2 % (c) The University of Glasgow 2006
3 % (c) The GRASP/AQUA Project, Glasgow University, 1992-1998
10 tcSimplifyInfer, tcSimplifyInferCheck,
11 tcSimplifyCheck, tcSimplifyRestricted,
12 tcSimplifyRuleLhs, tcSimplifyIPs,
13 tcSimplifySuperClasses,
14 tcSimplifyTop, tcSimplifyInteractive,
15 tcSimplifyBracket, tcSimplifyCheckPat,
17 tcSimplifyDeriv, tcSimplifyDefault,
23 #include "HsVersions.h"
25 import {-# SOURCE #-} TcUnify( unifyType )
29 import TcHsSyn ( hsLPatType )
37 import DsUtils -- Big-tuple functions
67 %************************************************************************
71 %************************************************************************
73 --------------------------------------
74 Notes on functional dependencies (a bug)
75 --------------------------------------
82 instance D a b => C a b -- Undecidable
83 -- (Not sure if it's crucial to this eg)
84 f :: C a b => a -> Bool
87 g :: C a b => a -> Bool
90 Here f typechecks, but g does not!! Reason: before doing improvement,
91 we reduce the (C a b1) constraint from the call of f to (D a b1).
93 Here is a more complicated example:
96 > class Foo a b | a->b
98 > class Bar a b | a->b
102 > instance Bar Obj Obj
104 > instance (Bar a b) => Foo a b
106 > foo:: (Foo a b) => a -> String
109 > runFoo:: (forall a b. (Foo a b) => a -> w) -> w
115 Could not deduce (Bar a b) from the context (Foo a b)
116 arising from use of `foo' at <interactive>:1
118 Add (Bar a b) to the expected type of an expression
119 In the first argument of `runFoo', namely `foo'
120 In the definition of `it': it = runFoo foo
122 Why all of the sudden does GHC need the constraint Bar a b? The
123 function foo didn't ask for that...
126 The trouble is that to type (runFoo foo), GHC has to solve the problem:
128 Given constraint Foo a b
129 Solve constraint Foo a b'
131 Notice that b and b' aren't the same. To solve this, just do
132 improvement and then they are the same. But GHC currently does
137 That is usually fine, but it isn't here, because it sees that Foo a b is
138 not the same as Foo a b', and so instead applies the instance decl for
139 instance Bar a b => Foo a b. And that's where the Bar constraint comes
142 The Right Thing is to improve whenever the constraint set changes at
143 all. Not hard in principle, but it'll take a bit of fiddling to do.
145 Note [Choosing which variables to quantify]
146 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
147 Suppose we are about to do a generalisation step. We have in our hand
150 T the type of the RHS
151 C the constraints from that RHS
153 The game is to figure out
155 Q the set of type variables over which to quantify
156 Ct the constraints we will *not* quantify over
157 Cq the constraints we will quantify over
159 So we're going to infer the type
163 and float the constraints Ct further outwards.
165 Here are the things that *must* be true:
167 (A) Q intersect fv(G) = EMPTY limits how big Q can be
168 (B) Q superset fv(Cq union T) \ oclose(fv(G),C) limits how small Q can be
170 (A) says we can't quantify over a variable that's free in the environment.
171 (B) says we must quantify over all the truly free variables in T, else
172 we won't get a sufficiently general type.
174 We do not *need* to quantify over any variable that is fixed by the
175 free vars of the environment G.
177 BETWEEN THESE TWO BOUNDS, ANY Q WILL DO!
179 Example: class H x y | x->y where ...
181 fv(G) = {a} C = {H a b, H c d}
184 (A) Q intersect {a} is empty
185 (B) Q superset {a,b,c,d} \ oclose({a}, C) = {a,b,c,d} \ {a,b} = {c,d}
187 So Q can be {c,d}, {b,c,d}
189 In particular, it's perfectly OK to quantify over more type variables
190 than strictly necessary; there is no need to quantify over 'b', since
191 it is determined by 'a' which is free in the envt, but it's perfectly
192 OK to do so. However we must not quantify over 'a' itself.
194 Other things being equal, however, we'd like to quantify over as few
195 variables as possible: smaller types, fewer type applications, more
196 constraints can get into Ct instead of Cq. Here's a good way to
199 Q = grow( fv(T), C ) \ oclose( fv(G), C )
201 That is, quantify over all variable that that MIGHT be fixed by the
202 call site (which influences T), but which aren't DEFINITELY fixed by
203 G. This choice definitely quantifies over enough type variables,
204 albeit perhaps too many.
206 Why grow( fv(T), C ) rather than fv(T)? Consider
208 class H x y | x->y where ...
213 If we used fv(T) = {c} we'd get the type
215 forall c. H c d => c -> b
217 And then if the fn was called at several different c's, each of
218 which fixed d differently, we'd get a unification error, because
219 d isn't quantified. Solution: quantify d. So we must quantify
220 everything that might be influenced by c.
222 Why not oclose( fv(T), C )? Because we might not be able to see
223 all the functional dependencies yet:
225 class H x y | x->y where ...
226 instance H x y => Eq (T x y) where ...
231 Now oclose(fv(T),C) = {c}, because the functional dependency isn't
232 apparent yet, and that's wrong. We must really quantify over d too.
234 There really isn't any point in quantifying over any more than
235 grow( fv(T), C ), because the call sites can't possibly influence
236 any other type variables.
240 -------------------------------------
242 -------------------------------------
244 It's very hard to be certain when a type is ambiguous. Consider
248 instance H x y => K (x,y)
250 Is this type ambiguous?
251 forall a b. (K (a,b), Eq b) => a -> a
253 Looks like it! But if we simplify (K (a,b)) we get (H a b) and
254 now we see that a fixes b. So we can't tell about ambiguity for sure
255 without doing a full simplification. And even that isn't possible if
256 the context has some free vars that may get unified. Urgle!
258 Here's another example: is this ambiguous?
259 forall a b. Eq (T b) => a -> a
260 Not if there's an insance decl (with no context)
261 instance Eq (T b) where ...
263 You may say of this example that we should use the instance decl right
264 away, but you can't always do that:
266 class J a b where ...
267 instance J Int b where ...
269 f :: forall a b. J a b => a -> a
271 (Notice: no functional dependency in J's class decl.)
272 Here f's type is perfectly fine, provided f is only called at Int.
273 It's premature to complain when meeting f's signature, or even
274 when inferring a type for f.
278 However, we don't *need* to report ambiguity right away. It'll always
279 show up at the call site.... and eventually at main, which needs special
280 treatment. Nevertheless, reporting ambiguity promptly is an excellent thing.
282 So here's the plan. We WARN about probable ambiguity if
284 fv(Cq) is not a subset of oclose(fv(T) union fv(G), C)
286 (all tested before quantification).
287 That is, all the type variables in Cq must be fixed by the the variables
288 in the environment, or by the variables in the type.
290 Notice that we union before calling oclose. Here's an example:
292 class J a b c | a b -> c
296 forall b c. (J a b c) => b -> b
298 Only if we union {a} from G with {b} from T before using oclose,
299 do we see that c is fixed.
301 It's a bit vague exactly which C we should use for this oclose call. If we
302 don't fix enough variables we might complain when we shouldn't (see
303 the above nasty example). Nothing will be perfect. That's why we can
304 only issue a warning.
307 Can we ever be *certain* about ambiguity? Yes: if there's a constraint
309 c in C such that fv(c) intersect (fv(G) union fv(T)) = EMPTY
311 then c is a "bubble"; there's no way it can ever improve, and it's
312 certainly ambiguous. UNLESS it is a constant (sigh). And what about
317 instance H x y => K (x,y)
319 Is this type ambiguous?
320 forall a b. (K (a,b), Eq b) => a -> a
322 Urk. The (Eq b) looks "definitely ambiguous" but it isn't. What we are after
323 is a "bubble" that's a set of constraints
325 Cq = Ca union Cq' st fv(Ca) intersect (fv(Cq') union fv(T) union fv(G)) = EMPTY
327 Hence another idea. To decide Q start with fv(T) and grow it
328 by transitive closure in Cq (no functional dependencies involved).
329 Now partition Cq using Q, leaving the definitely-ambiguous and probably-ok.
330 The definitely-ambiguous can then float out, and get smashed at top level
331 (which squashes out the constants, like Eq (T a) above)
334 --------------------------------------
335 Notes on principal types
336 --------------------------------------
341 f x = let g y = op (y::Int) in True
343 Here the principal type of f is (forall a. a->a)
344 but we'll produce the non-principal type
345 f :: forall a. C Int => a -> a
348 --------------------------------------
349 The need for forall's in constraints
350 --------------------------------------
352 [Exchange on Haskell Cafe 5/6 Dec 2000]
354 class C t where op :: t -> Bool
355 instance C [t] where op x = True
357 p y = (let f :: c -> Bool; f x = op (y >> return x) in f, y ++ [])
358 q y = (y ++ [], let f :: c -> Bool; f x = op (y >> return x) in f)
360 The definitions of p and q differ only in the order of the components in
361 the pair on their right-hand sides. And yet:
363 ghc and "Typing Haskell in Haskell" reject p, but accept q;
364 Hugs rejects q, but accepts p;
365 hbc rejects both p and q;
366 nhc98 ... (Malcolm, can you fill in the blank for us!).
368 The type signature for f forces context reduction to take place, and
369 the results of this depend on whether or not the type of y is known,
370 which in turn depends on which component of the pair the type checker
373 Solution: if y::m a, float out the constraints
374 Monad m, forall c. C (m c)
375 When m is later unified with [], we can solve both constraints.
378 --------------------------------------
379 Notes on implicit parameters
380 --------------------------------------
382 Note [Inheriting implicit parameters]
383 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
388 where f is *not* a top-level binding.
389 From the RHS of f we'll get the constraint (?y::Int).
390 There are two types we might infer for f:
394 (so we get ?y from the context of f's definition), or
396 f :: (?y::Int) => Int -> Int
398 At first you might think the first was better, becuase then
399 ?y behaves like a free variable of the definition, rather than
400 having to be passed at each call site. But of course, the WHOLE
401 IDEA is that ?y should be passed at each call site (that's what
402 dynamic binding means) so we'd better infer the second.
404 BOTTOM LINE: when *inferring types* you *must* quantify
405 over implicit parameters. See the predicate isFreeWhenInferring.
408 Note [Implicit parameters and ambiguity]
409 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
410 Only a *class* predicate can give rise to ambiguity
411 An *implicit parameter* cannot. For example:
412 foo :: (?x :: [a]) => Int
414 is fine. The call site will suppply a particular 'x'
416 Furthermore, the type variables fixed by an implicit parameter
417 propagate to the others. E.g.
418 foo :: (Show a, ?x::[a]) => Int
420 The type of foo looks ambiguous. But it isn't, because at a call site
422 let ?x = 5::Int in foo
423 and all is well. In effect, implicit parameters are, well, parameters,
424 so we can take their type variables into account as part of the
425 "tau-tvs" stuff. This is done in the function 'FunDeps.grow'.
428 Question 2: type signatures
429 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
430 BUT WATCH OUT: When you supply a type signature, we can't force you
431 to quantify over implicit parameters. For example:
435 This is perfectly reasonable. We do not want to insist on
437 (?x + 1) :: (?x::Int => Int)
439 That would be silly. Here, the definition site *is* the occurrence site,
440 so the above strictures don't apply. Hence the difference between
441 tcSimplifyCheck (which *does* allow implicit paramters to be inherited)
442 and tcSimplifyCheckBind (which does not).
444 What about when you supply a type signature for a binding?
445 Is it legal to give the following explicit, user type
446 signature to f, thus:
451 At first sight this seems reasonable, but it has the nasty property
452 that adding a type signature changes the dynamic semantics.
455 (let f x = (x::Int) + ?y
456 in (f 3, f 3 with ?y=5)) with ?y = 6
462 in (f 3, f 3 with ?y=5)) with ?y = 6
466 Indeed, simply inlining f (at the Haskell source level) would change the
469 Nevertheless, as Launchbury says (email Oct 01) we can't really give the
470 semantics for a Haskell program without knowing its typing, so if you
471 change the typing you may change the semantics.
473 To make things consistent in all cases where we are *checking* against
474 a supplied signature (as opposed to inferring a type), we adopt the
477 a signature does not need to quantify over implicit params.
479 [This represents a (rather marginal) change of policy since GHC 5.02,
480 which *required* an explicit signature to quantify over all implicit
481 params for the reasons mentioned above.]
483 But that raises a new question. Consider
485 Given (signature) ?x::Int
486 Wanted (inferred) ?x::Int, ?y::Bool
488 Clearly we want to discharge the ?x and float the ?y out. But
489 what is the criterion that distinguishes them? Clearly it isn't
490 what free type variables they have. The Right Thing seems to be
491 to float a constraint that
492 neither mentions any of the quantified type variables
493 nor any of the quantified implicit parameters
495 See the predicate isFreeWhenChecking.
498 Question 3: monomorphism
499 ~~~~~~~~~~~~~~~~~~~~~~~~
500 There's a nasty corner case when the monomorphism restriction bites:
504 The argument above suggests that we *must* generalise
505 over the ?y parameter, to get
506 z :: (?y::Int) => Int,
507 but the monomorphism restriction says that we *must not*, giving
509 Why does the momomorphism restriction say this? Because if you have
511 let z = x + ?y in z+z
513 you might not expect the addition to be done twice --- but it will if
514 we follow the argument of Question 2 and generalise over ?y.
517 Question 4: top level
518 ~~~~~~~~~~~~~~~~~~~~~
519 At the top level, monomorhism makes no sense at all.
522 main = let ?x = 5 in print foo
526 woggle :: (?x :: Int) => Int -> Int
529 We definitely don't want (foo :: Int) with a top-level implicit parameter
530 (?x::Int) becuase there is no way to bind it.
535 (A) Always generalise over implicit parameters
536 Bindings that fall under the monomorphism restriction can't
540 * Inlining remains valid
541 * No unexpected loss of sharing
542 * But simple bindings like
544 will be rejected, unless you add an explicit type signature
545 (to avoid the monomorphism restriction)
546 z :: (?y::Int) => Int
548 This seems unacceptable
550 (B) Monomorphism restriction "wins"
551 Bindings that fall under the monomorphism restriction can't
553 Always generalise over implicit parameters *except* for bindings
554 that fall under the monomorphism restriction
557 * Inlining isn't valid in general
558 * No unexpected loss of sharing
559 * Simple bindings like
561 accepted (get value of ?y from binding site)
563 (C) Always generalise over implicit parameters
564 Bindings that fall under the monomorphism restriction can't
565 be generalised, EXCEPT for implicit parameters
567 * Inlining remains valid
568 * Unexpected loss of sharing (from the extra generalisation)
569 * Simple bindings like
571 accepted (get value of ?y from occurrence sites)
576 None of these choices seems very satisfactory. But at least we should
577 decide which we want to do.
