Mercurial > cpdt > repo

annotate src/MoreDep.v @ 440:f923024bd284

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Vertical spacing pass, through end of Subset
author Adam Chlipala <adam@chlipala.net>
date Mon, 30 Jul 2012 16:50:02 -0400
parents 8077352044b2
children 97c60ffdb998
rev   line source
adam@371 1 (* Copyright (c) 2008-2012, Adam Chlipala
adamc@83 2 *
adamc@83 3 * This work is licensed under a
adamc@83 4 * Creative Commons Attribution-Noncommercial-No Derivative Works 3.0
adamc@83 5 * Unported License.
adamc@83 6 * The license text is available at:
adamc@83 7 * http://creativecommons.org/licenses/by-nc-nd/3.0/
adamc@83 8 *)
adamc@83 9
adamc@83 10 (* begin hide *)
adamc@85 11 Require Import Arith Bool List.
adamc@83 12
adam@314 13 Require Import CpdtTactics MoreSpecif.
adamc@83 14
adamc@83 15 Set Implicit Arguments.
adamc@83 16 (* end hide *)
adamc@83 17
adamc@83 18
adamc@83 19 (** %\chapter{More Dependent Types}% *)
adamc@83 20
adam@425 21 (** Subset types and their relatives help us integrate verification with programming. Though they reorganize the certified programmer's workflow, they tend not to have deep effects on proofs. We write largely the same proofs as we would for classical verification, with some of the structure moved into the programs themselves. It turns out that, when we use dependent types to their full potential, we warp the development and proving process even more than that, picking up "free theorems" to the extent that often a certified program is hardly more complex than its uncertified counterpart in Haskell or ML.
adamc@83 22
adam@338 23 In particular, we have only scratched the tip of the iceberg that is Coq's inductive definition mechanism. The inductive types we have seen so far have their counterparts in the other proof assistants that we surveyed in Chapter 1. This chapter explores the strange new world of dependent inductive datatypes (that is, dependent inductive types outside [Prop]), a possibility that sets Coq apart from all of the competition not based on type theory. *)
adamc@83 24
adamc@84 25
adamc@84 26 (** * Length-Indexed Lists *)
adamc@84 27
adam@338 28 (** Many introductions to dependent types start out by showing how to use them to eliminate array bounds checks%\index{array bounds checks}%. When the type of an array tells you how many elements it has, your compiler can detect out-of-bounds dereferences statically. Since we are working in a pure functional language, the next best thing is length-indexed lists%\index{length-indexed lists}%, which the following code defines. *)
adamc@84 29
adamc@84 30 Section ilist.
adamc@84 31 Variable A : Set.
adamc@84 32
adamc@84 33 Inductive ilist : nat -> Set :=
adamc@84 34 | Nil : ilist O
adamc@84 35 | Cons : forall n, A -> ilist n -> ilist (S n).
adamc@84 36
adamc@84 37 (** We see that, within its section, [ilist] is given type [nat -> Set]. Previously, every inductive type we have seen has either had plain [Set] as its type or has been a predicate with some type ending in [Prop]. The full generality of inductive definitions lets us integrate the expressivity of predicates directly into our normal programming.
adamc@84 38
adam@405 39 The [nat] argument to [ilist] tells us the length of the list. The types of [ilist]'s constructors tell us that a [Nil] list has length [O] and that a [Cons] list has length one greater than the length of its tail. We may apply [ilist] to any natural number, even natural numbers that are only known at runtime. It is this breaking of the%\index{phase distinction}% _phase distinction_ that characterizes [ilist] as _dependently typed_.
adamc@84 40
adamc@213 41 In expositions of list types, we usually see the length function defined first, but here that would not be a very productive function to code. Instead, let us implement list concatenation. *)
adamc@84 42
adamc@213 43 Fixpoint app n1 (ls1 : ilist n1) n2 (ls2 : ilist n2) : ilist (n1 + n2) :=
adamc@213 44 match ls1 with
adamc@213 45 | Nil => ls2
adamc@213 46 | Cons _ x ls1' => Cons x (app ls1' ls2)
adamc@213 47 end.
adamc@84 48
adam@338 49 (** Past Coq versions signalled an error for this definition. The code is still invalid within Coq's core language, but current Coq versions automatically add annotations to the original program, producing a valid core program. These are the annotations on [match] discriminees that we began to study in the previous chapter. We can rewrite [app] to give the annotations explicitly. *)
adamc@100 50
adamc@100 51 (* begin thide *)
adam@338 52 Fixpoint app' n1 (ls1 : ilist n1) n2 (ls2 : ilist n2) : ilist (n1 + n2) :=
adam@338 53 match ls1 in (ilist n1) return (ilist (n1 + n2)) with
adam@338 54 | Nil => ls2
adam@338 55 | Cons _ x ls1' => Cons x (app' ls1' ls2)
adam@338 56 end.
adamc@100 57 (* end thide *)
adamc@84 58
adam@398 59 (** Using [return] alone allowed us to express a dependency of the [match] result type on the _value_ of the discriminee. What %\index{Gallina terms!in}%[in] adds to our arsenal is a way of expressing a dependency on the _type_ of the discriminee. Specifically, the [n1] in the [in] clause above is a _binding occurrence_ whose scope is the [return] clause.
adamc@84 60
adam@398 61 We may use [in] clauses only to bind names for the arguments of an inductive type family. That is, each [in] clause must be an inductive type family name applied to a sequence of underscores and variable names of the proper length. The positions for _parameters_ to the type family must all be underscores. Parameters are those arguments declared with section variables or with entries to the left of the first colon in an inductive definition. They cannot vary depending on which constructor was used to build the discriminee, so Coq prohibits pointless matches on them. It is those arguments defined in the type to the right of the colon that we may name with [in] clauses.
adamc@84 62
adam@425 63 Our [app] function could be typed in so-called%\index{stratified type systems}% _stratified_ type systems, which avoid true dependency. That is, we could consider the length indices to lists to live in a separate, compile-time-only universe from the lists themselves. This stratification between a compile-time universe and a run-time universe, with no references to the latter in the former, gives rise to the terminology "stratified." Our next example would be harder to implement in a stratified system. We write an injection function from regular lists to length-indexed lists. A stratified implementation would need to duplicate the definition of lists across compile-time and run-time versions, and the run-time versions would need to be indexed by the compile-time versions. *)
adamc@84 64
adamc@100 65 (* EX: Implement injection from normal lists *)
adamc@100 66
adamc@100 67 (* begin thide *)
adamc@84 68 Fixpoint inject (ls : list A) : ilist (length ls) :=
adamc@213 69 match ls with
adamc@84 70 | nil => Nil
adamc@84 71 | h :: t => Cons h (inject t)
adamc@84 72 end.
adamc@84 73
adamc@84 74 (** We can define an inverse conversion and prove that it really is an inverse. *)
adamc@84 75
adamc@213 76 Fixpoint unject n (ls : ilist n) : list A :=
adamc@84 77 match ls with
adamc@84 78 | Nil => nil
adamc@84 79 | Cons _ h t => h :: unject t
adamc@84 80 end.
adamc@84 81
adamc@84 82 Theorem inject_inverse : forall ls, unject (inject ls) = ls.
