2 % (c) The GRASP/AQUA Project, Glasgow University, 1992-1998
5 \section[OccName]{@OccName@}
9 -- The NameSpace type; abstact
10 NameSpace, tcName, clsName, tcClsName, dataName, varName, ipName,
11 tvName, uvName, nameSpaceString,
14 OccName, -- Abstract, instance of Outputable
17 mkOccFS, mkSysOcc, mkSysOccFS, mkCCallOcc, mkVarOcc, mkKindOccFS,
18 mkSuperDictSelOcc, mkDFunOcc, mkForeignExportOcc,
19 mkDictOcc, mkIPOcc, mkWorkerOcc, mkMethodOcc, mkDefaultMethodOcc,
20 mkDerivedTyConOcc, mkClassTyConOcc, mkClassDataConOcc, mkSpecOcc,
23 isSysOcc, isTvOcc, isUvOcc, isDataOcc, isDataSymOcc, isSymOcc, isIPOcc, isValOcc,
25 occNameFS, occNameString, occNameUserString, occNameSpace, occNameFlavour,
29 TidyOccEnv, emptyTidyOccEnv, tidyOccName, initTidyOccEnv,
32 EncodedString, EncodedFS, UserString, UserFS, encode, encodeFS, decode, pprEncodedFS,
34 -- The basic form of names
35 isLexCon, isLexVar, isLexId, isLexSym,
36 isLexConId, isLexConSym, isLexVarId, isLexVarSym,
37 isLowerISO, isUpperISO
41 #include "HsVersions.h"
43 import Char ( isDigit, isUpper, isLower, ISALPHANUM, ord, chr, digitToInt )
44 import Util ( thenCmp )
45 import FiniteMap ( FiniteMap, emptyFM, lookupFM, addToFM, elemFM )
50 We hold both module names and identifier names in a 'Z-encoded' form
51 that makes them acceptable both as a C identifier and as a Haskell
54 They can always be decoded again when printing error messages
55 or anything else for the user, but it does make sense for it
56 to be represented here in encoded form, so that when generating
57 code the encoding operation is not performed on each occurrence.
59 These type synonyms help documentation.
62 type UserFS = FAST_STRING -- As the user typed it
63 type EncodedFS = FAST_STRING -- Encoded form
65 type UserString = String -- As the user typed it
66 type EncodedString = String -- Encoded form
69 pprEncodedFS :: EncodedFS -> SDoc
71 = getPprStyle $ \ sty ->
73 -- ptext (decodeFS fs) would needlessly pack the string again
74 then text (decode (_UNPK_ fs))
78 %************************************************************************
80 \subsection{Name space}
82 %************************************************************************
85 data NameSpace = VarName -- Variables
86 | IPName -- Implicit Parameters
87 | DataName -- Data constructors
88 | TvName -- Type variables
89 | UvName -- Usage variables
90 | TcClsName -- Type constructors and classes; Haskell has them
91 -- in the same name space for now.
94 -- Though type constructors and classes are in the same name space now,
95 -- the NameSpace type is abstract, so we can easily separate them later
96 tcName = TcClsName -- Type constructors
97 clsName = TcClsName -- Classes
98 tcClsName = TcClsName -- Not sure which!
107 nameSpaceString :: NameSpace -> String
108 nameSpaceString DataName = "Data constructor"
109 nameSpaceString VarName = "Variable"
110 nameSpaceString IPName = "Implicit Param"
111 nameSpaceString TvName = "Type variable"
112 nameSpaceString UvName = "Usage variable"
113 nameSpaceString TcClsName = "Type constructor or class"
117 %************************************************************************
119 \subsection[Name-pieces-datatypes]{The @OccName@ datatypes}
121 %************************************************************************
124 data OccName = OccName
131 instance Eq OccName where
132 (OccName sp1 s1) == (OccName sp2 s2) = s1 == s2 && sp1 == sp2
134 instance Ord OccName where
135 compare (OccName sp1 s1) (OccName sp2 s2) = (s1 `compare` s2) `thenCmp`
140 %************************************************************************
142 \subsection{Printing}
144 %************************************************************************
147 instance Outputable OccName where
150 pprOccName :: OccName -> SDoc
151 pprOccName (OccName sp occ) = pprEncodedFS occ
155 %************************************************************************
157 \subsection{Construction}
159 %************************************************************************
161 *Sys* things do no encoding; the caller should ensure that the thing is
165 mkSysOcc :: NameSpace -> EncodedString -> OccName
166 mkSysOcc occ_sp str = ASSERT2( alreadyEncoded str, text str )
167 OccName occ_sp (_PK_ str)
169 mkSysOccFS :: NameSpace -> EncodedFS -> OccName
170 mkSysOccFS occ_sp fs = ASSERT2( alreadyEncodedFS fs, ppr fs )
173 mkCCallOcc :: EncodedString -> OccName
174 -- This version of mkSysOcc doesn't check that the string is already encoded,
175 -- because it will be something like "{__ccall f dyn Int# -> Int#}"
176 -- This encodes a lot into something that then parses like an Id.