579 It's really not clear what is the Right Thing To Do. If you see
583 would you expect the value of ?y to be got from the *occurrence sites*
584 of 'z', or from the valuue of ?y at the *definition* of 'z'? In the
585 case of function definitions, the answer is clearly the former, but
586 less so in the case of non-fucntion definitions. On the other hand,
587 if we say that we get the value of ?y from the definition site of 'z',
588 then inlining 'z' might change the semantics of the program.
590 Choice (C) really says "the monomorphism restriction doesn't apply
591 to implicit parameters". Which is fine, but remember that every
592 innocent binding 'x = ...' that mentions an implicit parameter in
593 the RHS becomes a *function* of that parameter, called at each
594 use of 'x'. Now, the chances are that there are no intervening 'with'
595 clauses that bind ?y, so a decent compiler should common up all
596 those function calls. So I think I strongly favour (C). Indeed,
597 one could make a similar argument for abolishing the monomorphism
598 restriction altogether.
600 BOTTOM LINE: we choose (B) at present. See tcSimplifyRestricted
604 %************************************************************************
606 \subsection{tcSimplifyInfer}
608 %************************************************************************
610 tcSimplify is called when we *inferring* a type. Here's the overall game plan:
612 1. Compute Q = grow( fvs(T), C )
614 2. Partition C based on Q into Ct and Cq. Notice that ambiguous
615 predicates will end up in Ct; we deal with them at the top level
617 3. Try improvement, using functional dependencies
619 4. If Step 3 did any unification, repeat from step 1
620 (Unification can change the result of 'grow'.)
622 Note: we don't reduce dictionaries in step 2. For example, if we have
623 Eq (a,b), we don't simplify to (Eq a, Eq b). So Q won't be different
624 after step 2. However note that we may therefore quantify over more
625 type variables than we absolutely have to.
627 For the guts, we need a loop, that alternates context reduction and
628 improvement with unification. E.g. Suppose we have
630 class C x y | x->y where ...
632 and tcSimplify is called with:
634 Then improvement unifies a with b, giving
637 If we need to unify anything, we rattle round the whole thing all over
644 -> TcTyVarSet -- fv(T); type vars
646 -> TcM ([TcTyVar], -- Tyvars to quantify (zonked and quantified)
647 [Inst], -- Dict Ids that must be bound here (zonked)
648 TcDictBinds) -- Bindings
649 -- Any free (escaping) Insts are tossed into the environment
654 tcSimplifyInfer doc tau_tvs wanted
655 = do { tau_tvs1 <- zonkTcTyVarsAndFV (varSetElems tau_tvs)
656 ; wanted' <- mapM zonkInst wanted -- Zonk before deciding quantified tyvars
657 ; gbl_tvs <- tcGetGlobalTyVars
658 ; let preds1 = fdPredsOfInsts wanted'
659 gbl_tvs1 = oclose preds1 gbl_tvs
660 qtvs = grow preds1 tau_tvs1 `minusVarSet` gbl_tvs1
661 -- See Note [Choosing which variables to quantify]
663 -- To maximise sharing, remove from consideration any
664 -- constraints that don't mention qtvs at all
665 ; let (free, bound) = partition (isFreeWhenInferring qtvs) wanted'
668 -- To make types simple, reduce as much as possible
669 ; traceTc (text "infer" <+> (ppr preds1 $$ ppr (grow preds1 tau_tvs1) $$ ppr gbl_tvs $$
670 ppr gbl_tvs1 $$ ppr free $$ ppr bound))
671 ; (irreds1, binds1) <- tryHardCheckLoop doc bound
673 -- Note [Inference and implication constraints]
674 ; let want_dict d = tyVarsOfInst d `intersectsVarSet` qtvs
675 ; (irreds2, binds2) <- approximateImplications doc want_dict irreds1
677 -- Now work out all over again which type variables to quantify,
678 -- exactly in the same way as before, but starting from irreds2. Why?
679 -- a) By now improvment may have taken place, and we must *not*
680 -- quantify over any variable free in the environment
681 -- tc137 (function h inside g) is an example
683 -- b) Do not quantify over constraints that *now* do not
684 -- mention quantified type variables, because they are
685 -- simply ambiguous (or might be bound further out). Example:
686 -- f :: Eq b => a -> (a, b)
688 -- From the RHS of g we get the MethodInst f77 :: alpha -> (alpha, beta)
689 -- We decide to quantify over 'alpha' alone, but free1 does not include f77
690 -- because f77 mentions 'alpha'. Then reducing leaves only the (ambiguous)
691 -- constraint (Eq beta), which we dump back into the free set
692 -- See test tcfail181
694 -- c) irreds may contain type variables not previously mentioned,
695 -- e.g. instance D a x => Foo [a]
697 -- Then after simplifying we'll get (D a x), and x is fresh
698 -- We must quantify over x else it'll be totally unbound
699 ; tau_tvs2 <- zonkTcTyVarsAndFV (varSetElems tau_tvs1)
700 ; gbl_tvs2 <- zonkTcTyVarsAndFV (varSetElems gbl_tvs1)
701 -- Note that we start from gbl_tvs1
702 -- We use tcGetGlobalTyVars, then oclose wrt preds2, because
703 -- we've already put some of the original preds1 into frees
704 -- E.g. wanteds = C a b (where a->b)
707 -- Then b is fixed by gbl_tvs, so (C a b) will be in free, and
708 -- irreds2 will be empty. But we don't want to generalise over b!
709 ; let preds2 = fdPredsOfInsts irreds2 -- irreds2 is zonked
710 qtvs = grow preds2 tau_tvs2 `minusVarSet` oclose preds2 gbl_tvs2
711 ; let (free, irreds3) = partition (isFreeWhenInferring qtvs) irreds2
714 -- Turn the quantified meta-type variables into real type variables
715 ; qtvs2 <- zonkQuantifiedTyVars (varSetElems qtvs)
717 -- We can't abstract over any remaining unsolved
718 -- implications so instead just float them outwards. Ugh.
719 ; let (q_dicts0, implics) = partition isAbstractableInst irreds3
720 ; loc <- getInstLoc (ImplicOrigin doc)
721 ; implic_bind <- bindIrreds loc qtvs2 q_dicts0 implics
723 -- Prepare equality instances for quantification
724 ; let (q_eqs0,q_dicts) = partition isEqInst q_dicts0
725 ; q_eqs <- mapM finalizeEqInst q_eqs0
727 ; return (qtvs2, q_eqs ++ q_dicts, binds1 `unionBags` binds2 `unionBags` implic_bind) }
728 -- NB: when we are done, we might have some bindings, but
729 -- the final qtvs might be empty. See Note [NO TYVARS] below.
731 approximateImplications :: SDoc -> (Inst -> Bool) -> [Inst] -> TcM ([Inst], TcDictBinds)
732 -- Note [Inference and implication constraints]
733 -- Given a bunch of Dict and ImplicInsts, try to approximate the implications by
734 -- - fetching any dicts inside them that are free
735 -- - using those dicts as cruder constraints, to solve the implications
736 -- - returning the extra ones too
738 approximateImplications doc want_dict irreds
740 = return (irreds, emptyBag)
742 = do { extra_dicts' <- mapM cloneDict extra_dicts
743 ; tryHardCheckLoop doc (extra_dicts' ++ irreds) }
744 -- By adding extra_dicts', we make them
745 -- available to solve the implication constraints
747 extra_dicts = get_dicts (filter isImplicInst irreds)
749 get_dicts :: [Inst] -> [Inst] -- Returns only Dicts
750 -- Find the wanted constraints in implication constraints that satisfy
751 -- want_dict, and are not bound by forall's in the constraint itself
752 get_dicts ds = concatMap get_dict ds
754 get_dict d@(Dict {}) | want_dict d = [d]
756 get_dict (ImplicInst {tci_tyvars = tvs, tci_wanted = wanteds})
757 = [ d | let tv_set = mkVarSet tvs
758 , d <- get_dicts wanteds
759 , not (tyVarsOfInst d `intersectsVarSet` tv_set)]
760 get_dict i@(EqInst {}) | want_dict i = [i]
762 get_dict other = pprPanic "approximateImplications" (ppr other)
765 Note [Inference and implication constraints]
766 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
767 Suppose we have a wanted implication constraint (perhaps arising from
768 a nested pattern match) like
770 and we are now trying to quantify over 'a' when inferring the type for
771 a function. In principle it's possible that there might be an instance
772 instance (C a, E a) => D [a]
773 so the context (E a) would suffice. The Right Thing is to abstract over
774 the implication constraint, but we don't do that (a) because it'll be
775 surprising to programmers and (b) because we don't have the machinery to deal
776 with 'given' implications.
778 So our best approximation is to make (D [a]) part of the inferred
779 context, so we can use that to discharge the implication. Hence
780 the strange function get_dicts in approximateImplications.
782 The common cases are more clear-cut, when we have things like
784 Here, abstracting over (C b) is not an approximation at all -- but see
785 Note [Freeness and implications].
787 See Trac #1430 and test tc228.
791 -----------------------------------------------------------
792 -- tcSimplifyInferCheck is used when we know the constraints we are to simplify
793 -- against, but we don't know the type variables over which we are going to quantify.
794 -- This happens when we have a type signature for a mutually recursive group
797 -> TcTyVarSet -- fv(T)
800 -> TcM ([TyVar], -- Fully zonked, and quantified
801 TcDictBinds) -- Bindings
803 tcSimplifyInferCheck loc tau_tvs givens wanteds
804 = do { traceTc (text "tcSimplifyInferCheck <-" <+> ppr wanteds)
805 ; (irreds, binds) <- gentleCheckLoop loc givens wanteds
807 -- Figure out which type variables to quantify over
808 -- You might think it should just be the signature tyvars,
809 -- but in bizarre cases you can get extra ones
810 -- f :: forall a. Num a => a -> a
811 -- f x = fst (g (x, head [])) + 1
813 -- Here we infer g :: forall a b. a -> b -> (b,a)
814 -- We don't want g to be monomorphic in b just because
815 -- f isn't quantified over b.
816 ; let all_tvs = varSetElems (tau_tvs `unionVarSet` tyVarsOfInsts givens)
817 ; all_tvs <- zonkTcTyVarsAndFV all_tvs
818 ; gbl_tvs <- tcGetGlobalTyVars
819 ; let qtvs = varSetElems (all_tvs `minusVarSet` gbl_tvs)
820 -- We could close gbl_tvs, but its not necessary for
821 -- soundness, and it'll only affect which tyvars, not which
822 -- dictionaries, we quantify over
824 ; qtvs' <- zonkQuantifiedTyVars qtvs
826 -- Now we are back to normal (c.f. tcSimplCheck)
827 ; implic_bind <- bindIrreds loc qtvs' givens irreds
829 ; traceTc (text "tcSimplifyInferCheck ->" <+> ppr (implic_bind))
830 ; return (qtvs', binds `unionBags` implic_bind) }
833 Note [Squashing methods]
834 ~~~~~~~~~~~~~~~~~~~~~~~~~
835 Be careful if you want to float methods more:
836 truncate :: forall a. RealFrac a => forall b. Integral b => a -> b
837 From an application (truncate f i) we get
840 If we have also have a second occurrence of truncate, we get
843 When simplifying with i,f free, we might still notice that
844 t1=t3; but alas, the binding for t2 (which mentions t1)
845 may continue to float out!
850 class Y a b | a -> b where
853 instance Y [[a]] a where
856 k :: X a -> X a -> X a
858 g :: Num a => [X a] -> [X a]
861 h ys = ys ++ map (k (y [[0]])) xs
863 The excitement comes when simplifying the bindings for h. Initially
864 try to simplify {y @ [[t1]] t2, 0 @ t1}, with initial qtvs = {t2}.
865 From this we get t1:=:t2, but also various bindings. We can't forget
866 the bindings (because of [LOOP]), but in fact t1 is what g is
869 The net effect of [NO TYVARS]
872 isFreeWhenInferring :: TyVarSet -> Inst -> Bool
873 isFreeWhenInferring qtvs inst
874 = isFreeWrtTyVars qtvs inst -- Constrains no quantified vars
875 && isInheritableInst inst -- and no implicit parameter involved
876 -- see Note [Inheriting implicit parameters]
878 {- No longer used (with implication constraints)
879 isFreeWhenChecking :: TyVarSet -- Quantified tyvars
880 -> NameSet -- Quantified implicit parameters
882 isFreeWhenChecking qtvs ips inst
883 = isFreeWrtTyVars qtvs inst
884 && isFreeWrtIPs ips inst
887 isFreeWrtTyVars :: VarSet -> Inst -> Bool
888 isFreeWrtTyVars qtvs inst = tyVarsOfInst inst `disjointVarSet` qtvs
889 isFreeWrtIPs :: NameSet -> Inst -> Bool
890 isFreeWrtIPs ips inst = not (any (`elemNameSet` ips) (ipNamesOfInst inst))
894 %************************************************************************
896 \subsection{tcSimplifyCheck}
898 %************************************************************************
900 @tcSimplifyCheck@ is used when we know exactly the set of variables
901 we are going to quantify over. For example, a class or instance declaration.
904 -----------------------------------------------------------
905 -- tcSimplifyCheck is used when checking expression type signatures,
906 -- class decls, instance decls etc.
