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
66 %************************************************************************
70 %************************************************************************
72 --------------------------------------
73 Notes on functional dependencies (a bug)
74 --------------------------------------
81 instance D a b => C a b -- Undecidable
82 -- (Not sure if it's crucial to this eg)
83 f :: C a b => a -> Bool
86 g :: C a b => a -> Bool
89 Here f typechecks, but g does not!! Reason: before doing improvement,
90 we reduce the (C a b1) constraint from the call of f to (D a b1).
92 Here is a more complicated example:
95 > class Foo a b | a->b
97 > class Bar a b | a->b
101 > instance Bar Obj Obj
103 > instance (Bar a b) => Foo a b
105 > foo:: (Foo a b) => a -> String
108 > runFoo:: (forall a b. (Foo a b) => a -> w) -> w
114 Could not deduce (Bar a b) from the context (Foo a b)
115 arising from use of `foo' at <interactive>:1
117 Add (Bar a b) to the expected type of an expression
118 In the first argument of `runFoo', namely `foo'
119 In the definition of `it': it = runFoo foo
121 Why all of the sudden does GHC need the constraint Bar a b? The
122 function foo didn't ask for that...
125 The trouble is that to type (runFoo foo), GHC has to solve the problem:
127 Given constraint Foo a b
128 Solve constraint Foo a b'
130 Notice that b and b' aren't the same. To solve this, just do
131 improvement and then they are the same. But GHC currently does
136 That is usually fine, but it isn't here, because it sees that Foo a b is
137 not the same as Foo a b', and so instead applies the instance decl for
138 instance Bar a b => Foo a b. And that's where the Bar constraint comes
141 The Right Thing is to improve whenever the constraint set changes at
142 all. Not hard in principle, but it'll take a bit of fiddling to do.
144 Note [Choosing which variables to quantify]
145 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
146 Suppose we are about to do a generalisation step. We have in our hand
149 T the type of the RHS
150 C the constraints from that RHS
152 The game is to figure out
154 Q the set of type variables over which to quantify
155 Ct the constraints we will *not* quantify over
156 Cq the constraints we will quantify over
158 So we're going to infer the type
162 and float the constraints Ct further outwards.
164 Here are the things that *must* be true:
166 (A) Q intersect fv(G) = EMPTY limits how big Q can be
167 (B) Q superset fv(Cq union T) \ oclose(fv(G),C) limits how small Q can be
169 (A) says we can't quantify over a variable that's free in the environment.
170 (B) says we must quantify over all the truly free variables in T, else
171 we won't get a sufficiently general type.
173 We do not *need* to quantify over any variable that is fixed by the
174 free vars of the environment G.
176 BETWEEN THESE TWO BOUNDS, ANY Q WILL DO!
178 Example: class H x y | x->y where ...
180 fv(G) = {a} C = {H a b, H c d}
183 (A) Q intersect {a} is empty
184 (B) Q superset {a,b,c,d} \ oclose({a}, C) = {a,b,c,d} \ {a,b} = {c,d}
186 So Q can be {c,d}, {b,c,d}
188 In particular, it's perfectly OK to quantify over more type variables
189 than strictly necessary; there is no need to quantify over 'b', since
190 it is determined by 'a' which is free in the envt, but it's perfectly
191 OK to do so. However we must not quantify over 'a' itself.
193 Other things being equal, however, we'd like to quantify over as few
194 variables as possible: smaller types, fewer type applications, more
195 constraints can get into Ct instead of Cq. Here's a good way to
198 Q = grow( fv(T), C ) \ oclose( fv(G), C )
200 That is, quantify over all variable that that MIGHT be fixed by the
201 call site (which influences T), but which aren't DEFINITELY fixed by
202 G. This choice definitely quantifies over enough type variables,
203 albeit perhaps too many.
205 Why grow( fv(T), C ) rather than fv(T)? Consider
207 class H x y | x->y where ...
212 If we used fv(T) = {c} we'd get the type
214 forall c. H c d => c -> b
216 And then if the fn was called at several different c's, each of
217 which fixed d differently, we'd get a unification error, because
218 d isn't quantified. Solution: quantify d. So we must quantify
219 everything that might be influenced by c.
221 Why not oclose( fv(T), C )? Because we might not be able to see
222 all the functional dependencies yet:
224 class H x y | x->y where ...
225 instance H x y => Eq (T x y) where ...
230 Now oclose(fv(T),C) = {c}, because the functional dependency isn't
231 apparent yet, and that's wrong. We must really quantify over d too.
233 There really isn't any point in quantifying over any more than
234 grow( fv(T), C ), because the call sites can't possibly influence
235 any other type variables.
239 -------------------------------------
241 -------------------------------------
243 It's very hard to be certain when a type is ambiguous. Consider
247 instance H x y => K (x,y)
249 Is this type ambiguous?
250 forall a b. (K (a,b), Eq b) => a -> a
252 Looks like it! But if we simplify (K (a,b)) we get (H a b) and
253 now we see that a fixes b. So we can't tell about ambiguity for sure
254 without doing a full simplification. And even that isn't possible if
255 the context has some free vars that may get unified. Urgle!
257 Here's another example: is this ambiguous?
258 forall a b. Eq (T b) => a -> a
259 Not if there's an insance decl (with no context)
260 instance Eq (T b) where ...
262 You may say of this example that we should use the instance decl right
263 away, but you can't always do that:
265 class J a b where ...
266 instance J Int b where ...
268 f :: forall a b. J a b => a -> a
270 (Notice: no functional dependency in J's class decl.)
271 Here f's type is perfectly fine, provided f is only called at Int.
272 It's premature to complain when meeting f's signature, or even
273 when inferring a type for f.
277 However, we don't *need* to report ambiguity right away. It'll always
278 show up at the call site.... and eventually at main, which needs special
279 treatment. Nevertheless, reporting ambiguity promptly is an excellent thing.
281 So here's the plan. We WARN about probable ambiguity if
283 fv(Cq) is not a subset of oclose(fv(T) union fv(G), C)
285 (all tested before quantification).
286 That is, all the type variables in Cq must be fixed by the the variables
287 in the environment, or by the variables in the type.
289 Notice that we union before calling oclose. Here's an example:
291 class J a b c | a b -> c
295 forall b c. (J a b c) => b -> b
297 Only if we union {a} from G with {b} from T before using oclose,
298 do we see that c is fixed.
300 It's a bit vague exactly which C we should use for this oclose call. If we
301 don't fix enough variables we might complain when we shouldn't (see
302 the above nasty example). Nothing will be perfect. That's why we can
303 only issue a warning.
306 Can we ever be *certain* about ambiguity? Yes: if there's a constraint
308 c in C such that fv(c) intersect (fv(G) union fv(T)) = EMPTY
310 then c is a "bubble"; there's no way it can ever improve, and it's
311 certainly ambiguous. UNLESS it is a constant (sigh). And what about
316 instance H x y => K (x,y)
318 Is this type ambiguous?
319 forall a b. (K (a,b), Eq b) => a -> a
321 Urk. The (Eq b) looks "definitely ambiguous" but it isn't. What we are after
322 is a "bubble" that's a set of constraints
324 Cq = Ca union Cq' st fv(Ca) intersect (fv(Cq') union fv(T) union fv(G)) = EMPTY
326 Hence another idea. To decide Q start with fv(T) and grow it
327 by transitive closure in Cq (no functional dependencies involved).
328 Now partition Cq using Q, leaving the definitely-ambiguous and probably-ok.
329 The definitely-ambiguous can then float out, and get smashed at top level
330 (which squashes out the constants, like Eq (T a) above)
333 --------------------------------------
334 Notes on principal types
335 --------------------------------------
340 f x = let g y = op (y::Int) in True
342 Here the principal type of f is (forall a. a->a)
343 but we'll produce the non-principal type
344 f :: forall a. C Int => a -> a
347 --------------------------------------
348 The need for forall's in constraints
349 --------------------------------------
351 [Exchange on Haskell Cafe 5/6 Dec 2000]
353 class C t where op :: t -> Bool
354 instance C [t] where op x = True
356 p y = (let f :: c -> Bool; f x = op (y >> return x) in f, y ++ [])
357 q y = (y ++ [], let f :: c -> Bool; f x = op (y >> return x) in f)
359 The definitions of p and q differ only in the order of the components in
360 the pair on their right-hand sides. And yet:
362 ghc and "Typing Haskell in Haskell" reject p, but accept q;
363 Hugs rejects q, but accepts p;
364 hbc rejects both p and q;
365 nhc98 ... (Malcolm, can you fill in the blank for us!).
367 The type signature for f forces context reduction to take place, and
368 the results of this depend on whether or not the type of y is known,
369 which in turn depends on which component of the pair the type checker
372 Solution: if y::m a, float out the constraints
373 Monad m, forall c. C (m c)
374 When m is later unified with [], we can solve both constraints.
377 --------------------------------------
378 Notes on implicit parameters
379 --------------------------------------
381 Note [Inheriting implicit parameters]
382 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
387 where f is *not* a top-level binding.
388 From the RHS of f we'll get the constraint (?y::Int).
389 There are two types we might infer for f:
393 (so we get ?y from the context of f's definition), or
395 f :: (?y::Int) => Int -> Int
397 At first you might think the first was better, becuase then
398 ?y behaves like a free variable of the definition, rather than
399 having to be passed at each call site. But of course, the WHOLE
400 IDEA is that ?y should be passed at each call site (that's what
401 dynamic binding means) so we'd better infer the second.
403 BOTTOM LINE: when *inferring types* you *must* quantify
404 over implicit parameters. See the predicate isFreeWhenInferring.
407 Note [Implicit parameters and ambiguity]
408 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
409 Only a *class* predicate can give rise to ambiguity
410 An *implicit parameter* cannot. For example:
411 foo :: (?x :: [a]) => Int
413 is fine. The call site will suppply a particular 'x'
415 Furthermore, the type variables fixed by an implicit parameter
416 propagate to the others. E.g.
417 foo :: (Show a, ?x::[a]) => Int
419 The type of foo looks ambiguous. But it isn't, because at a call site
421 let ?x = 5::Int in foo
422 and all is well. In effect, implicit parameters are, well, parameters,
423 so we can take their type variables into account as part of the
424 "tau-tvs" stuff. This is done in the function 'FunDeps.grow'.
427 Question 2: type signatures
428 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
429 BUT WATCH OUT: When you supply a type signature, we can't force you
430 to quantify over implicit parameters. For example:
434 This is perfectly reasonable. We do not want to insist on
436 (?x + 1) :: (?x::Int => Int)
438 That would be silly. Here, the definition site *is* the occurrence site,
439 so the above strictures don't apply. Hence the difference between
440 tcSimplifyCheck (which *does* allow implicit paramters to be inherited)
441 and tcSimplifyCheckBind (which does not).
443 What about when you supply a type signature for a binding?
444 Is it legal to give the following explicit, user type
445 signature to f, thus:
450 At first sight this seems reasonable, but it has the nasty property
451 that adding a type signature changes the dynamic semantics.
454 (let f x = (x::Int) + ?y
455 in (f 3, f 3 with ?y=5)) with ?y = 6
461 in (f 3, f 3 with ?y=5)) with ?y = 6
465 Indeed, simply inlining f (at the Haskell source level) would change the
468 Nevertheless, as Launchbury says (email Oct 01) we can't really give the
469 semantics for a Haskell program without knowing its typing, so if you
470 change the typing you may change the semantics.
472 To make things consistent in all cases where we are *checking* against
473 a supplied signature (as opposed to inferring a type), we adopt the
476 a signature does not need to quantify over implicit params.
478 [This represents a (rather marginal) change of policy since GHC 5.02,
479 which *required* an explicit signature to quantify over all implicit
480 params for the reasons mentioned above.]
482 But that raises a new question. Consider
484 Given (signature) ?x::Int
485 Wanted (inferred) ?x::Int, ?y::Bool
487 Clearly we want to discharge the ?x and float the ?y out. But
488 what is the criterion that distinguishes them? Clearly it isn't
489 what free type variables they have. The Right Thing seems to be
490 to float a constraint that
491 neither mentions any of the quantified type variables
492 nor any of the quantified implicit parameters
494 See the predicate isFreeWhenChecking.
497 Question 3: monomorphism
498 ~~~~~~~~~~~~~~~~~~~~~~~~
499 There's a nasty corner case when the monomorphism restriction bites:
503 The argument above suggests that we *must* generalise
504 over the ?y parameter, to get
505 z :: (?y::Int) => Int,
506 but the monomorphism restriction says that we *must not*, giving
508 Why does the momomorphism restriction say this? Because if you have
510 let z = x + ?y in z+z
512 you might not expect the addition to be done twice --- but it will if
513 we follow the argument of Question 2 and generalise over ?y.
516 Question 4: top level
517 ~~~~~~~~~~~~~~~~~~~~~
518 At the top level, monomorhism makes no sense at all.
521 main = let ?x = 5 in print foo
525 woggle :: (?x :: Int) => Int -> Int
528 We definitely don't want (foo :: Int) with a top-level implicit parameter
529 (?x::Int) becuase there is no way to bind it.
534 (A) Always generalise over implicit parameters
535 Bindings that fall under the monomorphism restriction can't
539 * Inlining remains valid
540 * No unexpected loss of sharing
541 * But simple bindings like
543 will be rejected, unless you add an explicit type signature
544 (to avoid the monomorphism restriction)
545 z :: (?y::Int) => Int
547 This seems unacceptable
549 (B) Monomorphism restriction "wins"
550 Bindings that fall under the monomorphism restriction can't
552 Always generalise over implicit parameters *except* for bindings
553 that fall under the monomorphism restriction
556 * Inlining isn't valid in general
557 * No unexpected loss of sharing
558 * Simple bindings like
560 accepted (get value of ?y from binding site)
562 (C) Always generalise over implicit parameters
563 Bindings that fall under the monomorphism restriction can't
564 be generalised, EXCEPT for implicit parameters
566 * Inlining remains valid
567 * Unexpected loss of sharing (from the extra generalisation)
568 * Simple bindings like
570 accepted (get value of ?y from occurrence sites)
575 None of these choices seems very satisfactory. But at least we should
576 decide which we want to do.
