statement
stringlengths
1
1.39k
proof
stringlengths
0
4.58k
type
stringclasses
21 values
symbolic_name
stringlengths
1
49
library
stringclasses
18 values
filename
stringclasses
100 values
imports
listlengths
0
26
deps
listlengths
0
23
docstring
stringlengths
0
2.01k
source_url
stringclasses
1 value
commit
stringclasses
1 value
main
:= runTests [ qc "example" prop_example ].
Definition
main
test
test/mutation.v
[ "QuickChick", "Coq", "List", "String", "ExtrOcamlNatInt", "ListNotations" ]
[ "prop_example", "qc", "runTests" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
foo {A : Type}
:= | bar : A -> foo -> foo | baz : foo .
Inductive
foo
test
test/plugin.v
[ "QuickChick", "Coq", "Derive", "MonadNotation", "BindOptNotation", "mathcomp", "ssreflect", "ssrnat", "div" ]
[]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
qux : Type
:= | Qux: forall {A: Type}, A -> qux.
Inductive
qux
test
test/plugin.v
[ "QuickChick", "Coq", "Derive", "MonadNotation", "BindOptNotation", "mathcomp", "ssreflect", "ssrnat", "div" ]
[]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
quux: qux -> bool
:= fun a => match a with | Qux a => true end.
Definition
quux
test
test/plugin.v
[ "QuickChick", "Coq", "Derive", "MonadNotation", "BindOptNotation", "mathcomp", "ssreflect", "ssrnat", "div" ]
[ "qux" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
a : G nat
:= ret 1.
Definition
a
test
test/plugin.v
[ "QuickChick", "Coq", "Derive", "MonadNotation", "BindOptNotation", "mathcomp", "ssreflect", "ssrnat", "div" ]
[ "nat" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
b : G nat
:= v <- a ;; ret v.
Definition
b
test
test/plugin.v
[ "QuickChick", "Coq", "Derive", "MonadNotation", "BindOptNotation", "mathcomp", "ssreflect", "ssrnat", "div" ]
[ "nat" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
c : G (option nat)
:= ret (Some 42).
Definition
c
test
test/plugin.v
[ "QuickChick", "Coq", "Derive", "MonadNotation", "BindOptNotation", "mathcomp", "ssreflect", "ssrnat", "div" ]
[ "nat" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
d : G (option nat)
:= v <-- c;; ret (Some v).
Definition
d
test
test/plugin.v
[ "QuickChick", "Coq", "Derive", "MonadNotation", "BindOptNotation", "mathcomp", "ssreflect", "ssrnat", "div" ]
[ "nat" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
int
:= nat.
Definition
int
test
test/plugin.v
[ "QuickChick", "Coq", "Derive", "MonadNotation", "BindOptNotation", "mathcomp", "ssreflect", "ssrnat", "div" ]
[ "nat" ]
Test extraction hack (substitute type int = int)
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
teh
:= fun x : int => Nat.eqb x x.
Definition
teh
test
test/plugin.v
[ "QuickChick", "Coq", "Derive", "MonadNotation", "BindOptNotation", "mathcomp", "ssreflect", "ssrnat", "div" ]
[ "int" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
Tree A
:= | Leaf : Tree A | Node : A -> Tree A -> Tree A -> Tree A.
Inductive
Tree
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[]
Let's revisit our favorite datatype, binary trees:
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
Dec_Eq (A : Type)
:= { dec_eq : forall (x y : A), decidable (x = y) }.
Class
Dec_Eq
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[ "dec_eq" ]
The most common use of the Dec class is boolean equality testing. That is the purpose of the Dec_Eq typeclass.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
Dec_Eq_Tree {A} `{Dec_Eq A} : Dec_Eq (Tree A).
Proof. dec_eq. Defined.
Instance
Dec_Eq_Tree
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[ "Dec_Eq", "Tree", "dec_eq" ]
For the Dec_Eq class in particular, QuickChick provides a useful tactic for using the Coq-provided `decide equality` tactic in conjunction with existing Dec_Eq instances, to automate its construction. For example, for our Tree example we can invoke `dec_eq`, assuming our type A is also testable for equality --- note th...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
mirror {A : Type} (t : Tree A) : Tree A
:= match t with | Leaf => Leaf | Node x l r => Node x (mirror r) (mirror l) end.