177 -- But then alreadyEncoded complains about the braces!
178 mkCCallOcc str = OccName varName (_PK_ str)
180 -- Kind constructors get a speical function. Uniquely, they are not encoded,
181 -- so that they have names like '*'. This means that *even in interface files*
182 -- we'll get kinds like (* -> (* -> *)). We can't use mkSysOcc because it
183 -- has an ASSERT that doesn't hold.
184 mkKindOccFS :: NameSpace -> EncodedFS -> OccName
185 mkKindOccFS occ_sp fs = OccName occ_sp fs
188 *Source-code* things are encoded.
191 mkOccFS :: NameSpace -> UserFS -> OccName
192 mkOccFS occ_sp fs = mkSysOccFS occ_sp (encodeFS fs)
194 mkVarOcc :: UserFS -> OccName
195 mkVarOcc fs = mkSysOccFS varName (encodeFS fs)
200 %************************************************************************
202 \subsection{Predicates and taking them apart}
204 %************************************************************************
207 occNameFS :: OccName -> EncodedFS
208 occNameFS (OccName _ s) = s
210 occNameString :: OccName -> EncodedString
211 occNameString (OccName _ s) = _UNPK_ s
213 occNameUserString :: OccName -> UserString
214 occNameUserString occ = decode (occNameString occ)
216 occNameSpace :: OccName -> NameSpace
217 occNameSpace (OccName sp _) = sp
219 setOccNameSpace :: OccName -> NameSpace -> OccName
220 setOccNameSpace (OccName _ occ) sp = OccName sp occ
222 -- occNameFlavour is used only to generate good error messages
223 occNameFlavour :: OccName -> String
224 occNameFlavour (OccName sp _) = nameSpaceString sp
228 isTvOcc, isDataSymOcc, isSymOcc, isUvOcc :: OccName -> Bool
230 isTvOcc (OccName TvName _) = True
231 isTvOcc other = False
233 isUvOcc (OccName UvName _) = True
234 isUvOcc other = False
236 isValOcc (OccName VarName _) = True
237 isValOcc (OccName DataName _) = True
238 isValOcc other = False
240 -- Data constructor operator (starts with ':', or '[]')
241 -- Pretty inefficient!
242 isDataSymOcc (OccName DataName s) = isLexConSym (decodeFS s)
243 isDataSymOcc other = False
245 isDataOcc (OccName DataName _) = True
246 isDataOcc other = False
248 -- Any operator (data constructor or variable)
249 -- Pretty inefficient!
250 isSymOcc (OccName DataName s) = isLexConSym (decodeFS s)
251 isSymOcc (OccName VarName s) = isLexSym (decodeFS s)
253 isIPOcc (OccName IPName _) = True
258 %************************************************************************
260 \subsection{Making system names}
262 %************************************************************************
264 Here's our convention for splitting up the interface file name space:
266 d... dictionary identifiers
267 (local variables, so no name-clash worries)
269 $f... dict-fun identifiers (from inst decls)
270 $dm... default methods
271 $p... superclass selectors
273 $T... compiler-generated tycons for dictionaries
274 $D... ...ditto data cons
275 $sf.. specialised version of f
277 in encoded form these appear as Zdfxxx etc
279 :... keywords (export:, letrec: etc.)
281 This knowledge is encoded in the following functions.
284 @mk_deriv@ generates an @OccName@ from the one-char prefix and a string.