578 It's really not clear what is the Right Thing To Do. If you see
582 would you expect the value of ?y to be got from the *occurrence sites*
583 of 'z', or from the valuue of ?y at the *definition* of 'z'? In the
584 case of function definitions, the answer is clearly the former, but
585 less so in the case of non-fucntion definitions. On the other hand,
586 if we say that we get the value of ?y from the definition site of 'z',
587 then inlining 'z' might change the semantics of the program.
589 Choice (C) really says "the monomorphism restriction doesn't apply
590 to implicit parameters". Which is fine, but remember that every
591 innocent binding 'x = ...' that mentions an implicit parameter in
592 the RHS becomes a *function* of that parameter, called at each
593 use of 'x'. Now, the chances are that there are no intervening 'with'
594 clauses that bind ?y, so a decent compiler should common up all
595 those function calls. So I think I strongly favour (C). Indeed,
596 one could make a similar argument for abolishing the monomorphism
597 restriction altogether.
599 BOTTOM LINE: we choose (B) at present. See tcSimplifyRestricted
603 %************************************************************************
605 \subsection{tcSimplifyInfer}
607 %************************************************************************
609 tcSimplify is called when we *inferring* a type. Here's the overall game plan:
611 1. Compute Q = grow( fvs(T), C )
613 2. Partition C based on Q into Ct and Cq. Notice that ambiguous
614 predicates will end up in Ct; we deal with them at the top level
616 3. Try improvement, using functional dependencies
618 4. If Step 3 did any unification, repeat from step 1
619 (Unification can change the result of 'grow'.)
621 Note: we don't reduce dictionaries in step 2. For example, if we have
622 Eq (a,b), we don't simplify to (Eq a, Eq b). So Q won't be different
623 after step 2. However note that we may therefore quantify over more
624 type variables than we absolutely have to.
626 For the guts, we need a loop, that alternates context reduction and
627 improvement with unification. E.g. Suppose we have
629 class C x y | x->y where ...
631 and tcSimplify is called with:
633 Then improvement unifies a with b, giving
636 If we need to unify anything, we rattle round the whole thing all over
643 -> TcTyVarSet -- fv(T); type vars
645 -> TcM ([TcTyVar], -- Tyvars to quantify (zonked and quantified)
646 [Inst], -- Dict Ids that must be bound here (zonked)
647 TcDictBinds) -- Bindings
648 -- Any free (escaping) Insts are tossed into the environment
653 tcSimplifyInfer doc tau_tvs wanted
654 = do { tau_tvs1 <- zonkTcTyVarsAndFV (varSetElems tau_tvs)
655 ; wanted' <- mapM zonkInst wanted -- Zonk before deciding quantified tyvars
656 ; gbl_tvs <- tcGetGlobalTyVars
657 ; let preds1 = fdPredsOfInsts wanted'
658 gbl_tvs1 = oclose preds1 gbl_tvs
659 qtvs = grow preds1 tau_tvs1 `minusVarSet` gbl_tvs1
660 -- See Note [Choosing which variables to quantify]
662 -- To maximise sharing, remove from consideration any
663 -- constraints that don't mention qtvs at all
664 ; let (free, bound) = partition (isFreeWhenInferring qtvs) wanted'
667 -- To make types simple, reduce as much as possible
668 ; traceTc (text "infer" <+> (ppr preds1 $$ ppr (grow preds1 tau_tvs1) $$ ppr gbl_tvs $$
669 ppr gbl_tvs1 $$ ppr free $$ ppr bound))
670 ; (irreds1, binds1) <- tryHardCheckLoop doc bound
672 -- Note [Inference and implication constraints]
673 ; let want_dict d = tyVarsOfInst d `intersectsVarSet` qtvs
674 ; (irreds2, binds2) <- approximateImplications doc want_dict irreds1
676 -- Now work out all over again which type variables to quantify,
677 -- exactly in the same way as before, but starting from irreds2. Why?
678 -- a) By now improvment may have taken place, and we must *not*
679 -- quantify over any variable free in the environment
680 -- tc137 (function h inside g) is an example
682 -- b) Do not quantify over constraints that *now* do not
683 -- mention quantified type variables, because they are
684 -- simply ambiguous (or might be bound further out). Example:
685 -- f :: Eq b => a -> (a, b)
687 -- From the RHS of g we get the MethodInst f77 :: alpha -> (alpha, beta)
688 -- We decide to quantify over 'alpha' alone, but free1 does not include f77
689 -- because f77 mentions 'alpha'. Then reducing leaves only the (ambiguous)
690 -- constraint (Eq beta), which we dump back into the free set
691 -- See test tcfail181
693 -- c) irreds may contain type variables not previously mentioned,
694 -- e.g. instance D a x => Foo [a]
696 -- Then after simplifying we'll get (D a x), and x is fresh
697 -- We must quantify over x else it'll be totally unbound
698 ; tau_tvs2 <- zonkTcTyVarsAndFV (varSetElems tau_tvs1)
699 ; gbl_tvs2 <- zonkTcTyVarsAndFV (varSetElems gbl_tvs1)
700 -- Note that we start from gbl_tvs1
701 -- We use tcGetGlobalTyVars, then oclose wrt preds2, because
702 -- we've already put some of the original preds1 into frees
703 -- E.g. wanteds = C a b (where a->b)
706 -- Then b is fixed by gbl_tvs, so (C a b) will be in free, and
707 -- irreds2 will be empty. But we don't want to generalise over b!
708 ; let preds2 = fdPredsOfInsts irreds2 -- irreds2 is zonked
709 qtvs = grow preds2 tau_tvs2 `minusVarSet` oclose preds2 gbl_tvs2
710 ; let (free, irreds3) = partition (isFreeWhenInferring qtvs) irreds2
713 -- Turn the quantified meta-type variables into real type variables
714 ; qtvs2 <- zonkQuantifiedTyVars (varSetElems qtvs)
716 -- We can't abstract over any remaining unsolved
717 -- implications so instead just float them outwards. Ugh.
718 ; let (q_dicts0, implics) = partition isAbstractableInst irreds3
719 ; loc <- getInstLoc (ImplicOrigin doc)
720 ; implic_bind <- bindIrreds loc qtvs2 q_dicts0 implics
722 -- Prepare equality instances for quantification
723 ; let (q_eqs0,q_dicts) = partition isEqInst q_dicts0
724 ; q_eqs <- mapM finalizeEqInst q_eqs0
726 ; return (qtvs2, q_eqs ++ q_dicts, binds1 `unionBags` binds2 `unionBags` implic_bind) }
727 -- NB: when we are done, we might have some bindings, but
728 -- the final qtvs might be empty. See Note [NO TYVARS] below.
730 approximateImplications :: SDoc -> (Inst -> Bool) -> [Inst] -> TcM ([Inst], TcDictBinds)
731 -- Note [Inference and implication constraints]
732 -- Given a bunch of Dict and ImplicInsts, try to approximate the implications by
733 -- - fetching any dicts inside them that are free
734 -- - using those dicts as cruder constraints, to solve the implications
735 -- - returning the extra ones too
737 approximateImplications doc want_dict irreds
739 = return (irreds, emptyBag)
741 = do { extra_dicts' <- mapM cloneDict extra_dicts
742 ; tryHardCheckLoop doc (extra_dicts' ++ irreds) }
743 -- By adding extra_dicts', we make them
744 -- available to solve the implication constraints
746 extra_dicts = get_dicts (filter isImplicInst irreds)
748 get_dicts :: [Inst] -> [Inst] -- Returns only Dicts
749 -- Find the wanted constraints in implication constraints that satisfy
750 -- want_dict, and are not bound by forall's in the constraint itself
751 get_dicts ds = concatMap get_dict ds
753 get_dict d@(Dict {}) | want_dict d = [d]
755 get_dict (ImplicInst {tci_tyvars = tvs, tci_wanted = wanteds})
756 = [ d | let tv_set = mkVarSet tvs
757 , d <- get_dicts wanteds
758 , not (tyVarsOfInst d `intersectsVarSet` tv_set)]
759 get_dict i@(EqInst {}) | want_dict i = [i]
761 get_dict other = pprPanic "approximateImplications" (ppr other)
764 Note [Inference and implication constraints]
765 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
766 Suppose we have a wanted implication constraint (perhaps arising from
767 a nested pattern match) like
769 and we are now trying to quantify over 'a' when inferring the type for
770 a function. In principle it's possible that there might be an instance
771 instance (C a, E a) => D [a]
772 so the context (E a) would suffice. The Right Thing is to abstract over
773 the implication constraint, but we don't do that (a) because it'll be
774 surprising to programmers and (b) because we don't have the machinery to deal
775 with 'given' implications.
777 So our best approximation is to make (D [a]) part of the inferred
778 context, so we can use that to discharge the implication. Hence
779 the strange function get_dicts in approximateImplications.
781 The common cases are more clear-cut, when we have things like
783 Here, abstracting over (C b) is not an approximation at all -- but see
784 Note [Freeness and implications].
786 See Trac #1430 and test tc228.
790 -----------------------------------------------------------
791 -- tcSimplifyInferCheck is used when we know the constraints we are to simplify
792 -- against, but we don't know the type variables over which we are going to quantify.
793 -- This happens when we have a type signature for a mutually recursive group
796 -> TcTyVarSet -- fv(T)
799 -> TcM ([TyVar], -- Fully zonked, and quantified
800 TcDictBinds) -- Bindings
802 tcSimplifyInferCheck loc tau_tvs givens wanteds
803 = do { traceTc (text "tcSimplifyInferCheck <-" <+> ppr wanteds)
804 ; (irreds, binds) <- gentleCheckLoop loc givens wanteds
806 -- Figure out which type variables to quantify over
807 -- You might think it should just be the signature tyvars,
808 -- but in bizarre cases you can get extra ones
809 -- f :: forall a. Num a => a -> a
810 -- f x = fst (g (x, head [])) + 1
812 -- Here we infer g :: forall a b. a -> b -> (b,a)
813 -- We don't want g to be monomorphic in b just because
814 -- f isn't quantified over b.
815 ; let all_tvs = varSetElems (tau_tvs `unionVarSet` tyVarsOfInsts givens)
816 ; all_tvs <- zonkTcTyVarsAndFV all_tvs
817 ; gbl_tvs <- tcGetGlobalTyVars
818 ; let qtvs = varSetElems (all_tvs `minusVarSet` gbl_tvs)
819 -- We could close gbl_tvs, but its not necessary for
820 -- soundness, and it'll only affect which tyvars, not which
821 -- dictionaries, we quantify over
823 ; qtvs' <- zonkQuantifiedTyVars qtvs
825 -- Now we are back to normal (c.f. tcSimplCheck)
826 ; implic_bind <- bindIrreds loc qtvs' givens irreds
828 ; traceTc (text "tcSimplifyInferCheck ->" <+> ppr (implic_bind))
829 ; return (qtvs', binds `unionBags` implic_bind) }
832 Note [Squashing methods]
833 ~~~~~~~~~~~~~~~~~~~~~~~~~
834 Be careful if you want to float methods more:
835 truncate :: forall a. RealFrac a => forall b. Integral b => a -> b
836 From an application (truncate f i) we get
839 If we have also have a second occurrence of truncate, we get
842 When simplifying with i,f free, we might still notice that
843 t1=t3; but alas, the binding for t2 (which mentions t1)
844 may continue to float out!
849 class Y a b | a -> b where
852 instance Y [[a]] a where
855 k :: X a -> X a -> X a
857 g :: Num a => [X a] -> [X a]
860 h ys = ys ++ map (k (y [[0]])) xs
862 The excitement comes when simplifying the bindings for h. Initially
863 try to simplify {y @ [[t1]] t2, 0 @ t1}, with initial qtvs = {t2}.
864 From this we get t1:=:t2, but also various bindings. We can't forget
865 the bindings (because of [LOOP]), but in fact t1 is what g is
868 The net effect of [NO TYVARS]
871 isFreeWhenInferring :: TyVarSet -> Inst -> Bool
872 isFreeWhenInferring qtvs inst
873 = isFreeWrtTyVars qtvs inst -- Constrains no quantified vars
874 && isInheritableInst inst -- and no implicit parameter involved
875 -- see Note [Inheriting implicit parameters]
877 {- No longer used (with implication constraints)
878 isFreeWhenChecking :: TyVarSet -- Quantified tyvars
879 -> NameSet -- Quantified implicit parameters
881 isFreeWhenChecking qtvs ips inst
882 = isFreeWrtTyVars qtvs inst
883 && isFreeWrtIPs ips inst
886 isFreeWrtTyVars :: VarSet -> Inst -> Bool
887 isFreeWrtTyVars qtvs inst = tyVarsOfInst inst `disjointVarSet` qtvs
888 isFreeWrtIPs :: NameSet -> Inst -> Bool
889 isFreeWrtIPs ips inst = not (any (`elemNameSet` ips) (ipNamesOfInst inst))
893 %************************************************************************
895 \subsection{tcSimplifyCheck}
897 %************************************************************************
899 @tcSimplifyCheck@ is used when we know exactly the set of variables
900 we are going to quantify over. For example, a class or instance declaration.
903 -----------------------------------------------------------
904 -- tcSimplifyCheck is used when checking expression type signatures,
905 -- class decls, instance decls etc.