Fixpoint
mirror
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[ "Tree" ]
Armed with all these instances, we can now automatically test properties of trees. For example, in the BasicUsage tutorial we saw a `mirror` function:
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
faulty_mirrorP (t : Tree nat)
:= mirror t = t?.
Definition
faulty_mirrorP
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[ "Tree", "mirror", "nat" ]
Along with a faulty mirror property, specialized to nat for simpler testing:
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
balanced {A} : nat -> Tree A -> Prop
:= | B0 : balanced 0 Leaf | B1 : balanced 1 Leaf | BS : forall n x l r, balanced n l -> balanced n r -> balanced (S n) (Node x l r).
Inductive
balanced
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[ "Tree", "nat" ]
Another very common occurrence in Coq is to have complex inductive definitions that both constrain the inputs of theorems, and are used in the conclusion. For a complete example, we refer the user to the stlc tutorial. For here, let's consider the simpler case of balanced trees of height `n`, wher...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
insert {A} (x : A) (t : Tree A) : Tree A
:= match t with | Leaf => Node x Leaf Leaf | Node y l r => Node y (insert x l) r end.
Fixpoint
insert
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[ "Tree" ]
When implementing a data structure such as AVL trees, we would ensure that a balanced tree remains balanced after inserting an element with intricate rebalancing operations. Here, let's encode a very naive insertion function that always inserts elements on the leftmost path, and see how QuickChick can fig...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
all_trees_are_balanced (n : nat) (t : Tree nat)
:= balanced n t ?? 10.
Definition
all_trees_are_balanced
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[ "Tree", "balanced", "nat" ]
So let's try to check our first (obviously false) property using derived checkers: all trees (of natural numbers) are balanced.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
prop_gen_balanced_is_balanced
:= let fuel := 10 in (* Generate an arbitrary n *) forAll (choose (0,5)) (fun (n : nat) => (* Generate an arbitrary balanced tree of height n *) forAllMaybe (genSizedST (fun t => balanced n t) fuel) (fun (t : Tree nat) => (* Check the resulting tree is balanced. *) balanced n t ?? fuel)).
Definition
prop_gen_balanced_is_balanced
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[ "Tree", "balanced", "choose", "forAll", "forAllMaybe", "nat" ]
Now we can use this generator and the checker above, to sanity check that QuickChick has done the right thing:
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
balanced_preserves_balanced (fuel n x : nat) (t : Tree nat)
:= (balanced n t ?? fuel) ==>? (balanced n (insert x t) ?? fuel).
Definition
balanced_preserves_balanced
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[ "Tree", "balanced", "insert", "nat" ]
Perfect! Now let's try to write - and test - the property that insertion preserves balanced. We will use the '==>?' combinator which combines two option bools, treating failures in the precondition as a `None` - a discarded test.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
prop_balanced_preserves_balanced (n : nat)
:= let fuel := 10 in (* Generate an arbitrary balanced tree of height n *) forAllMaybe (genSizedST (fun t => balanced n t) fuel) (fun (t : Tree nat) => (* Generate an arbitrary integer x to insert *) forAll (choose (0,10)) (fun x => balanced_preserves_balanced fuel n x t)).
Definition
prop_balanced_preserves_balanced
tutorials
tutorials/Automation.v
[ "QuickChick", "mathcomp", "ssrbool" ]
[ "Tree", "balanced", "balanced_preserves_balanced", "choose", "forAll", "forAllMaybe", "nat" ]
Naturally, no balanced trees of height 5 could even be generated! However, we could use the derived constrained generators instead:
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
remove (x : nat) (l : list nat) : list nat
:= match l with | [] => [] | h::t => if Nat.eqb h x then t else h :: remove x t end.
Fixpoint
remove
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "nat" ]
It is not uncommon during a verification effort to spend many hours attempting to prove a slightly false theorem, only to result in frustration when the mistake is realized and one needs to start over. Other theorem provers have testing tools to quickly raise one's confidence before embarking on the bod...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
removeP (x : nat) (l : list nat) : bool
:= negb (existsb (fun y => x =? y) (remove x l)).
Definition
removeP
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "nat", "negb", "remove" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
remove_spec
:= forall x l, ~ In x (remove x l).
Definition
remove_spec
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "In", "remove" ]
Internally, the code is extracted to OCaml, compiled, and run. The following output is presented in your terminal, CoqIDE [Messages] pane, or Visual Studio Code [Info] pulldown menu tab: << 0 [0; 0] Failed! After 17 tests and 12 shrinks >> The output signifies that if we use an input where [x] is [0]...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
Color
:= Red | Green | Blue | Yellow.