adamc@84 83 induction ls; crush.
adamc@84 84 Qed.
adamc@100 85 (* end thide *)
adamc@100 86
adam@338 87 (* EX: Implement statically checked "car"/"hd" *)
adamc@84 88
adam@425 89 (** Now let us attempt a function that is surprisingly tricky to write. In ML, the list head function raises an exception when passed an empty list. With length-indexed lists, we can rule out such invalid calls statically, and here is a first attempt at doing so. We write [???] as a placeholder for a term that we do not know how to write, not for any real Coq notation like those introduced two chapters ago.
adamc@84 90
adamc@84 91 [[
adamc@84 92 Definition hd n (ls : ilist (S n)) : A :=
adamc@84 93 match ls with
adamc@84 94 | Nil => ???
adamc@84 95 | Cons _ h _ => h
adamc@84 96 end.
adamc@213 97
adamc@213 98 ]]
adamc@84 99
adamc@84 100 It is not clear what to write for the [Nil] case, so we are stuck before we even turn our function over to the type checker. We could try omitting the [Nil] case:
adamc@84 101
adamc@84 102 [[
adamc@84 103 Definition hd n (ls : ilist (S n)) : A :=
adamc@84 104 match ls with
adamc@84 105 | Cons _ h _ => h
adamc@84 106 end.
adam@338 107 ]]
adamc@84 108
adam@338 109 <<
adamc@84 110 Error: Non exhaustive pattern-matching: no clause found for pattern Nil
adam@338 111 >>
adamc@84 112
adam@425 113 Unlike in ML, we cannot use inexhaustive pattern matching, because there is no conception of a <<Match>> exception to be thrown. In fact, recent versions of Coq _do_ allow this, by implicit translation to a [match] that considers all constructors. It is educational to discover that encoding ourselves directly. We might try using an [in] clause somehow.
adamc@84 114
adamc@84 115 [[
adamc@84 116 Definition hd n (ls : ilist (S n)) : A :=
adamc@84 117 match ls in (ilist (S n)) with
adamc@84 118 | Cons _ h _ => h
adamc@84 119 end.
adamc@84 120 ]]
adamc@84 121
adam@338 122 <<
adam@338 123 Error: The reference n was not found in the current environment
adam@338 124 >>
adam@338 125
adam@398 126 In this and other cases, we feel like we want [in] clauses with type family arguments that are not variables. Unfortunately, Coq only supports variables in those positions. A completely general mechanism could only be supported with a solution to the problem of higher-order unification%~\cite{HOU}%, which is undecidable. There _are_ useful heuristics for handling non-variable indices which are gradually making their way into Coq, but we will spend some time in this and the next few chapters on effective pattern matching on dependent types using only the primitive [match] annotations.
adamc@84 127
adamc@84 128 Our final, working attempt at [hd] uses an auxiliary function and a surprising [return] annotation. *)
adamc@84 129
adamc@100 130 (* begin thide *)
adamc@84 131 Definition hd' n (ls : ilist n) :=
adamc@84 132 match ls in (ilist n) return (match n with O => unit | S _ => A end) with
adamc@84 133 | Nil => tt
adamc@84 134 | Cons _ h _ => h
adamc@84 135 end.
adamc@84 136
adam@283 137 Check hd'.
adam@283 138 (** %\vspace{-.15in}% [[
adam@283 139 hd'
adam@283 140 : forall n : nat, ilist n -> match n with
adam@283 141 | 0 => unit
adam@283 142 | S _ => A
adam@283 143 end
adam@283 144
adam@302 145 ]]
adam@302 146 *)
adam@283 147
adamc@84 148 Definition hd n (ls : ilist (S n)) : A := hd' ls.
adamc@100 149 (* end thide *)
adamc@84 150
adam@338 151 End ilist.
adam@338 152
adamc@84 153 (** We annotate our main [match] with a type that is itself a [match]. We write that the function [hd'] returns [unit] when the list is empty and returns the carried type [A] in all other cases. In the definition of [hd], we just call [hd']. Because the index of [ls] is known to be nonzero, the type checker reduces the [match] in the type of [hd'] to [A]. *)
adamc@84 154
adamc@85 155
adam@371 156 (** * The One Rule of Dependent Pattern Matching in Coq *)
adam@371 157
adam@405 158 (** The rest of this chapter will demonstrate a few other elegant applications of dependent types in Coq. Readers encountering such ideas for the first time often feel overwhelmed, concluding that there is some magic at work whereby Coq sometimes solves the halting problem for the programmer and sometimes does not, applying automated program understanding in a way far beyond what is found in conventional languages. The point of this section is to cut off that sort of thinking right now! Dependent type-checking in Coq follows just a few algorithmic rules. Chapters 10 and 12 introduce many of those rules more formally, and the main additional rule is centered on%\index{dependent pattern matching}% _dependent pattern matching_ of the kind we met in the previous section.
adam@371 159
adam@405 160 A dependent pattern match is a [match] expression where the type of the overall [match] is a function of the value and/or the type of the%\index{discriminee}% _discriminee_, the value being matched on. In other words, the [match] type _depends_ on the discriminee.
adam@371 161
adam@398 162 When exactly will Coq accept a dependent pattern match as well-typed? Some other dependently typed languages employ fancy decision procedures to determine when programs satisfy their very expressive types. The situation in Coq is just the opposite. Only very straightforward symbolic rules are applied. Such a design choice has its drawbacks, as it forces programmers to do more work to convince the type checker of program validity. However, the great advantage of a simple type checking algorithm is that its action on _invalid_ programs is easier to understand!
adam@371 163
adam@371 164 We come now to the one rule of dependent pattern matching in Coq. A general dependent pattern match assumes this form (with unnecessary parentheses included to make the syntax easier to parse):
adam@371 165 [[
adam@371 166 match E in (T x1 ... xn) as y return U with
adam@371 167 | C z1 ... zm => B
adam@371 168 | ...
adam@371 169 end
adam@371 170 ]]
adam@371 171
adam@371 172 The discriminee is a term [E], a value in some inductive type family [T], which takes [n] arguments. An %\index{in clause}%[in] clause binds an explicit name [xi] for the [i]th argument passed to [T] in the type of [E]. An %\index{as clause}%[as] clause binds the name [y] to refer to the discriminee [E].
adam@371 173
adam@371 174 We bind these new variables [xi] and [y] so that they may be referred to in [U], a type given in the %\index{return clause}%[return] clause. The overall type of the [match] will be [U], with [E] substituted for [y], and with each [xi] substituted by the actual argument appearing in that position within [E]'s type.
adam@371 175
adam@371 176 In general, each case of a [match] may have a pattern built up in several layers from the constructors of various inductive type families. To keep this exposition simple, we will focus on patterns that are just single applications of inductive type constructors to lists of variables. Coq actually compiles the more general kind of pattern matching into this more restricted kind automatically, so understanding the typing of [match] requires understanding the typing of [match]es lowered to match one constructor at a time.
adam@371 177
adam@371 178 The last piece of the typing rule tells how to type-check a [match] case. A generic constructor application [C z1 ... zm] has some type [T x1' ... xn'], an application of the type family used in [E]'s type, probably with occurrences of the [zi] variables. From here, a simple recipe determines what type we will require for the case body [B]. The type of [B] should be [U] with the following two substitutions applied: we replace [y] (the [as] clause variable) with [C z1 ... zm], and we replace each [xi] (the [in] clause variables) with [xi']. In other words, we specialize the result type based on what we learn based on which pattern has matched the discriminee.