907 tcSimplifyCheck :: InstLoc
908 -> [TcTyVar] -- Quantify over these
911 -> TcM TcDictBinds -- Bindings
912 tcSimplifyCheck loc qtvs givens wanteds
913 = ASSERT( all isTcTyVar qtvs && all isSkolemTyVar qtvs )
914 do { traceTc (text "tcSimplifyCheck")
915 ; (irreds, binds) <- gentleCheckLoop loc givens wanteds
916 ; implic_bind <- bindIrreds loc qtvs givens irreds
917 ; return (binds `unionBags` implic_bind) }
919 -----------------------------------------------------------
920 -- tcSimplifyCheckPat is used for existential pattern match
921 tcSimplifyCheckPat :: InstLoc
922 -> [TcTyVar] -- Quantify over these
925 -> TcM TcDictBinds -- Bindings
926 tcSimplifyCheckPat loc qtvs givens wanteds
927 = ASSERT( all isTcTyVar qtvs && all isSkolemTyVar qtvs )
928 do { traceTc (text "tcSimplifyCheckPat")
929 ; (irreds, binds) <- gentleCheckLoop loc givens wanteds
930 ; implic_bind <- bindIrredsR loc qtvs givens irreds
931 ; return (binds `unionBags` implic_bind) }
933 -----------------------------------------------------------
934 bindIrreds :: InstLoc -> [TcTyVar]
937 bindIrreds loc qtvs givens irreds
938 = bindIrredsR loc qtvs givens irreds
940 bindIrredsR :: InstLoc -> [TcTyVar] -> [Inst] -> [Inst] -> TcM TcDictBinds
941 -- Make a binding that binds 'irreds', by generating an implication
942 -- constraint for them, *and* throwing the constraint into the LIE
943 bindIrredsR loc qtvs givens irreds
947 = do { let givens' = filter isAbstractableInst givens
948 -- The givens can (redundantly) include methods
949 -- We want to retain both EqInsts and Dicts
950 -- There should be no implicadtion constraints
951 -- See Note [Pruning the givens in an implication constraint]
953 -- If there are no 'givens', then it's safe to
954 -- partition the 'wanteds' by their qtvs, thereby trimming irreds
955 -- See Note [Freeness and implications]
956 ; irreds' <- if null givens'
958 { let qtv_set = mkVarSet qtvs
959 (frees, real_irreds) = partition (isFreeWrtTyVars qtv_set) irreds
961 ; return real_irreds }
964 ; (implics, bind) <- makeImplicationBind loc qtvs givens' irreds'
965 -- This call does the real work
966 -- If irreds' is empty, it does something sensible
971 makeImplicationBind :: InstLoc -> [TcTyVar]
973 -> TcM ([Inst], TcDictBinds)
974 -- Make a binding that binds 'irreds', by generating an implication
975 -- constraint for them, *and* throwing the constraint into the LIE
976 -- The binding looks like
977 -- (ir1, .., irn) = f qtvs givens
978 -- where f is (evidence for) the new implication constraint
979 -- f :: forall qtvs. {reft} givens => (ir1, .., irn)
980 -- qtvs includes coercion variables
982 -- This binding must line up the 'rhs' in reduceImplication
983 makeImplicationBind loc all_tvs
984 givens -- Guaranteed all Dicts
987 | null irreds -- If there are no irreds, we are done
988 = return ([], emptyBag)
989 | otherwise -- Otherwise we must generate a binding
990 = do { uniq <- newUnique
991 ; span <- getSrcSpanM
992 ; let (eq_givens, dict_givens) = partition isEqInst givens
993 eq_tyvar_cos = mkTyVarTys (varSetElems $ tyVarsOfTypes $ map eqInstType eq_givens)
994 -- Urgh! See line 2187 or thereabouts. I believe that all these
995 -- 'givens' must be a simple CoVar. This MUST be cleaned up.
997 ; let name = mkInternalName uniq (mkVarOcc "ic") span
998 implic_inst = ImplicInst { tci_name = name,
999 tci_tyvars = all_tvs,
1000 tci_given = (eq_givens ++ dict_givens),
1001 tci_wanted = irreds, tci_loc = loc }
1002 ; let -- only create binder for dict_irreds
1003 (_, dict_irreds) = partition isEqInst irreds
1004 dict_irred_ids = map instToId dict_irreds
1005 lpat = mkBigLHsPatTup (map (L span . VarPat) dict_irred_ids)
1006 rhs = L span (mkHsWrap co (HsVar (instToId implic_inst)))
1007 co = mkWpApps (map instToId dict_givens)
1008 <.> mkWpTyApps eq_tyvar_cos
1009 <.> mkWpTyApps (mkTyVarTys all_tvs)
1010 bind | [dict_irred_id] <- dict_irred_ids = VarBind dict_irred_id rhs
1011 | otherwise = PatBind { pat_lhs = lpat,
1012 pat_rhs = unguardedGRHSs rhs,
1013 pat_rhs_ty = hsLPatType lpat,
1014 bind_fvs = placeHolderNames }
1015 ; traceTc $ text "makeImplicationBind" <+> ppr implic_inst
1016 ; return ([implic_inst], unitBag (L span bind))
1019 -----------------------------------------------------------
1020 tryHardCheckLoop :: SDoc
1022 -> TcM ([Inst], TcDictBinds)
1024 tryHardCheckLoop doc wanteds
1025 = do { (irreds,binds) <- checkLoop (mkRedEnv doc try_me []) wanteds
1026 ; return (irreds,binds)
1029 try_me _ = ReduceMe AddSCs
1030 -- Here's the try-hard bit
1032 -----------------------------------------------------------
1033 gentleCheckLoop :: InstLoc
1036 -> TcM ([Inst], TcDictBinds)
1038 gentleCheckLoop inst_loc givens wanteds
1039 = do { (irreds,binds) <- checkLoop env wanteds
1040 ; return (irreds,binds)
1043 env = mkRedEnv (pprInstLoc inst_loc) try_me givens
1045 try_me inst | isMethodOrLit inst = ReduceMe AddSCs
1047 -- When checking against a given signature
1048 -- we MUST be very gentle: Note [Check gently]
1050 gentleInferLoop :: SDoc -> [Inst]
1051 -> TcM ([Inst], TcDictBinds)
1052 gentleInferLoop doc wanteds
1053 = do { (irreds, binds) <- checkLoop env wanteds
1054 ; return (irreds, binds) }
1056 env = mkRedEnv doc try_me []
1057 try_me inst | isMethodOrLit inst = ReduceMe AddSCs
1062 ~~~~~~~~~~~~~~~~~~~~
1063 We have to very careful about not simplifying too vigorously
1068 f :: Show b => T b -> b
1069 f (MkT x) = show [x]
1071 Inside the pattern match, which binds (a:*, x:a), we know that
1073 Hence we have a dictionary for Show [a] available; and indeed we
1074 need it. We are going to build an implication contraint
1075 forall a. (b~[a]) => Show [a]
1076 Later, we will solve this constraint using the knowledge (Show b)
1078 But we MUST NOT reduce (Show [a]) to (Show a), else the whole
1079 thing becomes insoluble. So we simplify gently (get rid of literals
1080 and methods only, plus common up equal things), deferring the real
1081 work until top level, when we solve the implication constraint
1082 with tryHardCheckLooop.
1086 -----------------------------------------------------------
1089 -> TcM ([Inst], TcDictBinds)
1090 -- Precondition: givens are completely rigid
1091 -- Postcondition: returned Insts are zonked
1093 checkLoop env wanteds
1094 = go env wanteds (return ())
1095 where go env wanteds elim_skolems
1096 = do { -- We do need to zonk the givens; cf Note [Zonking RedEnv]
1097 ; env' <- zonkRedEnv env
1098 ; wanteds' <- zonkInsts wanteds
1100 ; (improved, binds, irreds, elim_more_skolems)
1101 <- reduceContext env' wanteds'
1102 ; let elim_skolems' = elim_skolems >> elim_more_skolems
1104 ; if not improved then
1105 elim_skolems' >> return (irreds, binds)
1108 -- If improvement did some unification, we go round again.
1109 -- We start again with irreds, not wanteds
1110 -- Using an instance decl might have introduced a fresh type
1111 -- variable which might have been unified, so we'd get an
1112 -- infinite loop if we started again with wanteds!
1114 { (irreds1, binds1) <- go env' irreds elim_skolems'
1115 ; return (irreds1, binds `unionBags` binds1) } }
1118 Note [Zonking RedEnv]
1119 ~~~~~~~~~~~~~~~~~~~~~
1120 It might appear as if the givens in RedEnv are always rigid, but that is not
1121 necessarily the case for programs involving higher-rank types that have class
1122 contexts constraining the higher-rank variables. An example from tc237 in the
1125 class Modular s a | s -> a
1127 wim :: forall a w. Integral a
1128 => a -> (forall s. Modular s a => M s w) -> w
1129 wim i k = error "urk"
1131 test5 :: (Modular s a, Integral a) => M s a
1134 test4 = wim 4 test4'
1136 Notice how the variable 'a' of (Modular s a) in the rank-2 type of wim is
1137 quantified further outside. When type checking test4, we have to check
1138 whether the signature of test5 is an instance of
1140 (forall s. Modular s a => M s w)
1142 Consequently, we will get (Modular s t_a), where t_a is a TauTv into the
1145 Given the FD of Modular in this example, class improvement will instantiate
1146 t_a to 'a', where 'a' is the skolem from test5's signatures (due to the
1147 Modular s a predicate in that signature). If we don't zonk (Modular s t_a) in
1148 the givens, we will get into a loop as improveOne uses the unification engine
1149 Unify.tcUnifyTys, which doesn't know about mutable type variables.
1154 class If b t e r | b t e -> r
1157 class Lte a b c | a b -> c where lte :: a -> b -> c
1159 instance (Lte a b l,If l b a c) => Max a b c
1161 Wanted: Max Z (S x) y
1163 Then we'll reduce using the Max instance to:
1164 (Lte Z (S x) l, If l (S x) Z y)
1165 and improve by binding l->T, after which we can do some reduction
1166 on both the Lte and If constraints. What we *can't* do is start again
1167 with (Max Z (S x) y)!
1171 %************************************************************************
1173 tcSimplifySuperClasses
1175 %************************************************************************
1177 Note [SUPERCLASS-LOOP 1]
1178 ~~~~~~~~~~~~~~~~~~~~~~~~
1179 We have to be very, very careful when generating superclasses, lest we
1180 accidentally build a loop. Here's an example:
1184 class S a => C a where { opc :: a -> a }
1185 class S b => D b where { opd :: b -> b }
1187 instance C Int where
1190 instance D Int where
1193 From (instance C Int) we get the constraint set {ds1:S Int, dd:D Int}
1194 Simplifying, we may well get:
1195 $dfCInt = :C ds1 (opd dd)
1198 Notice that we spot that we can extract ds1 from dd.
1200 Alas! Alack! We can do the same for (instance D Int):
1202 $dfDInt = :D ds2 (opc dc)
1206 And now we've defined the superclass in terms of itself.
1208 Solution: never generate a superclass selectors at all when
1209 satisfying the superclass context of an instance declaration.
1211 Two more nasty cases are in
1216 tcSimplifySuperClasses
1221 tcSimplifySuperClasses loc givens sc_wanteds
1222 = do { traceTc (text "tcSimplifySuperClasses")
1223 ; (irreds,binds1) <- checkLoop env sc_wanteds
1224 ; let (tidy_env, tidy_irreds) = tidyInsts irreds
1225 ; reportNoInstances tidy_env (Just (loc, givens)) tidy_irreds
1228 env = mkRedEnv (pprInstLoc loc) try_me givens
1229 try_me _ = ReduceMe NoSCs
1230 -- Like tryHardCheckLoop, but with NoSCs
1234 %************************************************************************
1236 \subsection{tcSimplifyRestricted}
1238 %************************************************************************
1240 tcSimplifyRestricted infers which type variables to quantify for a
1241 group of restricted bindings. This isn't trivial.
1244 We want to quantify over a to get id :: forall a. a->a
1247 We do not want to quantify over a, because there's an Eq a
1248 constraint, so we get eq :: a->a->Bool (notice no forall)
1251 RHS has type 'tau', whose free tyvars are tau_tvs
1252 RHS has constraints 'wanteds'
1255 Quantify over (tau_tvs \ ftvs(wanteds))
1256 This is bad. The constraints may contain (Monad (ST s))
1257 where we have instance Monad (ST s) where...
1258 so there's no need to be monomorphic in s!
1260 Also the constraint might be a method constraint,
1261 whose type mentions a perfectly innocent tyvar:
1262 op :: Num a => a -> b -> a
1263 Here, b is unconstrained. A good example would be
1265 We want to infer the polymorphic type
1266 foo :: forall b. b -> b
1269 Plan B (cunning, used for a long time up to and including GHC 6.2)
1270 Step 1: Simplify the constraints as much as possible (to deal
1271 with Plan A's problem). Then set
1272 qtvs = tau_tvs \ ftvs( simplify( wanteds ) )
1274 Step 2: Now simplify again, treating the constraint as 'free' if
1275 it does not mention qtvs, and trying to reduce it otherwise.
1276 The reasons for this is to maximise sharing.
1278 This fails for a very subtle reason. Suppose that in the Step 2
1279 a constraint (Foo (Succ Zero) (Succ Zero) b) gets thrown upstairs as 'free'.
1280 In the Step 1 this constraint might have been simplified, perhaps to
1281 (Foo Zero Zero b), AND THEN THAT MIGHT BE IMPROVED, to bind 'b' to 'T'.
1282 This won't happen in Step 2... but that in turn might prevent some other
1283 constraint (Baz [a] b) being simplified (e.g. via instance Baz [a] T where {..})
1284 and that in turn breaks the invariant that no constraints are quantified over.
1286 Test typecheck/should_compile/tc177 (which failed in GHC 6.2) demonstrates
1291 Step 1: Simplify the constraints as much as possible (to deal
1292 with Plan A's problem). Then set
1293 qtvs = tau_tvs \ ftvs( simplify( wanteds ) )
1294 Return the bindings from Step 1.
1297 A note about Plan C (arising from "bug" reported by George Russel March 2004)
1300 instance (HasBinary ty IO) => HasCodedValue ty
1302 foo :: HasCodedValue a => String -> IO a
1304 doDecodeIO :: HasCodedValue a => () -> () -> IO a
1305 doDecodeIO codedValue view
1306 = let { act = foo "foo" } in act
1308 You might think this should work becuase the call to foo gives rise to a constraint
1309 (HasCodedValue t), which can be satisfied by the type sig for doDecodeIO. But the
1310 restricted binding act = ... calls tcSimplifyRestricted, and PlanC simplifies the
1311 constraint using the (rather bogus) instance declaration, and now we are stuffed.
1313 I claim this is not really a bug -- but it bit Sergey as well as George. So here's
1317 Plan D (a variant of plan B)
1318 Step 1: Simplify the constraints as much as possible (to deal
1319 with Plan A's problem), BUT DO NO IMPROVEMENT. Then set
1320 qtvs = tau_tvs \ ftvs( simplify( wanteds ) )
1322 Step 2: Now simplify again, treating the constraint as 'free' if
1323 it does not mention qtvs, and trying to reduce it otherwise.