Inductive
Color
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[]
The [Show] typeclass contains a single function [show] from some type [A] to Coq's [string]. QuickChick provides default instances for [string]s (the identity function), [nat], [bool], [Z], etc. (via extraction to appropriate OCaml functions for efficiency), as well as some common compound datatypes: li...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
show_color : Show Color
:= {| show c := match c with | Red => "Red" | Green => "Green" | Blue => "Blue" | Yellow => "Yellow" end |}.
Instance
show_color
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Color", "Show" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
genColor : G Color
:= elems [ Red ; Green ; Blue ; Yellow ].
Definition
genColor
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Color" ]
Armed with [elems], we can write the following simple [Color] generator.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
showTree {A} `{_ : Show A} : Show (Tree A)
:= {| show := let fix aux t := match t with | Leaf => "Leaf" | Node x l r => "Node (" ++ show x ++ ") (" ++ aux l ++ ") (" ++ aux r ++ ")" end in aux |}.
Instance
showTree
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Show", "Tree" ]
Before getting to generators for trees, we give a simple [Show] instance. The rather inconvenient need for a local [let fix] declaration stems from the fact that Coq's typeclasses (unlike Haskell's) are not automatically recursive.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
genTreeSized {A} (sz : nat) (g : G A) : G (Tree A)
:= match sz with | O => returnGen Leaf | S sz' => oneOf [ returnGen Leaf ; liftGen3 Node g (genTreeSized sz' g) (genTreeSized sz' g) ] end.
Fixpoint
genTreeSized
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Tree", "liftGen3", "nat", "returnGen" ]
Of course, this fixpoint will not pass Coq's termination check. Attempting to justify this fixpoint to oneself, one might say that at some point the random generation will pick a [Leaf] so it will eventually terminate. Sadly, in this case the expected size of the generated Tree is infinite... The s...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
genTreeSized' {A} (sz : nat) (g : G A) : G (Tree A)
:= match sz with | O => returnGen Leaf | S sz' => freq [ (1, returnGen Leaf) ; (sz, liftGen3 Node g (genTreeSized' sz' g) (genTreeSized' sz' g)) ] end.
Fixpoint
genTreeSized'
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Tree", "liftGen3", "nat", "returnGen" ]
[freq] takes a list of generators, each one tagged with a natural number that serves as the weight of that generator. In the following example, a [Leaf] will be generated 1 / (sz + 1) of the time, while a [Node] the remaining sz / (sz + 1) of the time.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
mirror {A : Type} (t : Tree A) : Tree A
:= match t with | Leaf => Leaf | Node x l r => Node x (mirror r) (mirror l) end.
Fixpoint
mirror
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Tree" ]
To showcase this generator, we will use the notion of mirroring a tree: swapping its left and right subtrees recursively.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
eq_tree (t1 t2 : Tree nat) : bool
:= match t1, t2 with | Leaf, Leaf => true | Node x1 l1 r1, Node x2 l2 r2 => Nat.eqb x1 x2 && eq_tree l1 l2 && eq_tree r1 r2 | _, _ => false end.
Fixpoint
eq_tree
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Tree", "nat" ]
We also need a simple structural equality on trees
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
mirrorP (t : Tree nat)
:= eq_tree (mirror (mirror t)) t.
Definition
mirrorP
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Tree", "eq_tree", "mirror", "nat" ]
One expects that [mirror] should be unipotent; mirroring a tree twice yields the original tree.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
faultyMirrorP (t : Tree nat)
:= eq_tree (mirror t) t.
Definition
faultyMirrorP
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Tree", "eq_tree", "mirror", "nat" ]
QuickChick quickly responds that this property passed 10000 tests, so we gain some confidence in its truth. But what would happend if we had the *wrong* property?
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
shrinkTree {A} (s : A -> list A) (t : Tree A) : list (Tree A)
:= match t with | Leaf => [] | Node x l r => [l] ++ [r] ++ map (fun x' => Node x' l r) (s x) ++ map (fun l' => Node x l' r) (shrinkTree s l) ++ map (fun r' => Node x l r') (shrinkTree s r) end.
Fixpoint
shrinkTree
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Tree", "map" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
genTree {A} `{Gen A} : GenSized (Tree A)
:= {| arbitrarySized n := genTreeSized n arbitrary |}.