adam@371 179
adam@371 180 This is an exhaustive description of the ways to specify how to take advantage of which pattern has matched! No other mechanisms come into play. For instance, there is no way to specify that the types of certain free variables should be refined based on which pattern has matched. In the rest of the book, we will learn design patterns for achieving similar effects, where each technique leads to an encoding only in terms of [in], [as], and [return] clauses.
adam@371 181
adam@425 182 A few details have been omitted above. In Chapter 3, we learned that inductive type families may have both%\index{parameters}% _parameters_ and regular arguments. Within an [in] clause, a parameter position must have the wildcard [_] written, instead of a variable. (In general, Coq uses wildcard [_]'s either to indicate pattern variables that will not be mentioned again or to indicate positions where we would like type inference to infer the appropriate terms.) Furthermore, recent Coq versions are adding more and more heuristics to infer dependent [match] annotations in certain conditions. The general annotation inference problem is undecidable, so there will always be serious limitations on how much work these heuristics can do. When in doubt about why a particular dependent [match] is failing to type-check, add an explicit [return] annotation! At that point, the mechanical rule sketched in this section will provide a complete account of "what the type checker is thinking." Be sure to avoid the common pitfall of writing a [return] annotation that does not mention any variables bound by [in] or [as]; such a [match] will never refine typing requirements based on which pattern has matched. (One simple exception to this rule is that, when the discriminee is a variable, that same variable may be treated as if it were repeated as an [as] clause.) *)
adam@371 183
adam@371 184
adamc@85 185 (** * A Tagless Interpreter *)
adamc@85 186
adam@405 187 (** A favorite example for motivating the power of functional programming is implementation of a simple expression language interpreter. In ML and Haskell, such interpreters are often implemented using an algebraic datatype of values, where at many points it is checked that a value was built with the right constructor of the value type. With dependent types, we can implement a%\index{tagless interpreters}% _tagless_ interpreter that both removes this source of runtime inefficiency and gives us more confidence that our implementation is correct. *)
adamc@85 188
adamc@85 189 Inductive type : Set :=
adamc@85 190 | Nat : type
adamc@85 191 | Bool : type
adamc@85 192 | Prod : type -> type -> type.
adamc@85 193
adamc@85 194 Inductive exp : type -> Set :=
adamc@85 195 | NConst : nat -> exp Nat
adamc@85 196 | Plus : exp Nat -> exp Nat -> exp Nat
adamc@85 197 | Eq : exp Nat -> exp Nat -> exp Bool
adamc@85 198
adamc@85 199 | BConst : bool -> exp Bool
adamc@85 200 | And : exp Bool -> exp Bool -> exp Bool
adamc@85 201 | If : forall t, exp Bool -> exp t -> exp t -> exp t
adamc@85 202
adamc@85 203 | Pair : forall t1 t2, exp t1 -> exp t2 -> exp (Prod t1 t2)
adamc@85 204 | Fst : forall t1 t2, exp (Prod t1 t2) -> exp t1
adamc@85 205 | Snd : forall t1 t2, exp (Prod t1 t2) -> exp t2.
adamc@85 206
adamc@85 207 (** We have a standard algebraic datatype [type], defining a type language of naturals, booleans, and product (pair) types. Then we have the indexed inductive type [exp], where the argument to [exp] tells us the encoded type of an expression. In effect, we are defining the typing rules for expressions simultaneously with the syntax.
adamc@85 208
adam@398 209 We can give types and expressions semantics in a new style, based critically on the chance for _type-level computation_. *)
adamc@85 210
adamc@85 211 Fixpoint typeDenote (t : type) : Set :=
adamc@85 212 match t with
adamc@85 213 | Nat => nat
adamc@85 214 | Bool => bool
adamc@85 215 | Prod t1 t2 => typeDenote t1 * typeDenote t2
adamc@85 216 end%type.
adamc@85 217
adam@425 218 (** The [typeDenote] function compiles types of our object language into "native" Coq types. It is deceptively easy to implement. The only new thing we see is the [%]%\coqdocvar{%#<tt>#type#</tt>#%}% annotation, which tells Coq to parse the [match] expression using the notations associated with types. Without this annotation, the [*] would be interpreted as multiplication on naturals, rather than as the product type constructor. The token %\coqdocvar{%#<tt>#type#</tt>#%}% is one example of an identifer bound to a%\index{notation scope delimiter}% _notation scope delimiter_. In this book, we will not go into more detail on notation scopes, but the Coq manual can be consulted for more information.
adamc@85 219
adamc@85 220 We can define a function [expDenote] that is typed in terms of [typeDenote]. *)
adamc@85 221
adamc@213 222 Fixpoint expDenote t (e : exp t) : typeDenote t :=
adamc@213 223 match e with
adamc@85 224 | NConst n => n
adamc@85 225 | Plus e1 e2 => expDenote e1 + expDenote e2
adamc@85 226 | Eq e1 e2 => if eq_nat_dec (expDenote e1) (expDenote e2) then true else false
adamc@85 227
adamc@85 228 | BConst b => b
adamc@85 229 | And e1 e2 => expDenote e1 && expDenote e2
adamc@85 230 | If _ e' e1 e2 => if expDenote e' then expDenote e1 else expDenote e2
adamc@85 231
adamc@85 232 | Pair _ _ e1 e2 => (expDenote e1, expDenote e2)
adamc@85 233 | Fst _ _ e' => fst (expDenote e')
adamc@85 234 | Snd _ _ e' => snd (expDenote e')
adamc@85 235 end.
adamc@85 236
adam@437 237 (* begin hide *)
adam@437 238 (* begin thide *)
adam@437 239 Definition sumboool := sumbool.
adam@437 240 (* end thide *)
adam@437 241 (* end hide *)
adam@437 242
adamc@213 243 (** Despite the fancy type, the function definition is routine. In fact, it is less complicated than what we would write in ML or Haskell 98, since we do not need to worry about pushing final values in and out of an algebraic datatype. The only unusual thing is the use of an expression of the form [if E then true else false] in the [Eq] case. Remember that [eq_nat_dec] has a rich dependent type, rather than a simple boolean type. Coq's native [if] is overloaded to work on a test of any two-constructor type, so we can use [if] to build a simple boolean from the [sumbool] that [eq_nat_dec] returns.
adamc@85 244
adamc@85 245 We can implement our old favorite, a constant folding function, and prove it correct. It will be useful to write a function [pairOut] that checks if an [exp] of [Prod] type is a pair, returning its two components if so. Unsurprisingly, a first attempt leads to a type error.
adamc@85 246
adamc@85 247 [[
adamc@85 248 Definition pairOut t1 t2 (e : exp (Prod t1 t2)) : option (exp t1 * exp t2) :=
adamc@85 249 match e in (exp (Prod t1 t2)) return option (exp t1 * exp t2) with
adamc@85 250 | Pair _ _ e1 e2 => Some (e1, e2)
adamc@85 251 | _ => None
adamc@85 252 end.
