1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
|
-- Copyright 2022 Google LLC
--
-- Use of this source code is governed by a BSD-style
-- license that can be found in the LICENSE file or at
-- https://developers.google.com/open-source/licenses/bsd
module Lexing where
import Control.Monad.State.Strict
import Data.Char
import Data.HashSet qualified as HS
import qualified Data.Scientific as Scientific
import Data.String (fromString)
import Data.Text (Text)
import Data.Text qualified as T
import Data.Void
import Data.Word
import qualified Data.Map.Strict as M
import Text.Megaparsec hiding (Label, State)
import Text.Megaparsec.Char hiding (space, eol)
import qualified Text.Megaparsec.Char as MC
import qualified Text.Megaparsec.Char.Lexer as L
import Text.Megaparsec.Debug
import Err
import PPrint
import Types.Primitives
import Types.Source
import Util (toSnocList)
data ParseCtx = ParseCtx
{ curIndent :: Int -- used Reader-style (i.e. ask/local)
, canBreak :: Bool -- used Reader-style (i.e. ask/local)
, prevWhitespace :: Bool -- tracks whether we just consumed whitespace
, sourceIdCounter :: Int -- starts at 1 (0 is reserved for the root)
, curAtomicLexemes :: [SrcId]
, curLexemeInfo :: LexemeInfo } -- append to, writer-style
initParseCtx :: ParseCtx
initParseCtx = ParseCtx 0 False False 1 mempty mempty
type Parser = StateT ParseCtx (Parsec Void Text)
parseit :: Text -> Parser a -> Except a
parseit s p = case parse (fst <$> runStateT p initParseCtx) "" s of
Left e -> throw rootSrcId $ MiscParseErr $ errorBundlePretty e
Right x -> return x
mustParseit :: Text -> Parser a -> a
mustParseit s p = case parseit s p of
Success x -> x
Failure e -> error $ "This shouldn't happen:\n" ++ pprint e
debug :: (Show a) => String -> Parser a -> Parser a
debug lbl action = do
ctx <- get
lift $ dbg lbl $ fst <$> runStateT action ctx
-- === Lexemes ===
type Lexer = Parser
nextChar :: Lexer Char
nextChar = do
i <- getInput
guard $ not $ T.null i
return $ T.head i
{-# INLINE nextChar #-}
anyCaseName :: Lexer (WithSrc SourceName)
anyCaseName = label "name" $ lexeme LowerName anyCaseName' -- TODO: distinguish lowercase/uppercase
anyCaseName' :: Lexer SourceName
anyCaseName' =
liftM MkSourceName $ checkNotKeyword $ (:) <$> satisfy (\c -> isLower c || isUpper c) <*>
(T.unpack <$> takeWhileP Nothing (\c -> isAlphaNum c || c == '\'' || c == '_'))
anyName :: Lexer (WithSrc SourceName)
anyName = anyCaseName <|> symName
checkNotKeyword :: Parser String -> Parser String
checkNotKeyword p = try $ do
s <- p
failIf (s `HS.member` keyWordSet) $ show s ++ " is a reserved word"
return s
{-# INLINE checkNotKeyword #-}
data KeyWord = DefKW | ForKW | For_KW | RofKW | Rof_KW | CaseKW | OfKW
| DataKW | StructKW | InterfaceKW
| InstanceKW | GivenKW | WithKW | SatisfyingKW
| IfKW | ThenKW | ElseKW | DoKW
| ImportKW | ForeignKW | NamedInstanceKW
| EffectKW | HandlerKW | JmpKW | CtlKW | ReturnKW | ResumeKW
| CustomLinearizationKW | CustomLinearizationSymbolicKW | PassKW
deriving (Enum)
keyWordToken :: KeyWord -> String
keyWordToken = \case
DefKW -> "def"
ForKW -> "for"
RofKW -> "rof"
For_KW -> "for_"
Rof_KW -> "rof_"
CaseKW -> "case"
IfKW -> "if"
ThenKW -> "then"
ElseKW -> "else"
OfKW -> "of"
DataKW -> "data"
StructKW -> "struct"
InterfaceKW -> "interface"
InstanceKW -> "instance"
NamedInstanceKW -> "named-instance"
GivenKW -> "given"
WithKW -> "with"
SatisfyingKW -> "satisfying"
DoKW -> "do"
ImportKW -> "import"
ForeignKW -> "foreign"
EffectKW -> "effect"
HandlerKW -> "handler"
JmpKW -> "jmp"
CtlKW -> "ctl"
ReturnKW -> "return"
ResumeKW -> "resume"
CustomLinearizationKW -> "custom-linearization"
CustomLinearizationSymbolicKW -> "custom-linearization-symbolic"
PassKW -> "pass"
keyWord :: KeyWord -> Lexer ()
keyWord kw = atomicLexeme Keyword $ try $
string (fromString $ keyWordToken kw) >> notFollowedBy nameTailChar
where
nameTailChar :: Parser Char
nameTailChar = alphaNumChar <|> char '\'' <|> char '_'
keyWordSet :: HS.HashSet String
keyWordSet = HS.fromList keyWordStrs
keyWordStrs :: [String]
keyWordStrs = map keyWordToken [DefKW .. PassKW]
primName :: Lexer (WithSrc String)
primName = lexeme MiscLexeme $ try $ char '%' >> some alphaNumChar
charLit :: Lexer (WithSrc Char)
charLit = lexeme MiscLexeme $ char '\'' >> L.charLiteral <* char '\''
strLit :: Lexer (WithSrc String)
strLit = lexeme StringLiteralLexeme $ char '"' >> manyTill L.charLiteral (char '"')
natLit :: Lexer (WithSrc Word64)
natLit = lexeme LiteralLexeme $ try $ L.decimal <* notFollowedBy (char '.')