906 tcSimplifyCheck :: InstLoc
907 -> [TcTyVar] -- Quantify over these
910 -> TcM TcDictBinds -- Bindings
911 tcSimplifyCheck loc qtvs givens wanteds
912 = ASSERT( all isTcTyVar qtvs && all isSkolemTyVar qtvs )
913 do { traceTc (text "tcSimplifyCheck")
914 ; (irreds, binds) <- gentleCheckLoop loc givens wanteds
915 ; implic_bind <- bindIrreds loc qtvs givens irreds
916 ; return (binds `unionBags` implic_bind) }
918 -----------------------------------------------------------
919 -- tcSimplifyCheckPat is used for existential pattern match
920 tcSimplifyCheckPat :: InstLoc
921 -> [TcTyVar] -- Quantify over these
924 -> TcM TcDictBinds -- Bindings
925 tcSimplifyCheckPat loc qtvs givens wanteds
926 = ASSERT( all isTcTyVar qtvs && all isSkolemTyVar qtvs )
927 do { traceTc (text "tcSimplifyCheckPat")
928 ; (irreds, binds) <- gentleCheckLoop loc givens wanteds
929 ; implic_bind <- bindIrredsR loc qtvs givens irreds
930 ; return (binds `unionBags` implic_bind) }
932 -----------------------------------------------------------
933 bindIrreds :: InstLoc -> [TcTyVar]
936 bindIrreds loc qtvs givens irreds
937 = bindIrredsR loc qtvs givens irreds
939 bindIrredsR :: InstLoc -> [TcTyVar] -> [Inst] -> [Inst] -> TcM TcDictBinds
940 -- Make a binding that binds 'irreds', by generating an implication
941 -- constraint for them, *and* throwing the constraint into the LIE
942 bindIrredsR loc qtvs givens irreds
946 = do { let givens' = filter isAbstractableInst givens
947 -- The givens can (redundantly) include methods
948 -- We want to retain both EqInsts and Dicts
949 -- There should be no implicadtion constraints
950 -- See Note [Pruning the givens in an implication constraint]
952 -- If there are no 'givens', then it's safe to
953 -- partition the 'wanteds' by their qtvs, thereby trimming irreds
954 -- See Note [Freeness and implications]
955 ; irreds' <- if null givens'
957 { let qtv_set = mkVarSet qtvs
958 (frees, real_irreds) = partition (isFreeWrtTyVars qtv_set) irreds
960 ; return real_irreds }
963 ; (implics, bind) <- makeImplicationBind loc qtvs givens' irreds'
964 -- This call does the real work
965 -- If irreds' is empty, it does something sensible
970 makeImplicationBind :: InstLoc -> [TcTyVar]
972 -> TcM ([Inst], TcDictBinds)
973 -- Make a binding that binds 'irreds', by generating an implication
974 -- constraint for them, *and* throwing the constraint into the LIE
975 -- The binding looks like
976 -- (ir1, .., irn) = f qtvs givens
977 -- where f is (evidence for) the new implication constraint
978 -- f :: forall qtvs. {reft} givens => (ir1, .., irn)
979 -- qtvs includes coercion variables
981 -- This binding must line up the 'rhs' in reduceImplication
982 makeImplicationBind loc all_tvs
983 givens -- Guaranteed all Dicts
986 | null irreds -- If there are no irreds, we are done
987 = return ([], emptyBag)
988 | otherwise -- Otherwise we must generate a binding
989 = do { uniq <- newUnique
990 ; span <- getSrcSpanM
991 ; let (eq_givens, dict_givens) = partition isEqInst givens
992 eq_tyvar_cos = mkTyVarTys (varSetElems $ tyVarsOfTypes $ map eqInstType eq_givens)
993 -- Urgh! See line 2187 or thereabouts. I believe that all these
994 -- 'givens' must be a simple CoVar. This MUST be cleaned up.
996 ; let name = mkInternalName uniq (mkVarOcc "ic") span
997 implic_inst = ImplicInst { tci_name = name,
998 tci_tyvars = all_tvs,
999 tci_given = (eq_givens ++ dict_givens),
1000 tci_wanted = irreds, tci_loc = loc }
1001 ; let -- only create binder for dict_irreds
1002 (_, dict_irreds) = partition isEqInst irreds
1003 dict_irred_ids = map instToId dict_irreds
1004 lpat = mkBigLHsPatTup (map (L span . VarPat) dict_irred_ids)
1005 rhs = L span (mkHsWrap co (HsVar (instToId implic_inst)))
1006 co = mkWpApps (map instToId dict_givens)
1007 <.> mkWpTyApps eq_tyvar_cos
1008 <.> mkWpTyApps (mkTyVarTys all_tvs)
1009 bind | [dict_irred_id] <- dict_irred_ids = VarBind dict_irred_id rhs
1010 | otherwise = PatBind { pat_lhs = lpat,
1011 pat_rhs = unguardedGRHSs rhs,
1012 pat_rhs_ty = hsLPatType lpat,
1013 bind_fvs = placeHolderNames }
1014 ; traceTc $ text "makeImplicationBind" <+> ppr implic_inst
1015 ; return ([implic_inst], unitBag (L span bind))
1018 -----------------------------------------------------------
1019 tryHardCheckLoop :: SDoc
1021 -> TcM ([Inst], TcDictBinds)
1023 tryHardCheckLoop doc wanteds
1024 = do { (irreds,binds) <- checkLoop (mkRedEnv doc try_me []) wanteds
1025 ; return (irreds,binds)
1028 try_me _ = ReduceMe AddSCs
1029 -- Here's the try-hard bit
1031 -----------------------------------------------------------
1032 gentleCheckLoop :: InstLoc
1035 -> TcM ([Inst], TcDictBinds)
1037 gentleCheckLoop inst_loc givens wanteds
1038 = do { (irreds,binds) <- checkLoop env wanteds
1039 ; return (irreds,binds)
1042 env = mkRedEnv (pprInstLoc inst_loc) try_me givens
1044 try_me inst | isMethodOrLit inst = ReduceMe AddSCs
1046 -- When checking against a given signature
1047 -- we MUST be very gentle: Note [Check gently]
1049 gentleInferLoop :: SDoc -> [Inst]
1050 -> TcM ([Inst], TcDictBinds)
1051 gentleInferLoop doc wanteds
1052 = do { (irreds, binds) <- checkLoop env wanteds
1053 ; return (irreds, binds) }
1055 env = mkRedEnv doc try_me []
1056 try_me inst | isMethodOrLit inst = ReduceMe AddSCs
1061 ~~~~~~~~~~~~~~~~~~~~
1062 We have to very careful about not simplifying too vigorously
1067 f :: Show b => T b -> b
1068 f (MkT x) = show [x]
1070 Inside the pattern match, which binds (a:*, x:a), we know that
1072 Hence we have a dictionary for Show [a] available; and indeed we
1073 need it. We are going to build an implication contraint
1074 forall a. (b~[a]) => Show [a]
1075 Later, we will solve this constraint using the knowledge (Show b)
1077 But we MUST NOT reduce (Show [a]) to (Show a), else the whole
1078 thing becomes insoluble. So we simplify gently (get rid of literals
1079 and methods only, plus common up equal things), deferring the real
1080 work until top level, when we solve the implication constraint
1081 with tryHardCheckLooop.
1085 -----------------------------------------------------------
1088 -> TcM ([Inst], TcDictBinds)
1089 -- Precondition: givens are completely rigid
1090 -- Postcondition: returned Insts are zonked
1092 checkLoop env wanteds
1093 = go env wanteds (return ())
1094 where go env wanteds elim_skolems
1095 = do { -- We do need to zonk the givens; cf Note [Zonking RedEnv]
1096 ; env' <- zonkRedEnv env
1097 ; wanteds' <- zonkInsts wanteds
1099 ; (improved, binds, irreds, elim_more_skolems)
1100 <- reduceContext env' wanteds'
1101 ; let elim_skolems' = elim_skolems >> elim_more_skolems
1103 ; if not improved then
1104 elim_skolems' >> return (irreds, binds)
1107 -- If improvement did some unification, we go round again.
1108 -- We start again with irreds, not wanteds
1109 -- Using an instance decl might have introduced a fresh type
1110 -- variable which might have been unified, so we'd get an
1111 -- infinite loop if we started again with wanteds!
1113 { (irreds1, binds1) <- go env' irreds elim_skolems'
1114 ; return (irreds1, binds `unionBags` binds1) } }
1117 Note [Zonking RedEnv]
1118 ~~~~~~~~~~~~~~~~~~~~~
1119 It might appear as if the givens in RedEnv are always rigid, but that is not
1120 necessarily the case for programs involving higher-rank types that have class
1121 contexts constraining the higher-rank variables. An example from tc237 in the
1124 class Modular s a | s -> a
1126 wim :: forall a w. Integral a
1127 => a -> (forall s. Modular s a => M s w) -> w
1128 wim i k = error "urk"
1130 test5 :: (Modular s a, Integral a) => M s a
1133 test4 = wim 4 test4'
1135 Notice how the variable 'a' of (Modular s a) in the rank-2 type of wim is
1136 quantified further outside. When type checking test4, we have to check
1137 whether the signature of test5 is an instance of
1139 (forall s. Modular s a => M s w)
1141 Consequently, we will get (Modular s t_a), where t_a is a TauTv into the
1144 Given the FD of Modular in this example, class improvement will instantiate
1145 t_a to 'a', where 'a' is the skolem from test5's signatures (due to the
1146 Modular s a predicate in that signature). If we don't zonk (Modular s t_a) in
1147 the givens, we will get into a loop as improveOne uses the unification engine
1148 Unify.tcUnifyTys, which doesn't know about mutable type variables.
1153 class If b t e r | b t e -> r
1156 class Lte a b c | a b -> c where lte :: a -> b -> c
1158 instance (Lte a b l,If l b a c) => Max a b c
1160 Wanted: Max Z (S x) y
1162 Then we'll reduce using the Max instance to:
1163 (Lte Z (S x) l, If l (S x) Z y)
1164 and improve by binding l->T, after which we can do some reduction
1165 on both the Lte and If constraints. What we *can't* do is start again
1166 with (Max Z (S x) y)!
1170 %************************************************************************
1172 tcSimplifySuperClasses
1174 %************************************************************************
1176 Note [SUPERCLASS-LOOP 1]
1177 ~~~~~~~~~~~~~~~~~~~~~~~~
1178 We have to be very, very careful when generating superclasses, lest we
1179 accidentally build a loop. Here's an example:
1183 class S a => C a where { opc :: a -> a }
1184 class S b => D b where { opd :: b -> b }
1186 instance C Int where
1189 instance D Int where
1192 From (instance C Int) we get the constraint set {ds1:S Int, dd:D Int}
1193 Simplifying, we may well get:
1194 $dfCInt = :C ds1 (opd dd)
1197 Notice that we spot that we can extract ds1 from dd.
1199 Alas! Alack! We can do the same for (instance D Int):
1201 $dfDInt = :D ds2 (opc dc)
1205 And now we've defined the superclass in terms of itself.
1207 Solution: never generate a superclass selectors at all when
1208 satisfying the superclass context of an instance declaration.
1210 Two more nasty cases are in
1215 tcSimplifySuperClasses
1220 tcSimplifySuperClasses loc givens sc_wanteds
1221 = do { traceTc (text "tcSimplifySuperClasses")
1222 ; (irreds,binds1) <- checkLoop env sc_wanteds
1223 ; let (tidy_env, tidy_irreds) = tidyInsts irreds
1224 ; reportNoInstances tidy_env (Just (loc, givens)) tidy_irreds
1227 env = mkRedEnv (pprInstLoc loc) try_me givens
1228 try_me _ = ReduceMe NoSCs
1229 -- Like tryHardCheckLoop, but with NoSCs
1233 %************************************************************************
1235 \subsection{tcSimplifyRestricted}
1237 %************************************************************************
1239 tcSimplifyRestricted infers which type variables to quantify for a
1240 group of restricted bindings. This isn't trivial.
1243 We want to quantify over a to get id :: forall a. a->a
1246 We do not want to quantify over a, because there's an Eq a
1247 constraint, so we get eq :: a->a->Bool (notice no forall)
1250 RHS has type 'tau', whose free tyvars are tau_tvs
1251 RHS has constraints 'wanteds'
1254 Quantify over (tau_tvs \ ftvs(wanteds))
1255 This is bad. The constraints may contain (Monad (ST s))
1256 where we have instance Monad (ST s) where...
1257 so there's no need to be monomorphic in s!
1259 Also the constraint might be a method constraint,
1260 whose type mentions a perfectly innocent tyvar:
1261 op :: Num a => a -> b -> a
1262 Here, b is unconstrained. A good example would be
1264 We want to infer the polymorphic type
1265 foo :: forall b. b -> b
1268 Plan B (cunning, used for a long time up to and including GHC 6.2)
1269 Step 1: Simplify the constraints as much as possible (to deal
1270 with Plan A's problem). Then set
1271 qtvs = tau_tvs \ ftvs( simplify( wanteds ) )
1273 Step 2: Now simplify again, treating the constraint as 'free' if
1274 it does not mention qtvs, and trying to reduce it otherwise.
1275 The reasons for this is to maximise sharing.
1277 This fails for a very subtle reason. Suppose that in the Step 2
1278 a constraint (Foo (Succ Zero) (Succ Zero) b) gets thrown upstairs as 'free'.
1279 In the Step 1 this constraint might have been simplified, perhaps to
1280 (Foo Zero Zero b), AND THEN THAT MIGHT BE IMPROVED, to bind 'b' to 'T'.
1281 This won't happen in Step 2... but that in turn might prevent some other
1282 constraint (Baz [a] b) being simplified (e.g. via instance Baz [a] T where {..})
1283 and that in turn breaks the invariant that no constraints are quantified over.
1285 Test typecheck/should_compile/tc177 (which failed in GHC 6.2) demonstrates
1290 Step 1: Simplify the constraints as much as possible (to deal
1291 with Plan A's problem). Then set
1292 qtvs = tau_tvs \ ftvs( simplify( wanteds ) )
1293 Return the bindings from Step 1.
1296 A note about Plan C (arising from "bug" reported by George Russel March 2004)
1299 instance (HasBinary ty IO) => HasCodedValue ty
1301 foo :: HasCodedValue a => String -> IO a
1303 doDecodeIO :: HasCodedValue a => () -> () -> IO a
1304 doDecodeIO codedValue view
1305 = let { act = foo "foo" } in act
1307 You might think this should work becuase the call to foo gives rise to a constraint
1308 (HasCodedValue t), which can be satisfied by the type sig for doDecodeIO. But the
1309 restricted binding act = ... calls tcSimplifyRestricted, and PlanC simplifies the
1310 constraint using the (rather bogus) instance declaration, and now we are stuffed.
1312 I claim this is not really a bug -- but it bit Sergey as well as George. So here's
1316 Plan D (a variant of plan B)
1317 Step 1: Simplify the constraints as much as possible (to deal
1318 with Plan A's problem), BUT DO NO IMPROVEMENT. Then set
1319 qtvs = tau_tvs \ ftvs( simplify( wanteds ) )
1321 Step 2: Now simplify again, treating the constraint as 'free' if
1322 it does not mention qtvs, and trying to reduce it otherwise.
1324 The point here is that it's generally OK to have too few qtvs; that is,
1325 to make the thing more monomorphic than it could be. We don't want to
1326 do that in the common cases, but in wierd cases it's ok: the programmer
1327 can always add a signature.
