \= | \< | \< \=
:-
";" [.]* ; -- Toss single line comments
$white+ ;
@rsyms { tok (\p s -> PT p (eitherResIdent (TV . share) s)) }
($l | [\~ \! \@ \$ \% \^ \& \* \_ \- \+ \= \< \> \. \? \/]) ($l | $d | [\~ \! \@ \$ \% \^ \& \* \_ \- \+ \= \< \> \. \? \/]) * { tok (\p s -> PT p (eitherResIdent (T_Symbol . share) s)) }
$l $i* { tok (\p s -> PT p (eitherResIdent (TV . share) s)) }
$d+ { tok (\p s -> PT p (TI $ share s)) }
{
tok :: (Posn -> String -> Token) -> (Posn -> String -> Token)
tok f p s = f p s
share :: String -> String
share = id
data Tok =
TS !String !Int -- reserved words and symbols
| TL !String -- string literals
| TI !String -- integer literals
| TV !String -- identifiers
| TD !String -- double precision float literals
| TC !String -- character literals
| T_Symbol !String
deriving (Eq,Show,Ord)
data Token =
PT Posn Tok
| Err Posn
deriving (Eq,Show,Ord)
tokenPos :: [Token] -> String
tokenPos (PT (Pn _ l _) _ :_) = "line " ++ show l
tokenPos (Err (Pn _ l _) :_) = "line " ++ show l
tokenPos _ = "end of file"
tokenPosn :: Token -> Posn
tokenPosn (PT p _) = p
tokenPosn (Err p) = p
tokenLineCol :: Token -> (Int, Int)
tokenLineCol = posLineCol . tokenPosn
posLineCol :: Posn -> (Int, Int)
posLineCol (Pn _ l c) = (l,c)
mkPosToken :: Token -> ((Int, Int), String)
mkPosToken t@(PT p _) = (posLineCol p, prToken t)
prToken :: Token -> String
prToken t = case t of
PT _ (TS s _) -> s
PT _ (TL s) -> s
PT _ (TI s) -> s
PT _ (TV s) -> s
PT _ (TD s) -> s
PT _ (TC s) -> s
PT _ (T_Symbol s) -> s
data BTree = N | B String Tok BTree BTree deriving (Show)
eitherResIdent :: (String -> Tok) -> String -> Tok
eitherResIdent tv s = treeFind resWords
where
treeFind N = tv s
treeFind (B a t left right) | s < a = treeFind left
| s > a = treeFind right
| s == a = t
resWords :: BTree
resWords = b "check-sat" 20 (b ">" 10 (b "-" 5 (b "*" 3 (b ")" 2 (b "(" 1 N N) N) (b "+" 4 N N)) (b "=" 8 (b "<=" 7 (b "<" 6 N N) N) (b "=>" 9 N N))) (b "and" 15 (b "Bool" 13 (b "@" 12 (b ">=" 11 N N) N) (b "Int" 14 N N)) (b "assert-not" 18 (b "assert" 17 (b "as" 16 N N) N) (b "case" 19 N N)))) (b "forall" 30 (b "define-funs-rec" 25 (b "declare-sort" 23 (b "declare-fun" 22 (b "declare-datatypes" 21 N N) N) (b "default" 24 N N)) (b "exists" 28 (b "div" 27 (b "distinct" 26 N N) N) (b "false" 29 N N))) (b "mod" 35 (b "let" 33 (b "lambda" 32 (b "ite" 31 N N) N) (b "match" 34 N N)) (b "par" 38 (b "or" 37 (b "not" 36 N N) N) (b "true" 39 N N))))
where b s n = let bs = id s
in B bs (TS bs n)
unescapeInitTail :: String -> String
unescapeInitTail = id . unesc . tail . id where
unesc s = case s of
'\\':c:cs | elem c ['\"', '\\', '\''] -> c : unesc cs
'\\':'n':cs -> '\n' : unesc cs
'\\':'t':cs -> '\t' : unesc cs
'"':[] -> []
c:cs -> c : unesc cs
_ -> []
-------------------------------------------------------------------
-- Alex wrapper code.
-- A modified "posn" wrapper.
-------------------------------------------------------------------
data Posn = Pn !Int !Int !Int
deriving (Eq, Show,Ord)
alexStartPos :: Posn
alexStartPos = Pn 0 1 1
alexMove :: Posn -> Char -> Posn
alexMove (Pn a l c) '\t' = Pn (a+1) l (((c+7) `div` 8)*8+1)
alexMove (Pn a l c) '\n' = Pn (a+1) (l+1) 1
alexMove (Pn a l c) _ = Pn (a+1) l (c+1)
type Byte = Word8
type AlexInput = (Posn, -- current position,
Char, -- previous char
[Byte], -- pending bytes on the current char
String) -- current input string
tokens :: String -> [Token]
tokens str = go (alexStartPos, '\n', [], str)
where
go :: AlexInput -> [Token]
go inp@(pos, _, _, str) =
case alexScan inp 0 of
AlexEOF -> []
AlexError (pos, _, _, _) -> [Err pos]
AlexSkip inp' len -> go inp'
AlexToken inp' len act -> act pos (take len str) : (go inp')
alexGetByte :: AlexInput -> Maybe (Byte,AlexInput)
alexGetByte (p, c, (b:bs), s) = Just (b, (p, c, bs, s))
alexGetByte (p, _, [], s) =
case s of
[] -> Nothing
(c:s) ->
let p' = alexMove p c
(b:bs) = utf8Encode c
in p' `seq` Just (b, (p', c, bs, s))
alexInputPrevChar :: AlexInput -> Char
alexInputPrevChar (p, c, bs, s) = c
-- | Encode a Haskell String to a list of Word8 values, in UTF8 format.
utf8Encode :: Char -> [Word8]
utf8Encode = map fromIntegral . go . ord
where
go oc
| oc <= 0x7f = [oc]
| oc <= 0x7ff = [ 0xc0 + (oc `Data.Bits.shiftR` 6)
, 0x80 + oc Data.Bits..&. 0x3f
]
| oc <= 0xffff = [ 0xe0 + (oc `Data.Bits.shiftR` 12)
, 0x80 + ((oc `Data.Bits.shiftR` 6) Data.Bits..&. 0x3f)
, 0x80 + oc Data.Bits..&. 0x3f
]
| otherwise = [ 0xf0 + (oc `Data.Bits.shiftR` 18)
, 0x80 + ((oc `Data.Bits.shiftR` 12) Data.Bits..&. 0x3f)
, 0x80 + ((oc `Data.Bits.shiftR` 6) Data.Bits..&. 0x3f)
, 0x80 + oc Data.Bits..&. 0x3f
]
}