1325 The point here is that it's generally OK to have too few qtvs; that is,
1326 to make the thing more monomorphic than it could be. We don't want to
1327 do that in the common cases, but in wierd cases it's ok: the programmer
1328 can always add a signature.
1330 Too few qtvs => too many wanteds, which is what happens if you do less
1335 tcSimplifyRestricted -- Used for restricted binding groups
1336 -- i.e. ones subject to the monomorphism restriction
1339 -> [Name] -- Things bound in this group
1340 -> TcTyVarSet -- Free in the type of the RHSs
1341 -> [Inst] -- Free in the RHSs
1342 -> TcM ([TyVar], -- Tyvars to quantify (zonked and quantified)
1343 TcDictBinds) -- Bindings
1344 -- tcSimpifyRestricted returns no constraints to
1345 -- quantify over; by definition there are none.
1346 -- They are all thrown back in the LIE
1348 tcSimplifyRestricted doc top_lvl bndrs tau_tvs wanteds
1349 -- Zonk everything in sight
1350 = do { traceTc (text "tcSimplifyRestricted")
1351 ; wanteds' <- zonkInsts wanteds
1353 -- 'ReduceMe': Reduce as far as we can. Don't stop at
1354 -- dicts; the idea is to get rid of as many type
1355 -- variables as possible, and we don't want to stop
1356 -- at (say) Monad (ST s), because that reduces
1357 -- immediately, with no constraint on s.
1359 -- BUT do no improvement! See Plan D above
1360 -- HOWEVER, some unification may take place, if we instantiate
1361 -- a method Inst with an equality constraint
1362 ; let env = mkNoImproveRedEnv doc (\_ -> ReduceMe AddSCs)
1363 ; (_imp, _binds, constrained_dicts, elim_skolems)
1364 <- reduceContext env wanteds'
1367 -- Next, figure out the tyvars we will quantify over
1368 ; tau_tvs' <- zonkTcTyVarsAndFV (varSetElems tau_tvs)
1369 ; gbl_tvs' <- tcGetGlobalTyVars
1370 ; constrained_dicts' <- zonkInsts constrained_dicts
1372 ; let qtvs1 = tau_tvs' `minusVarSet` oclose (fdPredsOfInsts constrained_dicts) gbl_tvs'
1373 -- As in tcSimplifyInfer
1375 -- Do not quantify over constrained type variables:
1376 -- this is the monomorphism restriction
1377 constrained_tvs' = tyVarsOfInsts constrained_dicts'
1378 qtvs = qtvs1 `minusVarSet` constrained_tvs'
1379 pp_bndrs = pprWithCommas (quotes . ppr) bndrs
1382 ; warn_mono <- doptM Opt_WarnMonomorphism
1383 ; warnTc (warn_mono && (constrained_tvs' `intersectsVarSet` qtvs1))
1384 (vcat[ ptext (sLit "the Monomorphism Restriction applies to the binding")
1385 <> plural bndrs <+> ptext (sLit "for") <+> pp_bndrs,
1386 ptext (sLit "Consider giving a type signature for") <+> pp_bndrs])
1388 ; traceTc (text "tcSimplifyRestricted" <+> vcat [
1389 pprInsts wanteds, pprInsts constrained_dicts',
1391 ppr constrained_tvs', ppr tau_tvs', ppr qtvs ])
1393 -- The first step may have squashed more methods than
1394 -- necessary, so try again, this time more gently, knowing the exact
1395 -- set of type variables to quantify over.
1397 -- We quantify only over constraints that are captured by qtvs;
1398 -- these will just be a subset of non-dicts. This in contrast
1399 -- to normal inference (using isFreeWhenInferring) in which we quantify over
1400 -- all *non-inheritable* constraints too. This implements choice
1401 -- (B) under "implicit parameter and monomorphism" above.
1403 -- Remember that we may need to do *some* simplification, to
1404 -- (for example) squash {Monad (ST s)} into {}. It's not enough
1405 -- just to float all constraints
1407 -- At top level, we *do* squash methods becuase we want to
1408 -- expose implicit parameters to the test that follows
1409 ; let is_nested_group = isNotTopLevel top_lvl
1410 try_me inst | isFreeWrtTyVars qtvs inst,
1411 (is_nested_group || isDict inst) = Stop
1412 | otherwise = ReduceMe AddSCs
1413 env = mkNoImproveRedEnv doc try_me
1414 ; (_imp, binds, irreds, elim_skolems) <- reduceContext env wanteds'
1417 -- See "Notes on implicit parameters, Question 4: top level"
1418 ; ASSERT( all (isFreeWrtTyVars qtvs) irreds ) -- None should be captured
1419 if is_nested_group then
1421 else do { let (bad_ips, non_ips) = partition isIPDict irreds
1422 ; addTopIPErrs bndrs bad_ips
1423 ; extendLIEs non_ips }
1425 ; qtvs' <- zonkQuantifiedTyVars (varSetElems qtvs)
1426 ; return (qtvs', binds) }
1430 %************************************************************************
1434 %************************************************************************
1436 On the LHS of transformation rules we only simplify methods and constants,
1437 getting dictionaries. We want to keep all of them unsimplified, to serve
1438 as the available stuff for the RHS of the rule.
1440 Example. Consider the following left-hand side of a rule
1442 f (x == y) (y > z) = ...
1444 If we typecheck this expression we get constraints
1446 d1 :: Ord a, d2 :: Eq a
1448 We do NOT want to "simplify" to the LHS
1450 forall x::a, y::a, z::a, d1::Ord a.
1451 f ((==) (eqFromOrd d1) x y) ((>) d1 y z) = ...
1455 forall x::a, y::a, z::a, d1::Ord a, d2::Eq a.
1456 f ((==) d2 x y) ((>) d1 y z) = ...
1458 Here is another example:
1460 fromIntegral :: (Integral a, Num b) => a -> b
1461 {-# RULES "foo" fromIntegral = id :: Int -> Int #-}
1463 In the rule, a=b=Int, and Num Int is a superclass of Integral Int. But
1464 we *dont* want to get
1466 forall dIntegralInt.
1467 fromIntegral Int Int dIntegralInt (scsel dIntegralInt) = id Int
1469 because the scsel will mess up RULE matching. Instead we want
1471 forall dIntegralInt, dNumInt.
1472 fromIntegral Int Int dIntegralInt dNumInt = id Int
1476 g (x == y) (y == z) = ..
1478 where the two dictionaries are *identical*, we do NOT WANT
1480 forall x::a, y::a, z::a, d1::Eq a
1481 f ((==) d1 x y) ((>) d1 y z) = ...
1483 because that will only match if the dict args are (visibly) equal.
1484 Instead we want to quantify over the dictionaries separately.
1486 In short, tcSimplifyRuleLhs must *only* squash LitInst and MethInts, leaving
1487 all dicts unchanged, with absolutely no sharing. It's simpler to do this
1488 from scratch, rather than further parameterise simpleReduceLoop etc
1491 tcSimplifyRuleLhs :: [Inst] -> TcM ([Inst], TcDictBinds)
1492 tcSimplifyRuleLhs wanteds
1493 = go [] emptyBag wanteds
1496 = return (dicts, binds)
1497 go dicts binds (w:ws)
1499 = go (w:dicts) binds ws
1501 = do { w' <- zonkInst w -- So that (3::Int) does not generate a call
1502 -- to fromInteger; this looks fragile to me
1503 ; lookup_result <- lookupSimpleInst w'
1504 ; case lookup_result of
1506 go dicts (addInstToDictBind binds w rhs) (ws' ++ ws)
1507 NoInstance -> pprPanic "tcSimplifyRuleLhs" (ppr w)
1511 tcSimplifyBracket is used when simplifying the constraints arising from
1512 a Template Haskell bracket [| ... |]. We want to check that there aren't
1513 any constraints that can't be satisfied (e.g. Show Foo, where Foo has no
1514 Show instance), but we aren't otherwise interested in the results.
1515 Nor do we care about ambiguous dictionaries etc. We will type check
1516 this bracket again at its usage site.
1519 tcSimplifyBracket :: [Inst] -> TcM ()
1520 tcSimplifyBracket wanteds
1521 = do { tryHardCheckLoop doc wanteds
1524 doc = text "tcSimplifyBracket"
1528 %************************************************************************
1530 \subsection{Filtering at a dynamic binding}
1532 %************************************************************************
1537 we must discharge all the ?x constraints from B. We also do an improvement
1538 step; if we have ?x::t1 and ?x::t2 we must unify t1, t2.
1540 Actually, the constraints from B might improve the types in ?x. For example
1542 f :: (?x::Int) => Char -> Char
1545 then the constraint (?x::Int) arising from the call to f will
1546 force the binding for ?x to be of type Int.
1549 tcSimplifyIPs :: [Inst] -- The implicit parameters bound here
1552 -- We need a loop so that we do improvement, and then
1553 -- (next time round) generate a binding to connect the two
1555 -- Here the two ?x's have different types, and improvement
1556 -- makes them the same.
1558 tcSimplifyIPs given_ips wanteds
1559 = do { wanteds' <- zonkInsts wanteds
1560 ; given_ips' <- zonkInsts given_ips
1561 -- Unusually for checking, we *must* zonk the given_ips
1563 ; let env = mkRedEnv doc try_me given_ips'
1564 ; (improved, binds, irreds, elim_skolems) <- reduceContext env wanteds'
1567 ; if not improved then
1568 ASSERT( all is_free irreds )
1569 do { extendLIEs irreds
1572 tcSimplifyIPs given_ips wanteds }
1574 doc = text "tcSimplifyIPs" <+> ppr given_ips
1575 ip_set = mkNameSet (ipNamesOfInsts given_ips)
1576 is_free inst = isFreeWrtIPs ip_set inst
1578 -- Simplify any methods that mention the implicit parameter
1579 try_me inst | is_free inst = Stop
1580 | otherwise = ReduceMe NoSCs
1584 %************************************************************************
1586 \subsection[binds-for-local-funs]{@bindInstsOfLocalFuns@}
1588 %************************************************************************
1590 When doing a binding group, we may have @Insts@ of local functions.
1591 For example, we might have...
1593 let f x = x + 1 -- orig local function (overloaded)
1594 f.1 = f Int -- two instances of f
1599 The point is: we must drop the bindings for @f.1@ and @f.2@ here,
1600 where @f@ is in scope; those @Insts@ must certainly not be passed
1601 upwards towards the top-level. If the @Insts@ were binding-ified up
1602 there, they would have unresolvable references to @f@.
1604 We pass in an @init_lie@ of @Insts@ and a list of locally-bound @Ids@.
1605 For each method @Inst@ in the @init_lie@ that mentions one of the
1606 @Ids@, we create a binding. We return the remaining @Insts@ (in an
1607 @LIE@), as well as the @HsBinds@ generated.
1610 bindInstsOfLocalFuns :: [Inst] -> [TcId] -> TcM TcDictBinds
1611 -- Simlifies only MethodInsts, and generate only bindings of form
1613 -- We're careful not to even generate bindings of the form
1615 -- You'd think that'd be fine, but it interacts with what is
1616 -- arguably a bug in Match.tidyEqnInfo (see notes there)
1618 bindInstsOfLocalFuns wanteds local_ids
1619 | null overloaded_ids = do
1622 return emptyLHsBinds
1625 = do { (irreds, binds) <- gentleInferLoop doc for_me
1626 ; extendLIEs not_for_me
1630 doc = text "bindInsts" <+> ppr local_ids
1631 overloaded_ids = filter is_overloaded local_ids
1632 is_overloaded id = isOverloadedTy (idType id)
1633 (for_me, not_for_me) = partition (isMethodFor overloaded_set) wanteds
1635 overloaded_set = mkVarSet overloaded_ids -- There can occasionally be a lot of them
1636 -- so it's worth building a set, so that
1637 -- lookup (in isMethodFor) is faster
1641 %************************************************************************
1643 \subsection{Data types for the reduction mechanism}
1645 %************************************************************************
1647 The main control over context reduction is here
1651 = RedEnv { red_doc :: SDoc -- The context
1652 , red_try_me :: Inst -> WhatToDo
1653 , red_improve :: Bool -- True <=> do improvement
1654 , red_givens :: [Inst] -- All guaranteed rigid
1656 -- but see Note [Rigidity]
1657 , red_stack :: (Int, [Inst]) -- Recursion stack (for err msg)
1658 -- See Note [RedStack]
1662 -- The red_givens are rigid so far as cmpInst is concerned.
1663 -- There is one case where they are not totally rigid, namely in tcSimplifyIPs
1664 -- let ?x = e in ...
1665 -- Here, the given is (?x::a), where 'a' is not necy a rigid type
1666 -- But that doesn't affect the comparison, which is based only on mame.
1669 -- The red_stack pair (n,insts) pair is just used for error reporting.
1670 -- 'n' is always the depth of the stack.
1671 -- The 'insts' is the stack of Insts being reduced: to produce X
1672 -- I had to produce Y, to produce Y I had to produce Z, and so on.
1675 mkRedEnv :: SDoc -> (Inst -> WhatToDo) -> [Inst] -> RedEnv
1676 mkRedEnv doc try_me givens
1677 = RedEnv { red_doc = doc, red_try_me = try_me,
1678 red_givens = givens,
1680 red_improve = True }
1682 mkNoImproveRedEnv :: SDoc -> (Inst -> WhatToDo) -> RedEnv
1683 -- Do not do improvement; no givens
1684 mkNoImproveRedEnv doc try_me
1685 = RedEnv { red_doc = doc, red_try_me = try_me,
1688 red_improve = True }
1691 = ReduceMe WantSCs -- Try to reduce this
1692 -- If there's no instance, add the inst to the
1693 -- irreductible ones, but don't produce an error
1694 -- message of any kind.
1695 -- It might be quite legitimate such as (Eq a)!