1329 Too few qtvs => too many wanteds, which is what happens if you do less
1334 tcSimplifyRestricted -- Used for restricted binding groups
1335 -- i.e. ones subject to the monomorphism restriction
1338 -> [Name] -- Things bound in this group
1339 -> TcTyVarSet -- Free in the type of the RHSs
1340 -> [Inst] -- Free in the RHSs
1341 -> TcM ([TyVar], -- Tyvars to quantify (zonked and quantified)
1342 TcDictBinds) -- Bindings
1343 -- tcSimpifyRestricted returns no constraints to
1344 -- quantify over; by definition there are none.
1345 -- They are all thrown back in the LIE
1347 tcSimplifyRestricted doc top_lvl bndrs tau_tvs wanteds
1348 -- Zonk everything in sight
1349 = do { traceTc (text "tcSimplifyRestricted")
1350 ; wanteds' <- zonkInsts wanteds
1352 -- 'ReduceMe': Reduce as far as we can. Don't stop at
1353 -- dicts; the idea is to get rid of as many type
1354 -- variables as possible, and we don't want to stop
1355 -- at (say) Monad (ST s), because that reduces
1356 -- immediately, with no constraint on s.
1358 -- BUT do no improvement! See Plan D above
1359 -- HOWEVER, some unification may take place, if we instantiate
1360 -- a method Inst with an equality constraint
1361 ; let env = mkNoImproveRedEnv doc (\_ -> ReduceMe AddSCs)
1362 ; (_imp, _binds, constrained_dicts, elim_skolems)
1363 <- reduceContext env wanteds'
1366 -- Next, figure out the tyvars we will quantify over
1367 ; tau_tvs' <- zonkTcTyVarsAndFV (varSetElems tau_tvs)
1368 ; gbl_tvs' <- tcGetGlobalTyVars
1369 ; constrained_dicts' <- zonkInsts constrained_dicts
1371 ; let qtvs1 = tau_tvs' `minusVarSet` oclose (fdPredsOfInsts constrained_dicts) gbl_tvs'
1372 -- As in tcSimplifyInfer
1374 -- Do not quantify over constrained type variables:
1375 -- this is the monomorphism restriction
1376 constrained_tvs' = tyVarsOfInsts constrained_dicts'
1377 qtvs = qtvs1 `minusVarSet` constrained_tvs'
1378 pp_bndrs = pprWithCommas (quotes . ppr) bndrs
1381 ; warn_mono <- doptM Opt_WarnMonomorphism
1382 ; warnTc (warn_mono && (constrained_tvs' `intersectsVarSet` qtvs1))
1383 (vcat[ ptext (sLit "the Monomorphism Restriction applies to the binding")
1384 <> plural bndrs <+> ptext (sLit "for") <+> pp_bndrs,
1385 ptext (sLit "Consider giving a type signature for") <+> pp_bndrs])
1387 ; traceTc (text "tcSimplifyRestricted" <+> vcat [
1388 pprInsts wanteds, pprInsts constrained_dicts',
1390 ppr constrained_tvs', ppr tau_tvs', ppr qtvs ])
1392 -- The first step may have squashed more methods than
1393 -- necessary, so try again, this time more gently, knowing the exact
1394 -- set of type variables to quantify over.
1396 -- We quantify only over constraints that are captured by qtvs;
1397 -- these will just be a subset of non-dicts. This in contrast
1398 -- to normal inference (using isFreeWhenInferring) in which we quantify over
1399 -- all *non-inheritable* constraints too. This implements choice
1400 -- (B) under "implicit parameter and monomorphism" above.
1402 -- Remember that we may need to do *some* simplification, to
1403 -- (for example) squash {Monad (ST s)} into {}. It's not enough
1404 -- just to float all constraints
1406 -- At top level, we *do* squash methods becuase we want to
1407 -- expose implicit parameters to the test that follows
1408 ; let is_nested_group = isNotTopLevel top_lvl
1409 try_me inst | isFreeWrtTyVars qtvs inst,
1410 (is_nested_group || isDict inst) = Stop
1411 | otherwise = ReduceMe AddSCs
1412 env = mkNoImproveRedEnv doc try_me
1413 ; (_imp, binds, irreds, elim_skolems) <- reduceContext env wanteds'
1416 -- See "Notes on implicit parameters, Question 4: top level"
1417 ; ASSERT( all (isFreeWrtTyVars qtvs) irreds ) -- None should be captured
1418 if is_nested_group then
1420 else do { let (bad_ips, non_ips) = partition isIPDict irreds
1421 ; addTopIPErrs bndrs bad_ips
1422 ; extendLIEs non_ips }
1424 ; qtvs' <- zonkQuantifiedTyVars (varSetElems qtvs)
1425 ; return (qtvs', binds) }
1429 %************************************************************************
1433 %************************************************************************
1435 On the LHS of transformation rules we only simplify methods and constants,
1436 getting dictionaries. We want to keep all of them unsimplified, to serve
1437 as the available stuff for the RHS of the rule.
1439 Example. Consider the following left-hand side of a rule
1441 f (x == y) (y > z) = ...
1443 If we typecheck this expression we get constraints
1445 d1 :: Ord a, d2 :: Eq a
1447 We do NOT want to "simplify" to the LHS
1449 forall x::a, y::a, z::a, d1::Ord a.
1450 f ((==) (eqFromOrd d1) x y) ((>) d1 y z) = ...
1454 forall x::a, y::a, z::a, d1::Ord a, d2::Eq a.
1455 f ((==) d2 x y) ((>) d1 y z) = ...
1457 Here is another example:
1459 fromIntegral :: (Integral a, Num b) => a -> b
1460 {-# RULES "foo" fromIntegral = id :: Int -> Int #-}
1462 In the rule, a=b=Int, and Num Int is a superclass of Integral Int. But
1463 we *dont* want to get
1465 forall dIntegralInt.
1466 fromIntegral Int Int dIntegralInt (scsel dIntegralInt) = id Int
1468 because the scsel will mess up RULE matching. Instead we want
1470 forall dIntegralInt, dNumInt.
1471 fromIntegral Int Int dIntegralInt dNumInt = id Int
1475 g (x == y) (y == z) = ..
1477 where the two dictionaries are *identical*, we do NOT WANT
1479 forall x::a, y::a, z::a, d1::Eq a
1480 f ((==) d1 x y) ((>) d1 y z) = ...
1482 because that will only match if the dict args are (visibly) equal.
1483 Instead we want to quantify over the dictionaries separately.
1485 In short, tcSimplifyRuleLhs must *only* squash LitInst and MethInts, leaving
1486 all dicts unchanged, with absolutely no sharing. It's simpler to do this
1487 from scratch, rather than further parameterise simpleReduceLoop etc
1490 tcSimplifyRuleLhs :: [Inst] -> TcM ([Inst], TcDictBinds)
1491 tcSimplifyRuleLhs wanteds
1492 = go [] emptyBag wanteds
1495 = return (dicts, binds)
1496 go dicts binds (w:ws)
1498 = go (w:dicts) binds ws
1500 = do { w' <- zonkInst w -- So that (3::Int) does not generate a call
1501 -- to fromInteger; this looks fragile to me
1502 ; lookup_result <- lookupSimpleInst w'
1503 ; case lookup_result of
1505 go dicts (addInstToDictBind binds w rhs) (ws' ++ ws)
1506 NoInstance -> pprPanic "tcSimplifyRuleLhs" (ppr w)
1510 tcSimplifyBracket is used when simplifying the constraints arising from
1511 a Template Haskell bracket [| ... |]. We want to check that there aren't
1512 any constraints that can't be satisfied (e.g. Show Foo, where Foo has no
1513 Show instance), but we aren't otherwise interested in the results.
1514 Nor do we care about ambiguous dictionaries etc. We will type check
1515 this bracket again at its usage site.
1518 tcSimplifyBracket :: [Inst] -> TcM ()
1519 tcSimplifyBracket wanteds
1520 = do { tryHardCheckLoop doc wanteds
1523 doc = text "tcSimplifyBracket"
1527 %************************************************************************
1529 \subsection{Filtering at a dynamic binding}
1531 %************************************************************************
1536 we must discharge all the ?x constraints from B. We also do an improvement
1537 step; if we have ?x::t1 and ?x::t2 we must unify t1, t2.
1539 Actually, the constraints from B might improve the types in ?x. For example
1541 f :: (?x::Int) => Char -> Char
1544 then the constraint (?x::Int) arising from the call to f will
1545 force the binding for ?x to be of type Int.
1548 tcSimplifyIPs :: [Inst] -- The implicit parameters bound here
1551 -- We need a loop so that we do improvement, and then
1552 -- (next time round) generate a binding to connect the two
1554 -- Here the two ?x's have different types, and improvement
1555 -- makes them the same.
1557 tcSimplifyIPs given_ips wanteds
1558 = do { wanteds' <- zonkInsts wanteds
1559 ; given_ips' <- zonkInsts given_ips
1560 -- Unusually for checking, we *must* zonk the given_ips
1562 ; let env = mkRedEnv doc try_me given_ips'
1563 ; (improved, binds, irreds, elim_skolems) <- reduceContext env wanteds'
1566 ; if not improved then
1567 ASSERT( all is_free irreds )
1568 do { extendLIEs irreds
1571 tcSimplifyIPs given_ips wanteds }
1573 doc = text "tcSimplifyIPs" <+> ppr given_ips
1574 ip_set = mkNameSet (ipNamesOfInsts given_ips)
1575 is_free inst = isFreeWrtIPs ip_set inst
1577 -- Simplify any methods that mention the implicit parameter
1578 try_me inst | is_free inst = Stop
1579 | otherwise = ReduceMe NoSCs
1583 %************************************************************************
1585 \subsection[binds-for-local-funs]{@bindInstsOfLocalFuns@}
1587 %************************************************************************
1589 When doing a binding group, we may have @Insts@ of local functions.
1590 For example, we might have...
1592 let f x = x + 1 -- orig local function (overloaded)
1593 f.1 = f Int -- two instances of f
1598 The point is: we must drop the bindings for @f.1@ and @f.2@ here,
1599 where @f@ is in scope; those @Insts@ must certainly not be passed
1600 upwards towards the top-level. If the @Insts@ were binding-ified up
1601 there, they would have unresolvable references to @f@.
1603 We pass in an @init_lie@ of @Insts@ and a list of locally-bound @Ids@.
1604 For each method @Inst@ in the @init_lie@ that mentions one of the
1605 @Ids@, we create a binding. We return the remaining @Insts@ (in an
1606 @LIE@), as well as the @HsBinds@ generated.
1609 bindInstsOfLocalFuns :: [Inst] -> [TcId] -> TcM TcDictBinds
1610 -- Simlifies only MethodInsts, and generate only bindings of form
1612 -- We're careful not to even generate bindings of the form
1614 -- You'd think that'd be fine, but it interacts with what is
1615 -- arguably a bug in Match.tidyEqnInfo (see notes there)
1617 bindInstsOfLocalFuns wanteds local_ids
1618 | null overloaded_ids = do
1621 return emptyLHsBinds
1624 = do { (irreds, binds) <- gentleInferLoop doc for_me
1625 ; extendLIEs not_for_me
1629 doc = text "bindInsts" <+> ppr local_ids
1630 overloaded_ids = filter is_overloaded local_ids
1631 is_overloaded id = isOverloadedTy (idType id)
1632 (for_me, not_for_me) = partition (isMethodFor overloaded_set) wanteds
1634 overloaded_set = mkVarSet overloaded_ids -- There can occasionally be a lot of them
1635 -- so it's worth building a set, so that
1636 -- lookup (in isMethodFor) is faster
1640 %************************************************************************
1642 \subsection{Data types for the reduction mechanism}
1644 %************************************************************************
1646 The main control over context reduction is here
1650 = RedEnv { red_doc :: SDoc -- The context
1651 , red_try_me :: Inst -> WhatToDo
1652 , red_improve :: Bool -- True <=> do improvement
1653 , red_givens :: [Inst] -- All guaranteed rigid
1655 -- but see Note [Rigidity]
1656 , red_stack :: (Int, [Inst]) -- Recursion stack (for err msg)
1657 -- See Note [RedStack]
1661 -- The red_givens are rigid so far as cmpInst is concerned.
1662 -- There is one case where they are not totally rigid, namely in tcSimplifyIPs
1663 -- let ?x = e in ...
1664 -- Here, the given is (?x::a), where 'a' is not necy a rigid type
1665 -- But that doesn't affect the comparison, which is based only on mame.
1668 -- The red_stack pair (n,insts) pair is just used for error reporting.
1669 -- 'n' is always the depth of the stack.
1670 -- The 'insts' is the stack of Insts being reduced: to produce X
1671 -- I had to produce Y, to produce Y I had to produce Z, and so on.
1674 mkRedEnv :: SDoc -> (Inst -> WhatToDo) -> [Inst] -> RedEnv
1675 mkRedEnv doc try_me givens
1676 = RedEnv { red_doc = doc, red_try_me = try_me,
1677 red_givens = givens,
1679 red_improve = True }
1681 mkNoImproveRedEnv :: SDoc -> (Inst -> WhatToDo) -> RedEnv
1682 -- Do not do improvement; no givens
1683 mkNoImproveRedEnv doc try_me
1684 = RedEnv { red_doc = doc, red_try_me = try_me,
1687 red_improve = True }
1690 = ReduceMe WantSCs -- Try to reduce this
1691 -- If there's no instance, add the inst to the
1692 -- irreductible ones, but don't produce an error
1693 -- message of any kind.
1694 -- It might be quite legitimate such as (Eq a)!
1696 | Stop -- Return as irreducible unless it can
1697 -- be reduced to a constant in one step
1698 -- Do not add superclasses; see
1700 data WantSCs = NoSCs | AddSCs -- Tells whether we should add the superclasses
1701 -- of a predicate when adding it to the avails
1702 -- The reason for this flag is entirely the super-class loop problem
1703 -- Note [SUPER-CLASS LOOP 1]
1705 zonkRedEnv :: RedEnv -> TcM RedEnv
1707 = do { givens' <- mapM zonkInst (red_givens env)
1708 ; return $ env {red_givens = givens'}
1713 %************************************************************************
1715 \subsection[reduce]{@reduce@}
1717 %************************************************************************
1719 Note [Ancestor Equalities]
1720 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1721 During context reduction, we add to the wanted equalities also those
1722 equalities that (transitively) occur in superclass contexts of wanted
1723 class constraints. Consider the following code
1725 class a ~ Int => C a
1728 If (C a) is wanted, we want to add (a ~ Int), which will be discharged by
1729 substituting Int for a. Hence, we ultimately want (C Int), which we
1730 discharge with the explicit instance.