285 NB: The string must already be encoded!
288 mk_deriv :: NameSpace
289 -> String -- Distinguishes one sort of derived name from another
290 -> EncodedString -- Must be already encoded!! We don't want to encode it a
291 -- second time because encoding isn't itempotent
294 mk_deriv occ_sp sys_prefix str = mkSysOcc occ_sp (encode sys_prefix ++ str)
298 mkDictOcc, mkIPOcc, mkWorkerOcc, mkDefaultMethodOcc,
299 mkClassTyConOcc, mkClassDataConOcc, mkSpecOcc
300 :: OccName -> OccName
302 -- These derived variables have a prefix that no Haskell value could have
303 mkWorkerOcc = mk_simple_deriv varName "$w"
304 mkDefaultMethodOcc = mk_simple_deriv varName "$dm"
305 mkDerivedTyConOcc = mk_simple_deriv tcName ":" -- The : prefix makes sure it classifies
306 mkClassTyConOcc = mk_simple_deriv tcName ":T" -- as a tycon/datacon
307 mkClassDataConOcc = mk_simple_deriv dataName ":D" --
308 mkDictOcc = mk_simple_deriv varName "$d"
309 mkIPOcc = mk_simple_deriv varName "$i"
310 mkSpecOcc = mk_simple_deriv varName "$s"
311 mkForeignExportOcc = mk_simple_deriv varName "$f"
312 mkGenOcc1 = mk_simple_deriv varName "$gfrom" -- Generics
313 mkGenOcc2 = mk_simple_deriv varName "$gto" -- Generics
314 mk_simple_deriv sp px occ = mk_deriv sp px (occNameString occ)
317 isSysOcc :: OccName -> Bool -- True for all these '$' things
318 isSysOcc occ = case occNameUserString occ of
320 other -> False -- We don't care about the ':' ones
321 -- isSysOcc is only called for Ids anyway
325 mkSuperDictSelOcc :: Int -- Index of superclass, eg 3
326 -> OccName -- Class, eg "Ord"
327 -> OccName -- eg "p3Ord"
328 mkSuperDictSelOcc index cls_occ
329 = mk_deriv varName "$p" (show index ++ occNameString cls_occ)
334 mkDFunOcc :: EncodedString -- Typically the class and type glommed together e.g. "OrdMaybe"
335 -> Int -- Unique to distinguish dfuns which share the previous two
337 -- The requirement is that the (string,index) pair be unique in this module
339 -> OccName -- "$fOrdMaybe3"
341 mkDFunOcc string index
342 = mk_deriv VarName "$f" (show_index ++ string)
344 show_index | index == 0 = ""
345 | otherwise = show index
348 We used to add a '$m' to indicate a method, but that gives rise to bad
349 error messages from the type checker when we print the function name or pattern
350 of an instance-decl binding. Why? Because the binding is zapped
351 to use the method name in place of the selector name.
352 (See TcClassDcl.tcMethodBind)
354 The way it is now, -ddump-xx output may look confusing, but
355 you can always say -dppr-debug to get the uniques.
357 However, we *do* have to zap the first character to be lower case,
358 because overloaded constructors (blarg) generate methods too.
359 And convert to VarName space
361 e.g. a call to constructor MkFoo where
362 data (Ord a) => Foo a = MkFoo a
364 If this is necessary, we do it by prefixing '$m'. These
365 guys never show up in error messages. What a hack.
368 mkMethodOcc :: OccName -> OccName
369 mkMethodOcc occ@(OccName VarName fs) = occ
370 mkMethodOcc occ = mk_simple_deriv varName "$m" occ
374 %************************************************************************
376 \subsection{Tidying them up}
378 %************************************************************************
380 Before we print chunks of code we like to rename it so that
381 we don't have to print lots of silly uniques in it. But we mustn't
382 accidentally introduce name clashes! So the idea is that we leave the
383 OccName alone unless it accidentally clashes with one that is already
384 in scope; if so, we tack on '1' at the end and try again, then '2', and
385 so on till we find a unique one.
387 There's a wrinkle for operators. Consider '>>='. We can't use '>>=1'
388 because that isn't a single lexeme. So we encode it to 'lle' and *then*
389 tack on the '1', if necessary.
392 type TidyOccEnv = FiniteMap FAST_STRING Int -- The in-scope OccNames
393 emptyTidyOccEnv = emptyFM
395 initTidyOccEnv :: [OccName] -> TidyOccEnv -- Initialise with names to avoid!