Instance
genTree
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Gen", "GenSized", "Tree", "genTreeSized" ]
[sized] receives a function that given a number produces a generator, just like [genTreeSized'], and returns a generator that uses the size information inside the [G] monad. The [shrink] function is simply a shrinker like [shrinkTree].
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
shrTree {A} `{Shrink A} : Shrink (Tree A)
:= {| shrink x := shrinkTree shrink x |}.
Instance
shrTree
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Shrink", "Tree", "shrinkTree" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
size {A} (t : Tree A) : nat
:= match t with | Leaf => O | Node _ l r => 1 + size l + size r end.
Fixpoint
size
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Tree", "nat" ]
Earlier in this tutorial we claimed that [genTreeSized] produced "too many" [Leaf]s. But how can we justify that? Just looking at the result of [Sample] gives us an idea that something is going wrong but just observing a handful of samples cannot realistically provide statistical guarantees. That is whe...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
treeProp (g : nat -> G nat -> G (Tree nat)) n
:= forAll (g n (choose (0,n))) (fun t => collect (size t) true).
Definition
treeProp
tutorials
tutorials/BasicUsage.v
[ "QuickChick", "QcDefaultNotation", "List", "ZArith", "ListNotations", "String" ]
[ "Tree", "choose", "collect", "forAll", "nat", "size" ]
If we were to write a dummy property to check our generators and measure the size of generated trees, we could use [treeProp] below.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
string
:= list ascii.
Definition
string
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[]
This example is taken from the Logical Foundations Volume of Software Foundations textbook
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
reg_exp (T : Type) : Type
:= | EmptySet | EmptyStr | Char (t : T) | App (r1 r2 : reg_exp T) | Union (r1 r2 : reg_exp T) | Star (r : reg_exp T).
Inductive
reg_exp
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[]
We start with an inductive definion of the regular expression data type
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
EnumSizedreg_exp_SizedMonotonic T {_ : Enum T} : SizedMonotonic (@enumSized _ (@EnumSizedreg_exp T _)).
Proof. derive_enum_SizedMonotonic. Qed.
Instance
EnumSizedreg_exp_SizedMonotonic
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "Enum", "SizedMonotonic", "derive_enum_SizedMonotonic" ]
After automatically generating the [EnumSized] instance, we can generate correctness proofs. To do this proof we first define a "set-of-outcomes" semantics for our enumerator. In particular, the combinator [semEnumSize], with signature: semEnumSize : forall {A : Type}, E A -> nat -> set A maps an enum...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
EnumSizedreg_exp_SizeMonotonic T `{EnumMonotonic T}: forall s, SizeMonotonic (@enumSized _ (@EnumSizedreg_exp T _) s).
Proof. derive_enum_SizeMonotonic. Qed.
Instance
EnumSizedreg_exp_SizeMonotonic
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "EnumMonotonic", "SizeMonotonic", "derive_enum_SizeMonotonic" ]
Monotonicity in the internal size parameter -------------------------------------- We also prove that the enumerator is monotonic in the internal size parameter. That is, forall s s1 s2 : nat, s1 <= s2 -> semEnumSize (enumSized s1) s \subset semEnumSize (enumSized s2) s This property is captured...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
EnumSizedreg_expCorrect T `{EnumMonotonicCorrect T}: CorrectSized (@enumSized _ EnumSizedreg_exp).
Proof. derive_enum_Correct. Qed.
Instance
EnumSizedreg_expCorrect
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "CorrectSized", "EnumMonotonicCorrect", "derive_enum_Correct" ]
Correctness ----------- We use the two monotonicity properties to prove correctness. For simple enumerators like this one, correctness states: { x | exists s s', x \in semEnumSize (enumSized s) s' } <--> [set: A] That is, the set of elements that belongs to [semEnumSize (enumSized s) s'] for some si...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
app
:= @app.
Definition
app
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
exp_match {T: Type} : list T -> reg_exp T -> Prop
:= | MEmpty : [] =~ EmptyStr | MChar x : [x] =~ (Char x) | MApp s1 re1 s2 re2 : s1 =~ re1 -> s2 =~ re2 -> s1 ++ s2 =~ (App re1 re2) | MUnionL s1 re1 re2 : s1 =~ re1 -> s1 =~ (Union re1 re2) | MUnionR re1 s2 re2 : s2 =~ re2 -> s2 =~ (Union re1 re2) | MStar0 re : []...