1733 reduceContext :: RedEnv
1735 -> TcM (ImprovementDone,
1736 TcDictBinds, -- Dictionary bindings
1737 [Inst], -- Irreducible
1738 TcM ()) -- Undo skolems from SkolemOccurs
1740 reduceContext env wanteds
1741 = do { traceTc (text "reduceContext" <+> (vcat [
1742 text "----------------------",
1744 text "given" <+> ppr (red_givens env),
1745 text "wanted" <+> ppr wanteds,
1746 text "----------------------"
1750 ; let givens = red_givens env
1751 (given_eqs0, given_dicts0) = partition isEqInst givens
1752 (wanted_eqs0, wanted_non_eqs) = partition isEqInst wanteds
1753 (wanted_implics0, wanted_dicts) = partition isImplicInst wanted_non_eqs
1755 -- We want to add as wanted equalities those that (transitively)
1756 -- occur in superclass contexts of wanted class constraints.
1757 -- See Note [Ancestor Equalities]
1758 ; ancestor_eqs <- ancestorEqualities wanted_dicts
1759 ; let wanted_eqs = wanted_eqs0 ++ ancestor_eqs
1760 ; traceTc $ text "reduceContext: ancestor eqs" <+> ppr ancestor_eqs
1762 -- 1. Normalise the *given* *equality* constraints
1763 ; (given_eqs, eliminate_skolems) <- normaliseGivenEqs given_eqs0
1765 -- 2. Normalise the *given* *dictionary* constraints
1766 -- wrt. the toplevel and given equations
1767 ; (given_dicts, given_binds) <- normaliseGivenDicts given_eqs
1770 -- 5. Build the Avail mapping from "given_dicts"
1771 ; (init_state, _) <- getLIE $ do
1772 { init_state <- foldlM addGiven emptyAvails given_dicts
1776 -- !!! ToDo: what to do with the "extra_givens"? For the
1777 -- moment I'm simply discarding them, which is probably wrong
1779 -- 6. Solve the *wanted* *dictionary* constraints (not implications)
1780 -- This may expose some further equational constraints...
1781 ; (avails, extra_eqs) <- getLIE (reduceList env wanted_dicts init_state)
1782 ; (dict_binds, bound_dicts, dict_irreds)
1783 <- extractResults avails wanted_dicts
1784 ; traceTc $ text "reduceContext extractresults" <+> vcat
1785 [ppr avails, ppr wanted_dicts, ppr dict_binds]
1787 -- Solve the wanted *implications*. In doing so, we can provide
1788 -- as "given" all the dicts that were originally given,
1789 -- *or* for which we now have bindings,
1790 -- *or* which are now irreds
1791 ; let implic_env = env { red_givens = givens ++ bound_dicts
1793 ; (implic_binds_s, implic_irreds_s)
1794 <- mapAndUnzipM (reduceImplication implic_env) wanted_implics0
1795 ; let implic_binds = unionManyBags implic_binds_s
1796 implic_irreds = concat implic_irreds_s
1798 -- Normalise the wanted equality constraints
1799 ; eq_irreds <- normaliseWantedEqs given_eqs (wanted_eqs ++ extra_eqs)
1801 -- Normalise the wanted dictionaries
1802 ; let irreds = dict_irreds ++ implic_irreds
1803 eqs = eq_irreds ++ given_eqs
1804 ; (norm_irreds, normalise_binds) <- normaliseWantedDicts eqs irreds
1806 -- Figure out whether we should go round again. We do so in either
1808 -- (1) If any of the mutable tyvars in givens or irreds has been
1809 -- filled in by improvement, there is merit in going around
1810 -- again, because we may make further progress.
1811 -- (2) If we managed to normalise any dicts, there is merit in going
1812 -- around gain, because reduceList may be able to get further.
1814 -- ToDo: We may have exposed new
1815 -- equality constraints and should probably go round again
1816 -- then as well. But currently we are dropping them on the
1819 ; let all_irreds = norm_irreds ++ eq_irreds
1820 ; improvedMetaTy <- anyM isFilledMetaTyVar $ varSetElems $
1821 tyVarsOfInsts (givens ++ all_irreds)
1822 ; let improvedDicts = not $ isEmptyBag normalise_binds
1823 improved = improvedMetaTy || improvedDicts
1825 -- The old plan (fragile)
1826 -- improveed = availsImproved avails
1827 -- || (not $ isEmptyBag normalise_binds1)
1828 -- || (not $ isEmptyBag normalise_binds2)
1829 -- || (any isEqInst irreds)
1831 ; traceTc (text "reduceContext end" <+> (vcat [
1832 text "----------------------",
1834 text "given" <+> ppr givens,
1835 text "given_eqs" <+> ppr given_eqs,
1836 text "wanted" <+> ppr wanteds,
1837 text "wanted_dicts" <+> ppr wanted_dicts,
1839 text "avails" <+> pprAvails avails,
1840 text "improved =" <+> ppr improved,
1841 text "(all) irreds = " <+> ppr all_irreds,
1842 text "dict-binds = " <+> ppr dict_binds,
1843 text "implic-binds = " <+> ppr implic_binds,
1844 text "----------------------"
1848 given_binds `unionBags` normalise_binds
1849 `unionBags` dict_binds
1850 `unionBags` implic_binds,
1855 tcImproveOne :: Avails -> Inst -> TcM ImprovementDone
1856 tcImproveOne avails inst
1857 | not (isDict inst) = return False
1859 = do { inst_envs <- tcGetInstEnvs
1860 ; let eqns = improveOne (classInstances inst_envs)
1861 (dictPred inst, pprInstArising inst)
1862 [ (dictPred p, pprInstArising p)
1863 | p <- availsInsts avails, isDict p ]
1864 -- Avails has all the superclasses etc (good)
1865 -- It also has all the intermediates of the deduction (good)
1866 -- It does not have duplicates (good)
1867 -- NB that (?x::t1) and (?x::t2) will be held separately in avails
1868 -- so that improve will see them separate
1869 ; traceTc (text "improveOne" <+> ppr inst)
1872 unifyEqns :: [(Equation,(PredType,SDoc),(PredType,SDoc))]
1873 -> TcM ImprovementDone
1874 unifyEqns [] = return False
1876 = do { traceTc (ptext (sLit "Improve:") <+> vcat (map pprEquationDoc eqns))
1880 unify ((qtvs, pairs), what1, what2)
1881 = addErrCtxtM (mkEqnMsg what1 what2) $ do
1882 (_, _, tenv) <- tcInstTyVars (varSetElems qtvs)
1883 mapM_ (unif_pr tenv) pairs
1884 unif_pr tenv (ty1,ty2) = unifyType (substTy tenv ty1) (substTy tenv ty2)
1886 pprEquationDoc :: (Equation, (PredType, SDoc), (PredType, SDoc)) -> SDoc
1887 pprEquationDoc (eqn, (p1, _), (p2, _)) = vcat [pprEquation eqn, nest 2 (ppr p1), nest 2 (ppr p2)]
1889 mkEqnMsg :: (TcPredType, SDoc) -> (TcPredType, SDoc) -> TidyEnv
1890 -> TcM (TidyEnv, SDoc)
1891 mkEqnMsg (pred1,from1) (pred2,from2) tidy_env
1892 = do { pred1' <- zonkTcPredType pred1; pred2' <- zonkTcPredType pred2
1893 ; let { pred1'' = tidyPred tidy_env pred1'; pred2'' = tidyPred tidy_env pred2' }
1894 ; let msg = vcat [ptext (sLit "When using functional dependencies to combine"),
1895 nest 2 (sep [ppr pred1'' <> comma, nest 2 from1]),
1896 nest 2 (sep [ppr pred2'' <> comma, nest 2 from2])]
1897 ; return (tidy_env, msg) }
1900 The main context-reduction function is @reduce@. Here's its game plan.
1903 reduceList :: RedEnv -> [Inst] -> Avails -> TcM Avails
1904 reduceList env@(RedEnv {red_stack = (n,stk)}) wanteds state
1905 = do { traceTc (text "reduceList " <+> (ppr wanteds $$ ppr state))
1907 ; when (debugIsOn && (n > 8)) $ do
1908 debugDumpTcRn (hang (ptext (sLit "Interesting! Context reduction stack depth") <+> int n)
1909 2 (ifPprDebug (nest 2 (pprStack stk))))
1910 ; if n >= ctxtStkDepth dopts then
1911 failWithTc (reduceDepthErr n stk)
1915 go [] state = return state
1916 go (w:ws) state = do { state' <- reduce (env {red_stack = (n+1, w:stk)}) w state
1919 -- Base case: we're done!
1920 reduce :: RedEnv -> Inst -> Avails -> TcM Avails
1921 reduce env wanted avails
1922 -- It's the same as an existing inst, or a superclass thereof
1923 | Just _ <- findAvail avails wanted
1924 = do { traceTc (text "reduce: found " <+> ppr wanted)
1929 = do { traceTc (text "reduce" <+> ppr wanted $$ ppr avails)
1930 ; case red_try_me env wanted of {
1931 Stop -> try_simple (addIrred NoSCs);
1932 -- See Note [No superclasses for Stop]
1934 ReduceMe want_scs -> do -- It should be reduced
1935 { (avails, lookup_result) <- reduceInst env avails wanted
1936 ; case lookup_result of
1937 NoInstance -> addIrred want_scs avails wanted
1938 -- Add it and its superclasses
1940 GenInst [] rhs -> addWanted want_scs avails wanted rhs []
1942 GenInst wanteds' rhs
1943 -> do { avails1 <- addIrred NoSCs avails wanted
1944 ; avails2 <- reduceList env wanteds' avails1
1945 ; addWanted want_scs avails2 wanted rhs wanteds' } }
1946 -- Temporarily do addIrred *before* the reduceList,
1947 -- which has the effect of adding the thing we are trying
1948 -- to prove to the database before trying to prove the things it
1949 -- needs. See note [RECURSIVE DICTIONARIES]
1950 -- NB: we must not do an addWanted before, because that adds the
1951 -- superclasses too, and that can lead to a spurious loop; see
1952 -- the examples in [SUPERCLASS-LOOP]
1953 -- So we do an addIrred before, and then overwrite it afterwards with addWanted
1956 -- First, see if the inst can be reduced to a constant in one step
1957 -- Works well for literals (1::Int) and constant dictionaries (d::Num Int)
1958 -- Don't bother for implication constraints, which take real work
1959 try_simple do_this_otherwise
1960 = do { res <- lookupSimpleInst wanted
1962 GenInst [] rhs -> addWanted AddSCs avails wanted rhs []
1963 _ -> do_this_otherwise avails wanted }
1967 Note [SUPERCLASS-LOOP 2]
1968 ~~~~~~~~~~~~~~~~~~~~~~~~
1969 But the above isn't enough. Suppose we are *given* d1:Ord a,
1970 and want to deduce (d2:C [a]) where
1972 class Ord a => C a where
1973 instance Ord [a] => C [a] where ...
1975 Then we'll use the instance decl to deduce C [a] from Ord [a], and then add the
1976 superclasses of C [a] to avails. But we must not overwrite the binding
1977 for Ord [a] (which is obtained from Ord a) with a superclass selection or we'll just
1980 Here's another variant, immortalised in tcrun020
1981 class Monad m => C1 m
1982 class C1 m => C2 m x
1983 instance C2 Maybe Bool
1984 For the instance decl we need to build (C1 Maybe), and it's no good if
1985 we run around and add (C2 Maybe Bool) and its superclasses to the avails
1986 before we search for C1 Maybe.
1988 Here's another example
1989 class Eq b => Foo a b
1990 instance Eq a => Foo [a] a
1994 we'll first deduce that it holds (via the instance decl). We must not
1995 then overwrite the Eq t constraint with a superclass selection!
1997 At first I had a gross hack, whereby I simply did not add superclass constraints
1998 in addWanted, though I did for addGiven and addIrred. This was sub-optimal,
1999 becuase it lost legitimate superclass sharing, and it still didn't do the job:
2000 I found a very obscure program (now tcrun021) in which improvement meant the
2001 simplifier got two bites a the cherry... so something seemed to be an Stop
2002 first time, but reducible next time.
2004 Now we implement the Right Solution, which is to check for loops directly
2005 when adding superclasses. It's a bit like the occurs check in unification.
2008 Note [RECURSIVE DICTIONARIES]
2009 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2011 data D r = ZeroD | SuccD (r (D r));
2013 instance (Eq (r (D r))) => Eq (D r) where
2014 ZeroD == ZeroD = True
2015 (SuccD a) == (SuccD b) = a == b
2018 equalDC :: D [] -> D [] -> Bool;
2021 We need to prove (Eq (D [])). Here's how we go:
2025 by instance decl, holds if
2029 by instance decl of Eq, holds if
2031 where d2 = dfEqList d3
2034 But now we can "tie the knot" to give
2040 and it'll even run! The trick is to put the thing we are trying to prove
2041 (in this case Eq (D []) into the database before trying to prove its
2042 contributing clauses.
2045 %************************************************************************
2047 Reducing a single constraint
2049 %************************************************************************
2052 ---------------------------------------------
2053 reduceInst :: RedEnv -> Avails -> Inst -> TcM (Avails, LookupInstResult)
2054 reduceInst _ avails other_inst
2055 = do { result <- lookupSimpleInst other_inst
2056 ; return (avails, result) }
2059 Note [Equational Constraints in Implication Constraints]
2060 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2062 An implication constraint is of the form
2064 where Given and Wanted may contain both equational and dictionary
2065 constraints. The delay and reduction of these two kinds of constraints
2068 -) In the generated code, wanted Dictionary constraints are wrapped up in an
2069 implication constraint that is created at the code site where the wanted
2070 dictionaries can be reduced via a let-binding. This let-bound implication
2071 constraint is deconstructed at the use-site of the wanted dictionaries.
2073 -) While the reduction of equational constraints is also delayed, the delay
2074 is not manifest in the generated code. The required evidence is generated
2075 in the code directly at the use-site. There is no let-binding and deconstruction
2076 necessary. The main disadvantage is that we cannot exploit sharing as the
2077 same evidence may be generated at multiple use-sites. However, this disadvantage
2078 is limited because it only concerns coercions which are erased.
