{- (c) The University of Glasgow 2006 (c) The GRASP/AQUA Project, Glasgow University, 1992-1998 TcInstDecls: Typechecking instance declarations -} {-# LANGUAGE CPP #-} module TcInstDcls ( tcInstDecls1, tcInstDecls2 ) where #include "HsVersions.h" import HsSyn import TcBinds import TcTyClsDecls import TcClassDcl( tcClassDecl2, tcATDefault, HsSigFun, mkHsSigFun, findMethodBind, instantiateMethod ) import TcPat ( addInlinePrags, emptyPragEnv ) import TcRnMonad import TcValidity import TcHsSyn ( zonkTcTypeToTypes, emptyZonkEnv ) import TcMType import TcType import BuildTyCl import Inst import InstEnv import FamInst import FamInstEnv import TcDeriv import TcEnv import TcHsType import TcUnify import CoreSyn ( Expr(..), mkApps, mkVarApps, mkLams ) import MkCore ( nO_METHOD_BINDING_ERROR_ID ) import CoreUnfold ( mkInlineUnfolding, mkDFunUnfolding ) import Type import TcEvidence import TyCon import CoAxiom import DataCon import Class import Var import VarEnv import VarSet import PrelNames ( typeableClassName, genericClassNames ) import Bag import BasicTypes import DynFlags import ErrUtils import FastString import HscTypes ( isHsBootOrSig ) import Id import MkId import Name import NameSet import Outputable import SrcLoc import Util import BooleanFormula ( isUnsatisfied, pprBooleanFormulaNice ) import qualified GHC.LanguageExtensions as LangExt import Control.Monad import Maybes import Data.List ( partition ) {- Typechecking instance declarations is done in two passes. The first pass, made by @tcInstDecls1@, collects information to be used in the second pass. This pre-processed info includes the as-yet-unprocessed bindings inside the instance declaration. These are type-checked in the second pass, when the class-instance envs and GVE contain all the info from all the instance and value decls. Indeed that's the reason we need two passes over the instance decls. Note [How instance declarations are translated] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Here is how we translate instance declarations into Core Running example: class C a where op1, op2 :: Ix b => a -> b -> b op2 = <dm-rhs> instance C a => C [a] {-# INLINE [2] op1 #-} op1 = <rhs> ===> -- Method selectors op1,op2 :: forall a. C a => forall b. Ix b => a -> b -> b op1 = ... op2 = ... -- Default methods get the 'self' dictionary as argument -- so they can call other methods at the same type -- Default methods get the same type as their method selector $dmop2 :: forall a. C a => forall b. Ix b => a -> b -> b $dmop2 = /\a. \(d:C a). /\b. \(d2: Ix b). <dm-rhs> -- NB: type variables 'a' and 'b' are *both* in scope in <dm-rhs> -- Note [Tricky type variable scoping] -- A top-level definition for each instance method -- Here op1_i, op2_i are the "instance method Ids" -- The INLINE pragma comes from the user pragma {-# INLINE [2] op1_i #-} -- From the instance decl bindings op1_i, op2_i :: forall a. C a => forall b. Ix b => [a] -> b -> b op1_i = /\a. \(d:C a). let this :: C [a] this = df_i a d -- Note [Subtle interaction of recursion and overlap] local_op1 :: forall b. Ix b => [a] -> b -> b local_op1 = <rhs> -- Source code; run the type checker on this -- NB: Type variable 'a' (but not 'b') is in scope in <rhs> -- Note [Tricky type variable scoping] in local_op1 a d op2_i = /\a \d:C a. $dmop2 [a] (df_i a d) -- The dictionary function itself {-# NOINLINE CONLIKE df_i #-} -- Never inline dictionary functions df_i :: forall a. C a -> C [a] df_i = /\a. \d:C a. MkC (op1_i a d) (op2_i a d) -- But see Note [Default methods in instances] -- We can't apply the type checker to the default-method call -- Use a RULE to short-circuit applications of the class ops {-# RULE "op1@C[a]" forall a, d:C a. op1 [a] (df_i d) = op1_i a d #-} Note [Instances and loop breakers] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * Note that df_i may be mutually recursive with both op1_i and op2_i. It's crucial that df_i is not chosen as the loop breaker, even though op1_i has a (user-specified) INLINE pragma. * Instead the idea is to inline df_i into op1_i, which may then select methods from the MkC record, and thereby break the recursion with df_i, leaving a *self*-recurisve op1_i. (If op1_i doesn't call op at the same type, it won't mention df_i, so there won't be recursion in the first place.) * If op1_i is marked INLINE by the user there's a danger that we won't inline df_i in it, and that in turn means that (since it'll be a loop-breaker because df_i isn't), op1_i will ironically never be inlined. But this is OK: the recursion breaking happens by way of a RULE (the magic ClassOp rule above), and RULES work inside InlineRule unfoldings. See Note [RULEs enabled in SimplGently] in SimplUtils Note [ClassOp/DFun selection] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ One thing we see a lot is stuff like op2 (df d1 d2) where 'op2' is a ClassOp and 'df' is DFun. Now, we could inline *both* 'op2' and 'df' to get case (MkD ($cop1 d1 d2) ($cop2 d1 d2) ... of MkD _ op2 _ _ _ -> op2 And that will reduce to ($cop2 d1 d2) which is what we wanted. But it's tricky to make this work in practice, because it requires us to inline both 'op2' and 'df'. But neither is keen to inline without having seen the other's result; and it's very easy to get code bloat (from the big intermediate) if you inline a bit too much. Instead we use a cunning trick. * We arrange that 'df' and 'op2' NEVER inline. * We arrange that 'df' is ALWAYS defined in the sylised form df d1 d2 = MkD ($cop1 d1 d2) ($cop2 d1 d2) ... * We give 'df' a magical unfolding (DFunUnfolding [$cop1, $cop2, ..]) that lists its methods. * We make CoreUnfold.exprIsConApp_maybe spot a DFunUnfolding and return a suitable constructor application -- inlining df "on the fly" as it were. * ClassOp rules: We give the ClassOp 'op2' a BuiltinRule that extracts the right piece iff its argument satisfies exprIsConApp_maybe. This is done in MkId mkDictSelId * We make 'df' CONLIKE, so that shared uses still match; eg let d = df d1 d2 in ...(op2 d)...(op1 d)... Note [Single-method classes] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ If the class has just one method (or, more accurately, just one element of {superclasses + methods}), then we use a different strategy. class C a where op :: a -> a instance C a => C [a] where op = <blah> We translate the class decl into a newtype, which just gives a top-level axiom. The "constructor" MkC expands to a cast, as does the class-op selector. axiom Co:C a :: C a ~ (a->a) op :: forall a. C a -> (a -> a) op a d = d |> (Co:C a) MkC :: forall a. (a->a) -> C a MkC = /\a.\op. op |> (sym Co:C a) The clever RULE stuff doesn't work now, because ($df a d) isn't a constructor application, so exprIsConApp_maybe won't return Just <blah>. Instead, we simply rely on the fact that casts are cheap: $df :: forall a. C a => C [a] {-# INLINE df #-} -- NB: INLINE this $df = /\a. \d. MkC [a] ($cop_list a d) = $cop_list |> forall a. C a -> (sym (Co:C [a])) $cop_list :: forall a. C a => [a] -> [a] $cop_list = <blah> So if we see (op ($df a d)) we'll inline 'op' and '$df', since both are simply casts, and good things happen. Why do we use this different strategy? Because otherwise we end up with non-inlined dictionaries that look like $df = $cop |> blah which adds an extra indirection to every use, which seems stupid. See Trac #4138 for an example (although the regression reported there wasn't due to the indirection). There is an awkward wrinkle though: we want to be very careful when we have instance C a => C [a] where {-# INLINE op #-} op = ... then we'll get an INLINE pragma on $cop_list but it's important that $cop_list only inlines when it's applied to *two* arguments (the dictionary and the list argument). So we must not eta-expand $df above. We ensure that this doesn't happen by putting an INLINE pragma on the dfun itself; after all, it ends up being just a cast. There is one more dark corner to the INLINE story, even more deeply buried. Consider this (Trac #3772): class DeepSeq a => C a where gen :: Int -> a