1697 | Stop -- Return as irreducible unless it can
1698 -- be reduced to a constant in one step
1699 -- Do not add superclasses; see
1701 data WantSCs = NoSCs | AddSCs -- Tells whether we should add the superclasses
1702 -- of a predicate when adding it to the avails
1703 -- The reason for this flag is entirely the super-class loop problem
1704 -- Note [SUPER-CLASS LOOP 1]
1706 zonkRedEnv :: RedEnv -> TcM RedEnv
1708 = do { givens' <- mapM zonkInst (red_givens env)
1709 ; return $ env {red_givens = givens'}
1714 %************************************************************************
1716 \subsection[reduce]{@reduce@}
1718 %************************************************************************
1720 Note [Ancestor Equalities]
1721 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1722 During context reduction, we add to the wanted equalities also those
1723 equalities that (transitively) occur in superclass contexts of wanted
1724 class constraints. Consider the following code
1726 class a ~ Int => C a
1729 If (C a) is wanted, we want to add (a ~ Int), which will be discharged by
1730 substituting Int for a. Hence, we ultimately want (C Int), which we
1731 discharge with the explicit instance.
1734 reduceContext :: RedEnv
1736 -> TcM (ImprovementDone,
1737 TcDictBinds, -- Dictionary bindings
1738 [Inst], -- Irreducible
1739 TcM ()) -- Undo skolems from SkolemOccurs
1741 reduceContext env wanteds
1742 = do { traceTc (text "reduceContext" <+> (vcat [
1743 text "----------------------",
1745 text "given" <+> ppr (red_givens env),
1746 text "wanted" <+> ppr wanteds,
1747 text "----------------------"
1751 ; let givens = red_givens env
1752 (given_eqs0, given_dicts0) = partition isEqInst givens
1753 (wanted_eqs0, wanted_non_eqs) = partition isEqInst wanteds
1754 (wanted_implics0, wanted_dicts) = partition isImplicInst wanted_non_eqs
1756 -- We want to add as wanted equalities those that (transitively)
1757 -- occur in superclass contexts of wanted class constraints.
1758 -- See Note [Ancestor Equalities]
1759 ; ancestor_eqs <- ancestorEqualities wanted_dicts
1760 ; let wanted_eqs = wanted_eqs0 ++ ancestor_eqs
1761 ; traceTc $ text "reduceContext: ancestor eqs" <+> ppr ancestor_eqs
1763 -- 1. Normalise the *given* *equality* constraints
1764 ; (given_eqs, eliminate_skolems) <- normaliseGivenEqs given_eqs0
1766 -- 2. Normalise the *given* *dictionary* constraints
1767 -- wrt. the toplevel and given equations
1768 ; (given_dicts, given_binds) <- normaliseGivenDicts given_eqs
1771 -- 5. Build the Avail mapping from "given_dicts"
1772 ; (init_state, _) <- getLIE $ do
1773 { init_state <- foldlM addGiven emptyAvails given_dicts
1777 -- !!! ToDo: what to do with the "extra_givens"? For the
1778 -- moment I'm simply discarding them, which is probably wrong
1780 -- 6. Solve the *wanted* *dictionary* constraints (not implications)
1781 -- This may expose some further equational constraints...
1782 ; (avails, extra_eqs) <- getLIE (reduceList env wanted_dicts init_state)
1783 ; (dict_binds, bound_dicts, dict_irreds)
1784 <- extractResults avails wanted_dicts
1785 ; traceTc $ text "reduceContext extractresults" <+> vcat
1786 [ppr avails, ppr wanted_dicts, ppr dict_binds]
1788 -- Solve the wanted *implications*. In doing so, we can provide
1789 -- as "given" all the dicts that were originally given,
1790 -- *or* for which we now have bindings,
1791 -- *or* which are now irreds
1792 ; let implic_env = env { red_givens = givens ++ bound_dicts
1794 ; (implic_binds_s, implic_irreds_s)
1795 <- mapAndUnzipM (reduceImplication implic_env) wanted_implics0
1796 ; let implic_binds = unionManyBags implic_binds_s
1797 implic_irreds = concat implic_irreds_s
1799 -- Normalise the wanted equality constraints
1800 ; eq_irreds <- normaliseWantedEqs given_eqs (wanted_eqs ++ extra_eqs)
1802 -- Normalise the wanted dictionaries
1803 ; let irreds = dict_irreds ++ implic_irreds
1804 eqs = eq_irreds ++ given_eqs
1805 ; (norm_irreds, normalise_binds) <- normaliseWantedDicts eqs irreds
1807 -- Figure out whether we should go round again. We do so in either
1809 -- (1) If any of the mutable tyvars in givens or irreds has been
1810 -- filled in by improvement, there is merit in going around
1811 -- again, because we may make further progress.
1812 -- (2) If we managed to normalise any dicts, there is merit in going
1813 -- around gain, because reduceList may be able to get further.
1815 -- ToDo: We may have exposed new
1816 -- equality constraints and should probably go round again
1817 -- then as well. But currently we are dropping them on the
1820 ; let all_irreds = norm_irreds ++ eq_irreds
1821 ; improvedMetaTy <- anyM isFilledMetaTyVar $ varSetElems $
1822 tyVarsOfInsts (givens ++ all_irreds)
1823 ; let improvedDicts = not $ isEmptyBag normalise_binds
1824 improved = improvedMetaTy || improvedDicts
1826 -- The old plan (fragile)
1827 -- improveed = availsImproved avails
1828 -- || (not $ isEmptyBag normalise_binds1)
1829 -- || (not $ isEmptyBag normalise_binds2)
1830 -- || (any isEqInst irreds)
1832 ; traceTc (text "reduceContext end" <+> (vcat [
1833 text "----------------------",
1835 text "given" <+> ppr givens,
1836 text "given_eqs" <+> ppr given_eqs,
1837 text "wanted" <+> ppr wanteds,
1838 text "wanted_dicts" <+> ppr wanted_dicts,
1840 text "avails" <+> pprAvails avails,
1841 text "improved =" <+> ppr improved,
1842 text "(all) irreds = " <+> ppr all_irreds,
1843 text "dict-binds = " <+> ppr dict_binds,
1844 text "implic-binds = " <+> ppr implic_binds,
1845 text "----------------------"
1849 given_binds `unionBags` normalise_binds
1850 `unionBags` dict_binds
1851 `unionBags` implic_binds,
1856 tcImproveOne :: Avails -> Inst -> TcM ImprovementDone
1857 tcImproveOne avails inst
1858 | not (isDict inst) = return False
1860 = do { inst_envs <- tcGetInstEnvs
1861 ; let eqns = improveOne (classInstances inst_envs)
1862 (dictPred inst, pprInstArising inst)
1863 [ (dictPred p, pprInstArising p)
1864 | p <- availsInsts avails, isDict p ]
1865 -- Avails has all the superclasses etc (good)
1866 -- It also has all the intermediates of the deduction (good)
1867 -- It does not have duplicates (good)
1868 -- NB that (?x::t1) and (?x::t2) will be held separately in avails
1869 -- so that improve will see them separate
1870 ; traceTc (text "improveOne" <+> ppr inst)
1873 unifyEqns :: [(Equation,(PredType,SDoc),(PredType,SDoc))]
1874 -> TcM ImprovementDone
1875 unifyEqns [] = return False
1877 = do { traceTc (ptext (sLit "Improve:") <+> vcat (map pprEquationDoc eqns))
1881 unify ((qtvs, pairs), what1, what2)
1882 = addErrCtxtM (mkEqnMsg what1 what2) $ do
1883 (_, _, tenv) <- tcInstTyVars (varSetElems qtvs)
1884 mapM_ (unif_pr tenv) pairs
1885 unif_pr tenv (ty1,ty2) = unifyType (substTy tenv ty1) (substTy tenv ty2)
1887 pprEquationDoc :: (Equation, (PredType, SDoc), (PredType, SDoc)) -> SDoc
1888 pprEquationDoc (eqn, (p1, _), (p2, _)) = vcat [pprEquation eqn, nest 2 (ppr p1), nest 2 (ppr p2)]
1890 mkEqnMsg :: (TcPredType, SDoc) -> (TcPredType, SDoc) -> TidyEnv
1891 -> TcM (TidyEnv, SDoc)
1892 mkEqnMsg (pred1,from1) (pred2,from2) tidy_env
1893 = do { pred1' <- zonkTcPredType pred1; pred2' <- zonkTcPredType pred2
1894 ; let { pred1'' = tidyPred tidy_env pred1'; pred2'' = tidyPred tidy_env pred2' }
1895 ; let msg = vcat [ptext (sLit "When using functional dependencies to combine"),
1896 nest 2 (sep [ppr pred1'' <> comma, nest 2 from1]),
1897 nest 2 (sep [ppr pred2'' <> comma, nest 2 from2])]
1898 ; return (tidy_env, msg) }
1901 The main context-reduction function is @reduce@. Here's its game plan.
1904 reduceList :: RedEnv -> [Inst] -> Avails -> TcM Avails
1905 reduceList env@(RedEnv {red_stack = (n,stk)}) wanteds state
1906 = do { traceTc (text "reduceList " <+> (ppr wanteds $$ ppr state))
1908 ; when (debugIsOn && (n > 8)) $ do
1909 debugDumpTcRn (hang (ptext (sLit "Interesting! Context reduction stack depth") <+> int n)
1910 2 (ifPprDebug (nest 2 (pprStack stk))))
1911 ; if n >= ctxtStkDepth dopts then
1912 failWithTc (reduceDepthErr n stk)
1916 go [] state = return state
1917 go (w:ws) state = do { state' <- reduce (env {red_stack = (n+1, w:stk)}) w state
1920 -- Base case: we're done!
1921 reduce :: RedEnv -> Inst -> Avails -> TcM Avails
1922 reduce env wanted avails
1923 -- It's the same as an existing inst, or a superclass thereof
1924 | Just _ <- findAvail avails wanted
1925 = do { traceTc (text "reduce: found " <+> ppr wanted)
1930 = do { traceTc (text "reduce" <+> ppr wanted $$ ppr avails)
1931 ; case red_try_me env wanted of {
1932 Stop -> try_simple (addIrred NoSCs);
1933 -- See Note [No superclasses for Stop]
1935 ReduceMe want_scs -> do -- It should be reduced
1936 { (avails, lookup_result) <- reduceInst env avails wanted
1937 ; case lookup_result of
1938 NoInstance -> addIrred want_scs avails wanted
1939 -- Add it and its superclasses
1941 GenInst [] rhs -> addWanted want_scs avails wanted rhs []
1943 GenInst wanteds' rhs
1944 -> do { avails1 <- addIrred NoSCs avails wanted
1945 ; avails2 <- reduceList env wanteds' avails1
1946 ; addWanted want_scs avails2 wanted rhs wanteds' } }
1947 -- Temporarily do addIrred *before* the reduceList,
1948 -- which has the effect of adding the thing we are trying
1949 -- to prove to the database before trying to prove the things it
1950 -- needs. See note [RECURSIVE DICTIONARIES]
1951 -- NB: we must not do an addWanted before, because that adds the
1952 -- superclasses too, and that can lead to a spurious loop; see
1953 -- the examples in [SUPERCLASS-LOOP]
1954 -- So we do an addIrred before, and then overwrite it afterwards with addWanted
1957 -- First, see if the inst can be reduced to a constant in one step
1958 -- Works well for literals (1::Int) and constant dictionaries (d::Num Int)
1959 -- Don't bother for implication constraints, which take real work
1960 try_simple do_this_otherwise
1961 = do { res <- lookupSimpleInst wanted
1963 GenInst [] rhs -> addWanted AddSCs avails wanted rhs []
1964 _ -> do_this_otherwise avails wanted }
1968 Note [SUPERCLASS-LOOP 2]
1969 ~~~~~~~~~~~~~~~~~~~~~~~~
1970 But the above isn't enough. Suppose we are *given* d1:Ord a,
1971 and want to deduce (d2:C [a]) where
1973 class Ord a => C a where
1974 instance Ord [a] => C [a] where ...
1976 Then we'll use the instance decl to deduce C [a] from Ord [a], and then add the
1977 superclasses of C [a] to avails. But we must not overwrite the binding
1978 for Ord [a] (which is obtained from Ord a) with a superclass selection or we'll just
1981 Here's another variant, immortalised in tcrun020
1982 class Monad m => C1 m
1983 class C1 m => C2 m x
1984 instance C2 Maybe Bool
1985 For the instance decl we need to build (C1 Maybe), and it's no good if
1986 we run around and add (C2 Maybe Bool) and its superclasses to the avails
1987 before we search for C1 Maybe.
1989 Here's another example
1990 class Eq b => Foo a b
1991 instance Eq a => Foo [a] a
1995 we'll first deduce that it holds (via the instance decl). We must not
1996 then overwrite the Eq t constraint with a superclass selection!
1998 At first I had a gross hack, whereby I simply did not add superclass constraints
1999 in addWanted, though I did for addGiven and addIrred. This was sub-optimal,
2000 becuase it lost legitimate superclass sharing, and it still didn't do the job:
2001 I found a very obscure program (now tcrun021) in which improvement meant the
2002 simplifier got two bites a the cherry... so something seemed to be an Stop
2003 first time, but reducible next time.
2005 Now we implement the Right Solution, which is to check for loops directly
2006 when adding superclasses. It's a bit like the occurs check in unification.
2009 Note [RECURSIVE DICTIONARIES]
2010 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2012 data D r = ZeroD | SuccD (r (D r));
2014 instance (Eq (r (D r))) => Eq (D r) where
2015 ZeroD == ZeroD = True
2016 (SuccD a) == (SuccD b) = a == b
2019 equalDC :: D [] -> D [] -> Bool;
2022 We need to prove (Eq (D [])). Here's how we go:
2026 by instance decl, holds if
2030 by instance decl of Eq, holds if
2032 where d2 = dfEqList d3
2035 But now we can "tie the knot" to give
2041 and it'll even run! The trick is to put the thing we are trying to prove
2042 (in this case Eq (D []) into the database before trying to prove its
2043 contributing clauses.
2046 %************************************************************************
2048 Reducing a single constraint
2050 %************************************************************************
2053 ---------------------------------------------
2054 reduceInst :: RedEnv -> Avails -> Inst -> TcM (Avails, LookupInstResult)
2055 reduceInst _ avails other_inst
2056 = do { result <- lookupSimpleInst other_inst
2057 ; return (avails, result) }
2060 Note [Equational Constraints in Implication Constraints]
2061 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2063 An implication constraint is of the form
2065 where Given and Wanted may contain both equational and dictionary
2066 constraints. The delay and reduction of these two kinds of constraints
2069 -) In the generated code, wanted Dictionary constraints are wrapped up in an
2070 implication constraint that is created at the code site where the wanted
2071 dictionaries can be reduced via a let-binding. This let-bound implication
2072 constraint is deconstructed at the use-site of the wanted dictionaries.