2080 The different treatment is motivated by the different in representation. Dictionary
2081 constraints require manifest runtime dictionaries, while equations require coercions
2085 ---------------------------------------------
2086 reduceImplication :: RedEnv
2088 -> TcM (TcDictBinds, [Inst])
2091 Suppose we are simplifying the constraint
2092 forall bs. extras => wanted
2093 in the context of an overall simplification problem with givens 'givens'.
2096 * The 'givens' need not mention any of the quantified type variables
2097 e.g. forall {}. Eq a => Eq [a]
2098 forall {}. C Int => D (Tree Int)
2100 This happens when you have something like
2102 T1 :: Eq a => a -> T a
2105 f x = ...(case x of { T1 v -> v==v })...
2108 -- ToDo: should we instantiate tvs? I think it's not necessary
2110 -- Note on coercion variables:
2112 -- The extra given coercion variables are bound at two different sites:
2113 -- -) in the creation context of the implication constraint
2114 -- the solved equational constraints use these binders
2116 -- -) at the solving site of the implication constraint
2117 -- the solved dictionaries use these binders
2118 -- these binders are generated by reduceImplication
2120 reduceImplication env
2121 orig_implic@(ImplicInst { tci_name = name, tci_loc = inst_loc,
2123 tci_given = extra_givens, tci_wanted = wanteds })
2124 = do { -- Solve the sub-problem
2125 ; let try_me _ = ReduceMe AddSCs -- Note [Freeness and implications]
2126 env' = env { red_givens = extra_givens ++ red_givens env
2127 , red_doc = sep [ptext (sLit "reduceImplication for")
2129 nest 2 (parens $ ptext (sLit "within")
2131 , red_try_me = try_me }
2133 ; traceTc (text "reduceImplication" <+> vcat
2134 [ ppr (red_givens env), ppr extra_givens,
2136 ; (irreds, binds) <- checkLoop env' wanteds
2137 ; let (extra_eq_givens, extra_dict_givens) = partition isEqInst extra_givens
2138 -- SLPJ Sept 07: I think this is bogus; currently
2139 -- there are no Eqinsts in extra_givens
2140 dict_ids = map instToId extra_dict_givens
2142 -- Note [Reducing implication constraints]
2143 -- Tom -- update note, put somewhere!
2145 ; traceTc (text "reduceImplication result" <+> vcat
2146 [ppr irreds, ppr binds])
2148 ; -- extract superclass binds
2149 -- (sc_binds,_) <- extractResults avails []
2150 -- ; traceTc (text "reduceImplication sc_binds" <+> vcat
2151 -- [ppr sc_binds, ppr avails])
2154 -- SLPJ Sept 07: what if improvement happened inside the checkLoop?
2155 -- Then we must iterate the outer loop too!
2157 ; traceTc (text "reduceImplication condition" <+> ppr ((isEmptyLHsBinds binds) || (null irreds)))
2159 -- Progress is no longer measered by the number of bindings
2160 ; if (isEmptyLHsBinds binds) && (not $ null irreds) then -- No progress
2161 -- If there are any irreds, we back off and do nothing
2162 return (emptyBag, [orig_implic])
2164 { (simpler_implic_insts, bind)
2165 <- makeImplicationBind inst_loc tvs extra_givens irreds
2166 -- This binding is useless if the recursive simplification
2167 -- made no progress; but currently we don't try to optimise that
2168 -- case. After all, we only try hard to reduce at top level, or
2169 -- when inferring types.
2171 ; let dict_wanteds = filter (not . isEqInst) wanteds
2172 -- TOMDO: given equational constraints bug!
2173 -- we need a different evidence for given
2174 -- equations depending on whether we solve
2175 -- dictionary constraints or equational constraints
2177 eq_tyvars = varSetElems $ tyVarsOfTypes $ map eqInstType extra_eq_givens
2178 -- SLPJ Sept07: this looks Utterly Wrong to me, but I think
2179 -- that current extra_givens has no EqInsts, so
2180 -- it makes no difference
2181 co = wrap_inline -- Note [Always inline implication constraints]
2183 <.> mkWpLams eq_tyvars
2184 <.> mkWpLams dict_ids
2185 <.> WpLet (binds `unionBags` bind)
2186 wrap_inline | null dict_ids = idHsWrapper
2187 | otherwise = WpInline
2188 rhs = mkLHsWrap co payload
2189 loc = instLocSpan inst_loc
2190 payload = mkBigLHsTup (map (L loc . HsVar . instToId) dict_wanteds)
2193 ; traceTc (vcat [text "reduceImplication" <+> ppr name,
2194 ppr simpler_implic_insts,
2195 text "->" <+> ppr rhs])
2196 ; return (unitBag (L loc (VarBind (instToId orig_implic) rhs)),
2197 simpler_implic_insts)
2200 reduceImplication _ i = pprPanic "reduceImplication" (ppr i)
2203 Note [Always inline implication constraints]
2204 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2205 Suppose an implication constraint floats out of an INLINE function.
2206 Then although the implication has a single call site, it won't be
2207 inlined. And that is bad because it means that even if there is really
2208 *no* overloading (type signatures specify the exact types) there will
2209 still be dictionary passing in the resulting code. To avert this,
2210 we mark the implication constraints themselves as INLINE, at least when
2211 there is no loss of sharing as a result.
2213 Note [Freeness and implications]
2214 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2215 It's hard to say when an implication constraint can be floated out. Consider
2216 forall {} Eq a => Foo [a]
2217 The (Foo [a]) doesn't mention any of the quantified variables, but it
2218 still might be partially satisfied by the (Eq a).
2220 There is a useful special case when it *is* easy to partition the
2221 constraints, namely when there are no 'givens'. Consider
2222 forall {a}. () => Bar b
2223 There are no 'givens', and so there is no reason to capture (Bar b).
2224 We can let it float out. But if there is even one constraint we
2225 must be much more careful:
2226 forall {a}. C a b => Bar (m b)
2227 because (C a b) might have a superclass (D b), from which we might
2228 deduce (Bar [b]) when m later gets instantiated to []. Ha!
2230 Here is an even more exotic example
2232 Now consider the constraint
2233 forall b. D Int b => C Int
2234 We can satisfy the (C Int) from the superclass of D, so we don't want
2235 to float the (C Int) out, even though it mentions no type variable in
2238 One more example: the constraint
2240 instance (C a, E c) => E (a,c)
2242 constraint: forall b. D Int b => E (Int,c)
2244 You might think that the (D Int b) can't possibly contribute
2245 to solving (E (Int,c)), since the latter mentions 'c'. But
2246 in fact it can, because solving the (E (Int,c)) constraint needs
2249 and the (C Int) can be satisfied from the superclass of (D Int b).
2250 So we must still not float (E (Int,c)) out.
2252 To think about: special cases for unary type classes?
2254 Note [Pruning the givens in an implication constraint]
2255 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2256 Suppose we are about to form the implication constraint
2257 forall tvs. Eq a => Ord b
2258 The (Eq a) cannot contribute to the (Ord b), because it has no access to
2259 the type variable 'b'. So we could filter out the (Eq a) from the givens.
2260 But BE CAREFUL of the examples above in [Freeness and implications].
2262 Doing so would be a bit tidier, but all the implication constraints get
2263 simplified away by the optimiser, so it's no great win. So I don't take
2264 advantage of that at the moment.
2266 If you do, BE CAREFUL of wobbly type variables.
2269 %************************************************************************
2271 Avails and AvailHow: the pool of evidence
2273 %************************************************************************
2277 data Avails = Avails !ImprovementDone !AvailEnv
2279 type ImprovementDone = Bool -- True <=> some unification has happened
2280 -- so some Irreds might now be reducible
2281 -- keys that are now
2283 type AvailEnv = FiniteMap Inst AvailHow
2285 = IsIrred -- Used for irreducible dictionaries,
2286 -- which are going to be lambda bound
2288 | Given Inst -- Used for dictionaries for which we have a binding
2289 -- e.g. those "given" in a signature
2291 | Rhs -- Used when there is a RHS
2292 (LHsExpr TcId) -- The RHS
2293 [Inst] -- Insts free in the RHS; we need these too
2295 instance Outputable Avails where
2298 pprAvails :: Avails -> SDoc
2299 pprAvails (Avails imp avails)
2300 = vcat [ ptext (sLit "Avails") <> (if imp then ptext (sLit "[improved]") else empty)
2302 vcat [ sep [ppr inst, nest 2 (equals <+> ppr avail)]
2303 | (inst,avail) <- fmToList avails ]]
2305 instance Outputable AvailHow where
2308 -------------------------
2309 pprAvail :: AvailHow -> SDoc
2310 pprAvail IsIrred = text "Irred"
2311 pprAvail (Given x) = text "Given" <+> ppr x
2312 pprAvail (Rhs rhs bs) = sep [text "Rhs" <+> ppr bs,
2315 -------------------------
2316 extendAvailEnv :: AvailEnv -> Inst -> AvailHow -> AvailEnv
2317 extendAvailEnv env inst avail = addToFM env inst avail
2319 findAvailEnv :: AvailEnv -> Inst -> Maybe AvailHow
2320 findAvailEnv env wanted = lookupFM env wanted
2321 -- NB 1: the Ord instance of Inst compares by the class/type info
2322 -- *not* by unique. So
2323 -- d1::C Int == d2::C Int
2325 emptyAvails :: Avails
2326 emptyAvails = Avails False emptyFM
2328 findAvail :: Avails -> Inst -> Maybe AvailHow
2329 findAvail (Avails _ avails) wanted = findAvailEnv avails wanted
2331 elemAvails :: Inst -> Avails -> Bool
2332 elemAvails wanted (Avails _ avails) = wanted `elemFM` avails
2334 extendAvails :: Avails -> Inst -> AvailHow -> TcM Avails
2336 extendAvails avails@(Avails imp env) inst avail
2337 = do { imp1 <- tcImproveOne avails inst -- Do any improvement
2338 ; return (Avails (imp || imp1) (extendAvailEnv env inst avail)) }
2340 availsInsts :: Avails -> [Inst]
2341 availsInsts (Avails _ avails) = keysFM avails
2343 _availsImproved :: Avails -> ImprovementDone
2344 _availsImproved (Avails imp _) = imp
2347 Extracting the bindings from a bunch of Avails.
2348 The bindings do *not* come back sorted in dependency order.
2349 We assume that they'll be wrapped in a big Rec, so that the
2350 dependency analyser can sort them out later
2353 type DoneEnv = FiniteMap Inst [Id]
2354 -- Tracks which things we have evidence for
2356 extractResults :: Avails
2358 -> TcM (TcDictBinds, -- Bindings
2359 [Inst], -- The insts bound by the bindings
2360 [Inst]) -- Irreducible ones
2361 -- Note [Reducing implication constraints]
2363 extractResults (Avails _ avails) wanteds
2364 = go emptyBag [] [] emptyFM wanteds
2366 go :: TcDictBinds -- Bindings for dicts
2367 -> [Inst] -- Bound by the bindings
2369 -> DoneEnv -- Has an entry for each inst in the above three sets
2371 -> TcM (TcDictBinds, [Inst], [Inst])
2372 go binds bound_dicts irreds _ []
2373 = return (binds, bound_dicts, irreds)
2375 go binds bound_dicts irreds done (w:ws)
2376 | Just done_ids@(done_id : rest_done_ids) <- lookupFM done w
2377 = if w_id `elem` done_ids then
2378 go binds bound_dicts irreds done ws
2380 go (add_bind (nlHsVar done_id)) bound_dicts irreds
2381 (addToFM done w (done_id : w_id : rest_done_ids)) ws
2383 | otherwise -- Not yet done
2384 = case findAvailEnv avails w of
2385 Nothing -> pprTrace "Urk: extractResults" (ppr w) $
2386 go binds bound_dicts irreds done ws
2388 Just IsIrred -> go binds bound_dicts (w:irreds) done' ws
2390 Just (Rhs rhs ws') -> go (add_bind rhs) (w:bound_dicts) irreds done' (ws' ++ ws)
2392 Just (Given g) -> go binds' bound_dicts irreds (addToFM done w [g_id]) ws
2395 binds' | w_id == g_id = binds
2396 | otherwise = add_bind (nlHsVar g_id)
2399 done' = addToFM done w [w_id]
2400 add_bind rhs = addInstToDictBind binds w rhs
2404 Note [No superclasses for Stop]
2405 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2406 When we decide not to reduce an Inst -- the 'WhatToDo' --- we still
2407 add it to avails, so that any other equal Insts will be commoned up
2408 right here. However, we do *not* add superclasses. If we have
2411 but a is not bound here, then we *don't* want to derive dn from df
2412 here lest we lose sharing.
2415 addWanted :: WantSCs -> Avails -> Inst -> LHsExpr TcId -> [Inst] -> TcM Avails
2416 addWanted want_scs avails wanted rhs_expr wanteds
2417 = addAvailAndSCs want_scs avails wanted avail
2419 avail = Rhs rhs_expr wanteds
2421 addGiven :: Avails -> Inst -> TcM Avails
2422 addGiven avails given = addAvailAndSCs AddSCs avails given (Given given)
2423 -- Always add superclasses for 'givens'
2425 -- No ASSERT( not (given `elemAvails` avails) ) because in an instance
2426 -- decl for Ord t we can add both Ord t and Eq t as 'givens',
2427 -- so the assert isn't true
2431 addIrred :: WantSCs -> Avails -> Inst -> TcM Avails
2432 addIrred want_scs avails irred = ASSERT2( not (irred `elemAvails` avails), ppr irred $$ ppr avails )
2433 addAvailAndSCs want_scs avails irred IsIrred
2435 addAvailAndSCs :: WantSCs -> Avails -> Inst -> AvailHow -> TcM Avails
2436 addAvailAndSCs want_scs avails inst avail
2437 | not (isClassDict inst) = extendAvails avails inst avail
2438 | NoSCs <- want_scs = extendAvails avails inst avail
2439 | otherwise = do { traceTc (text "addAvailAndSCs" <+> vcat [ppr inst, ppr deps])
2440 ; avails' <- extendAvails avails inst avail
2441 ; addSCs is_loop avails' inst }
2443 is_loop pred = any (`tcEqType` mkPredTy pred) dep_tys
2444 -- Note: this compares by *type*, not by Unique
2445 deps = findAllDeps (unitVarSet (instToVar inst)) avail
2446 dep_tys = map idType (varSetElems deps)
2448 findAllDeps :: IdSet -> AvailHow -> IdSet
2449 -- Find all the Insts that this one depends on
2450 -- See Note [SUPERCLASS-LOOP 2]
2451 -- Watch out, though. Since the avails may contain loops
2452 -- (see Note [RECURSIVE DICTIONARIES]), so we need to track the ones we've seen so far
2453 findAllDeps so_far (Rhs _ kids) = foldl find_all so_far kids
2454 findAllDeps so_far _ = so_far
2456 find_all :: IdSet -> Inst -> IdSet
2458 | isEqInst kid = so_far
2459 | kid_id `elemVarSet` so_far = so_far
2460 | Just avail <- findAvail avails kid = findAllDeps so_far' avail
2461 | otherwise = so_far'
2463 so_far' = extendVarSet so_far kid_id -- Add the new kid to so_far
2464 kid_id = instToId kid
2466 addSCs :: (TcPredType -> Bool) -> Avails -> Inst -> TcM Avails
2467 -- Add all the superclasses of the Inst to Avails
2468 -- The first param says "don't do this because the original thing
2469 -- depends on this one, so you'd build a loop"
2470 -- Invariant: the Inst is already in Avails.