adam@338 253 ]]
adamc@85 254
adam@338 255 <<
adamc@85 256 Error: The reference t2 was not found in the current environment
adam@338 257 >>
adamc@85 258
adamc@85 259 We run again into the problem of not being able to specify non-variable arguments in [in] clauses. The problem would just be hopeless without a use of an [in] clause, though, since the result type of the [match] depends on an argument to [exp]. Our solution will be to use a more general type, as we did for [hd]. First, we define a type-valued function to use in assigning a type to [pairOut]. *)
adamc@85 260
adamc@100 261 (* EX: Define a function [pairOut : forall t1 t2, exp (Prod t1 t2) -> option (exp t1 * exp t2)] *)
adamc@100 262
adamc@100 263 (* begin thide *)
adamc@85 264 Definition pairOutType (t : type) :=
adamc@85 265 match t with
adamc@85 266 | Prod t1 t2 => option (exp t1 * exp t2)
adamc@85 267 | _ => unit
adamc@85 268 end.
adamc@85 269
adamc@85 270 (** When passed a type that is a product, [pairOutType] returns our final desired type. On any other input type, [pairOutType] returns [unit], since we do not care about extracting components of non-pairs. Now we can write another helper function to provide the default behavior of [pairOut], which we will apply for inputs that are not literal pairs. *)
adamc@85 271
adamc@85 272 Definition pairOutDefault (t : type) :=
adamc@85 273 match t return (pairOutType t) with
adamc@85 274 | Prod _ _ => None
adamc@85 275 | _ => tt
adamc@85 276 end.
adamc@85 277
adamc@85 278 (** Now [pairOut] is deceptively easy to write. *)
adamc@85 279
adamc@85 280 Definition pairOut t (e : exp t) :=
adamc@85 281 match e in (exp t) return (pairOutType t) with
adamc@85 282 | Pair _ _ e1 e2 => Some (e1, e2)
adamc@85 283 | _ => pairOutDefault _
adamc@85 284 end.
adamc@100 285 (* end thide *)
adamc@85 286
adam@338 287 (** There is one important subtlety in this definition. Coq allows us to use convenient ML-style pattern matching notation, but, internally and in proofs, we see that patterns are expanded out completely, matching one level of inductive structure at a time. Thus, the default case in the [match] above expands out to one case for each constructor of [exp] besides [Pair], and the underscore in [pairOutDefault _] is resolved differently in each case. From an ML or Haskell programmer's perspective, what we have here is type inference determining which code is run (returning either [None] or [tt]), which goes beyond what is possible with type inference guiding parametric polymorphism in Hindley-Milner languages%\index{Hindley-Milner}%, but is similar to what goes on with Haskell type classes%\index{type classes}%.
adamc@85 288
adam@417 289 With [pairOut] available, we can write [cfold] in a straightforward way. There are really no surprises beyond that Coq verifies that this code has such an expressive type, given the small annotation burden. In some places, we see that Coq's [match] annotation inference is too smart for its own good, and we have to turn that inference with explicit [return] clauses. *)
adamc@85 290
adamc@204 291 Fixpoint cfold t (e : exp t) : exp t :=
adamc@204 292 match e with
adamc@85 293 | NConst n => NConst n
adamc@85 294 | Plus e1 e2 =>
adamc@85 295 let e1' := cfold e1 in
adamc@85 296 let e2' := cfold e2 in
adam@417 297 match e1', e2' return exp Nat with
adamc@85 298 | NConst n1, NConst n2 => NConst (n1 + n2)
adamc@85 299 | _, _ => Plus e1' e2'
adamc@85 300 end
adamc@85 301 | Eq e1 e2 =>
adamc@85 302 let e1' := cfold e1 in
adamc@85 303 let e2' := cfold e2 in
adam@417 304 match e1', e2' return exp Bool with
adamc@85 305 | NConst n1, NConst n2 => BConst (if eq_nat_dec n1 n2 then true else false)
adamc@85 306 | _, _ => Eq e1' e2'
adamc@85 307 end
adamc@85 308
adamc@85 309 | BConst b => BConst b
adamc@85 310 | And e1 e2 =>
adamc@85 311 let e1' := cfold e1 in
adamc@85 312 let e2' := cfold e2 in
adam@417 313 match e1', e2' return exp Bool with
adamc@85 314 | BConst b1, BConst b2 => BConst (b1 && b2)
adamc@85 315 | _, _ => And e1' e2'
adamc@85 316 end
adamc@85 317 | If _ e e1 e2 =>
adamc@85 318 let e' := cfold e in
adamc@85 319 match e' with
adamc@85 320 | BConst true => cfold e1
adamc@85 321 | BConst false => cfold e2
adamc@85 322 | _ => If e' (cfold e1) (cfold e2)
adamc@85 323 end
adamc@85 324
adamc@85 325 | Pair _ _ e1 e2 => Pair (cfold e1) (cfold e2)
adamc@85 326 | Fst _ _ e =>
adamc@85 327 let e' := cfold e in
adamc@85 328 match pairOut e' with
adamc@85 329 | Some p => fst p
adamc@85 330 | None => Fst e'
adamc@85 331 end
adamc@85 332 | Snd _ _ e =>
adamc@85 333 let e' := cfold e in
adamc@85 334 match pairOut e' with
adamc@85 335 | Some p => snd p
adamc@85 336 | None => Snd e'
adamc@85 337 end
adamc@85 338 end.
adamc@85 339
adamc@85 340 (** The correctness theorem for [cfold] turns out to be easy to prove, once we get over one serious hurdle. *)
adamc@85 341
adamc@85 342 Theorem cfold_correct : forall t (e : exp t), expDenote e = expDenote (cfold e).
adamc@100 343 (* begin thide *)
adamc@85 344 induction e; crush.
adamc@85 345
adamc@85 346 (** The first remaining subgoal is:
adamc@85 347
adamc@85 348 [[
adamc@85 349 expDenote (cfold e1) + expDenote (cfold e2) =
adamc@85 350 expDenote
adamc@85 351 match cfold e1 with
adamc@85 352 | NConst n1 =>
adamc@85 353 match cfold e2 with
adamc@85 354 | NConst n2 => NConst (n1 + n2)
adamc@85 355 | Plus _ _ => Plus (cfold e1) (cfold e2)
adamc@85 356 | Eq _ _ => Plus (cfold e1) (cfold e2)
adamc@85 357 | BConst _ => Plus (cfold e1) (cfold e2)
adamc@85 358 | And _ _ => Plus (cfold e1) (cfold e2)
adamc@85 359 | If _ _ _ _ => Plus (cfold e1) (cfold e2)
adamc@85 360 | Pair _ _ _ _ => Plus (cfold e1) (cfold e2)
adamc@85 361 | Fst _ _ _ => Plus (cfold e1) (cfold e2)
adamc@85 362 | Snd _ _ _ => Plus (cfold e1) (cfold e2)
adamc@85 363 end
adamc@85 364 | Plus _ _ => Plus (cfold e1) (cfold e2)
adamc@85 365 | Eq _ _ => Plus (cfold e1) (cfold e2)
adamc@85 366 | BConst _ => Plus (cfold e1) (cfold e2)
adamc@85 367 | And _ _ => Plus (cfold e1) (cfold e2)
adamc@85 368 | If _ _ _ _ => Plus (cfold e1) (cfold e2)
adamc@85 369 | Pair _ _ _ _ => Plus (cfold e1) (cfold e2)
adamc@85 370 | Fst _ _ _ => Plus (cfold e1) (cfold e2)
adamc@85 371 | Snd _ _ _ => Plus (cfold e1) (cfold e2)
adamc@85 372 end
adamc@213 373
adamc@85 374 ]]
adamc@85 375
adamc@85 376 We would like to do a case analysis on [cfold e1], and we attempt that in the way that has worked so far.
