Polymorphic Perplexion
Polymorphism plays a vital role in automating verification in LH. However, thanks to its ubiquity, we often take it for granted, and it can be quite baffling to figure out why verification fails with monomorphic signatures. Let me explain why, using a simple example that has puzzled me and other users several times.
22: 23: module PolymorphicPerplexion where
A Type for Ordered Lists¶
Previously we have seen how you can use LH to define a type of lists whose values are in increasing (ok, non-decreasing!) order.
First, we define an IncList a
type, with Emp
("empty")
and :<
("cons") constructors.
38: data IncList a = Emp 39: | (:<) { hd :: a, tl :: IncList a } 40: 41: infixr 9 :<
Next, we refine the type to specify that each "cons" :<
constructor takes as input a hd
and a tl
which must
be an IncList a
of values v
each of which is greater
than hd
.
50: {-@ data IncList a = Emp 51: | (:<) { hd :: a, tl :: IncList {v:a | hd <= v}} 52: @-}
We can confirm that the above definition ensures that the only legal values are increasingly ordered lists, as LH accepts the first list below, but rejects the second where the elements are out of order.
61: legalList :: IncList Int 62: (PolymorphicPerplexion.IncList GHC.Types.Int)legalList = GHC.Types.Int0 :< GHC.Types.Int1 :< GHC.Types.Int2 :< GHC.Types.Int3 :< {VV : forall a . (PolymorphicPerplexion.IncList a) | VV == Emp}Emp 63: 64: illegalList :: IncList Int 65: (PolymorphicPerplexion.IncList GHC.Types.Int)illegalList = GHC.Types.Int0 :< GHC.Types.Int1 :< GHC.Types.Int3 :< GHC.Types.Int2 :< {VV : forall a . (PolymorphicPerplexion.IncList a) | VV == Emp}Emp
A Polymorphic Insertion Sort¶
Next, lets write a simple insertion-sort function that
takes a plain unordered list of [a]
and returns the elements
in increasing order:
76: insertSortP :: (Ord a) => [a] -> IncList a 77: forall a . (GHC.Classes.Ord<[]> a) => [a] -> (PolymorphicPerplexion.IncList a)insertSortP [a]xs = foldr a -> (PolymorphicPerplexion.IncList a) -> (PolymorphicPerplexion.IncList a)insertP {VV : forall a . (PolymorphicPerplexion.IncList a) | VV == Emp}Emp {v : [a] | len v >= 0 && v == xs}xs 78: 79: insertP :: (Ord a) => a -> IncList a -> IncList a 80: forall a . (GHC.Classes.Ord<[]> a) => a -> (PolymorphicPerplexion.IncList a) -> (PolymorphicPerplexion.IncList a)insertP ay Emp = {VV : a | VV == y}y :< {VV : forall a . (PolymorphicPerplexion.IncList a) | VV == Emp}Emp 81: insertP y (x :< xs) 82: | {VV : a | VV == y}y <= {VV : a | VV == x}x = {VV : a | VV == y}y :< {VV : a | VV == x}x :< {v : (PolymorphicPerplexion.IncList {VV : a | x <= VV}) | v == xs}xs 83: | otherwise = {VV : a | VV == x}x :< (PolymorphicPerplexion.IncList a)insertP {VV : a | VV == y}y {v : (PolymorphicPerplexion.IncList {VV : a | x <= VV}) | v == xs}xs
Happily, LH is able to verify the above code without any trouble!
(If that seemed too easy, don't worry: if you mess up the comparison,
e.g. change the guard to x <= y
LH will complain about it.)
A Monomorphic Insertion Sort¶
However, lets take the exact same code as above but change
the type signatures to make the functions monomorphic, here,
specialized to Int
lists.
99: insertSortM :: [Int] -> IncList Int 100: [GHC.Types.Int] -> (PolymorphicPerplexion.IncList GHC.Types.Int)insertSortM [GHC.Types.Int]xs = foldr GHC.Types.Int -> (PolymorphicPerplexion.IncList GHC.Types.Int) -> (PolymorphicPerplexion.IncList GHC.Types.Int)insertM {VV : forall a . (PolymorphicPerplexion.IncList a) | VV == Emp}Emp {v : [GHC.Types.Int] | len v >= 0 && v == xs}xs 101: 102: insertM :: Int -> IncList Int -> IncList Int 103: GHC.Types.Int -> (PolymorphicPerplexion.IncList GHC.Types.Int) -> (PolymorphicPerplexion.IncList GHC.Types.Int)insertM GHC.Types.Inty Emp = {v : GHC.Types.Int | v == y}y :< {VV : forall a . (PolymorphicPerplexion.IncList a) | VV == Emp}Emp 104: insertM y (x :< xs) 105: | {v : GHC.Types.Int | v == y}y <= {v : GHC.Types.Int | v == x}x = {v : GHC.Types.Int | v == y}y :< {v : GHC.Types.Int | v == x}x :< {v : (PolymorphicPerplexion.IncList {v : GHC.Types.Int | x <= v}) | v == xs}xs 106: | otherwise = {v : GHC.Types.Int | v == x}x :< (PolymorphicPerplexion.IncList GHC.Types.Int)insertM {v : GHC.Types.Int | v == y}y {v : (PolymorphicPerplexion.IncList {v : GHC.Types.Int | x <= v}) | v == xs}xs
Huh? Now LH appears to be unhappy with the code! How is this possible?