2472 addSCs is_loop avails dict
2473 = ASSERT( isDict dict )
2474 do { sc_dicts <- newDictBndrs (instLoc dict) sc_theta'
2475 ; foldlM add_sc avails (zipEqual "add_scs" sc_dicts sc_sels) }
2477 (clas, tys) = getDictClassTys dict
2478 (tyvars, sc_theta, sc_sels, _) = classBigSig clas
2479 sc_theta' = filter (not . isEqPred) $
2480 substTheta (zipTopTvSubst tyvars tys) sc_theta
2482 add_sc avails (sc_dict, sc_sel)
2483 | is_loop (dictPred sc_dict) = return avails -- See Note [SUPERCLASS-LOOP 2]
2484 | is_given sc_dict = return avails
2485 | otherwise = do { avails' <- extendAvails avails sc_dict (Rhs sc_sel_rhs [dict])
2486 ; addSCs is_loop avails' sc_dict }
2488 sc_sel_rhs = L (instSpan dict) (HsWrap co_fn (HsVar sc_sel))
2489 co_fn = WpApp (instToVar dict) <.> mkWpTyApps tys
2491 is_given :: Inst -> Bool
2492 is_given sc_dict = case findAvail avails sc_dict of
2493 Just (Given _) -> True -- Given is cheaper than superclass selection
2496 -- From the a set of insts obtain all equalities that (transitively) occur in
2497 -- superclass contexts of class constraints (aka the ancestor equalities).
2499 ancestorEqualities :: [Inst] -> TcM [Inst]
2501 = mapM mkWantedEqInst -- turn only equality predicates..
2502 . filter isEqPred -- ..into wanted equality insts
2504 . addAEsToBag emptyBag -- collect the superclass constraints..
2505 . map dictPred -- ..of all predicates in a bag
2506 . filter isClassDict
2508 addAEsToBag :: Bag PredType -> [PredType] -> Bag PredType
2509 addAEsToBag bag [] = bag
2510 addAEsToBag bag (pred:preds)
2511 | pred `elemBag` bag = addAEsToBag bag preds
2512 | isEqPred pred = addAEsToBag bagWithPred preds
2513 | isClassPred pred = addAEsToBag bagWithPred predsWithSCs
2514 | otherwise = addAEsToBag bag preds
2516 bagWithPred = bag `snocBag` pred
2517 predsWithSCs = preds ++ substTheta (zipTopTvSubst tyvars tys) sc_theta
2519 (tyvars, sc_theta, _, _) = classBigSig clas
2520 (clas, tys) = getClassPredTys pred
2524 %************************************************************************
2526 \section{tcSimplifyTop: defaulting}
2528 %************************************************************************
2531 @tcSimplifyTop@ is called once per module to simplify all the constant
2532 and ambiguous Insts.
2534 We need to be careful of one case. Suppose we have
2536 instance Num a => Num (Foo a b) where ...
2538 and @tcSimplifyTop@ is given a constraint (Num (Foo x y)). Then it'll simplify
2539 to (Num x), and default x to Int. But what about y??
2541 It's OK: the final zonking stage should zap y to (), which is fine.
2545 tcSimplifyTop, tcSimplifyInteractive :: [Inst] -> TcM TcDictBinds
2546 tcSimplifyTop wanteds
2547 = tc_simplify_top doc False wanteds
2549 doc = text "tcSimplifyTop"
2551 tcSimplifyInteractive wanteds
2552 = tc_simplify_top doc True wanteds
2554 doc = text "tcSimplifyInteractive"
2556 -- The TcLclEnv should be valid here, solely to improve
2557 -- error message generation for the monomorphism restriction
2558 tc_simplify_top :: SDoc -> Bool -> [Inst] -> TcM (Bag (LHsBind TcId))
2559 tc_simplify_top doc interactive wanteds
2560 = do { dflags <- getDOpts
2561 ; wanteds <- zonkInsts wanteds
2562 ; mapM_ zonkTopTyVar (varSetElems (tyVarsOfInsts wanteds))
2564 ; traceTc (text "tc_simplify_top 0: " <+> ppr wanteds)
2565 ; (irreds1, binds1) <- tryHardCheckLoop doc1 wanteds
2566 -- ; (irreds1, binds1) <- gentleInferLoop doc1 wanteds
2567 ; traceTc (text "tc_simplify_top 1: " <+> ppr irreds1)
2568 ; (irreds2, binds2) <- approximateImplications doc2 (\_ -> True) irreds1
2569 ; traceTc (text "tc_simplify_top 2: " <+> ppr irreds2)
2571 -- Use the defaulting rules to do extra unification
2572 -- NB: irreds2 are already zonked
2573 ; (irreds3, binds3) <- disambiguate doc3 interactive dflags irreds2
2575 -- Deal with implicit parameters
2576 ; let (bad_ips, non_ips) = partition isIPDict irreds3
2577 (ambigs, others) = partition isTyVarDict non_ips
2579 ; topIPErrs bad_ips -- Can arise from f :: Int -> Int
2581 ; addNoInstanceErrs others
2582 ; addTopAmbigErrs ambigs
2584 ; return (binds1 `unionBags` binds2 `unionBags` binds3) }
2586 doc1 = doc <+> ptext (sLit "(first round)")
2587 doc2 = doc <+> ptext (sLit "(approximate)")
2588 doc3 = doc <+> ptext (sLit "(disambiguate)")
2591 If a dictionary constrains a type variable which is
2592 * not mentioned in the environment
2593 * and not mentioned in the type of the expression
2594 then it is ambiguous. No further information will arise to instantiate
2595 the type variable; nor will it be generalised and turned into an extra
2596 parameter to a function.
2598 It is an error for this to occur, except that Haskell provided for
2599 certain rules to be applied in the special case of numeric types.
2601 * at least one of its classes is a numeric class, and
2602 * all of its classes are numeric or standard
2603 then the type variable can be defaulted to the first type in the
2604 default-type list which is an instance of all the offending classes.
2606 So here is the function which does the work. It takes the ambiguous
2607 dictionaries and either resolves them (producing bindings) or
2608 complains. It works by splitting the dictionary list by type
2609 variable, and using @disambigOne@ to do the real business.
2611 @disambigOne@ assumes that its arguments dictionaries constrain all
2612 the same type variable.
2614 ADR Comment 20/6/94: I've changed the @CReturnable@ case to default to
2615 @()@ instead of @Int@. I reckon this is the Right Thing to do since
2616 the most common use of defaulting is code like:
2618 _ccall_ foo `seqPrimIO` bar
2620 Since we're not using the result of @foo@, the result if (presumably)
2624 disambiguate :: SDoc -> Bool -> DynFlags -> [Inst] -> TcM ([Inst], TcDictBinds)
2625 -- Just does unification to fix the default types
2626 -- The Insts are assumed to be pre-zonked
2627 disambiguate doc interactive dflags insts
2629 = return (insts, emptyBag)
2631 | null defaultable_groups
2632 = do { traceTc (text "disambigutate, no defaultable groups" <+> vcat [ppr unaries, ppr insts, ppr bad_tvs, ppr defaultable_groups])
2633 ; return (insts, emptyBag) }
2636 = do { -- Figure out what default types to use
2637 default_tys <- getDefaultTys extended_defaulting ovl_strings
2639 ; traceTc (text "disambiguate1" <+> vcat [ppr insts, ppr unaries, ppr bad_tvs, ppr defaultable_groups])
2640 ; mapM_ (disambigGroup default_tys) defaultable_groups
2642 -- disambigGroup does unification, hence try again
2643 ; tryHardCheckLoop doc insts }
2646 extended_defaulting = interactive || dopt Opt_ExtendedDefaultRules dflags
2647 ovl_strings = dopt Opt_OverloadedStrings dflags
2649 unaries :: [(Inst, Class, TcTyVar)] -- (C tv) constraints
2650 bad_tvs :: TcTyVarSet -- Tyvars mentioned by *other* constraints
2651 (unaries, bad_tvs_s) = partitionWith find_unary insts
2652 bad_tvs = unionVarSets bad_tvs_s
2654 -- Finds unary type-class constraints
2655 find_unary d@(Dict {tci_pred = ClassP cls [ty]})
2656 | Just tv <- tcGetTyVar_maybe ty = Left (d,cls,tv)
2657 find_unary inst = Right (tyVarsOfInst inst)
2659 -- Group by type variable
2660 defaultable_groups :: [[(Inst,Class,TcTyVar)]]
2661 defaultable_groups = filter defaultable_group (equivClasses cmp_tv unaries)
2662 cmp_tv (_,_,tv1) (_,_,tv2) = tv1 `compare` tv2
2664 defaultable_group :: [(Inst,Class,TcTyVar)] -> Bool
2665 defaultable_group ds@((_,_,tv):_)
2666 = isTyConableTyVar tv -- Note [Avoiding spurious errors]
2667 && not (tv `elemVarSet` bad_tvs)
2668 && defaultable_classes [c | (_,c,_) <- ds]
2669 defaultable_group [] = panic "defaultable_group"
2671 defaultable_classes clss
2672 | extended_defaulting = any isInteractiveClass clss
2673 | otherwise = all is_std_class clss && (any is_num_class clss)
2675 -- In interactive mode, or with -XExtendedDefaultRules,
2676 -- we default Show a to Show () to avoid graututious errors on "show []"
2677 isInteractiveClass cls
2678 = is_num_class cls || (classKey cls `elem` [showClassKey, eqClassKey, ordClassKey])
2680 is_num_class cls = isNumericClass cls || (ovl_strings && (cls `hasKey` isStringClassKey))
2681 -- is_num_class adds IsString to the standard numeric classes,
2682 -- when -foverloaded-strings is enabled
2684 is_std_class cls = isStandardClass cls || (ovl_strings && (cls `hasKey` isStringClassKey))
2685 -- Similarly is_std_class
2687 -----------------------
2688 disambigGroup :: [Type] -- The default types
2689 -> [(Inst,Class,TcTyVar)] -- All standard classes of form (C a)
2690 -> TcM () -- Just does unification, to fix the default types
2692 disambigGroup default_tys dicts
2693 = try_default default_tys
2695 (_,_,tyvar) = ASSERT(not (null dicts)) head dicts -- Should be non-empty
2696 classes = [c | (_,c,_) <- dicts]
2698 try_default [] = return ()
2699 try_default (default_ty : default_tys)
2700 = tryTcLIE_ (try_default default_tys) $
2701 do { tcSimplifyDefault [mkClassPred clas [default_ty] | clas <- classes]
2702 -- This may fail; then the tryTcLIE_ kicks in
2703 -- Failure here is caused by there being no type in the
2704 -- default list which can satisfy all the ambiguous classes.
2705 -- For example, if Real a is reqd, but the only type in the
2706 -- default list is Int.
2708 -- After this we can't fail
2709 ; warnDefault dicts default_ty
2710 ; unifyType default_ty (mkTyVarTy tyvar)
2711 ; return () -- TOMDO: do something with the coercion
2715 -----------------------
2716 getDefaultTys :: Bool -> Bool -> TcM [Type]
2717 getDefaultTys extended_deflts ovl_strings
2718 = do { mb_defaults <- getDeclaredDefaultTys
2719 ; case mb_defaults of {
2720 Just tys -> return tys ; -- User-supplied defaults
2723 -- No use-supplied default
2724 -- Use [Integer, Double], plus modifications
2725 { integer_ty <- tcMetaTy integerTyConName
2726 ; checkWiredInTyCon doubleTyCon
2727 ; string_ty <- tcMetaTy stringTyConName
2728 ; return (opt_deflt extended_deflts unitTy
2729 -- Note [Default unitTy]
2731 [integer_ty,doubleTy]
2733 opt_deflt ovl_strings string_ty) } } }
2735 opt_deflt True ty = [ty]
2736 opt_deflt False _ = []
2739 Note [Default unitTy]
2740 ~~~~~~~~~~~~~~~~~~~~~
2741 In interative mode (or with -XExtendedDefaultRules) we add () as the first type we
2742 try when defaulting. This has very little real impact, except in the following case.
2744 Text.Printf.printf "hello"
2745 This has type (forall a. IO a); it prints "hello", and returns 'undefined'. We don't
2746 want the GHCi repl loop to try to print that 'undefined'. The neatest thing is to
2747 default the 'a' to (), rather than to Integer (which is what would otherwise happen;
2748 and then GHCi doesn't attempt to print the (). So in interactive mode, we add
2749 () to the list of defaulting types. See Trac #1200.
2751 Note [Avoiding spurious errors]
2752 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2753 When doing the unification for defaulting, we check for skolem
2754 type variables, and simply don't default them. For example:
2755 f = (*) -- Monomorphic
2756 g :: Num a => a -> a
2758 Here, we get a complaint when checking the type signature for g,
2759 that g isn't polymorphic enough; but then we get another one when
2760 dealing with the (Num a) context arising from f's definition;
2761 we try to unify a with Int (to default it), but find that it's
2762 already been unified with the rigid variable from g's type sig
2765 %************************************************************************
2767 \subsection[simple]{@Simple@ versions}
2769 %************************************************************************
2771 Much simpler versions when there are no bindings to make!
2773 @tcSimplifyThetas@ simplifies class-type constraints formed by
2774 @deriving@ declarations and when specialising instances. We are
2775 only interested in the simplified bunch of class/type constraints.