adamc@85 377
adamc@85 378 [[
adamc@85 379 destruct (cfold e1).
adam@338 380 ]]
adamc@85 381
adam@338 382 <<
adamc@85 383 User error: e1 is used in hypothesis e
adam@338 384 >>
adamc@85 385
adamc@85 386 Coq gives us another cryptic error message. Like so many others, this one basically means that Coq is not able to build some proof about dependent types. It is hard to generate helpful and specific error messages for problems like this, since that would require some kind of understanding of the dependency structure of a piece of code. We will encounter many examples of case-specific tricks for recovering from errors like this one.
adamc@85 387
adam@350 388 For our current proof, we can use a tactic [dep_destruct]%\index{tactics!dep\_destruct}% defined in the book [CpdtTactics] module. General elimination/inversion of dependently typed hypotheses is undecidable, since it must be implemented with [match] expressions that have the restriction on [in] clauses that we have already discussed. The tactic [dep_destruct] makes a best effort to handle some common cases, relying upon the more primitive %\index{tactics!dependent destruction}%[dependent destruction] tactic that comes with Coq. In a future chapter, we will learn about the explicit manipulation of equality proofs that is behind [dep_destruct]'s implementation in Ltac, but for now, we treat it as a useful black box. (In Chapter 12, we will also see how [dependent destruction] forces us to make a larger philosophical commitment about our logic than we might like, and we will see some workarounds.) *)
adamc@85 389
adamc@85 390 dep_destruct (cfold e1).
adamc@85 391
adamc@85 392 (** This successfully breaks the subgoal into 5 new subgoals, one for each constructor of [exp] that could produce an [exp Nat]. Note that [dep_destruct] is successful in ruling out the other cases automatically, in effect automating some of the work that we have done manually in implementing functions like [hd] and [pairOut].
adamc@85 393
adam@405 394 This is the only new trick we need to learn to complete the proof. We can back up and give a short, automated proof. *)
adamc@85 395
adamc@85 396 Restart.
adamc@85 397
adamc@85 398 induction e; crush;
adamc@85 399 repeat (match goal with
adam@405 400 | [ |- context[match cfold ?E with NConst _ => _ | _ => _ end] ] =>
adamc@213 401 dep_destruct (cfold E)
adamc@213 402 | [ |- context[match pairOut (cfold ?E) with Some _ => _
adamc@213 403 | None => _ end] ] =>
adamc@213 404 dep_destruct (cfold E)
adamc@85 405 | [ |- (if ?E then _ else _) = _ ] => destruct E
adamc@85 406 end; crush).
adamc@85 407 Qed.
adamc@100 408 (* end thide *)
adamc@86 409
adam@405 410 (** With this example, we get a first taste of how to build automated proofs that adapt automatically to changes in function definitions. *)
adam@405 411
adamc@86 412
adam@338 413 (** * Dependently Typed Red-Black Trees *)
adamc@94 414
adam@338 415 (** Red-black trees are a favorite purely functional data structure with an interesting invariant. We can use dependent types to enforce that operations on red-black trees preserve the invariant. For simplicity, we specialize our red-black trees to represent sets of [nat]s. *)
adamc@100 416
adamc@94 417 Inductive color : Set := Red | Black.
adamc@94 418
adamc@94 419 Inductive rbtree : color -> nat -> Set :=
adamc@94 420 | Leaf : rbtree Black 0
adamc@214 421 | RedNode : forall n, rbtree Black n -> nat -> rbtree Black n -> rbtree Red n
adamc@94 422 | BlackNode : forall c1 c2 n, rbtree c1 n -> nat -> rbtree c2 n -> rbtree Black (S n).
adamc@94 423
adamc@214 424 (** A value of type [rbtree c d] is a red-black tree node whose root has color [c] and that has black depth [d]. The latter property means that there are no more than [d] black-colored nodes on any path from the root to a leaf. *)
adamc@214 425
adamc@214 426 (** At first, it can be unclear that this choice of type indices tracks any useful property. To convince ourselves, we will prove that every red-black tree is balanced. We will phrase our theorem in terms of a depth calculating function that ignores the extra information in the types. It will be useful to parameterize this function over a combining operation, so that we can re-use the same code to calculate the minimum or maximum height among all paths from root to leaf. *)
adamc@214 427
adamc@100 428 (* EX: Prove that every [rbtree] is balanced. *)
adamc@100 429
adamc@100 430 (* begin thide *)
adamc@95 431 Require Import Max Min.
adamc@95 432
adamc@95 433 Section depth.
adamc@95 434 Variable f : nat -> nat -> nat.
adamc@95 435
adamc@214 436 Fixpoint depth c n (t : rbtree c n) : nat :=
adamc@95 437 match t with
adamc@95 438 | Leaf => 0
adamc@95 439 | RedNode _ t1 _ t2 => S (f (depth t1) (depth t2))
adamc@95 440 | BlackNode _ _ _ t1 _ t2 => S (f (depth t1) (depth t2))
adamc@95 441 end.
adamc@95 442 End depth.
adamc@95 443
adam@338 444 (** Our proof of balanced-ness decomposes naturally into a lower bound and an upper bound. We prove the lower bound first. Unsurprisingly, a tree's black depth provides such a bound on the minimum path length. We use the richly typed procedure [min_dec] to do case analysis on whether [min X Y] equals [X] or [Y]. *)
adamc@214 445
adam@283 446 Check min_dec.
adam@283 447 (** %\vspace{-.15in}% [[
adam@283 448 min_dec
adam@283 449 : forall n m : nat, {min n m = n} + {min n m = m}
adam@302 450 ]]
adam@302 451 *)
adam@283 452
adamc@95 453 Theorem depth_min : forall c n (t : rbtree c n), depth min t >= n.
adamc@95 454 induction t; crush;
adamc@95 455 match goal with
adamc@95 456 | [ |- context[min ?X ?Y] ] => destruct (min_dec X Y)
adamc@95 457 end; crush.
adamc@95 458 Qed.
adamc@95 459
adamc@214 460 (** There is an analogous upper-bound theorem based on black depth. Unfortunately, a symmetric proof script does not suffice to establish it. *)
adamc@214 461
adamc@214 462 Theorem depth_max : forall c n (t : rbtree c n), depth max t <= 2 * n + 1.
adamc@214 463 induction t; crush;
adamc@214 464 match goal with
adamc@214 465 | [ |- context[max ?X ?Y] ] => destruct (max_dec X Y)
adamc@214 466 end; crush.