2074 -) While the reduction of equational constraints is also delayed, the delay
2075 is not manifest in the generated code. The required evidence is generated
2076 in the code directly at the use-site. There is no let-binding and deconstruction
2077 necessary. The main disadvantage is that we cannot exploit sharing as the
2078 same evidence may be generated at multiple use-sites. However, this disadvantage
2079 is limited because it only concerns coercions which are erased.
2081 The different treatment is motivated by the different in representation. Dictionary
2082 constraints require manifest runtime dictionaries, while equations require coercions
2086 ---------------------------------------------
2087 reduceImplication :: RedEnv
2089 -> TcM (TcDictBinds, [Inst])
2092 Suppose we are simplifying the constraint
2093 forall bs. extras => wanted
2094 in the context of an overall simplification problem with givens 'givens'.
2097 * The 'givens' need not mention any of the quantified type variables
2098 e.g. forall {}. Eq a => Eq [a]
2099 forall {}. C Int => D (Tree Int)
2101 This happens when you have something like
2103 T1 :: Eq a => a -> T a
2106 f x = ...(case x of { T1 v -> v==v })...
2109 -- ToDo: should we instantiate tvs? I think it's not necessary
2111 -- Note on coercion variables:
2113 -- The extra given coercion variables are bound at two different sites:
2114 -- -) in the creation context of the implication constraint
2115 -- the solved equational constraints use these binders
2117 -- -) at the solving site of the implication constraint
2118 -- the solved dictionaries use these binders
2119 -- these binders are generated by reduceImplication
2121 reduceImplication env
2122 orig_implic@(ImplicInst { tci_name = name, tci_loc = inst_loc,
2124 tci_given = extra_givens, tci_wanted = wanteds })
2125 = do { -- Solve the sub-problem
2126 ; let try_me _ = ReduceMe AddSCs -- Note [Freeness and implications]
2127 env' = env { red_givens = extra_givens ++ red_givens env
2128 , red_doc = sep [ptext (sLit "reduceImplication for")
2130 nest 2 (parens $ ptext (sLit "within")
2132 , red_try_me = try_me }
2134 ; traceTc (text "reduceImplication" <+> vcat
2135 [ ppr (red_givens env), ppr extra_givens,
2137 ; (irreds, binds) <- checkLoop env' wanteds
2138 ; let (extra_eq_givens, extra_dict_givens) = partition isEqInst extra_givens
2139 -- SLPJ Sept 07: I think this is bogus; currently
2140 -- there are no Eqinsts in extra_givens
2141 dict_ids = map instToId extra_dict_givens
2143 -- Note [Reducing implication constraints]
2144 -- Tom -- update note, put somewhere!
2146 ; traceTc (text "reduceImplication result" <+> vcat
2147 [ppr irreds, ppr binds])
2149 ; -- extract superclass binds
2150 -- (sc_binds,_) <- extractResults avails []
2151 -- ; traceTc (text "reduceImplication sc_binds" <+> vcat
2152 -- [ppr sc_binds, ppr avails])
2155 -- SLPJ Sept 07: what if improvement happened inside the checkLoop?
2156 -- Then we must iterate the outer loop too!
2158 ; traceTc (text "reduceImplication condition" <+> ppr ((isEmptyLHsBinds binds) || (null irreds)))
2160 -- Progress is no longer measered by the number of bindings
2161 ; if (isEmptyLHsBinds binds) && (not $ null irreds) then -- No progress
2162 -- If there are any irreds, we back off and do nothing
2163 return (emptyBag, [orig_implic])
2165 { (simpler_implic_insts, bind)
2166 <- makeImplicationBind inst_loc tvs extra_givens irreds
2167 -- This binding is useless if the recursive simplification
2168 -- made no progress; but currently we don't try to optimise that
2169 -- case. After all, we only try hard to reduce at top level, or
2170 -- when inferring types.
2172 ; let dict_wanteds = filter (not . isEqInst) wanteds
2173 -- TOMDO: given equational constraints bug!
2174 -- we need a different evidence for given
2175 -- equations depending on whether we solve
2176 -- dictionary constraints or equational constraints
2178 eq_tyvars = varSetElems $ tyVarsOfTypes $ map eqInstType extra_eq_givens
2179 -- SLPJ Sept07: this looks Utterly Wrong to me, but I think
2180 -- that current extra_givens has no EqInsts, so
2181 -- it makes no difference
2182 co = wrap_inline -- Note [Always inline implication constraints]
2184 <.> mkWpLams eq_tyvars
2185 <.> mkWpLams dict_ids
2186 <.> WpLet (binds `unionBags` bind)
2187 wrap_inline | null dict_ids = idHsWrapper
2188 | otherwise = WpInline
2189 rhs = mkLHsWrap co payload
2190 loc = instLocSpan inst_loc
2191 payload = mkBigLHsTup (map (L loc . HsVar . instToId) dict_wanteds)
2194 ; traceTc (vcat [text "reduceImplication" <+> ppr name,
2195 ppr simpler_implic_insts,
2196 text "->" <+> ppr rhs])
2197 ; return (unitBag (L loc (VarBind (instToId orig_implic) rhs)),
2198 simpler_implic_insts)
2201 reduceImplication _ i = pprPanic "reduceImplication" (ppr i)
2204 Note [Always inline implication constraints]
2205 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2206 Suppose an implication constraint floats out of an INLINE function.
2207 Then although the implication has a single call site, it won't be
2208 inlined. And that is bad because it means that even if there is really
2209 *no* overloading (type signatures specify the exact types) there will
2210 still be dictionary passing in the resulting code. To avert this,
2211 we mark the implication constraints themselves as INLINE, at least when
2212 there is no loss of sharing as a result.
2214 Note [Freeness and implications]
2215 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2216 It's hard to say when an implication constraint can be floated out. Consider
2217 forall {} Eq a => Foo [a]
2218 The (Foo [a]) doesn't mention any of the quantified variables, but it
2219 still might be partially satisfied by the (Eq a).
2221 There is a useful special case when it *is* easy to partition the
2222 constraints, namely when there are no 'givens'. Consider
2223 forall {a}. () => Bar b
2224 There are no 'givens', and so there is no reason to capture (Bar b).
2225 We can let it float out. But if there is even one constraint we
2226 must be much more careful:
2227 forall {a}. C a b => Bar (m b)
2228 because (C a b) might have a superclass (D b), from which we might
2229 deduce (Bar [b]) when m later gets instantiated to []. Ha!
2231 Here is an even more exotic example
2233 Now consider the constraint
2234 forall b. D Int b => C Int
2235 We can satisfy the (C Int) from the superclass of D, so we don't want
2236 to float the (C Int) out, even though it mentions no type variable in
2239 One more example: the constraint
2241 instance (C a, E c) => E (a,c)
2243 constraint: forall b. D Int b => E (Int,c)
2245 You might think that the (D Int b) can't possibly contribute
2246 to solving (E (Int,c)), since the latter mentions 'c'. But
2247 in fact it can, because solving the (E (Int,c)) constraint needs
2250 and the (C Int) can be satisfied from the superclass of (D Int b).
2251 So we must still not float (E (Int,c)) out.
2253 To think about: special cases for unary type classes?
2255 Note [Pruning the givens in an implication constraint]
2256 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2257 Suppose we are about to form the implication constraint
2258 forall tvs. Eq a => Ord b
2259 The (Eq a) cannot contribute to the (Ord b), because it has no access to
2260 the type variable 'b'. So we could filter out the (Eq a) from the givens.
2261 But BE CAREFUL of the examples above in [Freeness and implications].
2263 Doing so would be a bit tidier, but all the implication constraints get
2264 simplified away by the optimiser, so it's no great win. So I don't take
2265 advantage of that at the moment.
2267 If you do, BE CAREFUL of wobbly type variables.
2270 %************************************************************************
2272 Avails and AvailHow: the pool of evidence
2274 %************************************************************************
2278 data Avails = Avails !ImprovementDone !AvailEnv
2280 type ImprovementDone = Bool -- True <=> some unification has happened
2281 -- so some Irreds might now be reducible
2282 -- keys that are now
2284 type AvailEnv = FiniteMap Inst AvailHow
2286 = IsIrred -- Used for irreducible dictionaries,
2287 -- which are going to be lambda bound
2289 | Given Inst -- Used for dictionaries for which we have a binding
2290 -- e.g. those "given" in a signature
2292 | Rhs -- Used when there is a RHS
2293 (LHsExpr TcId) -- The RHS
2294 [Inst] -- Insts free in the RHS; we need these too
2296 instance Outputable Avails where
2299 pprAvails :: Avails -> SDoc
2300 pprAvails (Avails imp avails)
2301 = vcat [ ptext (sLit "Avails") <> (if imp then ptext (sLit "[improved]") else empty)
2303 vcat [ sep [ppr inst, nest 2 (equals <+> ppr avail)]
2304 | (inst,avail) <- fmToList avails ]]
2306 instance Outputable AvailHow where
2309 -------------------------
2310 pprAvail :: AvailHow -> SDoc
2311 pprAvail IsIrred = text "Irred"
2312 pprAvail (Given x) = text "Given" <+> ppr x
2313 pprAvail (Rhs rhs bs) = sep [text "Rhs" <+> ppr bs,
2316 -------------------------
2317 extendAvailEnv :: AvailEnv -> Inst -> AvailHow -> AvailEnv
2318 extendAvailEnv env inst avail = addToFM env inst avail
2320 findAvailEnv :: AvailEnv -> Inst -> Maybe AvailHow
2321 findAvailEnv env wanted = lookupFM env wanted
2322 -- NB 1: the Ord instance of Inst compares by the class/type info
2323 -- *not* by unique. So
2324 -- d1::C Int == d2::C Int
2326 emptyAvails :: Avails
2327 emptyAvails = Avails False emptyFM
2329 findAvail :: Avails -> Inst -> Maybe AvailHow
2330 findAvail (Avails _ avails) wanted = findAvailEnv avails wanted
2332 elemAvails :: Inst -> Avails -> Bool
2333 elemAvails wanted (Avails _ avails) = wanted `elemFM` avails
2335 extendAvails :: Avails -> Inst -> AvailHow -> TcM Avails
2337 extendAvails avails@(Avails imp env) inst avail
2338 = do { imp1 <- tcImproveOne avails inst -- Do any improvement
2339 ; return (Avails (imp || imp1) (extendAvailEnv env inst avail)) }
2341 availsInsts :: Avails -> [Inst]
2342 availsInsts (Avails _ avails) = keysFM avails
2344 _availsImproved :: Avails -> ImprovementDone
2345 _availsImproved (Avails imp _) = imp
2348 Extracting the bindings from a bunch of Avails.
2349 The bindings do *not* come back sorted in dependency order.
2350 We assume that they'll be wrapped in a big Rec, so that the
2351 dependency analyser can sort them out later
2354 type DoneEnv = FiniteMap Inst [Id]
2355 -- Tracks which things we have evidence for
2357 extractResults :: Avails
2359 -> TcM (TcDictBinds, -- Bindings
2360 [Inst], -- The insts bound by the bindings
2361 [Inst]) -- Irreducible ones
2362 -- Note [Reducing implication constraints]
2364 extractResults (Avails _ avails) wanteds
2365 = go emptyBag [] [] emptyFM wanteds
2367 go :: TcDictBinds -- Bindings for dicts
2368 -> [Inst] -- Bound by the bindings
2370 -> DoneEnv -- Has an entry for each inst in the above three sets
2372 -> TcM (TcDictBinds, [Inst], [Inst])
2373 go binds bound_dicts irreds _ []
2374 = return (binds, bound_dicts, irreds)
2376 go binds bound_dicts irreds done (w:ws)
2377 | Just done_ids@(done_id : rest_done_ids) <- lookupFM done w
2378 = if w_id `elem` done_ids then
2379 go binds bound_dicts irreds done ws
2381 go (add_bind (nlHsVar done_id)) bound_dicts irreds
2382 (addToFM done w (done_id : w_id : rest_done_ids)) ws
2384 | otherwise -- Not yet done
2385 = case findAvailEnv avails w of
2386 Nothing -> pprTrace "Urk: extractResults" (ppr w) $
2387 go binds bound_dicts irreds done ws
2389 Just IsIrred -> go binds bound_dicts (w:irreds) done' ws
2391 Just (Rhs rhs ws') -> go (add_bind rhs) (w:bound_dicts) irreds done' (ws' ++ ws)
2393 Just (Given g) -> go binds' bound_dicts irreds (addToFM done w [g_id]) ws
2396 binds' | w_id == g_id = binds
2397 | otherwise = add_bind (nlHsVar g_id)
2400 done' = addToFM done w [w_id]
2401 add_bind rhs = addInstToDictBind binds w rhs
2405 Note [No superclasses for Stop]
2406 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2407 When we decide not to reduce an Inst -- the 'WhatToDo' --- we still
2408 add it to avails, so that any other equal Insts will be commoned up
2409 right here. However, we do *not* add superclasses. If we have
2412 but a is not bound here, then we *don't* want to derive dn from df
2413 here lest we lose sharing.
2416 addWanted :: WantSCs -> Avails -> Inst -> LHsExpr TcId -> [Inst] -> TcM Avails
2417 addWanted want_scs avails wanted rhs_expr wanteds
2418 = addAvailAndSCs want_scs avails wanted avail
2420 avail = Rhs rhs_expr wanteds
2422 addGiven :: Avails -> Inst -> TcM Avails
2423 addGiven avails given = addAvailAndSCs AddSCs avails given (Given given)
2424 -- Always add superclasses for 'givens'
2426 -- No ASSERT( not (given `elemAvails` avails) ) because in an instance
2427 -- decl for Ord t we can add both Ord t and Eq t as 'givens',
2428 -- so the assert isn't true
2432 addIrred :: WantSCs -> Avails -> Inst -> TcM Avails
2433 addIrred want_scs avails irred = ASSERT2( not (irred `elemAvails` avails), ppr irred $$ ppr avails )
2434 addAvailAndSCs want_scs avails irred IsIrred
2436 addAvailAndSCs :: WantSCs -> Avails -> Inst -> AvailHow -> TcM Avails
2437 addAvailAndSCs want_scs avails inst avail
2438 | not (isClassDict inst) = extendAvails avails inst avail
2439 | NoSCs <- want_scs = extendAvails avails inst avail
2440 | otherwise = do { traceTc (text "addAvailAndSCs" <+> vcat [ppr inst, ppr deps])
2441 ; avails' <- extendAvails avails inst avail
2442 ; addSCs is_loop avails' inst }
2444 is_loop pred = any (`tcEqType` mkPredTy pred) dep_tys
2445 -- Note: this compares by *type*, not by Unique
2446 deps = findAllDeps (unitVarSet (instToVar inst)) avail
2447 dep_tys = map idType (varSetElems deps)
2449 findAllDeps :: IdSet -> AvailHow -> IdSet
2450 -- Find all the Insts that this one depends on
2451 -- See Note [SUPERCLASS-LOOP 2]
2452 -- Watch out, though. Since the avails may contain loops
2453 -- (see Note [RECURSIVE DICTIONARIES]), so we need to track the ones we've seen so far
2454 findAllDeps so_far (Rhs _ kids) = foldl find_all so_far kids
2455 findAllDeps so_far _ = so_far
2457 find_all :: IdSet -> Inst -> IdSet
2459 | isEqInst kid = so_far
2460 | kid_id `elemVarSet` so_far = so_far
2461 | Just avail <- findAvail avails kid = findAllDeps so_far' avail
2462 | otherwise = so_far'
2464 so_far' = extendVarSet so_far kid_id -- Add the new kid to so_far
2465 kid_id = instToId kid
2467 addSCs :: (TcPredType -> Bool) -> Avails -> Inst -> TcM Avails
2468 -- Add all the superclasses of the Inst to Avails
2469 -- The first param says "don't do this because the original thing
2470 -- depends on this one, so you'd build a loop"
2471 -- Invariant: the Inst is already in Avails.