2777 It simplifies to constraints of the form (C a b c) where
2778 a,b,c are type variables. This is required for the context of
2779 instance declarations.
2782 tcSimplifyDeriv :: InstOrigin
2784 -> ThetaType -- Wanted
2785 -> TcM ThetaType -- Needed
2786 -- Given instance (wanted) => C inst_ty
2787 -- Simplify 'wanted' as much as possible
2789 tcSimplifyDeriv orig tyvars theta
2790 = do { (tvs, _, tenv) <- tcInstTyVars tyvars
2791 -- The main loop may do unification, and that may crash if
2792 -- it doesn't see a TcTyVar, so we have to instantiate. Sigh
2793 -- ToDo: what if two of them do get unified?
2794 ; wanteds <- newDictBndrsO orig (substTheta tenv theta)
2795 ; (irreds, _) <- tryHardCheckLoop doc wanteds
2797 ; let (tv_dicts, others) = partition ok irreds
2798 ; addNoInstanceErrs others
2799 -- See Note [Exotic derived instance contexts] in TcMType
2801 ; let rev_env = zipTopTvSubst tvs (mkTyVarTys tyvars)
2802 simpl_theta = substTheta rev_env (map dictPred tv_dicts)
2803 -- This reverse-mapping is a pain, but the result
2804 -- should mention the original TyVars not TcTyVars
2806 ; return simpl_theta }
2808 doc = ptext (sLit "deriving classes for a data type")
2810 ok dict | isDict dict = validDerivPred (dictPred dict)
2815 @tcSimplifyDefault@ just checks class-type constraints, essentially;
2816 used with \tr{default} declarations. We are only interested in
2817 whether it worked or not.
2820 tcSimplifyDefault :: ThetaType -- Wanted; has no type variables in it
2823 tcSimplifyDefault theta = do
2824 wanteds <- newDictBndrsO DefaultOrigin theta
2825 (irreds, _) <- tryHardCheckLoop doc wanteds
2826 addNoInstanceErrs irreds
2830 traceTc (ptext (sLit "tcSimplifyDefault failing")) >> failM
2832 doc = ptext (sLit "default declaration")
2836 %************************************************************************
2838 \section{Errors and contexts}
2840 %************************************************************************
2842 ToDo: for these error messages, should we note the location as coming
2843 from the insts, or just whatever seems to be around in the monad just
2847 groupErrs :: ([Inst] -> TcM ()) -- Deal with one group
2848 -> [Inst] -- The offending Insts
2850 -- Group together insts with the same origin
2851 -- We want to report them together in error messages
2855 groupErrs report_err (inst:insts)
2856 = do { do_one (inst:friends)
2857 ; groupErrs report_err others }
2859 -- (It may seem a bit crude to compare the error messages,
2860 -- but it makes sure that we combine just what the user sees,
2861 -- and it avoids need equality on InstLocs.)
2862 (friends, others) = partition is_friend insts
2863 loc_msg = showSDoc (pprInstLoc (instLoc inst))
2864 is_friend friend = showSDoc (pprInstLoc (instLoc friend)) == loc_msg
2865 do_one insts = addInstCtxt (instLoc (head insts)) (report_err insts)
2866 -- Add location and context information derived from the Insts
2868 -- Add the "arising from..." part to a message about bunch of dicts
2869 addInstLoc :: [Inst] -> Message -> Message
2870 addInstLoc insts msg = msg $$ nest 2 (pprInstArising (head insts))
2872 addTopIPErrs :: [Name] -> [Inst] -> TcM ()
2875 addTopIPErrs bndrs ips
2876 = do { dflags <- getDOpts
2877 ; addErrTcM (tidy_env, mk_msg dflags tidy_ips) }
2879 (tidy_env, tidy_ips) = tidyInsts ips
2881 = vcat [sep [ptext (sLit "Implicit parameters escape from"),
2882 nest 2 (ptext (sLit "the monomorphic top-level binding")
2883 <> plural bndrs <+> ptext (sLit "of")
2884 <+> pprBinders bndrs <> colon)],
2885 nest 2 (vcat (map ppr_ip ips)),
2886 monomorphism_fix dflags]
2887 ppr_ip ip = pprPred (dictPred ip) <+> pprInstArising ip
2889 topIPErrs :: [Inst] -> TcM ()
2891 = groupErrs report tidy_dicts
2893 (tidy_env, tidy_dicts) = tidyInsts dicts
2894 report dicts = addErrTcM (tidy_env, mk_msg dicts)
2895 mk_msg dicts = addInstLoc dicts (ptext (sLit "Unbound implicit parameter") <>
2896 plural tidy_dicts <+> pprDictsTheta tidy_dicts)
2898 addNoInstanceErrs :: [Inst] -- Wanted (can include implications)
2900 addNoInstanceErrs insts
2901 = do { let (tidy_env, tidy_insts) = tidyInsts insts
2902 ; reportNoInstances tidy_env Nothing tidy_insts }
2906 -> Maybe (InstLoc, [Inst]) -- Context
2907 -- Nothing => top level
2908 -- Just (d,g) => d describes the construct
2910 -> [Inst] -- What is wanted (can include implications)
2913 reportNoInstances tidy_env mb_what insts
2914 = groupErrs (report_no_instances tidy_env mb_what) insts
2916 report_no_instances :: TidyEnv -> Maybe (InstLoc, [Inst]) -> [Inst] -> TcM ()
2917 report_no_instances tidy_env mb_what insts
2918 = do { inst_envs <- tcGetInstEnvs
2919 ; let (implics, insts1) = partition isImplicInst insts
2920 (insts2, overlaps) = partitionWith (check_overlap inst_envs) insts1
2921 (eqInsts, insts3) = partition isEqInst insts2
2922 ; traceTc (text "reportNoInstances" <+> vcat
2923 [ppr insts, ppr implics, ppr insts1, ppr insts2])
2924 ; mapM_ complain_implic implics
2925 ; mapM_ (\doc -> addErrTcM (tidy_env, doc)) overlaps
2926 ; groupErrs complain_no_inst insts3
2927 ; mapM_ (addErrTcM . mk_eq_err) eqInsts
2930 complain_no_inst insts = addErrTcM (tidy_env, mk_no_inst_err insts)
2932 complain_implic inst -- Recurse!
2933 = reportNoInstances tidy_env
2934 (Just (tci_loc inst, tci_given inst))
2937 check_overlap :: (InstEnv,InstEnv) -> Inst -> Either Inst SDoc
2938 -- Right msg => overlap message
2939 -- Left inst => no instance
2940 check_overlap inst_envs wanted
2941 | not (isClassDict wanted) = Left wanted
2943 = case lookupInstEnv inst_envs clas tys of
2944 ([], _) -> Left wanted -- No match
2945 -- The case of exactly one match and no unifiers means a
2946 -- successful lookup. That can't happen here, because dicts
2947 -- only end up here if they didn't match in Inst.lookupInst
2949 | debugIsOn -> pprPanic "reportNoInstance" (ppr wanted)
2950 res -> Right (mk_overlap_msg wanted res)
2952 (clas,tys) = getDictClassTys wanted
2954 mk_overlap_msg dict (matches, unifiers)
2955 = ASSERT( not (null matches) )
2956 vcat [ addInstLoc [dict] ((ptext (sLit "Overlapping instances for")
2957 <+> pprPred (dictPred dict))),
2958 sep [ptext (sLit "Matching instances") <> colon,
2959 nest 2 (vcat [pprInstances ispecs, pprInstances unifiers])],
2960 if not (isSingleton matches)
2961 then -- Two or more matches
2963 else -- One match, plus some unifiers
2964 ASSERT( not (null unifiers) )
2965 parens (vcat [ptext (sLit "The choice depends on the instantiation of") <+>
2966 quotes (pprWithCommas ppr (varSetElems (tyVarsOfInst dict))),
2967 ptext (sLit "To pick the first instance above, use -XIncoherentInstances"),
2968 ptext (sLit "when compiling the other instance declarations")])]
2970 ispecs = [ispec | (ispec, _) <- matches]
2972 mk_eq_err :: Inst -> (TidyEnv, SDoc)
2973 mk_eq_err inst = misMatchMsg tidy_env (eqInstTys inst)
2975 mk_no_inst_err insts
2976 | null insts = empty
2978 | Just (loc, givens) <- mb_what, -- Nested (type signatures, instance decls)
2979 not (isEmptyVarSet (tyVarsOfInsts insts))
2980 = vcat [ addInstLoc insts $
2981 sep [ ptext (sLit "Could not deduce") <+> pprDictsTheta insts
2982 , nest 2 $ ptext (sLit "from the context") <+> pprDictsTheta givens]
2983 , show_fixes (fix1 loc : fixes2) ]
2985 | otherwise -- Top level
2986 = vcat [ addInstLoc insts $
2987 ptext (sLit "No instance") <> plural insts
2988 <+> ptext (sLit "for") <+> pprDictsTheta insts
2989 , show_fixes fixes2 ]
2992 fix1 loc = sep [ ptext (sLit "add") <+> pprDictsTheta insts
2993 <+> ptext (sLit "to the context of"),
2994 nest 2 (ppr (instLocOrigin loc)) ]
2995 -- I'm not sure it helps to add the location
2996 -- nest 2 (ptext (sLit "at") <+> ppr (instLocSpan loc)) ]
2998 fixes2 | null instance_dicts = []
2999 | otherwise = [sep [ptext (sLit "add an instance declaration for"),
3000 pprDictsTheta instance_dicts]]
3001 instance_dicts = [d | d <- insts, isClassDict d, not (isTyVarDict d)]
3002 -- Insts for which it is worth suggesting an adding an instance declaration
3003 -- Exclude implicit parameters, and tyvar dicts
3005 show_fixes :: [SDoc] -> SDoc
3006 show_fixes [] = empty
3007 show_fixes (f:fs) = sep [ptext (sLit "Possible fix:"),
3008 nest 2 (vcat (f : map (ptext (sLit "or") <+>) fs))]
3010 addTopAmbigErrs :: [Inst] -> TcRn ()
3011 addTopAmbigErrs dicts
3012 -- Divide into groups that share a common set of ambiguous tyvars
3013 = ifErrsM (return ()) $ -- Only report ambiguity if no other errors happened
3014 -- See Note [Avoiding spurious errors]
3015 mapM_ report (equivClasses cmp [(d, tvs_of d) | d <- tidy_dicts])
3017 (tidy_env, tidy_dicts) = tidyInsts dicts
3019 tvs_of :: Inst -> [TcTyVar]
3020 tvs_of d = varSetElems (tyVarsOfInst d)
3021 cmp (_,tvs1) (_,tvs2) = tvs1 `compare` tvs2
3023 report :: [(Inst,[TcTyVar])] -> TcM ()
3024 report pairs@((inst,tvs) : _) = do -- The pairs share a common set of ambiguous tyvars
3025 (tidy_env, mono_msg) <- mkMonomorphismMsg tidy_env tvs
3026 setSrcSpan (instSpan inst) $
3027 -- the location of the first one will do for the err message
3028 addErrTcM (tidy_env, msg $$ mono_msg)
3030 dicts = map fst pairs
3031 msg = sep [text "Ambiguous type variable" <> plural tvs <+>
3032 pprQuotedList tvs <+> in_msg,
3033 nest 2 (pprDictsInFull dicts)]
3034 in_msg = text "in the constraint" <> plural dicts <> colon
3035 report [] = panic "addTopAmbigErrs"
3038 mkMonomorphismMsg :: TidyEnv -> [TcTyVar] -> TcM (TidyEnv, Message)
3039 -- There's an error with these Insts; if they have free type variables
3040 -- it's probably caused by the monomorphism restriction.
3041 -- Try to identify the offending variable
3042 -- ASSUMPTION: the Insts are fully zonked
3043 mkMonomorphismMsg tidy_env inst_tvs
3044 = do { dflags <- getDOpts
3045 ; (tidy_env, docs) <- findGlobals (mkVarSet inst_tvs) tidy_env
3046 ; return (tidy_env, mk_msg dflags docs) }
3048 mk_msg _ _ | any isRuntimeUnk inst_tvs
3049 = vcat [ptext (sLit "Cannot resolve unknown runtime types:") <+>
3050 (pprWithCommas ppr inst_tvs),
3051 ptext (sLit "Use :print or :force to determine these types")]
3052 mk_msg _ [] = ptext (sLit "Probable fix: add a type signature that fixes these type variable(s)")
3053 -- This happens in things like
3054 -- f x = show (read "foo")
3055 -- where monomorphism doesn't play any role
3057 = vcat [ptext (sLit "Possible cause: the monomorphism restriction applied to the following:"),
3059 monomorphism_fix dflags]
3061 monomorphism_fix :: DynFlags -> SDoc
3062 monomorphism_fix dflags
3063 = ptext (sLit "Probable fix:") <+> vcat
3064 [ptext (sLit "give these definition(s) an explicit type signature"),
3065 if dopt Opt_MonomorphismRestriction dflags
3066 then ptext (sLit "or use -XNoMonomorphismRestriction")
3067 else empty] -- Only suggest adding "-XNoMonomorphismRestriction"
3068 -- if it is not already set!
3070 warnDefault :: [(Inst, Class, Var)] -> Type -> TcM ()
3071 warnDefault ups default_ty = do
3072 warn_flag <- doptM Opt_WarnTypeDefaults
3073 addInstCtxt (instLoc (head (dicts))) (warnTc warn_flag warn_msg)
3075 dicts = [d | (d,_,_) <- ups]
3078 (_, tidy_dicts) = tidyInsts dicts
3079 warn_msg = vcat [ptext (sLit "Defaulting the following constraint(s) to type") <+>
3080 quotes (ppr default_ty),
3081 pprDictsInFull tidy_dicts]
3083 reduceDepthErr :: Int -> [Inst] -> SDoc
3084 reduceDepthErr n stack
3085 = vcat [ptext (sLit "Context reduction stack overflow; size =") <+> int n,
3086 ptext (sLit "Use -fcontext-stack=N to increase stack size to N"),
3087 nest 4 (pprStack stack)]
3089 pprStack :: [Inst] -> SDoc
3090 pprStack stack = vcat (map pprInstInFull stack)