adamc@214 467
adamc@214 468 (** Two subgoals remain. One of them is: [[
adamc@214 469 n : nat
adamc@214 470 t1 : rbtree Black n
adamc@214 471 n0 : nat
adamc@214 472 t2 : rbtree Black n
adamc@214 473 IHt1 : depth max t1 <= n + (n + 0) + 1
adamc@214 474 IHt2 : depth max t2 <= n + (n + 0) + 1
adamc@214 475 e : max (depth max t1) (depth max t2) = depth max t1
adamc@214 476 ============================
adamc@214 477 S (depth max t1) <= n + (n + 0) + 1
adamc@214 478
adamc@214 479 ]]
adamc@214 480
adam@398 481 We see that [IHt1] is _almost_ the fact we need, but it is not quite strong enough. We will need to strengthen our induction hypothesis to get the proof to go through. *)
adamc@214 482
adamc@214 483 Abort.
adamc@214 484
adamc@214 485 (** In particular, we prove a lemma that provides a stronger upper bound for trees with black root nodes. We got stuck above in a case about a red root node. Since red nodes have only black children, our IH strengthening will enable us to finish the proof. *)
adamc@214 486
adamc@95 487 Lemma depth_max' : forall c n (t : rbtree c n), match c with
adamc@95 488 | Red => depth max t <= 2 * n + 1
adamc@95 489 | Black => depth max t <= 2 * n
adamc@95 490 end.
adamc@95 491 induction t; crush;
adamc@95 492 match goal with
adamc@95 493 | [ |- context[max ?X ?Y] ] => destruct (max_dec X Y)
adamc@100 494 end; crush;
adamc@100 495 repeat (match goal with
adamc@214 496 | [ H : context[match ?C with Red => _ | Black => _ end] |- _ ] =>
adamc@214 497 destruct C
adamc@100 498 end; crush).
adamc@95 499 Qed.
adamc@95 500
adam@338 501 (** The original theorem follows easily from the lemma. We use the tactic %\index{tactics!generalize}%[generalize pf], which, when [pf] proves the proposition [P], changes the goal from [Q] to [P -> Q]. This transformation is useful because it makes the truth of [P] manifest syntactically, so that automation machinery can rely on [P], even if that machinery is not smart enough to establish [P] on its own. *)
adamc@214 502
adamc@95 503 Theorem depth_max : forall c n (t : rbtree c n), depth max t <= 2 * n + 1.
adamc@95 504 intros; generalize (depth_max' t); destruct c; crush.
adamc@95 505 Qed.
adamc@95 506
adamc@214 507 (** The final balance theorem establishes that the minimum and maximum path lengths of any tree are within a factor of two of each other. *)
adamc@214 508
adamc@95 509 Theorem balanced : forall c n (t : rbtree c n), 2 * depth min t + 1 >= depth max t.
adamc@95 510 intros; generalize (depth_min t); generalize (depth_max t); crush.
adamc@95 511 Qed.
adamc@100 512 (* end thide *)
adamc@95 513
adamc@214 514 (** Now we are ready to implement an example operation on our trees, insertion. Insertion can be thought of as breaking the tree invariants locally but then rebalancing. In particular, in intermediate states we find red nodes that may have red children. The type [rtree] captures the idea of such a node, continuing to track black depth as a type index. *)
adamc@95 515
adamc@94 516 Inductive rtree : nat -> Set :=
adamc@94 517 | RedNode' : forall c1 c2 n, rbtree c1 n -> nat -> rbtree c2 n -> rtree n.
adamc@94 518
adam@338 519 (** Before starting to define [insert], we define predicates capturing when a data value is in the set represented by a normal or possibly invalid tree. *)
adamc@214 520
adamc@96 521 Section present.
adamc@96 522 Variable x : nat.
adamc@96 523
adamc@214 524 Fixpoint present c n (t : rbtree c n) : Prop :=
adamc@96 525 match t with
adamc@96 526 | Leaf => False
adamc@96 527 | RedNode _ a y b => present a \/ x = y \/ present b
adamc@96 528 | BlackNode _ _ _ a y b => present a \/ x = y \/ present b
adamc@96 529 end.
adamc@96 530
adamc@96 531 Definition rpresent n (t : rtree n) : Prop :=
adamc@96 532 match t with
adamc@96 533 | RedNode' _ _ _ a y b => present a \/ x = y \/ present b
adamc@96 534 end.
adamc@96 535 End present.
adamc@96 536
adam@338 537 (** Insertion relies on two balancing operations. It will be useful to give types to these operations using a relative of the subset types from last chapter. While subset types let us pair a value with a proof about that value, here we want to pair a value with another non-proof dependently typed value. The %\index{Gallina terms!sigT}%[sigT] type fills this role. *)
adamc@214 538
adamc@100 539 Locate "{ _ : _ & _ }".
adamc@214 540 (** [[
adamc@214 541 Notation Scope
adamc@214 542 "{ x : A & P }" := sigT (fun x : A => P)
adam@302 543 ]]
adam@302 544 *)
adamc@214 545
adamc@100 546 Print sigT.
adamc@214 547 (** [[
adamc@214 548 Inductive sigT (A : Type) (P : A -> Type) : Type :=
adamc@214 549 existT : forall x : A, P x -> sigT P
adam@302 550 ]]
adam@302 551 *)
adamc@214 552
adamc@214 553 (** It will be helpful to define a concise notation for the constructor of [sigT]. *)
adamc@100 554
adamc@94 555 Notation "{< x >}" := (existT _ _ x).
adamc@94 556
adamc@214 557 (** Each balance function is used to construct a new tree whose keys include the keys of two input trees, as well as a new key. One of the two input trees may violate the red-black alternation invariant (that is, it has an [rtree] type), while the other tree is known to be valid. Crucially, the two input trees have the same black depth.
adamc@214 558
adam@338 559 A balance operation may return a tree whose root is of either color. Thus, we use a [sigT] type to package the result tree with the color of its root. Here is the definition of the first balance operation, which applies when the possibly invalid [rtree] belongs to the left of the valid [rbtree].
adam@338 560
adam@425 561 A quick word of encouragement: After writing this code, even I do not understand the precise details of how balancing works! I consulted Chris Okasaki's paper "Red-Black Trees in a Functional Setting" %\cite{Okasaki} %and transcribed the code to use dependent types. Luckily, the details are not so important here; types alone will tell us that insertion preserves balanced-ness, and we will prove that insertion produces trees containing the right keys.*)
adamc@214 562
adamc@94 563 Definition balance1 n (a : rtree n) (data : nat) c2 :=
adamc@214 564 match a in rtree n return rbtree c2 n
adamc@214 565 -> { c : color & rbtree c (S n) } with
adam@380 566 | RedNode' _ c0 _ t1 y t2 =>
adam@380 567 match t1 in rbtree c n return rbtree c0 n -> rbtree c2 n
adamc@214 568 -> { c : color & rbtree c (S n) } with
adamc@214 569 | RedNode _ a x b => fun c d =>
adamc@214 570 {<RedNode (BlackNode a x b) y (BlackNode c data d)>}
adamc@94 571 | t1' => fun t2 =>
adam@380 572 match t2 in rbtree c n return rbtree Black n -> rbtree c2 n
adamc@214 573 -> { c : color & rbtree c (S n) } with
adamc@214 574 | RedNode _ b x c => fun a d =>
adamc@214 575 {<RedNode (BlackNode a y b) x (BlackNode c data d)>}
adamc@95 576 | b => fun a t => {<BlackNode (RedNode a y b) data t>}
adamc@94 577 end t1'
adamc@94 578 end t2
adamc@94 579 end.