instance C a => C [a] where gen n = ... class DeepSeq a where deepSeq :: a -> b -> b instance DeepSeq a => DeepSeq [a] where {-# INLINE deepSeq #-} deepSeq xs b = foldr deepSeq b xs That gives rise to these defns: $cdeepSeq :: DeepSeq a -> [a] -> b -> b -- User INLINE( 3 args )! $cdeepSeq a (d:DS a) b (x:[a]) (y:b) = ... $fDeepSeq[] :: DeepSeq a -> DeepSeq [a] -- DFun (with auto INLINE pragma) $fDeepSeq[] a d = $cdeepSeq a d |> blah $cp1 a d :: C a => DeepSep [a] -- We don't want to eta-expand this, lest -- $cdeepSeq gets inlined in it! $cp1 a d = $fDeepSep[] a (scsel a d) $fC[] :: C a => C [a] -- Ordinary DFun $fC[] a d = MkC ($cp1 a d) ($cgen a d) Here $cp1 is the code that generates the superclass for C [a]. The issue is this: we must not eta-expand $cp1 either, or else $fDeepSeq[] and then $cdeepSeq will inline there, which is definitely wrong. Like on the dfun, we solve this by adding an INLINE pragma to $cp1. Note [Subtle interaction of recursion and overlap] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider this class C a where { op1,op2 :: a -> a } instance C a => C [a] where op1 x = op2 x ++ op2 x op2 x = ... instance C [Int] where ... When type-checking the C [a] instance, we need a C [a] dictionary (for the call of op2). If we look up in the instance environment, we find an overlap. And in *general* the right thing is to complain (see Note [Overlapping instances] in InstEnv). But in *this* case it's wrong to complain, because we just want to delegate to the op2 of this same instance. Why is this justified? Because we generate a (C [a]) constraint in a context in which 'a' cannot be instantiated to anything that matches other overlapping instances, or else we would not be executing this version of op1 in the first place. It might even be a bit disguised: nullFail :: C [a] => [a] -> [a] nullFail x = op2 x ++ op2 x instance C a => C [a] where op1 x = nullFail x Precisely this is used in package 'regex-base', module Context.hs. See the overlapping instances for RegexContext, and the fact that they call 'nullFail' just like the example above. The DoCon package also does the same thing; it shows up in module Fraction.hs. Conclusion: when typechecking the methods in a C [a] instance, we want to treat the 'a' as an *existential* type variable, in the sense described by Note [Binding when looking up instances]. That is why isOverlappableTyVar responds True to an InstSkol, which is the kind of skolem we use in tcInstDecl2. Note [Tricky type variable scoping] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In our example class C a where op1, op2 :: Ix b => a -> b -> b op2 = <dm-rhs> instance C a => C [a] {-# INLINE [2] op1 #-} op1 = <rhs> note that 'a' and 'b' are *both* in scope in <dm-rhs>, but only 'a' is in scope in <rhs>. In particular, we must make sure that 'b' is in scope when typechecking <dm-rhs>. This is achieved by subFunTys, which brings appropriate tyvars into scope. This happens for both <dm-rhs> and for <rhs>, but that doesn't matter: the *renamer* will have complained if 'b' is mentioned in <rhs>. ************************************************************************ * * \subsection{Extracting instance decls} * * ************************************************************************ Gather up the instance declarations from their various sources -} tcInstDecls1 -- Deal with both source-code and imported instance decls :: [TyClGroup Name] -- For deriving stuff -> [LInstDecl Name] -- Source code instance decls -> [LDerivDecl Name] -- Source code stand-alone deriving decls -> TcM (TcGblEnv, -- The full inst env [InstInfo Name], -- Source-code instance decls to process; -- contains all dfuns for this module HsValBinds Name) -- Supporting bindings for derived instances tcInstDecls1 tycl_decls inst_decls deriv_decls = checkNoErrs $ do { -- Stop if addInstInfos etc discovers any errors -- (they recover, so that we get more than one error each -- round) -- Do class and family instance declarations ; stuff <- mapAndRecoverM tcLocalInstDecl inst_decls ; let (local_infos_s, fam_insts_s, datafam_deriv_infos) = unzip3 stuff fam_insts = concat fam_insts_s local_infos' = concat local_infos_s -- Handwritten instances of the poly-kinded Typeable class are -- forbidden, so we handle those separately (typeable_instances, local_infos) = partition bad_typeable_instance local_infos' ; addClsInsts local_infos $ addFamInsts fam_insts $ do { -- Compute instances from "deriving" clauses; -- This stuff computes a context for the derived instance -- decl, so it needs to know about all the instances possible -- NB: class instance declarations can contain derivings as -- part of associated data type declarations failIfErrsM -- If the addInsts stuff gave any errors, don't -- try the deriving stuff, because that may give -- more errors still ; traceTc "tcDeriving" Outputable.empty ; th_stage <- getStage -- See Note [Deriving inside TH brackets ] ; (gbl_env, deriv_inst_info, deriv_binds) <- if isBrackStage th_stage then do { gbl_env <- getGblEnv ; return (gbl_env, emptyBag, emptyValBindsOut) } else do { data_deriv_infos <- mkDerivInfos tycl_decls ; let deriv_infos = concat datafam_deriv_infos ++ data_deriv_infos ; tcDeriving deriv_infos deriv_decls } -- Fail if there are any handwritten instance of poly-kinded Typeable ; mapM_ typeable_err typeable_instances -- Check that if the module is compiled with -XSafe, there are no -- hand written instances of old Typeable as then unsafe casts could be -- performed. Derived instances are OK. ; dflags <- getDynFlags ; when (safeLanguageOn dflags) $ forM_ local_infos $ \x -> case x of _ | genInstCheck x -> addErrAt (getSrcSpan $ iSpec x) (genInstErr x) _ -> return () -- As above but for Safe Inference mode. ; when (safeInferOn dflags) $ forM_ local_infos $ \x -> case x of _ | genInstCheck x -> recordUnsafeInfer emptyBag _ -> return () ; return ( gbl_env , bagToList deriv_inst_info ++ local_infos , deriv_binds ) }} where -- Separate the Typeable instances from the rest bad_typeable_instance i = typeableClassName == is_cls_nm (iSpec i) -- Check for hand-written Generic instances (disallowed in Safe Haskell) genInstCheck ty = is_cls_nm (iSpec ty) `elem` genericClassNames genInstErr i = hang (text ("Generic instances can only be " ++ "derived in Safe Haskell.") $+$ text "Replace the following instance:") 2 (pprInstanceHdr (iSpec i)) -- Report an error or a warning for a Typeable instances. -- If we are working on an .hs-boot file, we just report a warning, -- and ignore the instance. We do this, to give users a chance to fix -- their code. typeable_err i = setSrcSpan (getSrcSpan (iSpec i)) $ do env <- getGblEnv if isHsBootOrSig (tcg_src env) then do warn <- woptM Opt_WarnDerivingTypeable when warn $ addWarnTc (Reason Opt_WarnDerivingTypeable) $ vcat [ ppTypeable <+> text "instances in .hs-boot files are ignored" , text "This warning will become an error in future versions of the compiler" ] else addErrTc $ text "Class" <+> ppTypeable <+> text "does not support user-specified instances" ppTypeable :: SDoc ppTypeable = quotes (ppr typeableClassName) addClsInsts :: [InstInfo Name] -> TcM a -> TcM a addClsInsts infos thing_inside = tcExtendLocalInstEnv (map iSpec infos) thing_inside addFamInsts :: [FamInst] -> TcM a -> TcM a -- Extend (a) the family instance envt -- (b) the type envt with stuff from data type decls addFamInsts fam_insts thing_inside = tcExtendLocalFamInstEnv fam_insts $ tcExtendGlobalEnv axioms $ tcExtendTyConEnv data_rep_tycons $ do { traceTc "addFamInsts" (pprFamInsts fam_insts) ; tcg_env <- tcAddImplicits data_rep_tycons -- Does not add its axiom; that comes from -- adding the 'axioms' above ; setGblEnv tcg_env thing_inside } where axioms = map (ACoAxiom . toBranchedAxiom . famInstAxiom) fam_insts data_rep_tycons = famInstsRepTyCons fam_insts -- The representation tycons for 'data instances' declarations {- Note [Deriving inside TH brackets] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Given a declaration bracket [d| data T = A | B