doubleLit :: Lexer (WithSrc Double)
doubleLit = lexeme LiteralLexeme $
try L.float
<|> try (fromIntegral <$> (L.decimal :: Parser Int) <* char '.')
<|> try do
s <- L.scientific
case Scientific.toBoundedRealFloat s of
Right f -> return f
Left _ -> fail "Non-representable floating point literal"
knownSymStrs :: HS.HashSet String
knownSymStrs = HS.fromList
[ ".", ":", "::", "!", "=", "-", "+", "||", "&&"
, "$", "&>", "|", ",", ",>", "<-", "+=", ":="
, "->", "->>", "=>", "?->", "?=>", "--o", "--", "<<<", ">>>"
, "..", "<..", "..<", "..<", "<..<", "?", "#", "##", "#?", "#&", "#|", "@"]
sym :: Text -> Lexer ()
sym s = atomicLexeme Symbol $ sym' s
symWithId :: Text -> Lexer SrcId
symWithId s = liftM srcPos $ lexeme Symbol $ sym' s
-- string must be in `knownSymStrs`
sym' :: Text -> Lexer ()
sym' s = void $ try $ string s >> notFollowedBy symChar
anySym :: Lexer (WithSrc String)
anySym = lexeme Symbol $ try $ do
s <- some symChar
failIf (s `HS.member` knownSymStrs) ""
return s
symName :: Lexer (WithSrc SourceName)
symName = label "symbol name" $ lexeme Symbol $ try $ do
s <- between (char '(') (char ')') $ some symChar
return $ MkSourceName $ "(" <> s <> ")"
backquoteName :: Lexer (WithSrc SourceName)
backquoteName = label "backquoted name" $
lexeme Symbol $ try $ between (char '`') (char '`') anyCaseName'
-- brackets and punctuation
-- (can't treat as sym because e.g. `((` is two separate lexemes)
lParen, rParen, lBracket, rBracket, lBrace, rBrace, semicolon, underscore :: Lexer ()
lParen = charLexeme '('
rParen = charLexeme ')'
lBracket = charLexeme '['
rBracket = charLexeme ']'
lBrace = charLexeme '{'
rBrace = charLexeme '}'
semicolon = charLexeme ';'
underscore = charLexeme '_'
charLexeme :: Char -> Parser ()
charLexeme c = atomicLexeme Symbol $ void $ char c
symChar :: Parser Char
symChar = token (\c -> if HS.member c symChars then Just c else Nothing) mempty
symChars :: HS.HashSet Char
symChars = HS.fromList ".,!$^&*:-~+/=<>|?\\@#"
-- XXX: unlike other lexemes, this doesn't consume trailing whitespace
dot :: Parser SrcId
dot = srcPos <$> lexeme' (return ()) Symbol (void $ char '.')