adamc@94 580
adam@405 581 (** We apply a trick that I call the%\index{convoy pattern}% _convoy pattern_. Recall that [match] annotations only make it possible to describe a dependence of a [match] _result type_ on the discriminee. There is no automatic refinement of the types of free variables. However, it is possible to effect such a refinement by finding a way to encode free variable type dependencies in the [match] result type, so that a [return] clause can express the connection.
adamc@214 582
adam@425 583 In particular, we can extend the [match] to return _functions over the free variables whose types we want to refine_. In the case of [balance1], we only find ourselves wanting to refine the type of one tree variable at a time. We match on one subtree of a node, and we want the type of the other subtree to be refined based on what we learn. We indicate this with a [return] clause starting like [rbtree _ n -> ...], where [n] is bound in an [in] pattern. Such a [match] expression is applied immediately to the "old version" of the variable to be refined, and the type checker is happy.
adamc@214 584
adam@338 585 Here is the symmetric function [balance2], for cases where the possibly invalid tree appears on the right rather than on the left. *)
adamc@214 586
adamc@94 587 Definition balance2 n (a : rtree n) (data : nat) c2 :=
adamc@94 588 match a in rtree n return rbtree c2 n -> { c : color & rbtree c (S n) } with
adam@380 589 | RedNode' _ c0 _ t1 z t2 =>
adam@380 590 match t1 in rbtree c n return rbtree c0 n -> rbtree c2 n
adamc@214 591 -> { c : color & rbtree c (S n) } with
adamc@214 592 | RedNode _ b y c => fun d a =>
adamc@214 593 {<RedNode (BlackNode a data b) y (BlackNode c z d)>}
adamc@94 594 | t1' => fun t2 =>
adam@380 595 match t2 in rbtree c n return rbtree Black n -> rbtree c2 n
adamc@214 596 -> { c : color & rbtree c (S n) } with
adamc@214 597 | RedNode _ c z' d => fun b a =>
adamc@214 598 {<RedNode (BlackNode a data b) z (BlackNode c z' d)>}
adamc@95 599 | b => fun a t => {<BlackNode t data (RedNode a z b)>}
adamc@94 600 end t1'
adamc@94 601 end t2
adamc@94 602 end.
adamc@94 603
adamc@214 604 (** Now we are almost ready to get down to the business of writing an [insert] function. First, we enter a section that declares a variable [x], for the key we want to insert. *)
adamc@214 605
adamc@94 606 Section insert.
adamc@94 607 Variable x : nat.
adamc@94 608
adamc@214 609 (** Most of the work of insertion is done by a helper function [ins], whose return types are expressed using a type-level function [insResult]. *)
adamc@214 610
adamc@94 611 Definition insResult c n :=
adamc@94 612 match c with
adamc@94 613 | Red => rtree n
adamc@94 614 | Black => { c' : color & rbtree c' n }
adamc@94 615 end.
adamc@94 616
adam@338 617 (** That is, inserting into a tree with root color [c] and black depth [n], the variety of tree we get out depends on [c]. If we started with a red root, then we get back a possibly invalid tree of depth [n]. If we started with a black root, we get back a valid tree of depth [n] with a root node of an arbitrary color.
adamc@214 618
adamc@214 619 Here is the definition of [ins]. Again, we do not want to dwell on the functional details. *)
adamc@214 620
adamc@214 621 Fixpoint ins c n (t : rbtree c n) : insResult c n :=
adamc@214 622 match t with
adamc@94 623 | Leaf => {< RedNode Leaf x Leaf >}
adamc@94 624 | RedNode _ a y b =>
adamc@94 625 if le_lt_dec x y
adamc@94 626 then RedNode' (projT2 (ins a)) y b
adamc@94 627 else RedNode' a y (projT2 (ins b))
adamc@94 628 | BlackNode c1 c2 _ a y b =>
adamc@94 629 if le_lt_dec x y
adamc@94 630 then
adamc@94 631 match c1 return insResult c1 _ -> _ with
adamc@94 632 | Red => fun ins_a => balance1 ins_a y b
adamc@94 633 | _ => fun ins_a => {< BlackNode (projT2 ins_a) y b >}
adamc@94 634 end (ins a)
adamc@94 635 else
adamc@94 636 match c2 return insResult c2 _ -> _ with
adamc@94 637 | Red => fun ins_b => balance2 ins_b y a
adamc@94 638 | _ => fun ins_b => {< BlackNode a y (projT2 ins_b) >}
adamc@94 639 end (ins b)
adamc@94 640 end.
adamc@94 641
adam@398 642 (** The one new trick is a variation of the convoy pattern. In each of the last two pattern matches, we want to take advantage of the typing connection between the trees [a] and [b]. We might naively apply the convoy pattern directly on [a] in the first [match] and on [b] in the second. This satisfies the type checker per se, but it does not satisfy the termination checker. Inside each [match], we would be calling [ins] recursively on a locally bound variable. The termination checker is not smart enough to trace the dataflow into that variable, so the checker does not know that this recursive argument is smaller than the original argument. We make this fact clearer by applying the convoy pattern on _the result of a recursive call_, rather than just on that call's argument.
adamc@214 643
adamc@214 644 Finally, we are in the home stretch of our effort to define [insert]. We just need a few more definitions of non-recursive functions. First, we need to give the final characterization of [insert]'s return type. Inserting into a red-rooted tree gives a black-rooted tree where black depth has increased, and inserting into a black-rooted tree gives a tree where black depth has stayed the same and where the root is an arbitrary color. *)
adamc@214 645
adamc@94 646 Definition insertResult c n :=
adamc@94 647 match c with
adamc@94 648 | Red => rbtree Black (S n)
adamc@94 649 | Black => { c' : color & rbtree c' n }
adamc@94 650 end.
adamc@94 651
adamc@214 652 (** A simple clean-up procedure translates [insResult]s into [insertResult]s. *)
adamc@214 653
adamc@97 654 Definition makeRbtree c n : insResult c n -> insertResult c n :=
adamc@214 655 match c with
adamc@94 656 | Red => fun r =>
adamc@214 657 match r with
adamc@94 658 | RedNode' _ _ _ a x b => BlackNode a x b
adamc@94 659 end
adamc@94 660 | Black => fun r => r
adamc@94 661 end.
adamc@94 662
adamc@214 663 (** We modify Coq's default choice of implicit arguments for [makeRbtree], so that we do not need to specify the [c] and [n] arguments explicitly in later calls. *)
adamc@214 664
adamc@97 665 Implicit Arguments makeRbtree [c n].
adamc@94 666
adamc@214 667 (** Finally, we define [insert] as a simple composition of [ins] and [makeRbtree]. *)
adamc@214 668
adamc@94 669 Definition insert c n (t : rbtree c n) : insertResult c n :=
adamc@97 670 makeRbtree (ins t).
adamc@94 671
adamc@214 672 (** As we noted earlier, the type of [insert] guarantees that it outputs balanced trees whose depths have not increased too much. We also want to know that [insert] operates correctly on trees interpreted as finite sets, so we finish this section with a proof of that fact. *)
adamc@214 673
adamc@95 674 Section present.