Lets look at the type error:
114: /Users/rjhala/PerplexingPolymorphicProperties.lhs:80:27-38: Error: Liquid Type Mismatch 115: 116: 80 | | otherwise = x :< insertM y xs 117: ^^^^^^^^^^^^ 118: Inferred type 119: VV : Int 120: 121: not a subtype of Required type 122: VV : {VV : Int | x <= VV} 123: 124: In Context 125: x : Int
LH expects that since we're using the "cons" operator :<
the "tail"
value insertM y xs
must contain values VV
that are greater than the
"head" x
. The error says that, LH cannot prove this requirement of
actual list insertM y xs
.
Hmm, well thats a puzzler. Two questions that should come to mind.
-
Why does the above fact hold in the first place?
-
How is LH able to deduce this fact with the polymorphic signature but not the monomorphic one?
Lets ponder the first question: why is every element
of insert y xs
in fact larger than x
? For three reasons:
(a) every element in xs
is larger than x
, as the
list x :< xs
was ordered,
(b) y
is larger than x
as established by the otherwise
and crucially
(c) the elements returned by insert y xs
are either y
or from xs
!
Now onto the second question: how does LH verify the polymorphic code,
but not the monomorphic one? The reason is the fact (c)! LH is a modular
verifier, meaning that the only information that it has about the behavior
of insert
at a call-site is the information captured in the (refinement)
type specification for insert
. The polymorphic signature:
156: insertP :: (Ord a) => a -> IncList a -> IncList a
via parametricity, implicitly states fact (c). That is, if at a call-site
insertP y xs
we pass in a value that is greater an x
and a list of values
greater than x
then via polymorphic instantiation at the call-site, LH
infers that the returned value must also be a list of elements greater than x
!
However, the monomorphic signature
167: insertM :: Int -> IncList Int -> IncList Int
offers no such insight. It simply says the function takes in an Int
and another
ordered list of Int
and returns another ordered list, whose actual elements could
be arbitrary Int
. Specifically, at the call-site insertP y xs
LH has no way to
conclude the the returned elements are indeed greater than x
and hence rejects
the monomorphic code.
Perplexity¶
While parametricity is all very nice, and LH's polymorphic instanatiation is very
clever and useful, it can also be quite mysterious. For example, q curious user
OisÃn pointed out
that while the code below is rejected that if you uncomment the type signature
for go
then it is accepted by LH!
187: insertSortP' :: (Ord a) => [a] -> IncList a 188: forall a . (GHC.Classes.Ord<[]> a) => [a] -> (PolymorphicPerplexion.IncList a)insertSortP' = (PolymorphicPerplexion.IncList a)foldr a -> (PolymorphicPerplexion.IncList a) -> (PolymorphicPerplexion.IncList a)go {VV : forall a . (PolymorphicPerplexion.IncList a) | VV == Emp}Emp 189: where 190: -- go :: (Ord a) => a -> IncList a -> IncList a 191: a -> (PolymorphicPerplexion.IncList a) -> (PolymorphicPerplexion.IncList a)go ay Emp = {VV : a | VV == y}y :< {VV : forall a . (PolymorphicPerplexion.IncList a) | VV == Emp}Emp 192: go y (x :< xs) 193: | {VV : a | VV == y}y <= {VV : a | VV == x}x = {VV : a | VV == y}y :< {VV : a | VV == x}x :< {v : (PolymorphicPerplexion.IncList {VV : a | x <= VV}) | v == xs}xs 194: | otherwise = {VV : a | VV == x}x :< (PolymorphicPerplexion.IncList a)go {VV : a | VV == y}y {v : (PolymorphicPerplexion.IncList {VV : a | x <= VV}) | v == xs}xs
This is thoroughly perplexing, but again, is explained by the absence of
parametricity. When we remove the type signature, GHC defaults to giving
go
a monomorphic signature where the a
is not universally quantified,
and which roughly captures the same specification as the monomorphic insertM
above causing verification to fail!
Restoring the signature provides LH with the polymorphic specification,
which can be instantiated at the call-site to recover the fact (c)
that is crucial for verification.
Moral¶
I hope that example illustrates two points.
First, parametric polymorphism lets type specifications say a lot more than they immediately let on: so do write polymorphic signatures whenever possible.
Second, on a less happy note, explaining why fancy type checkers fail remains a vexing problem, whose difficulty is compounded by increasing the cleverness of the type system.
We'd love to hear any ideas you might have to solve the explanation problem!