deriving( Show ) |] there is really no point in generating the derived code for deriving( Show) and then type-checking it. This will happen at the call site anyway, and the type check should never fail! Moreover (Trac #6005) the scoping of the generated code inside the bracket does not seem to work out. The easy solution is simply not to generate the derived instances at all. (A less brutal solution would be to generate them with no bindings.) This will become moot when we shift to the new TH plan, so the brutal solution will do. -} tcLocalInstDecl :: LInstDecl Name -> TcM ([InstInfo Name], [FamInst], [DerivInfo]) -- A source-file instance declaration -- Type-check all the stuff before the "where" -- -- We check for respectable instance type, and context tcLocalInstDecl (L loc (TyFamInstD { tfid_inst = decl })) = do { fam_inst <- tcTyFamInstDecl Nothing (L loc decl) ; return ([], [fam_inst], []) } tcLocalInstDecl (L loc (DataFamInstD { dfid_inst = decl })) = do { (fam_inst, m_deriv_info) <- tcDataFamInstDecl Nothing (L loc decl) ; return ([], [fam_inst], maybeToList m_deriv_info) } tcLocalInstDecl (L loc (ClsInstD { cid_inst = decl })) = do { (insts, fam_insts, deriv_infos) <- tcClsInstDecl (L loc decl) ; return (insts, fam_insts, deriv_infos) } tcClsInstDecl :: LClsInstDecl Name -> TcM ([InstInfo Name], [FamInst], [DerivInfo]) -- the returned DerivInfos are for any associated data families tcClsInstDecl (L loc (ClsInstDecl { cid_poly_ty = poly_ty, cid_binds = binds , cid_sigs = uprags, cid_tyfam_insts = ats , cid_overlap_mode = overlap_mode , cid_datafam_insts = adts })) = setSrcSpan loc $ addErrCtxt (instDeclCtxt1 poly_ty) $ do { is_boot <- tcIsHsBootOrSig ; checkTc (not is_boot || (isEmptyLHsBinds binds && null uprags)) badBootDeclErr ; (tyvars, theta, clas, inst_tys) <- tcHsClsInstType InstDeclCtxt poly_ty ; let mini_env = mkVarEnv (classTyVars clas `zip` inst_tys) mini_subst = mkTvSubst (mkInScopeSet (mkVarSet tyvars)) mini_env mb_info = Just (clas, mini_env) -- Next, process any associated types. ; traceTc "tcLocalInstDecl" (ppr poly_ty) ; tyfam_insts0 <- tcExtendTyVarEnv tyvars $ mapAndRecoverM (tcTyFamInstDecl mb_info) ats ; datafam_stuff <- tcExtendTyVarEnv tyvars $ mapAndRecoverM (tcDataFamInstDecl mb_info) adts ; let (datafam_insts, m_deriv_infos) = unzip datafam_stuff deriv_infos = catMaybes m_deriv_infos -- Check for missing associated types and build them -- from their defaults (if available) ; let defined_ats = mkNameSet (map (tyFamInstDeclName . unLoc) ats) `unionNameSet` mkNameSet (map (unLoc . dfid_tycon . unLoc) adts) ; tyfam_insts1 <- mapM (tcATDefault True loc mini_subst defined_ats) (classATItems clas) -- Finally, construct the Core representation of the instance. -- (This no longer includes the associated types.) ; dfun_name <- newDFunName clas inst_tys (getLoc (hsSigType poly_ty)) -- Dfun location is that of instance *header* ; ispec <- newClsInst (fmap unLoc overlap_mode) dfun_name tyvars theta clas inst_tys ; let inst_info = InstInfo { iSpec = ispec , iBinds = InstBindings { ib_binds = binds , ib_tyvars = map Var.varName tyvars -- Scope over bindings , ib_pragmas = uprags , ib_extensions = [] , ib_derived = False } } ; return ( [inst_info], tyfam_insts0 ++ concat tyfam_insts1 ++ datafam_insts , deriv_infos ) } {- ************************************************************************ * * Type checking family instances * * ************************************************************************ Family instances are somewhat of a hybrid. They are processed together with class instance heads, but can contain data constructors and hence they share a lot of kinding and type checking code with ordinary algebraic data types (and GADTs). -} tcFamInstDeclCombined :: Maybe ClsInfo -> Located Name -> TcM TyCon tcFamInstDeclCombined mb_clsinfo fam_tc_lname = do { -- Type family instances require -XTypeFamilies -- and can't (currently) be in an hs-boot file ; traceTc "tcFamInstDecl" (ppr fam_tc_lname) ; type_families <- xoptM LangExt.TypeFamilies ; is_boot <- tcIsHsBootOrSig -- Are we compiling an hs-boot file? ; checkTc type_families $ badFamInstDecl fam_tc_lname ; checkTc (not is_boot) $ badBootFamInstDeclErr -- Look up the family TyCon and check for validity including -- check that toplevel type instances are not for associated types. ; fam_tc <- tcLookupLocatedTyCon fam_tc_lname ; when (isNothing mb_clsinfo && -- Not in a class decl isTyConAssoc fam_tc) -- but an associated type (addErr $ assocInClassErr fam_tc_lname) ; return fam_tc } tcTyFamInstDecl :: Maybe ClsInfo -> LTyFamInstDecl Name -> TcM FamInst -- "type instance" tcTyFamInstDecl mb_clsinfo (L loc decl@(TyFamInstDecl { tfid_eqn = eqn })) = setSrcSpan loc $ tcAddTyFamInstCtxt decl $ do { let fam_lname = tfe_tycon (unLoc eqn) ; fam_tc <- tcFamInstDeclCombined mb_clsinfo fam_lname -- (0) Check it's an open type family ; checkTc (isFamilyTyCon fam_tc) (notFamily fam_tc) ; checkTc (isTypeFamilyTyCon fam_tc) (wrongKindOfFamily fam_tc) ; checkTc (isOpenTypeFamilyTyCon fam_tc) (notOpenFamily fam_tc) -- (1) do the work of verifying the synonym group ; co_ax_branch <- tcTyFamInstEqn (famTyConShape fam_tc) mb_clsinfo eqn -- (2) check for validity ; checkValidCoAxBranch mb_clsinfo fam_tc co_ax_branch -- (3) construct coercion axiom ; rep_tc_name <- newFamInstAxiomName fam_lname [coAxBranchLHS co_ax_branch] ; let axiom = mkUnbranchedCoAxiom rep_tc_name fam_tc co_ax_branch ; newFamInst SynFamilyInst axiom } tcDataFamInstDecl :: Maybe ClsInfo -> LDataFamInstDecl Name -> TcM (FamInst, Maybe DerivInfo) -- "newtype instance" and "data instance" tcDataFamInstDecl mb_clsinfo (L loc decl@(DataFamInstDecl { dfid_pats = pats , dfid_tycon = fam_tc_name , dfid_defn = defn@HsDataDefn { dd_ND = new_or_data, dd_cType = cType , dd_ctxt = ctxt, dd_cons = cons , dd_derivs = derivs } })) = setSrcSpan loc $ tcAddDataFamInstCtxt decl $ do { fam_tc <- tcFamInstDeclCombined mb_clsinfo fam_tc_name -- Check that the family declaration is for the right kind ; checkTc (isFamilyTyCon fam_tc) (notFamily fam_tc) ; checkTc (isDataFamilyTyCon fam_tc) (wrongKindOfFamily fam_tc) -- Kind check type patterns ; tcFamTyPats (famTyConShape fam_tc) mb_clsinfo pats (kcDataDefn (unLoc fam_tc_name) pats defn) $ \tvs' pats' res_kind -> do { -- Check that left-hand sides are ok (mono-types, no type families, -- consistent instantiations, etc) ; checkValidFamPats mb_clsinfo fam_tc tvs' [] pats' -- Result kind must be '*' (otherwise, we have too few patterns) ; checkTc (isLiftedTypeKind res_kind) $ tooFewParmsErr (tyConArity fam_tc) ; stupid_theta <- solveEqualities $ tcHsContext ctxt ; stupid_theta <- zonkTcTypeToTypes emptyZonkEnv stupid_theta ; gadt_syntax <- dataDeclChecks (tyConName fam_tc) new_or_data stupid_theta cons -- Construct representation tycon ; rep_tc_name <- newFamInstTyConName fam_tc_name pats' ; axiom_name <- newFamInstAxiomName fam_tc_name [pats'] ; let (eta_pats, etad_tvs) = eta_reduce pats' eta_tvs = filterOut (`elem` etad_tvs) tvs' full_tvs = eta_tvs ++ etad_tvs -- Put the eta-removed tyvars at the end -- Remember, tvs' is in arbitrary order (except kind vars are -- first, so there is no reason to suppose that the etad_tvs -- (obtained from the pats) are at the end (Trac #11148) orig_res_ty = mkTyConApp fam_tc pats' ; (rep_tc, axiom) <- fixM $ \ ~(rec_rep_tc, _) -> do { let ty_binders = mkTyBindersPreferAnon full_tvs liftedTypeKind ; data_cons <- tcConDecls rec_rep_tc (full_tvs, ty_binders, orig_res_ty) cons ; tc_rhs <- case new_or_data of DataType -> return (mkDataTyConRhs data_cons) NewType -> ASSERT( not (null data_cons) ) mkNewTyConRhs rep_tc_name rec_rep_tc (head data_cons) -- freshen tyvars ; let axiom = mkSingleCoAxiom Representational axiom_name eta_tvs [] fam_tc eta_pats (mkTyConApp rep_tc (mkTyVarTys eta_tvs)) parent = DataFamInstTyCon axiom fam_tc pats' -- NB: Use the full_tvs from the pats. See bullet toward -- the end of Note [Data type families] in TyCon rep_tc = mkAlgTyCon rep_tc_name ty_binders liftedTypeKind full_tvs (map (const