2473 addSCs is_loop avails dict
2474 = ASSERT( isDict dict )
2475 do { sc_dicts <- newDictBndrs (instLoc dict) sc_theta'
2476 ; foldlM add_sc avails (zipEqual "add_scs" sc_dicts sc_sels) }
2478 (clas, tys) = getDictClassTys dict
2479 (tyvars, sc_theta, sc_sels, _) = classBigSig clas
2480 sc_theta' = filter (not . isEqPred) $
2481 substTheta (zipTopTvSubst tyvars tys) sc_theta
2483 add_sc avails (sc_dict, sc_sel)
2484 | is_loop (dictPred sc_dict) = return avails -- See Note [SUPERCLASS-LOOP 2]
2485 | is_given sc_dict = return avails
2486 | otherwise = do { avails' <- extendAvails avails sc_dict (Rhs sc_sel_rhs [dict])
2487 ; addSCs is_loop avails' sc_dict }
2489 sc_sel_rhs = L (instSpan dict) (HsWrap co_fn (HsVar sc_sel))
2490 co_fn = WpApp (instToVar dict) <.> mkWpTyApps tys
2492 is_given :: Inst -> Bool
2493 is_given sc_dict = case findAvail avails sc_dict of
2494 Just (Given _) -> True -- Given is cheaper than superclass selection
2497 -- From the a set of insts obtain all equalities that (transitively) occur in
2498 -- superclass contexts of class constraints (aka the ancestor equalities).
2500 ancestorEqualities :: [Inst] -> TcM [Inst]
2502 = mapM mkWantedEqInst -- turn only equality predicates..
2503 . filter isEqPred -- ..into wanted equality insts
2505 . addAEsToBag emptyBag -- collect the superclass constraints..
2506 . map dictPred -- ..of all predicates in a bag
2507 . filter isClassDict
2509 addAEsToBag :: Bag PredType -> [PredType] -> Bag PredType
2510 addAEsToBag bag [] = bag
2511 addAEsToBag bag (pred:preds)
2512 | pred `elemBag` bag = addAEsToBag bag preds
2513 | isEqPred pred = addAEsToBag bagWithPred preds
2514 | isClassPred pred = addAEsToBag bagWithPred predsWithSCs
2515 | otherwise = addAEsToBag bag preds
2517 bagWithPred = bag `snocBag` pred
2518 predsWithSCs = preds ++ substTheta (zipTopTvSubst tyvars tys) sc_theta
2520 (tyvars, sc_theta, _, _) = classBigSig clas
2521 (clas, tys) = getClassPredTys pred
2525 %************************************************************************
2527 \section{tcSimplifyTop: defaulting}
2529 %************************************************************************
2532 @tcSimplifyTop@ is called once per module to simplify all the constant
2533 and ambiguous Insts.
2535 We need to be careful of one case. Suppose we have
2537 instance Num a => Num (Foo a b) where ...
2539 and @tcSimplifyTop@ is given a constraint (Num (Foo x y)). Then it'll simplify
2540 to (Num x), and default x to Int. But what about y??
2542 It's OK: the final zonking stage should zap y to (), which is fine.
2546 tcSimplifyTop, tcSimplifyInteractive :: [Inst] -> TcM TcDictBinds
2547 tcSimplifyTop wanteds
2548 = tc_simplify_top doc False wanteds
2550 doc = text "tcSimplifyTop"
2552 tcSimplifyInteractive wanteds
2553 = tc_simplify_top doc True wanteds
2555 doc = text "tcSimplifyInteractive"
2557 -- The TcLclEnv should be valid here, solely to improve
2558 -- error message generation for the monomorphism restriction
2559 tc_simplify_top :: SDoc -> Bool -> [Inst] -> TcM (Bag (LHsBind TcId))
2560 tc_simplify_top doc interactive wanteds
2561 = do { dflags <- getDOpts
2562 ; wanteds <- zonkInsts wanteds
2563 ; mapM_ zonkTopTyVar (varSetElems (tyVarsOfInsts wanteds))
2565 ; traceTc (text "tc_simplify_top 0: " <+> ppr wanteds)
2566 ; (irreds1, binds1) <- tryHardCheckLoop doc1 wanteds
2567 -- ; (irreds1, binds1) <- gentleInferLoop doc1 wanteds
2568 ; traceTc (text "tc_simplify_top 1: " <+> ppr irreds1)
2569 ; (irreds2, binds2) <- approximateImplications doc2 (\_ -> True) irreds1
2570 ; traceTc (text "tc_simplify_top 2: " <+> ppr irreds2)
2572 -- Use the defaulting rules to do extra unification
2573 -- NB: irreds2 are already zonked
2574 ; (irreds3, binds3) <- disambiguate doc3 interactive dflags irreds2
2576 -- Deal with implicit parameters
2577 ; let (bad_ips, non_ips) = partition isIPDict irreds3
2578 (ambigs, others) = partition isTyVarDict non_ips
2580 ; topIPErrs bad_ips -- Can arise from f :: Int -> Int
2582 ; addNoInstanceErrs others
2583 ; addTopAmbigErrs ambigs
2585 ; return (binds1 `unionBags` binds2 `unionBags` binds3) }
2587 doc1 = doc <+> ptext (sLit "(first round)")
2588 doc2 = doc <+> ptext (sLit "(approximate)")
2589 doc3 = doc <+> ptext (sLit "(disambiguate)")
2592 If a dictionary constrains a type variable which is
2593 * not mentioned in the environment
2594 * and not mentioned in the type of the expression
2595 then it is ambiguous. No further information will arise to instantiate
2596 the type variable; nor will it be generalised and turned into an extra
2597 parameter to a function.
2599 It is an error for this to occur, except that Haskell provided for
2600 certain rules to be applied in the special case of numeric types.
2602 * at least one of its classes is a numeric class, and
2603 * all of its classes are numeric or standard
2604 then the type variable can be defaulted to the first type in the
2605 default-type list which is an instance of all the offending classes.
2607 So here is the function which does the work. It takes the ambiguous
2608 dictionaries and either resolves them (producing bindings) or
2609 complains. It works by splitting the dictionary list by type
2610 variable, and using @disambigOne@ to do the real business.
2612 @disambigOne@ assumes that its arguments dictionaries constrain all
2613 the same type variable.
2615 ADR Comment 20/6/94: I've changed the @CReturnable@ case to default to
2616 @()@ instead of @Int@. I reckon this is the Right Thing to do since
2617 the most common use of defaulting is code like:
2619 _ccall_ foo `seqPrimIO` bar
2621 Since we're not using the result of @foo@, the result if (presumably)
2625 disambiguate :: SDoc -> Bool -> DynFlags -> [Inst] -> TcM ([Inst], TcDictBinds)
2626 -- Just does unification to fix the default types
2627 -- The Insts are assumed to be pre-zonked
2628 disambiguate doc interactive dflags insts
2630 = return (insts, emptyBag)
2632 | null defaultable_groups
2633 = do { traceTc (text "disambigutate, no defaultable groups" <+> vcat [ppr unaries, ppr insts, ppr bad_tvs, ppr defaultable_groups])
2634 ; return (insts, emptyBag) }
2637 = do { -- Figure out what default types to use
2638 default_tys <- getDefaultTys extended_defaulting ovl_strings
2640 ; traceTc (text "disambiguate1" <+> vcat [ppr insts, ppr unaries, ppr bad_tvs, ppr defaultable_groups])
2641 ; mapM_ (disambigGroup default_tys) defaultable_groups
2643 -- disambigGroup does unification, hence try again
2644 ; tryHardCheckLoop doc insts }
2647 extended_defaulting = interactive || dopt Opt_ExtendedDefaultRules dflags
2648 ovl_strings = dopt Opt_OverloadedStrings dflags
2650 unaries :: [(Inst, Class, TcTyVar)] -- (C tv) constraints
2651 bad_tvs :: TcTyVarSet -- Tyvars mentioned by *other* constraints
2652 (unaries, bad_tvs_s) = partitionWith find_unary insts
2653 bad_tvs = unionVarSets bad_tvs_s
2655 -- Finds unary type-class constraints
2656 find_unary d@(Dict {tci_pred = ClassP cls [ty]})
2657 | Just tv <- tcGetTyVar_maybe ty = Left (d,cls,tv)
2658 find_unary inst = Right (tyVarsOfInst inst)
2660 -- Group by type variable
2661 defaultable_groups :: [[(Inst,Class,TcTyVar)]]
2662 defaultable_groups = filter defaultable_group (equivClasses cmp_tv unaries)
2663 cmp_tv (_,_,tv1) (_,_,tv2) = tv1 `compare` tv2
2665 defaultable_group :: [(Inst,Class,TcTyVar)] -> Bool
2666 defaultable_group ds@((_,_,tv):_)
2667 = isTyConableTyVar tv -- Note [Avoiding spurious errors]
2668 && not (tv `elemVarSet` bad_tvs)
2669 && defaultable_classes [c | (_,c,_) <- ds]
2670 defaultable_group [] = panic "defaultable_group"
2672 defaultable_classes clss
2673 | extended_defaulting = any isInteractiveClass clss
2674 | otherwise = all is_std_class clss && (any is_num_class clss)
2676 -- In interactive mode, or with -XExtendedDefaultRules,
2677 -- we default Show a to Show () to avoid graututious errors on "show []"
2678 isInteractiveClass cls
2679 = is_num_class cls || (classKey cls `elem` [showClassKey, eqClassKey, ordClassKey])
2681 is_num_class cls = isNumericClass cls || (ovl_strings && (cls `hasKey` isStringClassKey))
2682 -- is_num_class adds IsString to the standard numeric classes,
2683 -- when -foverloaded-strings is enabled
2685 is_std_class cls = isStandardClass cls || (ovl_strings && (cls `hasKey` isStringClassKey))
2686 -- Similarly is_std_class
2688 -----------------------
2689 disambigGroup :: [Type] -- The default types
2690 -> [(Inst,Class,TcTyVar)] -- All standard classes of form (C a)
2691 -> TcM () -- Just does unification, to fix the default types
2693 disambigGroup default_tys dicts
2694 = try_default default_tys
2696 (_,_,tyvar) = ASSERT(not (null dicts)) head dicts -- Should be non-empty
2697 classes = [c | (_,c,_) <- dicts]
2699 try_default [] = return ()
2700 try_default (default_ty : default_tys)
2701 = tryTcLIE_ (try_default default_tys) $
2702 do { tcSimplifyDefault [mkClassPred clas [default_ty] | clas <- classes]
2703 -- This may fail; then the tryTcLIE_ kicks in
2704 -- Failure here is caused by there being no type in the
2705 -- default list which can satisfy all the ambiguous classes.
2706 -- For example, if Real a is reqd, but the only type in the
2707 -- default list is Int.
2709 -- After this we can't fail
2710 ; warnDefault dicts default_ty
2711 ; unifyType default_ty (mkTyVarTy tyvar)
2712 ; return () -- TOMDO: do something with the coercion
2716 -----------------------
2717 getDefaultTys :: Bool -> Bool -> TcM [Type]
2718 getDefaultTys extended_deflts ovl_strings
2719 = do { mb_defaults <- getDeclaredDefaultTys
2720 ; case mb_defaults of {
2721 Just tys -> return tys ; -- User-supplied defaults
2724 -- No use-supplied default
2725 -- Use [Integer, Double], plus modifications
2726 { integer_ty <- tcMetaTy integerTyConName
2727 ; checkWiredInTyCon doubleTyCon
2728 ; string_ty <- tcMetaTy stringTyConName
2729 ; return (opt_deflt extended_deflts unitTy
2730 -- Note [Default unitTy]
2732 [integer_ty,doubleTy]
2734 opt_deflt ovl_strings string_ty) } } }
2736 opt_deflt True ty = [ty]
2737 opt_deflt False _ = []
2740 Note [Default unitTy]
2741 ~~~~~~~~~~~~~~~~~~~~~
2742 In interative mode (or with -XExtendedDefaultRules) we add () as the first type we
2743 try when defaulting. This has very little real impact, except in the following case.
2745 Text.Printf.printf "hello"
2746 This has type (forall a. IO a); it prints "hello", and returns 'undefined'. We don't
2747 want the GHCi repl loop to try to print that 'undefined'. The neatest thing is to
2748 default the 'a' to (), rather than to Integer (which is what would otherwise happen;
2749 and then GHCi doesn't attempt to print the (). So in interactive mode, we add
2750 () to the list of defaulting types. See Trac #1200.
2752 Note [Avoiding spurious errors]
2753 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2754 When doing the unification for defaulting, we check for skolem
2755 type variables, and simply don't default them. For example:
2756 f = (*) -- Monomorphic
2757 g :: Num a => a -> a
2759 Here, we get a complaint when checking the type signature for g,
2760 that g isn't polymorphic enough; but then we get another one when
2761 dealing with the (Num a) context arising from f's definition;
2762 we try to unify a with Int (to default it), but find that it's
2763 already been unified with the rigid variable from g's type sig
2766 %************************************************************************
2768 \subsection[simple]{@Simple@ versions}
2770 %************************************************************************
2772 Much simpler versions when there are no bindings to make!
2774 @tcSimplifyThetas@ simplifies class-type constraints formed by
2775 @deriving@ declarations and when specialising instances. We are
2776 only interested in the simplified bunch of class/type constraints.