396 initTidyOccEnv = foldl (\env (OccName _ fs) -> addToFM env fs 1) emptyTidyOccEnv
398 tidyOccName :: TidyOccEnv -> OccName -> (TidyOccEnv, OccName)
400 tidyOccName in_scope occ@(OccName occ_sp fs)
401 | not (fs `elemFM` in_scope)
402 = (addToFM in_scope fs 1, occ) -- First occurrence
404 | otherwise -- Already occurs
405 = go in_scope (_UNPK_ fs)
408 go in_scope str = case lookupFM in_scope pk_str of
409 Just n -> go (addToFM in_scope pk_str (n+1)) (str ++ show n)
410 -- Need to go round again, just in case "t3" (say)
411 -- clashes with a "t3" that's already in scope
413 Nothing -> (addToFM in_scope pk_str 1, mkSysOccFS occ_sp pk_str)
420 %************************************************************************
422 \subsection{The 'Z' encoding}
424 %************************************************************************
426 This is the main name-encoding and decoding function. It encodes any
427 string into a string that is acceptable as a C name. This is the name
428 by which things are known right through the compiler.
430 The basic encoding scheme is this.
432 * Tuples (,,,) are coded as Z3T
434 * Alphabetic characters (upper and lower) and digits
435 all translate to themselves;
436 except 'Z', which translates to 'ZZ'
437 and 'z', which translates to 'zz'
438 We need both so that we can preserve the variable/tycon distinction
440 * Most other printable characters translate to 'zx' or 'Zx' for some
441 alphabetic character x
443 * The others translate as 'znnnU' where 'nnn' is the decimal number
447 --------------------------
462 -- alreadyEncoded is used in ASSERTs to check for encoded
463 -- strings. It isn't fail-safe, of course, because, say 'zh' might
464 -- be encoded or not.
465 alreadyEncoded :: String -> Bool
466 alreadyEncoded s = all ok s
469 -- This is a bit of a lie; if we really wanted spaces
470 -- in names we'd have to encode them. But we do put
471 -- spaces in ccall "occurrences", and we don't want to
473 ok ch = ISALPHANUM ch
475 alreadyEncodedFS :: FAST_STRING -> Bool
476 alreadyEncodedFS fs = alreadyEncoded (_UNPK_ fs)
478 encode :: UserString -> EncodedString
479 encode cs = case maybe_tuple cs of
480 Just n -> 'Z' : show n ++ "T" -- Tuples go to Z2T etc
484 go (c:cs) = encode_ch c ++ go cs
486 -- ToDo: Unboxed tuples too, perhaps?
487 maybe_tuple ('(' : cs) = check_tuple (0::Int) cs
488 maybe_tuple other = Nothing
490 check_tuple :: Int -> String -> Maybe Int
491 check_tuple n (',' : cs) = check_tuple (n+1) cs
492 check_tuple n ")" = Just n
493 check_tuple n other = Nothing
495 encodeFS :: UserFS -> EncodedFS
496 encodeFS fast_str | all unencodedChar str = fast_str
497 | otherwise = _PK_ (encode str)
499 str = _UNPK_ fast_str
501 unencodedChar :: Char -> Bool -- True for chars that don't need encoding
502 unencodedChar 'Z' = False
503 unencodedChar 'z' = False
504 unencodedChar c = ISALPHANUM c
506 encode_ch :: Char -> EncodedString
507 encode_ch c | unencodedChar c = [c] -- Common case first
510 encode_ch '(' = "ZL" -- Needed for things like (,), and (->)
511 encode_ch ')' = "ZR" -- For symmetry with (
531 encode_ch '\'' = "zq"
532 encode_ch '\\' = "zr"
537 encode_ch c = 'z' : shows (ord c) "U"