Nominal) full_tvs) (fmap unLoc cType) stupid_theta tc_rhs parent Recursive gadt_syntax -- We always assume that indexed types are recursive. Why? -- (1) Due to their open nature, we can never be sure that a -- further instance might not introduce a new recursive -- dependency. (2) They are always valid loop breakers as -- they involve a coercion. ; return (rep_tc, axiom) } -- Remember to check validity; no recursion to worry about here ; checkValidTyCon rep_tc ; let m_deriv_info = case derivs of Nothing -> Nothing Just (L _ preds) -> Just $ DerivInfo { di_rep_tc = rep_tc , di_preds = preds , di_ctxt = tcMkDataFamInstCtxt decl } ; fam_inst <- newFamInst (DataFamilyInst rep_tc) axiom ; return (fam_inst, m_deriv_info) } } where eta_reduce :: [Type] -> ([Type], [TyVar]) -- See Note [Eta reduction for data families] in FamInstEnv -- Splits the incoming patterns into two: the [TyVar] -- are the patterns that can be eta-reduced away. -- e.g. T [a] Int a d c ==> (T [a] Int a, [d,c]) -- -- NB: quadratic algorithm, but types are small here eta_reduce pats = go (reverse pats) [] go (pat:pats) etad_tvs | Just tv <- getTyVar_maybe pat , not (tv `elemVarSet` tyCoVarsOfTypes pats) = go pats (tv : etad_tvs) go pats etad_tvs = (reverse pats, etad_tvs) {- ********************************************************************* * * Type-checking instance declarations, pass 2 * * ********************************************************************* -} tcInstDecls2 :: [LTyClDecl Name] -> [InstInfo Name] -> TcM (LHsBinds Id) -- (a) From each class declaration, -- generate any default-method bindings -- (b) From each instance decl -- generate the dfun binding tcInstDecls2 tycl_decls inst_decls = do { -- (a) Default methods from class decls let class_decls = filter (isClassDecl . unLoc) tycl_decls ; dm_binds_s <- mapM tcClassDecl2 class_decls ; let dm_binds = unionManyBags dm_binds_s -- (b) instance declarations ; let dm_ids = collectHsBindsBinders dm_binds -- Add the default method Ids (again) -- See Note [Default methods in the type environment] ; inst_binds_s <- tcExtendLetEnv TopLevel dm_ids $ mapM tcInstDecl2 inst_decls -- Done ; return (dm_binds `unionBags` unionManyBags inst_binds_s) } {- Note [Default methods in the type environment] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The default method Ids are already in the type environment (see Note [Default method Ids and Template Haskell] in TcTyClsDcls), BUT they don't have their InlinePragmas yet. Usually that would not matter, because the simplifier propagates information from binding site to use. But, unusually, when compiling instance decls we *copy* the INLINE pragma from the default method to the method for that particular operation (see Note [INLINE and default methods] below). So right here in tcInstDecls2 we must re-extend the type envt with the default method Ids replete with their INLINE pragmas. Urk. -} tcInstDecl2 :: InstInfo Name -> TcM (LHsBinds Id) -- Returns a binding for the dfun tcInstDecl2 (InstInfo { iSpec = ispec, iBinds = ibinds }) = recoverM (return emptyLHsBinds) $ setSrcSpan loc $ addErrCtxt (instDeclCtxt2 (idType dfun_id)) $ do { -- Instantiate the instance decl with skolem constants ; (inst_tyvars, dfun_theta, inst_head) <- tcSkolDFunType (idType dfun_id) ; dfun_ev_vars <- newEvVars dfun_theta -- We instantiate the dfun_id with superSkolems. -- See Note [Subtle interaction of recursion and overlap] -- and Note [Binding when looking up instances] ; let (clas, inst_tys) = tcSplitDFunHead inst_head (class_tyvars, sc_theta, _, op_items) = classBigSig clas sc_theta' = substTheta (zipTvSubst class_tyvars inst_tys) sc_theta ; traceTc "tcInstDecl2" (vcat [ppr inst_tyvars, ppr inst_tys, ppr dfun_theta, ppr sc_theta']) -- Deal with 'SPECIALISE instance' pragmas -- See Note [SPECIALISE instance pragmas] ; spec_inst_info@(spec_inst_prags,_) <- tcSpecInstPrags dfun_id ibinds -- Typecheck superclasses and methods -- See Note [Typechecking plan for instance declarations] ; dfun_ev_binds_var <- newTcEvBinds ; let dfun_ev_binds = TcEvBinds dfun_ev_binds_var ; ((sc_meth_ids, sc_meth_binds, sc_meth_implics), tclvl) <- pushTcLevelM $ do { fam_envs <- tcGetFamInstEnvs ; (sc_ids, sc_binds, sc_implics) <- tcSuperClasses dfun_id clas inst_tyvars dfun_ev_vars inst_tys dfun_ev_binds fam_envs sc_theta' -- Typecheck the methods ; (meth_ids, meth_binds, meth_implics) <- tcMethods dfun_id clas inst_tyvars dfun_ev_vars inst_tys dfun_ev_binds spec_inst_info op_items ibinds ; return ( sc_ids ++ meth_ids , sc_binds `unionBags` meth_binds , sc_implics `unionBags` meth_implics ) } ; env <- getLclEnv ; emitImplication $ Implic { ic_tclvl = tclvl , ic_skols = inst_tyvars , ic_no_eqs = False , ic_given = dfun_ev_vars , ic_wanted = mkImplicWC sc_meth_implics , ic_status = IC_Unsolved , ic_binds = Just dfun_ev_binds_var , ic_env = env , ic_info = InstSkol } -- Create the result bindings ; self_dict <- newDict clas inst_tys ; let class_tc = classTyCon clas [dict_constr] = tyConDataCons class_tc dict_bind = mkVarBind self_dict (L loc con_app_args) -- We don't produce a binding for the dict_constr; instead we -- rely on the simplifier to unfold this saturated application -- We do this rather than generate an HsCon directly, because -- it means that the special cases (e.g. dictionary with only one -- member) are dealt with by the common MkId.mkDataConWrapId -- code rather than needing to be repeated here. -- con_app_tys = MkD ty1 ty2 -- con_app_scs = MkD ty1 ty2 sc1 sc2 -- con_app_args = MkD ty1 ty2 sc1 sc2 op1 op2 con_app_tys = wrapId (mkWpTyApps inst_tys) (dataConWrapId dict_constr) -- NB: We *can* have covars in inst_tys, in the case of -- promoted GADT constructors. con_app_args = foldl app_to_meth con_app_tys sc_meth_ids app_to_meth :: HsExpr Id -> Id -> HsExpr Id app_to_meth fun meth_id = L loc fun `HsApp` L loc (wrapId arg_wrapper meth_id) inst_tv_tys = mkTyVarTys inst_tyvars arg_wrapper = mkWpEvVarApps dfun_ev_vars <.> mkWpTyApps inst_tv_tys is_newtype = isNewTyCon class_tc dfun_id_w_prags = addDFunPrags dfun_id dict_constr is_newtype inst_tyvars dfun_ev_vars inst_tys sc_meth_ids dfun_spec_prags | is_newtype = SpecPrags [] | otherwise = SpecPrags spec_inst_prags -- Newtype dfuns just inline unconditionally, -- so don't attempt to specialise them export = ABE { abe_wrap = idHsWrapper , abe_poly = dfun_id_w_prags , abe_mono = self_dict , abe_prags = dfun_spec_prags } -- NB: see Note [SPECIALISE instance pragmas] main_bind = AbsBinds { abs_tvs = inst_tyvars , abs_ev_vars = dfun_ev_vars , abs_exports = [export] , abs_ev_binds = [] , abs_binds = unitBag dict_bind } ; return (unitBag (L loc main_bind) `unionBags` sc_meth_binds) } where dfun_id = instanceDFunId ispec loc = getSrcSpan dfun_id addDFunPrags :: DFunId -> DataCon -> Bool -> [TyVar] -> [Id] -> [Type] -> [Id] -> DFunId -- DFuns need a special Unfolding and InlinePrag -- See Note [ClassOp/DFun selection] -- and Note [Single-method classes] -- It's easiest to create those unfoldings right here, where -- have all the pieces in hand, even though we are messing with -- Core at this point, which the typechecker doesn't usually do addDFunPrags dfun_id dict_con is_newtype dfun_tvs dfun_evs inst_tys sc_meth_ids | is_newtype = dfun_id `setIdUnfolding` mkInlineUnfolding (Just 0) con_app `setInlinePragma` alwaysInlinePragma { inl_sat = Just 0 } | otherwise = dfun_id `setIdUnfolding` mkDFunUnfolding dfun_bndrs dict_con dict_args `setInlinePragma` dfunInlinePragma where dfun_bndrs = dfun_tvs ++ dfun_evs dict_args = map Type inst_tys ++ [mkVarApps (Var id) dfun_bndrs | id <- sc_meth_ids] con_app = mkLams dfun_bndrs $ mkApps (Var (dataConWrapId dict_con)) dict_args wrapId :: HsWrapper -> id -> HsExpr id wrapId wrapper id = mkHsWrap wrapper (HsVar (noLoc id)) {- Note [Typechecking plan for instance declarations] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ For intance declarations we generate the following bindings and implication constraints. Example: instance Ord a => Ord [a] where compare = <compare-rhs> generates this: Bindings: -- Method bindings $ccompare :: forall a. Ord a => a -> a -> Ordering $ccompare = /\a \(d:Ord a). let <meth-ev-binds> in ... -- Superclass bindings $cp1Ord :: forall a. Ord a => Eq [a] $cp1Ord = /\a \(d:Ord a). let <sc-ev-binds> in dfEqList (dw :: Eq a) Constraints: forall a. Ord a => -- Method constraint (forall. (empty) => <constraints from compare-rhs>) -- Superclass constraint /\ (forall. (empty) => dw :: Eq a) Notice that * Per-meth/sc implication. There is one inner implication per superclass or method, with no skolem variables or givens. The only reason for this one is to gather the evidence bindings privately for this superclass or method. This implication is generated by checkInstConstraints. * Overall instance implication. There is an overall enclosing implication for the whole instance declaratation, with the expected skolems and givens. We need this to get the correct "redundant constraint" warnings, gathering all the uses from all the methods and superclasses. See TcSimplify Note [Tracking redundant constraints] * The given constraints in the outer implication may generate evidence, notably by superclass selection. Since the method and superclass bindings are top-level, we want that evidence copied into *every* method or superclass definition. (Some of it will be usused in some, but dead-code elimination will drop it.) We achieve this by putting the the evidence variable for the overall instance implicaiton into the AbsBinds for each method/superclass. Hence the 'dfun_ev_binds' passed into tcMethods and tcSuperClasses. (And that in turn is why the abs_ev_binds field of AbBinds is a [TcEvBinds] rather than simply TcEvBinds. This is a bit of a hack, but works very nicely in practice. * Note that if a method has a locally-polymorphic binding, there will be yet another implication for that, generated by tcPolyCheck in tcMethodBody. E.g. class C a where foo :: forall b. Ord b => blah ************************************************************************ * * Type-checking superclases * * ************************************************************************ -} tcSuperClasses :: DFunId -> Class -> [TcTyVar] -> [EvVar] -> [TcType] -> TcEvBinds -> FamInstEnvs -> TcThetaType -> TcM ([EvVar], LHsBinds Id, Bag Implication) -- Make a new top-level function binding for each superclass, -- something like -- $Ordp1 :: forall a. Ord a => Eq [a] -- $Ordp1 = /\a \(d:Ord a). dfunEqList a (sc_sel d) -- -- See Note [Recursive superclasses] for why this is so hard! -- In effect, be build a special-purpose solver for the first step -- of solving each superclass constraint tcSuperClasses dfun_id cls tyvars dfun_evs inst_tys dfun_ev_binds _fam_envs sc_theta = do { (ids, binds, implics) <- mapAndUnzip3M tc_super (zip sc_theta [fIRST_TAG..]) ; return (ids, listToBag binds, listToBag implics) } where loc = getSrcSpan dfun_id size = sizeTypes inst_tys tc_super (sc_pred, n) = do { (sc_implic, ev_binds_var, sc_ev_tm) <- checkInstConstraints $ emitWanted (ScOrigin size) sc_pred ; sc_top_name <- newName (mkSuperDictAuxOcc n (getOccName cls)) ; sc_ev_id <- newEvVar sc_pred ; addTcEvBind ev_binds_var $ mkWantedEvBind sc_ev_id sc_ev_tm ; let sc_top_ty = mkInvForAllTys tyvars (mkPiTypes dfun_evs sc_pred) sc_top_id = mkLocalId sc_top_name sc_top_ty export = ABE { abe_wrap = idHsWrapper , abe_poly = sc_top_id , abe_mono = sc_ev_id , abe_prags = noSpecPrags } local_ev_binds = TcEvBinds ev_binds_var bind = AbsBinds { abs_tvs = tyvars , abs_ev_vars = dfun_evs , abs_exports = [export] , abs_ev_binds = [dfun_ev_binds, local_ev_binds] , abs_binds = emptyBag } ; return (sc_top_id, L loc bind, sc_implic) } ------------------- checkInstConstraints :: TcM result -> TcM (Implication, EvBindsVar, result) -- See Note [Typechecking plan for instance declarations] checkInstConstraints thing_inside = do { (tclvl, wanted, result) <- pushLevelAndCaptureConstraints $ thing_inside ; ev_binds_var <- newTcEvBinds ; env <- getLclEnv ; let implic = Implic { ic_tclvl = tclvl , ic_skols = [] , ic_no_eqs = False , ic_given = [] , ic_wanted = wanted , ic_status = IC_Unsolved , ic_binds = Just ev_binds_var , ic_env = env , ic_info = InstSkol } ; return (implic, ev_binds_var, result) } {- Note [Recursive superclasses] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ See Trac #3731, #4809, #5751, #5913, #6117, #6161, which all describe somewhat more complicated situations, but ones encountered in practice. See also tests tcrun020, tcrun021, tcrun033, and Trac #11427. ----- THE PROBLEM -------- The problem is that it is all too easy to create a class whose superclass is bottom when it should not be. Consider the following (extreme) situation: class C a => D a where ... instance D [a] => D [a] where ... (dfunD) instance C [a] => C [a] where ... (dfunC) Although this looks wrong (assume D [a] to prove D [a]), it is only a more extreme case of what happens with recursive dictionaries, and it can, just about, make sense because the methods do some work before recursing. To implement the dfunD we must generate code for the superclass C [a], which we had better not get by superclass selection from the supplied argument: dfunD :: forall a. D [a] -> D [a] dfunD = \d::D [a] -> MkD (scsel d) .. Otherwise if we later encounter a situation where we have a [Wanted] dw::D [a] we might solve it thus: dw := dfunD dw Which is all fine except that now ** the superclass C is bottom **! The instance we want is: dfunD :: forall a. D [a] -> D [a] dfunD = \d::D [a] -> MkD (dfunC (scsel d)) ... ----- THE SOLUTION -------- The basic solution is simple: be very careful about using superclass selection to generate a superclass witness in a dictionary function definition. More precisely: Superclass Invariant: in every class dictionary, every superclass dictionary field is non-bottom To achieve the Superclass Invariant, in a dfun definition we can generate a guaranteed-non-bottom superclass witness from: (sc1) one of the dictionary arguments itself (all non-bottom) (sc2) an immediate superclass of a smaller dictionary (sc3) a call of a dfun (always returns a dictionary constructor) The tricky case is (sc2). We proceed by induction on the size of the (type of) the dictionary, defined by TcValidity.sizeTypes. Let's suppose we are building a dictionary of size 3, and suppose the Superclass Invariant holds of smaller dictionaries. Then if we have a smaller dictionary, its immediate superclasses will be non-bottom by induction. What does "we have a smaller dictionary" mean? It might be one of the arguments of the instance, or one of its superclasses. Here is an example, taken from CmmExpr: class Ord r => UserOfRegs r a where ... (i1) instance UserOfRegs r a => UserOfRegs r (Maybe a) where (i2) instance (Ord r, UserOfRegs r CmmReg) => UserOfRegs r CmmExpr where For (i1) we can get the (Ord r) superclass by selection from (UserOfRegs r a), since it is smaller than the thing we are building (UserOfRegs r (Maybe a). But for (i2) that isn't the case, so we must add an explicit, and perhaps surprising, (Ord r) argument to the instance declaration. Here's another example from Trac #6161: class Super a => Duper a where ... class Duper (Fam a) => Foo a where ... (i3) instance Foo a => Duper (Fam a) where ... (i4) instance Foo Float where ... It would be horribly wrong to define dfDuperFam :: Foo a -> Duper (Fam a) -- from (i3) dfDuperFam d = MkDuper (sc_sel1 (sc_sel2 d)) ... dfFooFloat :: Foo Float -- from (i4) dfFooFloat = MkFoo (dfDuperFam dfFooFloat) ... Now the Super superclass of Duper is definitely bottom! This won't happen because when processing (i3) we can use the superclasses of (Foo a), which is smaller, namely Duper (Fam a). But that is *not* smaller than the target so we can't take *its* superclasses. As a result the program is rightly rejected, unless you add (Super (Fam a)) to the context of (i3). Note [Solving superclass constraints] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ How do we ensure that every superclass witness is generated by one of (sc1) (sc2) or (sc3) in Note [Recursive superclases]. Answer: * Superclass "wanted" constraints have CtOrigin of (ScOrigin size) where 'size' is the size of the instance declaration. e.g. class C a => D a where... instance blah => D [a] where ... The wanted superclass constraint for C [a] has origin ScOrigin size, where size = size( D [a] ). * (sc1) When we rewrite such a wanted constraint, it retains its origin. But if we apply an instance declaration, we can set the origin to (ScOrigin infinity), thus lifting any restrictions by making prohibitedSuperClassSolve return False. * (sc2) ScOrigin wanted constraints can't be solved from a superclass selection, except at a smaller type. This test is implemented by TcInteract.prohibitedSuperClassSolve * The "given" constraints of an instance decl have CtOrigin GivenOrigin InstSkol. * When we make a superclass selection from InstSkol we use a SkolemInfo of (InstSC size), where 'size' is the size of the constraint whose superclass we are taking. An similarly when taking the superclass of an InstSC. This is implemented in TcCanonical.newSCWorkFromFlavored Note [Silent superclass arguments] (historical interest only) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ NB1: this note describes our *old* solution to the recursive-superclass problem. I'm keeping the Note for now, just as institutional memory. However, the code for silent superclass arguments was removed in late Dec 2014 NB2: the silent-superclass solution introduced new problems of its own, in the form of instance overlap. Tests SilentParametersOverlapping, T5051, and T7862 are examples NB3: the silent-superclass solution also generated tons of extra dictionaries. For example, in monad-transformer code, when constructing a Monad dictionary you had to pass an Applicative dictionary; and to construct that you neede a Functor dictionary. Yet these extra dictionaries were often never used. Test T3064 compiled *far* faster after silent superclasses were eliminated. Our solution to this problem "silent superclass arguments". We pass to each dfun some ``silent superclass arguments’’, which are the immediate superclasses of the dictionary we are trying to construct. In our example: dfun :: forall a. C [a] -> D [a] -> D [a] dfun = \(dc::C [a]) (dd::D [a]) -> DOrd dc ... Notice the extra (dc :: C [a]) argument compared to the previous version. This gives us: ----------------------------------------------------------- DFun Superclass Invariant ~~~~~~~~~~~~~~~~~~~~~~~~ In the body of a DFun, every superclass argument to the returned dictionary is either * one of the arguments of the DFun, or * constant, bound at top level ----------------------------------------------------------- This net effect is that it is safe to treat a dfun application as wrapping a dictionary constructor around its arguments (in particular, a dfun never picks superclasses from the arguments under the dictionary constructor). No superclass is hidden inside a dfun application. The extra arguments required to satisfy the DFun Superclass Invariant always come first, and are called the "silent" arguments. You can find out how many silent arguments there are using Id.dfunNSilent; and then you can just drop that number of arguments to see the ones that were in the original instance declaration. DFun types are built (only) by MkId.mkDictFunId, so that is where we decide what silent arguments are to be added. -} {- ************************************************************************ * * Type-checking an instance method * * ************************************************************************ tcMethod - Make the method bindings, as a [(NonRec, HsBinds)], one per method - Remembering to use fresh Name (the instance method Name) as the binder - Bring the instance method Ids into scope, for the benefit of tcInstSig - Use sig_fn mapping instance method Name -> instance tyvars - Ditto prag_fn - Use tcValBinds to do the checking -} tcMethods :: DFunId -> Class -> [TcTyVar] -> [EvVar] -> [TcType] -> TcEvBinds -> ([Located TcSpecPrag], TcPragEnv) -> [ClassOpItem] -> InstBindings Name -> TcM ([Id], LHsBinds Id, Bag Implication) -- The returned inst_meth_ids all have types starting -- forall tvs. theta => ... tcMethods dfun_id clas tyvars dfun_ev_vars inst_tys dfun_ev_binds (spec_inst_prags, prag_fn) op_items (InstBindings { ib_binds = binds , ib_tyvars = lexical_tvs , ib_pragmas = sigs , ib_extensions = exts , ib_derived = is_derived }) = tcExtendTyVarEnv2 (lexical_tvs `zip` tyvars) $ -- The lexical_tvs scope over the 'where' part do { traceTc "tcInstMeth" (ppr sigs $$ ppr binds) ; checkMinimalDefinition ; (ids, binds, mb_implics) <- set_exts exts $ mapAndUnzip3M tc_item op_items ; return (ids, listToBag binds, listToBag (catMaybes mb_implics)) } where set_exts :: [LangExt.Extension] -> TcM a -> TcM a set_exts es thing = foldr setXOptM thing es hs_sig_fn = mkHsSigFun sigs inst_loc = getSrcSpan dfun_id ---------------------- tc_item :: ClassOpItem -> TcM (Id, LHsBind Id, Maybe Implication) tc_item (sel_id, dm_info) | Just (user_bind, bndr_loc, prags) <- findMethodBind (idName sel_id) binds prag_fn = tcMethodBody clas tyvars dfun_ev_vars inst_tys dfun_ev_binds is_derived hs_sig_fn spec_inst_prags prags sel_id user_bind bndr_loc | otherwise = do { traceTc "tc_def" (ppr sel_id) ; tc_default sel_id dm_info } ---------------------- tc_default :: Id -> DefMethInfo -> TcM (TcId, LHsBind Id, Maybe Implication) tc_default sel_id (Just (dm_name, _)) = do { (meth_bind, inline_prags) <- mkDefMethBind clas inst_tys sel_id dm_name ; tcMethodBody clas tyvars dfun_ev_vars inst_tys dfun_ev_binds is_derived hs_sig_fn spec_inst_prags inline_prags sel_id meth_bind inst_loc } tc_default sel_id Nothing -- No default method at all = do { traceTc "tc_def: warn" (ppr sel_id) ; (meth_id, _) <- mkMethIds clas tyvars dfun_ev_vars inst_tys sel_id ; dflags <- getDynFlags ; let meth_bind = mkVarBind meth_id $ mkLHsWrap lam_wrapper (error_rhs dflags) ; return (meth_id, meth_bind, Nothing) } where error_rhs dflags = L inst_loc $ HsApp error_fun (error_msg dflags) error_fun = L inst_loc $ wrapId (mkWpTyApps [ getRuntimeRep "tcInstanceMethods.tc_default" meth_tau , meth_tau]) nO_METHOD_BINDING_ERROR_ID error_msg dflags = L inst_loc (HsLit (HsStringPrim "" (unsafeMkByteString (error_string dflags)))) meth_tau = funResultTy (piResultTys (idType sel_id) inst_tys) error_string dflags = showSDoc dflags (hcat [ppr inst_loc, vbar, ppr sel_id ]) lam_wrapper = mkWpTyLams tyvars <.> mkWpLams dfun_ev_vars ---------------------- -- Check if one of the minimal complete definitions is satisfied checkMinimalDefinition = whenIsJust (isUnsatisfied methodExists (classMinimalDef clas)) $ warnUnsatisfiedMinimalDefinition methodExists meth = isJust (findMethodBind meth binds prag_fn) ------------------------ tcMethodBody :: Class -> [TcTyVar] -> [EvVar] -> [TcType] -> TcEvBinds -> Bool -> HsSigFun -> [LTcSpecPrag] -> [LSig Name] -> Id -> LHsBind Name -> SrcSpan -> TcM (TcId, LHsBind Id, Maybe Implication) tcMethodBody clas tyvars dfun_ev_vars inst_tys dfun_ev_binds is_derived sig_fn spec_inst_prags prags sel_id (L bind_loc meth_bind) bndr_loc = add_meth_ctxt $ do { traceTc "tcMethodBody" (ppr sel_id <+> ppr (idType sel_id)) ; (global_meth_id, local_meth_id) <- setSrcSpan bndr_loc $ mkMethIds clas tyvars dfun_ev_vars inst_tys sel_id ; let lm_bind = meth_bind { fun_id = L bndr_loc (idName local_meth_id) } -- Substitute the local_meth_name for the binder -- NB: the binding is always a FunBind ; global_meth_id <- addInlinePrags global_meth_id prags ; spec_prags <- tcSpecPrags global_meth_id prags -- taking instance signature into account might change the type of -- the local_meth_id ; (meth_implic, ev_binds_var, tc_bind) <- checkInstConstraints $ tcMethodBodyHelp sig_fn sel_id local_meth_id (L bind_loc lm_bind) ; let specs = mk_meth_spec_prags global_meth_id spec_inst_prags spec_prags export = ABE { abe_poly = global_meth_id , abe_mono = local_meth_id , abe_wrap = idHsWrapper , abe_prags = specs } local_ev_binds = TcEvBinds ev_binds_var full_bind = AbsBinds { abs_tvs = tyvars , abs_ev_vars = dfun_ev_vars , abs_exports = [export] , abs_ev_binds = [dfun_ev_binds, local_ev_binds] , abs_binds = tc_bind } ; return (global_meth_id, L bind_loc full_bind, Just meth_implic) } where -- For instance decls that come from deriving clauses -- we want to print out the full source code if there's an error -- because otherwise the user won't see the code at all add_meth_ctxt thing | is_derived = addLandmarkErrCtxt (derivBindCtxt sel_id clas inst_tys) thing | otherwise = thing tcMethodBodyHelp :: HsSigFun -> Id -> TcId -> LHsBind Name -> TcM (LHsBinds TcId) tcMethodBodyHelp hs_sig_fn sel_id local_meth_id meth_bind | Just hs_sig_ty <- hs_sig_fn sel_name -- There is a signature in the instance -- See Note [Instance method signatures] = do { (sig_ty, hs_wrap) <- setSrcSpan (getLoc (hsSigType hs_sig_ty)) $ do { inst_sigs <- xoptM LangExt.InstanceSigs ; checkTc inst_sigs (misplacedInstSig sel_name hs_sig_ty) ; sig_ty <- tcHsSigType (FunSigCtxt sel_name False) hs_sig_ty ; let local_meth_ty = idType local_meth_id ; hs_wrap <- addErrCtxtM (methSigCtxt sel_name sig_ty local_meth_ty) $ tcSubType ctxt (Just sel_id) sig_ty (mkCheckExpType local_meth_ty) ; return (sig_ty, hs_wrap) } ; inner_meth_name <- newName (nameOccName sel_name) ; tc_sig <- instTcTySig ctxt hs_sig_ty sig_ty inner_meth_name ; (tc_bind, [inner_id]) <- tcPolyCheck NonRecursive no_prag_fn tc_sig meth_bind ; let export = ABE { abe_poly = local_meth_id , abe_mono = inner_id , abe_wrap = hs_wrap , abe_prags = noSpecPrags } ; return (unitBag $ L (getLoc meth_bind) $ AbsBinds { abs_tvs = [], abs_ev_vars = [] , abs_exports = [export] , abs_binds = tc_bind, abs_ev_binds = [] }) } | otherwise -- No instance signature = do { tc_sig <- instTcTySigFromId local_meth_id -- Absent a type sig, there are no new scoped type variables here -- Only the ones from the instance decl itself, which are already -- in scope. Example: -- class C a where { op :: forall b. Eq b => ... } -- instance C [c] where { op = <rhs> } -- In <rhs>, 'c' is scope but 'b' is not! ; (tc_bind, _) <- tcPolyCheck NonRecursive no_prag_fn tc_sig meth_bind ; return tc_bind } where ctxt = FunSigCtxt sel_name True sel_name = idName sel_id no_prag_fn = emptyPragEnv -- No pragmas for local_meth_id; -- they are all for meth_id ------------------------ mkMethIds :: Class -> [TcTyVar] -> [EvVar] -> [TcType] -> Id -> TcM (TcId, TcId) -- returns (poly_id, local_id), but ignoring any instance signature -- See Note [Instance method signatures] mkMethIds clas tyvars dfun_ev_vars inst_tys sel_id = do { poly_meth_name <- newName (mkClassOpAuxOcc sel_occ) ; local_meth_name <- newName sel_occ -- Base the local_meth_name on the selector name, because -- type errors from tcMethodBody come from here ; let poly_meth_id = mkLocalId poly_meth_name poly_meth_ty local_meth_id = mkLocalId local_meth_name local_meth_ty ; return (poly_meth_id, local_meth_id) } where sel_name = idName sel_id sel_occ = nameOccName sel_name local_meth_ty = instantiateMethod clas sel_id inst_tys poly_meth_ty = mkSpecSigmaTy tyvars theta local_meth_ty theta = map idType dfun_ev_vars methSigCtxt :: Name -> TcType -> TcType -> TidyEnv -> TcM (TidyEnv, MsgDoc) methSigCtxt sel_name sig_ty meth_ty env0 = do { (env1, sig_ty) <- zonkTidyTcType env0 sig_ty ; (env2, meth_ty) <- zonkTidyTcType env1 meth_ty ; let msg = hang (text "When checking that instance signature for" <+> quotes (ppr sel_name)) 2 (vcat [ text "is more general than its signature in the class" , text "Instance sig:" <+> ppr sig_ty , text " Class sig:" <+> ppr meth_ty ]) ; return (env2, msg) } misplacedInstSig :: Name -> LHsSigType Name -> SDoc misplacedInstSig name hs_ty = vcat [ hang (text "Illegal type signature in instance declaration:") 2 (hang (pprPrefixName name) 2 (dcolon <+> ppr hs_ty)) , text "(Use InstanceSigs to allow this)" ] {- Note [Instance method signatures] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ With -XInstanceSigs we allow the user to supply a signature for the method in an instance declaration. Here is an artificial example: data T a = MkT a instance Ord a => Ord (T a) where (>) :: forall b. b -> b -> Bool (>) = error "You can't compare Ts" The instance signature can be *more* polymorphic than the instantiated class method (in this case: Age -> Age -> Bool), but it cannot be less polymorphic. Moreover, if a signature is given, the implementation code should match the signature, and type variables bound in the singature should scope over the method body. We achieve this by building a TcSigInfo for the method, whether or not there is an instance method signature, and using that to typecheck the declaration (in tcMethodBody). That means, conveniently, that the type variables bound in the signature will scope over the body. What about the check that the instance method signature is more polymorphic than the instantiated class method type? We just do a tcSubType call in tcMethodBodyHelp, and generate a nested AbsBind, like this (for the example above AbsBind { abs_tvs = [a], abs_ev_vars = [d:Ord a] , abs_exports = ABExport { (>) :: forall a. Ord a => T a -> T a -> Bool , gr_lcl :: T a -> T a -> Bool } , abs_binds = AbsBind { abs_tvs = [], abs_ev_vars = [] , abs_exports = ABExport { gr_lcl :: T a -> T a -> Bool , gr_inner :: forall b. b -> b -> Bool } , abs_binds = AbsBind { abs_tvs = [b], abs_ev_vars = [] , ..etc.. } } } Wow! Three nested AbsBinds! * The outer one abstracts over the tyvars and dicts for the instance * The middle one is only present if there is an instance signature, and does the impedance matching for that signature * The inner one is for the method binding itself against either the signature from the class, or the the instance signature. -} ---------------------- mk_meth_spec_prags :: Id -> [LTcSpecPrag] -> [LTcSpecPrag] -> TcSpecPrags -- Adapt the 'SPECIALISE instance' pragmas to work for this method Id -- There are two sources: -- * spec_prags_for_me: {-# SPECIALISE op :: <blah> #-} -- * spec_prags_from_inst: derived from {-# SPECIALISE instance :: <blah> #-} -- These ones have the dfun inside, but [perhaps surprisingly] -- the correct wrapper. -- See Note [Handling SPECIALISE pragmas] in TcBinds mk_meth_spec_prags meth_id spec_inst_prags spec_prags_for_me = SpecPrags (spec_prags_for_me ++ spec_prags_from_inst) where spec_prags_from_inst | isInlinePragma (idInlinePragma meth_id) = [] -- Do not inherit SPECIALISE from the instance if the -- method is marked INLINE, because then it'll be inlined -- and the specialisation would do nothing. (Indeed it'll provoke -- a warning from the desugarer | otherwise = [ L inst_loc (SpecPrag meth_id wrap inl) | L inst_loc (SpecPrag _ wrap inl) <- spec_inst_prags] mkDefMethBind :: Class -> [Type] -> Id -> Name -> TcM (LHsBind Name, [LSig Name]) -- The is a default method (vanailla or generic) defined in the class -- So make a binding op = $dmop @t1 @t2 -- where $dmop is the name of the default method in the class, -- and t1,t2 are the instace types. -- See Note [Default methods in instances] for why we use -- visible type application here mkDefMethBind clas inst_tys sel_id dm_name = do { dflags <- getDynFlags ; dm_id <- tcLookupId dm_name ; let inline_prag = idInlinePragma dm_id inline_prags | isAnyInlinePragma inline_prag = [noLoc (InlineSig fn inline_prag)] | otherwise = [] -- Copy the inline pragma (if any) from the default method -- to this version. Note [INLINE and default methods] fn = noLoc (idName sel_id) visible_inst_tys = [ ty | (tcb, ty) <- tyConBinders (classTyCon clas) `zip` inst_tys , binderVisibility tcb /= Invisible ] rhs = foldl mk_vta (nlHsVar dm_name) visible_inst_tys bind = noLoc $ mkTopFunBind Generated fn $ [mkSimpleMatch [] rhs] ; liftIO (dumpIfSet_dyn dflags Opt_D_dump_deriv "Filling in method body" (vcat [ppr clas <+> ppr inst_tys, nest 2 (ppr sel_id <+> equals <+> ppr rhs)])) ; return (bind, inline_prags) } where mk_vta :: LHsExpr Name -> Type -> LHsExpr Name mk_vta fun ty = noLoc (HsAppType fun (mkEmptyWildCardBndrs $ noLoc $ HsCoreTy ty)) -- NB: use visible type application -- See Note [Default methods in instances] ---------------------- derivBindCtxt :: Id -> Class -> [Type ] -> SDoc derivBindCtxt sel_id clas tys = vcat [ text "When typechecking the code for" <+> quotes (ppr sel_id) , nest 2 (text "in a derived instance for" <+> quotes (pprClassPred clas tys) <> colon) , nest 2 $ text "To see the code I am typechecking, use -ddump-deriv" ] warnUnsatisfiedMinimalDefinition :: ClassMinimalDef -> TcM () warnUnsatisfiedMinimalDefinition mindef = do { warn <- woptM Opt_WarnMissingMethods ; warnTc (Reason Opt_WarnMissingMethods) warn message } where message = vcat [text "No explicit implementation for" ,nest 2 $ pprBooleanFormulaNice mindef ] {- Note [Export helper functions] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ We arrange to export the "helper functions" of an instance declaration, so that they are not subject to preInlineUnconditionally, even if their RHS is trivial. Reason: they are mentioned in the DFunUnfolding of the dict fun as Ids, not as CoreExprs, so we can't substitute a non-variable for them. We could change this by making DFunUnfoldings have CoreExprs, but it seems a bit simpler this way. Note [Default methods in instances] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider this class Baz v x where foo :: x -> x foo y = <blah> instance Baz Int Int From the class decl we get $dmfoo :: forall v x. Baz v x => x -> x $dmfoo y = <blah> Notice that the type is ambiguous. So we use Visible Type Application to disambiguate: $dBazIntInt = MkBaz fooIntInt fooIntInt = $dmfoo @Int @Int Lacking VTA we'd get ambiguity errors involving the default method. This applies equally to vanilla default methods (Trac #1061) and generic default methods (Trac #12220). Historical note: before we had VTA we had to generate post-type-checked code, which took a lot more code, and didn't work for generic default methods. Note [INLINE and default methods] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Default methods need special case. They are supposed to behave rather like macros. For exmample class Foo a where op1, op2 :: Bool -> a -> a {-# INLINE op1 #-} op1 b x = op2 (not b) x instance Foo Int where -- op1 via default method op2 b x = <blah> The instance declaration should behave just as if 'op1' had been defined with the code, and INLINE pragma, from its original definition. That is, just as if you'd written instance Foo Int where op2 b x = <blah> {-# INLINE op1 #-} op1 b x = op2 (not b) x So for the above example we generate: {-# INLINE $dmop1 #-} -- $dmop1 has an InlineCompulsory unfolding $dmop1 d b x = op2 d (not b) x $fFooInt = MkD $cop1 $cop2 {-# INLINE $cop1 #-} $cop1 = $dmop1 $fFooInt $cop2 = <blah> Note carefully: * We *copy* any INLINE pragma from the default method $dmop1 to the instance $cop1. Otherwise we'll just inline the former in the latter and stop, which isn't what the user expected * Regardless of its pragma, we give the default method an unfolding with an InlineCompulsory source. That means that it'll be inlined at every use site, notably in each instance declaration, such as $cop1. This inlining must happen even though a) $dmop1 is not saturated in $cop1 b) $cop1 itself has an INLINE pragma It's vital that $dmop1 *is* inlined in this way, to allow the mutual recursion between $fooInt and $cop1 to be broken * To communicate the need for an InlineCompulsory to the desugarer (which makes the Unfoldings), we use the IsDefaultMethod constructor in TcSpecPrags. ************************************************************************ * * Specialise instance pragmas * * ************************************************************************ Note [SPECIALISE instance pragmas] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider instance (Ix a, Ix b) => Ix (a,b) where {-# SPECIALISE instance Ix (Int,Int) #-} range (x,y) = ... We make a specialised version of the dictionary function, AND specialised versions of each *method*. Thus we should generate something like this: $dfIxPair :: (Ix a, Ix b) => Ix (a,b) {-# DFUN [$crangePair, ...] #-} {-# SPECIALISE $dfIxPair :: Ix (Int,Int) #-} $dfIxPair da db = Ix ($crangePair da db) (...other methods...) $crange :: (Ix a, Ix b) -> ((a,b),(a,b)) -> [(a,b)] {-# SPECIALISE $crange :: ((Int,Int),(Int,Int)) -> [(Int,Int)] #-} $crange da db = <blah> The SPECIALISE pragmas are acted upon by the desugarer, which generate dii :: Ix Int dii = ... $s$dfIxPair :: Ix ((Int,Int),(Int,Int)) {-# DFUN [$crangePair di di, ...] #-} $s$dfIxPair = Ix ($crangePair di di) (...) {-# RULE forall (d1,d2:Ix Int). $dfIxPair Int Int d1 d2 = $s$dfIxPair #-} $s$crangePair :: ((Int,Int),(Int,Int)) -> [(Int,Int)] $c$crangePair = ...specialised RHS of $crangePair... {-# RULE forall (d1,d2:Ix Int). $crangePair Int Int d1 d2 = $s$crangePair #-} Note that * The specialised dictionary $s$dfIxPair is very much needed, in case we call a function that takes a dictionary, but in a context where the specialised dictionary can be used. See Trac #7797. * The ClassOp rule for 'range' works equally well on $s$dfIxPair, because it still has a DFunUnfolding. See Note [ClassOp/DFun selection] * A call (range ($dfIxPair Int Int d1 d2)) might simplify two ways: --> {ClassOp rule for range} $crangePair Int Int d1 d2 --> {SPEC rule for $crangePair} $s$crangePair or thus: --> {SPEC rule for $dfIxPair} range $s$dfIxPair --> {ClassOpRule for range} $s$crangePair It doesn't matter which way. * We want to specialise the RHS of both $dfIxPair and $crangePair, but the SAME HsWrapper will do for both! We can call tcSpecPrag just once, and pass the result (in spec_inst_info) to tcMethods. -} tcSpecInstPrags :: DFunId -> InstBindings Name -> TcM ([Located TcSpecPrag], TcPragEnv) tcSpecInstPrags dfun_id (InstBindings { ib_binds = binds, ib_pragmas = uprags }) = do { spec_inst_prags <- mapM (wrapLocM (tcSpecInst dfun_id)) $ filter isSpecInstLSig uprags -- The filter removes the pragmas for methods ; return (spec_inst_prags, mkPragEnv uprags binds) } ------------------------------ tcSpecInst :: Id -> Sig Name -> TcM TcSpecPrag tcSpecInst dfun_id prag@(SpecInstSig _ hs_ty) = addErrCtxt (spec_ctxt prag) $ do { (tyvars, theta, clas, tys) <- tcHsClsInstType SpecInstCtxt hs_ty ; let spec_dfun_ty = mkDictFunTy tyvars theta clas tys ; co_fn <- tcSpecWrapper SpecInstCtxt (idType dfun_id) spec_dfun_ty ; return (SpecPrag dfun_id co_fn defaultInlinePragma) } where spec_ctxt prag = hang (text "In the SPECIALISE pragma") 2 (ppr prag) tcSpecInst _ _ = panic "tcSpecInst" {- ************************************************************************ * * \subsection{Error messages} * * ************************************************************************ -} instDeclCtxt1 :: LHsSigType Name -> SDoc instDeclCtxt1 hs_inst_ty = inst_decl_ctxt (ppr (getLHsInstDeclHead hs_inst_ty)) instDeclCtxt2 :: Type -> SDoc instDeclCtxt2 dfun_ty = inst_decl_ctxt (ppr (mkClassPred cls tys)) where (_,_,cls,tys) = tcSplitDFunTy dfun_ty inst_decl_ctxt :: SDoc -> SDoc inst_decl_ctxt doc = hang (text "In the instance declaration for") 2 (quotes doc) badBootFamInstDeclErr :: SDoc badBootFamInstDeclErr = text "Illegal family instance in hs-boot file" notFamily :: TyCon -> SDoc notFamily tycon = vcat [ text "Illegal family instance for" <+> quotes (ppr tycon) , nest 2 $ parens (ppr tycon <+> text "is not an indexed type family")] tooFewParmsErr :: Arity -> SDoc tooFewParmsErr arity = text "Family instance has too few parameters; expected" <+> ppr arity assocInClassErr :: Located Name -> SDoc assocInClassErr name = text "Associated type" <+> quotes (ppr name) <+> text "must be inside a class instance" badFamInstDecl :: Located Name -> SDoc badFamInstDecl tc_name = vcat [ text "Illegal family instance for" <+> quotes (ppr tc_name) , nest 2 (parens $ text "Use TypeFamilies to allow indexed type families") ] notOpenFamily :: TyCon -> SDoc notOpenFamily tc = text "Illegal instance for closed family" <+> quotes (ppr tc)