Inductive
exp_match
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "reg_exp" ]
The following inductive relation holds whenever a string of characters drawn from a set [T] matches a regular expression.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
DecOptexp_match_monotonic {T} `{_ : Dec_Eq T} `{_ : EnumMonotonic T} (m : list T) n : DecOptSizeMonotonic (exp_match m n).
Proof. derive_mon. Qed.
Instance
DecOptexp_match_monotonic
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "DecOptSizeMonotonic", "Dec_Eq", "EnumMonotonic", "derive_mon", "exp_match" ]
We can now prove correctness of the derived checker. Monotonicity ------------ As before, we need to show monotonicity. In particular, we show that if the validity of a proposition has been decided, then the decision will not change by providing more fuel. In particular: forall (s1 s2 : nat) (b : b...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
DecOptexp_match_sound {T} `{_ : Dec_Eq T} `{_ : EnumMonotonicCorrect T} (m : list T) n : DecOptSoundPos (exp_match m n).
Proof. derive_sound. Qed.
Instance
DecOptexp_match_sound
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "DecOptSoundPos", "Dec_Eq", "EnumMonotonicCorrect", "derive_sound", "exp_match" ]
Soundness and Completeness -------------------------- Using monotonicity we can prove soundness and completeness. Soundness states: forall (P : Prop) (H : DecOpt P) (s : nat), decOpt s = Some true -> P That is, is [decOpt s] is [true] for some [s], then [P] holds. It is captured by the [DecOptSou...
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
DecOptexp_match_complete {T} `{_ : Dec_Eq T} `{_ : EnumMonotonicCorrect T} (m : list T) n : DecOptCompletePos (exp_match m n).
Proof. derive_complete. Qed.
Instance
DecOptexp_match_complete
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "DecOptCompletePos", "Dec_Eq", "EnumMonotonicCorrect", "derive_complete", "exp_match" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
EnumSizedSuchThateq_SizedMonotonic X {_ : Enum X} {_ : Dec_Eq X} (n : X) : SizedMonotonicOptFP (@enumSizeST _ _ (EnumSizedSuchThateq n)).
Proof. derive_enumST_SizedMonotonicFP. Qed.
Instance
EnumSizedSuchThateq_SizedMonotonic
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "Dec_Eq", "Enum", "SizedMonotonicOptFP", "derive_enumST_SizedMonotonicFP" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
EnumSizedSuchThateq_SizeMonotonic X `{_ : EnumMonotonic X} {_ : Dec_Eq X} (n : X) : forall s, SizeMonotonicOptFP (@enumSizeST _ _ (EnumSizedSuchThateq n) s).
Proof. derive_enumST_SizeMonotonicFP. Qed.
Instance
EnumSizedSuchThateq_SizeMonotonic
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "Dec_Eq", "EnumMonotonic", "SizeMonotonicOptFP", "derive_enumST_SizeMonotonicFP" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
EnumSizedSuchThatexp_match_SizedMonotonic {T} `{_ : Dec_Eq T} `{_ : EnumMonotonic T} e: SizedMonotonicOptFP (@enumSizeST _ _ (EnumSizedSuchThatexp_match e)).
Proof. derive_enumST_SizedMonotonicFP. Qed.
Instance
EnumSizedSuchThatexp_match_SizedMonotonic
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "Dec_Eq", "EnumMonotonic", "SizedMonotonicOptFP", "derive_enumST_SizedMonotonicFP" ]
As with the simple enumeration, before deriving correctness, we need to derive monotonicity.
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129
EnumSizedSuchThatexp_match_SizeMonotonic {T} `{_ : Dec_Eq T} `{_ : EnumMonotonic T} e : forall s, SizeMonotonicOptFP (@enumSizeST _ _ (EnumSizedSuchThatexp_match e) s).
Proof. derive_enumST_SizeMonotonicFP. Qed.
Instance
EnumSizedSuchThatexp_match_SizeMonotonic
tutorials
tutorials/DerivingProofs.v
[ "Coq", "Init.Nat", "List", "ListNotations", "QuickChick", "QcNotation", "QcDefaultNotation", "Coq.Strings.Ascii", "EnumProofs", "CheckerProofs" ]
[ "Dec_Eq", "EnumMonotonic", "SizeMonotonicOptFP", "derive_enumST_SizeMonotonicFP" ]
https://github.com/QuickChick/QuickChick
5a6c291b11e9affe059c0e1812612bc474d0a129