-- === Util ===
sc :: Parser ()
sc = (skipSome s >> recordWhitespace) <|> return ()
where s = hidden space <|> hidden lineComment
lineComment :: Parser ()
lineComment = do
try $ string "--" >> notFollowedBy (void (char 'o'))
void (takeWhileP (Just "char") (/= '\n'))
outputLines :: Parser ()
outputLines = void $ many (symbol ">" >> takeWhileP Nothing (/= '\n') >> ((eol >> return ()) <|> eof))
space :: Parser ()
space = gets canBreak >>= \case
True -> space1
False -> void $ takeWhile1P (Just "white space") (`elem` (" \t" :: String))
setCanBreakLocally :: Bool -> Parser a -> Parser a
setCanBreakLocally brLocal p = do
brPrev <- gets canBreak
modify \ctx -> ctx {canBreak = brLocal}
ans <- p
modify \ctx -> ctx {canBreak = brPrev}
return ans
{-# INLINE setCanBreakLocally #-}
mayBreak :: Parser a -> Parser a
mayBreak p = setCanBreakLocally True p
{-# INLINE mayBreak #-}
mayNotBreak :: Parser a -> Parser a
mayNotBreak p = setCanBreakLocally False p
{-# INLINE mayNotBreak #-}
precededByWhitespace :: Parser Bool
precededByWhitespace = gets prevWhitespace
{-# INLINE precededByWhitespace #-}
recordWhitespace :: Parser ()
recordWhitespace = modify \ctx -> ctx { prevWhitespace = True }
{-# INLINE recordWhitespace #-}
recordNonWhitespace :: Parser ()
recordNonWhitespace = modify \ctx -> ctx { prevWhitespace = False }
{-# INLINE recordNonWhitespace #-}
nameString :: Parser String
nameString = lexemeIgnoreSrcId LowerName . try $ (:) <$> lowerChar <*> many alphaNumChar
thisNameString :: Text -> Parser ()
thisNameString s = lexemeIgnoreSrcId MiscLexeme $ try $ string s >> notFollowedBy alphaNumChar
bracketed :: Parser () -> Parser () -> Parser a -> Parser a
bracketed left right p = do
left
ans <- mayBreak $ sc >> p
right
return ans
{-# INLINE bracketed #-}
braces :: Parser a -> Parser a
braces p = bracketed lBrace rBrace p
{-# INLINE braces #-}
nextLine :: Parser ()
nextLine = do
eol
n <- curIndent <$> get
void $ mayNotBreak $ many $ try (sc >> eol)
void $ replicateM n (char ' ')
withSource :: Parser a -> Parser (Text, a)
withSource p = do
s <- getInput
start <- getOffset
x <- p
end <- getOffset
return (T.take (end - start) s, x)
{-# INLINE withSource #-}
withIndent :: Parser a -> Parser a
withIndent p = do
nextLine
indent <- T.length <$> takeWhileP (Just "space") (==' ')
when (indent <= 0) empty
locallyExtendCurIndent indent $ mayNotBreak p
{-# INLINE withIndent #-}
locallyExtendCurIndent :: Int -> Parser a -> Parser a
locallyExtendCurIndent n p = do
indentPrev <- gets curIndent
modify \ctx -> ctx { curIndent = indentPrev + n }
ans <- p
modify \ctx -> ctx { curIndent = indentPrev }
return ans
eol :: Parser ()
eol = void MC.eol
eolf :: Parser ()
eolf = eol <|> eof
failIf :: Bool -> String -> Parser ()
failIf True s = fail s
failIf False _ = return ()
freshSrcId :: Parser SrcId
freshSrcId = do
c <- gets sourceIdCounter
modify \ctx -> ctx { sourceIdCounter = c + 1 }
return $ SrcId c
withLexemeInfo :: Parser a -> Parser (LexemeInfo, a)
withLexemeInfo cont = do
smPrev <- gets curLexemeInfo
modify \ctx -> ctx { curLexemeInfo = mempty }
result <- cont
sm <- gets curLexemeInfo
modify \ctx -> ctx { curLexemeInfo = smPrev }
return (sm, result)
emitLexemeInfo :: LexemeInfo -> Parser ()
emitLexemeInfo m = modify \ctx -> ctx { curLexemeInfo = curLexemeInfo ctx <> m }
lexemeIgnoreSrcId :: LexemeType -> Parser a -> Parser a
lexemeIgnoreSrcId lexemeType p = withoutSrc <$> lexeme lexemeType p
symbol :: Text -> Parser ()
symbol s = void $ L.symbol sc s
lexeme :: LexemeType -> Parser a -> Parser (WithSrc a)
lexeme lexemeType p = lexeme' sc lexemeType p
{-# INLINE lexeme #-}
lexeme' :: Parser () -> LexemeType -> Parser a -> Parser (WithSrc a)
lexeme' sc' lexemeType p = do
start <- getOffset
ans <- p
end <- getOffset
recordNonWhitespace
sc'
sid <- freshSrcId
emitLexemeInfo $ mempty
{ lexemeList = toSnocList [sid]
, lexemeInfo = M.singleton sid (lexemeType, (start, end)) }
return $ WithSrc sid ans
{-# INLINE lexeme' #-}
atomicLexeme :: LexemeType -> Parser () -> Parser ()
atomicLexeme lexemeType p = do
WithSrc sid () <- lexeme lexemeType p
modify \ctx -> ctx { curAtomicLexemes = curAtomicLexemes ctx ++ [sid] }
{-# INLINE atomicLexeme #-}
collectAtomicLexemeIds :: Parser a -> Parser ([SrcId], a)
collectAtomicLexemeIds p = do
prevAtomicLexemes <- gets curAtomicLexemes
modify \ctx -> ctx { curAtomicLexemes = [] }
ans <- p
localLexemes <- gets curAtomicLexemes
modify \ctx -> ctx { curAtomicLexemes = prevAtomicLexemes }
return (localLexemes, ans)
|