540 Decode is used for user printing.
543 decodeFS :: FAST_STRING -> FAST_STRING
544 decodeFS fs = _PK_ (decode (_UNPK_ fs))
546 decode :: EncodedString -> UserString
548 decode ('Z' : rest) = decode_escape rest
549 decode ('z' : rest) = decode_escape rest
550 decode (c : rest) = c : decode rest
552 decode_escape :: EncodedString -> UserString
554 decode_escape ('L' : rest) = '(' : decode rest
555 decode_escape ('R' : rest) = ')' : decode rest
556 decode_escape ('M' : rest) = '[' : decode rest
557 decode_escape ('N' : rest) = ']' : decode rest
558 decode_escape ('C' : rest) = ':' : decode rest
559 decode_escape ('Z' : rest) = 'Z' : decode rest
561 decode_escape ('z' : rest) = 'z' : decode rest
562 decode_escape ('a' : rest) = '&' : decode rest
563 decode_escape ('b' : rest) = '|' : decode rest
564 decode_escape ('c' : rest) = '^' : decode rest
565 decode_escape ('d' : rest) = '$' : decode rest
566 decode_escape ('e' : rest) = '=' : decode rest
567 decode_escape ('g' : rest) = '>' : decode rest
568 decode_escape ('h' : rest) = '#' : decode rest
569 decode_escape ('i' : rest) = '.' : decode rest
570 decode_escape ('l' : rest) = '<' : decode rest
571 decode_escape ('m' : rest) = '-' : decode rest
572 decode_escape ('n' : rest) = '!' : decode rest
573 decode_escape ('p' : rest) = '+' : decode rest
574 decode_escape ('q' : rest) = '\'' : decode rest
575 decode_escape ('r' : rest) = '\\' : decode rest
576 decode_escape ('s' : rest) = '/' : decode rest
577 decode_escape ('t' : rest) = '*' : decode rest
578 decode_escape ('u' : rest) = '_' : decode rest
579 decode_escape ('v' : rest) = '%' : decode rest
581 -- Tuples are coded as Z23T
582 -- Characters not having a specific code are coded as z224U
583 decode_escape (c : rest)
584 | isDigit c = go (digitToInt c) rest
586 go n (c : rest) | isDigit c = go (10*n + digitToInt c) rest
587 go n ('T' : rest) = '(' : replicate n ',' ++ ')' : decode rest
588 go n ('U' : rest) = chr n : decode rest
589 go n other = pprPanic "decode_escape" (ppr n <+> text (c:rest))
591 decode_escape (c : rest) = pprTrace "decode_escape" (char c) (decode rest)
595 %************************************************************************
597 n\subsection{Lexical categories}
599 %************************************************************************
601 These functions test strings to see if they fit the lexical categories
602 defined in the Haskell report.
605 isLexCon, isLexVar, isLexId, isLexSym :: FAST_STRING -> Bool
606 isLexConId, isLexConSym, isLexVarId, isLexVarSym :: FAST_STRING -> Bool
608 isLexCon cs = isLexConId cs || isLexConSym cs
609 isLexVar cs = isLexVarId cs || isLexVarSym cs
611 isLexId cs = isLexConId cs || isLexVarId cs
612 isLexSym cs = isLexConSym cs || isLexVarSym cs
616 isLexConId cs -- Prefix type or data constructors
617 | _NULL_ cs = False -- e.g. "Foo", "[]", "(,)"
618 | cs == SLIT("[]") = True
619 | otherwise = startsConId (_HEAD_ cs)
621 isLexVarId cs -- Ordinary prefix identifiers
622 | _NULL_ cs = False -- e.g. "x", "_x"
623 | otherwise = startsVarId (_HEAD_ cs)
625 isLexConSym cs -- Infix type or data constructors
626 | _NULL_ cs = False -- e.g. ":-:", ":", "->"
627 | cs == SLIT("->") = True
628 | otherwise = startsConSym (_HEAD_ cs)
630 isLexVarSym cs -- Infix identifiers
631 | _NULL_ cs = False -- e.g. "+"
632 | otherwise = startsVarSym (_HEAD_ cs)
635 startsVarSym, startsVarId, startsConSym, startsConId :: Char -> Bool
636 startsVarSym c = isSymbolASCII c || isSymbolISO c -- Infix Ids
637 startsConSym c = c == ':' -- Infix data constructors
638 startsVarId c = isLower c || isLowerISO c || c == '_' -- Ordinary Ids
639 startsConId c = isUpper c || isUpperISO c || c == '(' -- Ordinary type constructors and data constructors
642 isSymbolASCII c = c `elem` "!#$%&*+./<=>?@\\^|~-"
643 isSymbolISO c = ord c `elem` (0xd7 : 0xf7 : [0xa1 .. 0xbf])
644 isUpperISO (C# c#) = c# `geChar#` '\xc0'# && c# `leChar#` '\xde'# && c# `neChar#` '\xd7'#
645 --0xc0 <= oc && oc <= 0xde && oc /= 0xd7 where oc = ord c
646 isLowerISO (C# c#) = c# `geChar#` '\xdf'# && c# `leChar#` '\xff'# && c# `neChar#` '\xf7'#
647 --0xdf <= oc && oc <= 0xff && oc /= 0xf7 where oc = ord c