2778 It simplifies to constraints of the form (C a b c) where
2779 a,b,c are type variables. This is required for the context of
2780 instance declarations.
2783 tcSimplifyDeriv :: InstOrigin
2785 -> ThetaType -- Wanted
2786 -> TcM ThetaType -- Needed
2787 -- Given instance (wanted) => C inst_ty
2788 -- Simplify 'wanted' as much as possible
2790 tcSimplifyDeriv orig tyvars theta
2791 = do { (tvs, _, tenv) <- tcInstTyVars tyvars
2792 -- The main loop may do unification, and that may crash if
2793 -- it doesn't see a TcTyVar, so we have to instantiate. Sigh
2794 -- ToDo: what if two of them do get unified?
2795 ; wanteds <- newDictBndrsO orig (substTheta tenv theta)
2796 ; (irreds, _) <- tryHardCheckLoop doc wanteds
2798 ; let (tv_dicts, others) = partition ok irreds
2799 ; addNoInstanceErrs others
2800 -- See Note [Exotic derived instance contexts] in TcMType
2802 ; let rev_env = zipTopTvSubst tvs (mkTyVarTys tyvars)
2803 simpl_theta = substTheta rev_env (map dictPred tv_dicts)
2804 -- This reverse-mapping is a pain, but the result
2805 -- should mention the original TyVars not TcTyVars
2807 ; return simpl_theta }
2809 doc = ptext (sLit "deriving classes for a data type")
2811 ok dict | isDict dict = validDerivPred (dictPred dict)
2816 @tcSimplifyDefault@ just checks class-type constraints, essentially;
2817 used with \tr{default} declarations. We are only interested in
2818 whether it worked or not.
2821 tcSimplifyDefault :: ThetaType -- Wanted; has no type variables in it
2824 tcSimplifyDefault theta = do
2825 wanteds <- newDictBndrsO DefaultOrigin theta
2826 (irreds, _) <- tryHardCheckLoop doc wanteds
2827 addNoInstanceErrs irreds
2831 traceTc (ptext (sLit "tcSimplifyDefault failing")) >> failM
2833 doc = ptext (sLit "default declaration")
2837 %************************************************************************
2839 \section{Errors and contexts}
2841 %************************************************************************
2843 ToDo: for these error messages, should we note the location as coming
2844 from the insts, or just whatever seems to be around in the monad just
2848 groupErrs :: ([Inst] -> TcM ()) -- Deal with one group
2849 -> [Inst] -- The offending Insts
2851 -- Group together insts with the same origin
2852 -- We want to report them together in error messages
2856 groupErrs report_err (inst:insts)
2857 = do { do_one (inst:friends)
2858 ; groupErrs report_err others }
2860 -- (It may seem a bit crude to compare the error messages,
2861 -- but it makes sure that we combine just what the user sees,
2862 -- and it avoids need equality on InstLocs.)
2863 (friends, others) = partition is_friend insts
2864 loc_msg = showSDoc (pprInstLoc (instLoc inst))
2865 is_friend friend = showSDoc (pprInstLoc (instLoc friend)) == loc_msg
2866 do_one insts = addInstCtxt (instLoc (head insts)) (report_err insts)
2867 -- Add location and context information derived from the Insts
2869 -- Add the "arising from..." part to a message about bunch of dicts
2870 addInstLoc :: [Inst] -> Message -> Message
2871 addInstLoc insts msg = msg $$ nest 2 (pprInstArising (head insts))
2873 addTopIPErrs :: [Name] -> [Inst] -> TcM ()
2876 addTopIPErrs bndrs ips
2877 = do { dflags <- getDOpts
2878 ; addErrTcM (tidy_env, mk_msg dflags tidy_ips) }
2880 (tidy_env, tidy_ips) = tidyInsts ips
2882 = vcat [sep [ptext (sLit "Implicit parameters escape from"),
2883 nest 2 (ptext (sLit "the monomorphic top-level binding")
2884 <> plural bndrs <+> ptext (sLit "of")
2885 <+> pprBinders bndrs <> colon)],
2886 nest 2 (vcat (map ppr_ip ips)),
2887 monomorphism_fix dflags]
2888 ppr_ip ip = pprPred (dictPred ip) <+> pprInstArising ip
2890 topIPErrs :: [Inst] -> TcM ()
2892 = groupErrs report tidy_dicts
2894 (tidy_env, tidy_dicts) = tidyInsts dicts
2895 report dicts = addErrTcM (tidy_env, mk_msg dicts)
2896 mk_msg dicts = addInstLoc dicts (ptext (sLit "Unbound implicit parameter") <>
2897 plural tidy_dicts <+> pprDictsTheta tidy_dicts)
2899 addNoInstanceErrs :: [Inst] -- Wanted (can include implications)
2901 addNoInstanceErrs insts
2902 = do { let (tidy_env, tidy_insts) = tidyInsts insts
2903 ; reportNoInstances tidy_env Nothing tidy_insts }
2907 -> Maybe (InstLoc, [Inst]) -- Context
2908 -- Nothing => top level
2909 -- Just (d,g) => d describes the construct
2911 -> [Inst] -- What is wanted (can include implications)
2914 reportNoInstances tidy_env mb_what insts
2915 = groupErrs (report_no_instances tidy_env mb_what) insts
2917 report_no_instances :: TidyEnv -> Maybe (InstLoc, [Inst]) -> [Inst] -> TcM ()
2918 report_no_instances tidy_env mb_what insts
2919 = do { inst_envs <- tcGetInstEnvs
2920 ; let (implics, insts1) = partition isImplicInst insts
2921 (insts2, overlaps) = partitionWith (check_overlap inst_envs) insts1
2922 (eqInsts, insts3) = partition isEqInst insts2
2923 ; traceTc (text "reportNoInstances" <+> vcat
2924 [ppr insts, ppr implics, ppr insts1, ppr insts2])
2925 ; mapM_ complain_implic implics
2926 ; mapM_ (\doc -> addErrTcM (tidy_env, doc)) overlaps
2927 ; groupErrs complain_no_inst insts3
2928 ; mapM_ (addErrTcM . mk_eq_err) eqInsts
2931 complain_no_inst insts = addErrTcM (tidy_env, mk_no_inst_err insts)
2933 complain_implic inst -- Recurse!
2934 = reportNoInstances tidy_env
2935 (Just (tci_loc inst, tci_given inst))
2938 check_overlap :: (InstEnv,InstEnv) -> Inst -> Either Inst SDoc
2939 -- Right msg => overlap message
2940 -- Left inst => no instance
2941 check_overlap inst_envs wanted
2942 | not (isClassDict wanted) = Left wanted
2944 = case lookupInstEnv inst_envs clas tys of
2945 ([], _) -> Left wanted -- No match
2946 -- The case of exactly one match and no unifiers means a
2947 -- successful lookup. That can't happen here, because dicts
2948 -- only end up here if they didn't match in Inst.lookupInst
2950 | debugIsOn -> pprPanic "reportNoInstance" (ppr wanted)
2951 res -> Right (mk_overlap_msg wanted res)
2953 (clas,tys) = getDictClassTys wanted
2955 mk_overlap_msg dict (matches, unifiers)
2956 = ASSERT( not (null matches) )
2957 vcat [ addInstLoc [dict] ((ptext (sLit "Overlapping instances for")
2958 <+> pprPred (dictPred dict))),
2959 sep [ptext (sLit "Matching instances") <> colon,
2960 nest 2 (vcat [pprInstances ispecs, pprInstances unifiers])],
2961 if not (isSingleton matches)
2962 then -- Two or more matches
2964 else -- One match, plus some unifiers
2965 ASSERT( not (null unifiers) )
2966 parens (vcat [ptext (sLit "The choice depends on the instantiation of") <+>
2967 quotes (pprWithCommas ppr (varSetElems (tyVarsOfInst dict))),
2968 ptext (sLit "To pick the first instance above, use -XIncoherentInstances"),
2969 ptext (sLit "when compiling the other instance declarations")])]
2971 ispecs = [ispec | (ispec, _) <- matches]
2973 mk_eq_err :: Inst -> (TidyEnv, SDoc)
2974 mk_eq_err inst = misMatchMsg tidy_env (eqInstTys inst)
2976 mk_no_inst_err insts
2977 | null insts = empty
2979 | Just (loc, givens) <- mb_what, -- Nested (type signatures, instance decls)
2980 not (isEmptyVarSet (tyVarsOfInsts insts))
2981 = vcat [ addInstLoc insts $
2982 sep [ ptext (sLit "Could not deduce") <+> pprDictsTheta insts
2983 , nest 2 $ ptext (sLit "from the context") <+> pprDictsTheta givens]
2984 , show_fixes (fix1 loc : fixes2) ]
2986 | otherwise -- Top level
2987 = vcat [ addInstLoc insts $
2988 ptext (sLit "No instance") <> plural insts
2989 <+> ptext (sLit "for") <+> pprDictsTheta insts
2990 , show_fixes fixes2 ]
2993 fix1 loc = sep [ ptext (sLit "add") <+> pprDictsTheta insts
2994 <+> ptext (sLit "to the context of"),
2995 nest 2 (ppr (instLocOrigin loc)) ]
2996 -- I'm not sure it helps to add the location
2997 -- nest 2 (ptext (sLit "at") <+> ppr (instLocSpan loc)) ]
2999 fixes2 | null instance_dicts = []
3000 | otherwise = [sep [ptext (sLit "add an instance declaration for"),
3001 pprDictsTheta instance_dicts]]
3002 instance_dicts = [d | d <- insts, isClassDict d, not (isTyVarDict d)]
3003 -- Insts for which it is worth suggesting an adding an instance declaration
3004 -- Exclude implicit parameters, and tyvar dicts
3006 show_fixes :: [SDoc] -> SDoc
3007 show_fixes [] = empty
3008 show_fixes (f:fs) = sep [ptext (sLit "Possible fix:"),
3009 nest 2 (vcat (f : map (ptext (sLit "or") <+>) fs))]
3011 addTopAmbigErrs :: [Inst] -> TcRn ()
3012 addTopAmbigErrs dicts
3013 -- Divide into groups that share a common set of ambiguous tyvars
3014 = ifErrsM (return ()) $ -- Only report ambiguity if no other errors happened
3015 -- See Note [Avoiding spurious errors]
3016 mapM_ report (equivClasses cmp [(d, tvs_of d) | d <- tidy_dicts])
3018 (tidy_env, tidy_dicts) = tidyInsts dicts
3020 tvs_of :: Inst -> [TcTyVar]
3021 tvs_of d = varSetElems (tyVarsOfInst d)
3022 cmp (_,tvs1) (_,tvs2) = tvs1 `compare` tvs2
3024 report :: [(Inst,[TcTyVar])] -> TcM ()
3025 report pairs@((inst,tvs) : _) = do -- The pairs share a common set of ambiguous tyvars
3026 (tidy_env, mono_msg) <- mkMonomorphismMsg tidy_env tvs
3027 setSrcSpan (instSpan inst) $
3028 -- the location of the first one will do for the err message
3029 addErrTcM (tidy_env, msg $$ mono_msg)
3031 dicts = map fst pairs
3032 msg = sep [text "Ambiguous type variable" <> plural tvs <+>
3033 pprQuotedList tvs <+> in_msg,
3034 nest 2 (pprDictsInFull dicts)]
3035 in_msg = text "in the constraint" <> plural dicts <> colon
3036 report [] = panic "addTopAmbigErrs"
3039 mkMonomorphismMsg :: TidyEnv -> [TcTyVar] -> TcM (TidyEnv, Message)
3040 -- There's an error with these Insts; if they have free type variables
3041 -- it's probably caused by the monomorphism restriction.
3042 -- Try to identify the offending variable
3043 -- ASSUMPTION: the Insts are fully zonked
3044 mkMonomorphismMsg tidy_env inst_tvs
3045 = do { dflags <- getDOpts
3046 ; (tidy_env, docs) <- findGlobals (mkVarSet inst_tvs) tidy_env
3047 ; return (tidy_env, mk_msg dflags docs) }
3049 mk_msg _ _ | any isRuntimeUnk inst_tvs
3050 = vcat [ptext (sLit "Cannot resolve unknown runtime types:") <+>
3051 (pprWithCommas ppr inst_tvs),
3052 ptext (sLit "Use :print or :force to determine these types")]
3053 mk_msg _ [] = ptext (sLit "Probable fix: add a type signature that fixes these type variable(s)")
3054 -- This happens in things like
3055 -- f x = show (read "foo")
3056 -- where monomorphism doesn't play any role
3058 = vcat [ptext (sLit "Possible cause: the monomorphism restriction applied to the following:"),
3060 monomorphism_fix dflags]
3062 monomorphism_fix :: DynFlags -> SDoc
3063 monomorphism_fix dflags
3064 = ptext (sLit "Probable fix:") <+> vcat
3065 [ptext (sLit "give these definition(s) an explicit type signature"),
3066 if dopt Opt_MonomorphismRestriction dflags
3067 then ptext (sLit "or use -XNoMonomorphismRestriction")
3068 else empty] -- Only suggest adding "-XNoMonomorphismRestriction"
3069 -- if it is not already set!
3071 warnDefault :: [(Inst, Class, Var)] -> Type -> TcM ()
3072 warnDefault ups default_ty = do
3073 warn_flag <- doptM Opt_WarnTypeDefaults
3074 addInstCtxt (instLoc (head (dicts))) (warnTc warn_flag warn_msg)
3076 dicts = [d | (d,_,_) <- ups]
3079 (_, tidy_dicts) = tidyInsts dicts
3080 warn_msg = vcat [ptext (sLit "Defaulting the following constraint(s) to type") <+>
3081 quotes (ppr default_ty),
3082 pprDictsInFull tidy_dicts]
3084 reduceDepthErr :: Int -> [Inst] -> SDoc
3085 reduceDepthErr n stack
3086 = vcat [ptext (sLit "Context reduction stack overflow; size =") <+> int n,
3087 ptext (sLit "Use -fcontext-stack=N to increase stack size to N"),
3088 nest 4 (pprStack stack)]
3090 pprStack :: [Inst] -> SDoc
3091 pprStack stack = vcat (map pprInstInFull stack)