adamc@95 675 Variable z : nat.
adamc@95 676
adamc@214 677 (** The variable [z] stands for an arbitrary key. We will reason about [z]'s presence in particular trees. As usual, outside the section the theorems we prove will quantify over all possible keys, giving us the facts we wanted.
adamc@214 678
adam@367 679 We start by proving the correctness of the balance operations. It is useful to define a custom tactic [present_balance] that encapsulates the reasoning common to the two proofs. We use the keyword %\index{Vernacular commands!Ltac}%[Ltac] to assign a name to a proof script. This particular script just iterates between [crush] and identification of a tree that is being pattern-matched on and should be destructed. *)
adamc@214 680
adamc@98 681 Ltac present_balance :=
adamc@98 682 crush;
adamc@98 683 repeat (match goal with
adam@425 684 | [ _ : context[match ?T with Leaf => _ | _ => _ end] |- _ ] =>
adam@425 685 dep_destruct T
adam@405 686 | [ |- context[match ?T with Leaf => _ | _ => _ end] ] => dep_destruct T
adamc@98 687 end; crush).
adamc@98 688
adamc@214 689 (** The balance correctness theorems are simple first-order logic equivalences, where we use the function [projT2] to project the payload of a [sigT] value. *)
adamc@214 690
adam@294 691 Lemma present_balance1 : forall n (a : rtree n) (y : nat) c2 (b : rbtree c2 n),
adamc@95 692 present z (projT2 (balance1 a y b))
adamc@95 693 <-> rpresent z a \/ z = y \/ present z b.
adamc@98 694 destruct a; present_balance.
adamc@95 695 Qed.
adamc@95 696
adamc@213 697 Lemma present_balance2 : forall n (a : rtree n) (y : nat) c2 (b : rbtree c2 n),
adamc@95 698 present z (projT2 (balance2 a y b))
adamc@95 699 <-> rpresent z a \/ z = y \/ present z b.
adamc@98 700 destruct a; present_balance.
adamc@95 701 Qed.
adamc@95 702
adamc@214 703 (** To state the theorem for [ins], it is useful to define a new type-level function, since [ins] returns different result types based on the type indices passed to it. Recall that [x] is the section variable standing for the key we are inserting. *)
adamc@214 704
adamc@95 705 Definition present_insResult c n :=
adamc@95 706 match c return (rbtree c n -> insResult c n -> Prop) with
adamc@95 707 | Red => fun t r => rpresent z r <-> z = x \/ present z t
adamc@95 708 | Black => fun t r => present z (projT2 r) <-> z = x \/ present z t
adamc@95 709 end.
adamc@95 710
adamc@214 711 (** Now the statement and proof of the [ins] correctness theorem are straightforward, if verbose. We proceed by induction on the structure of a tree, followed by finding case analysis opportunities on expressions we see being analyzed in [if] or [match] expressions. After that, we pattern-match to find opportunities to use the theorems we proved about balancing. Finally, we identify two variables that are asserted by some hypothesis to be equal, and we use that hypothesis to replace one variable with the other everywhere. *)
adamc@214 712
adamc@95 713 Theorem present_ins : forall c n (t : rbtree c n),
adamc@95 714 present_insResult t (ins t).
adamc@95 715 induction t; crush;
adamc@95 716 repeat (match goal with
adam@338 717 | [ _ : context[if ?E then _ else _] |- _ ] => destruct E
adamc@95 718 | [ |- context[if ?E then _ else _] ] => destruct E
adam@338 719 | [ _ : context[match ?C with Red => _ | Black => _ end]
adamc@214 720 |- _ ] => destruct C
adamc@95 721 end; crush);
adamc@95 722 try match goal with
adam@338 723 | [ _ : context[balance1 ?A ?B ?C] |- _ ] =>
adamc@95 724 generalize (present_balance1 A B C)
adamc@95 725 end;
adamc@95 726 try match goal with
adam@338 727 | [ _ : context[balance2 ?A ?B ?C] |- _ ] =>
adamc@95 728 generalize (present_balance2 A B C)
adamc@95 729 end;
adamc@95 730 try match goal with
adamc@95 731 | [ |- context[balance1 ?A ?B ?C] ] =>
adamc@95 732 generalize (present_balance1 A B C)
adamc@95 733 end;
adamc@95 734 try match goal with
adamc@95 735 | [ |- context[balance2 ?A ?B ?C] ] =>
adamc@95 736 generalize (present_balance2 A B C)
adamc@95 737 end;
adamc@214 738 crush;
adamc@95 739 match goal with
adamc@95 740 | [ z : nat, x : nat |- _ ] =>
adamc@95 741 match goal with
adamc@95 742 | [ H : z = x |- _ ] => rewrite H in *; clear H
adamc@95 743 end
adamc@95 744 end;
adamc@95 745 tauto.
adamc@95 746 Qed.
adamc@95 747
adamc@214 748 (** The hard work is done. The most readable way to state correctness of [insert] involves splitting the property into two color-specific theorems. We write a tactic to encapsulate the reasoning steps that work to establish both facts. *)
adamc@214 749
adamc@213 750 Ltac present_insert :=
adamc@213 751 unfold insert; intros n t; inversion t;
adamc@97 752 generalize (present_ins t); simpl;
adamc@97 753 dep_destruct (ins t); tauto.
adamc@97 754
adamc@95 755 Theorem present_insert_Red : forall n (t : rbtree Red n),
adamc@95 756 present z (insert t)
adamc@95 757 <-> (z = x \/ present z t).
adamc@213 758 present_insert.
adamc@95 759 Qed.
adamc@95 760
adamc@95 761 Theorem present_insert_Black : forall n (t : rbtree Black n),
adamc@95 762 present z (projT2 (insert t))
adamc@95 763 <-> (z = x \/ present z t).
adamc@213 764 present_insert.
adamc@95 765 Qed.
adamc@95 766 End present.
adamc@94 767 End insert.
adamc@94 768
adam@425 769 (** We can generate executable OCaml code with the command %\index{Vernacular commands!Recursive Extraction}%[Recursive Extraction insert], which also automatically outputs the OCaml versions of all of [insert]'s dependencies. In our previous extractions, we wound up with clean OCaml code. Here, we find uses of %\index{Obj.magic}%<<Obj.magic>>, OCaml's unsafe cast operator for tweaking the apparent type of an expression in an arbitrary way. Casts appear for this example because the return type of [insert] depends on the _value_ of the function's argument, a pattern which OCaml cannot handle. Since Coq's type system is much more expressive than OCaml's, such casts are unavoidable in general. Since the OCaml type-checker is no longer checking full safety of programs, we must rely on Coq's extractor to use casts only in provably safe ways. *)
adam@338 770
adam@338 771 (* begin hide *)
adam@338 772 Recursive Extraction insert.
adam@338 773 (* end hide *)
adam@283 774
adamc@94 775
adamc@86 776 (** * A Certified Regular Expression Matcher *)
adamc@86 777
adamc@93 778 (** Another interesting example is regular expressions with dependent types that express which predicates over strings particular regexps implement. We can then assign a dependent type to a regular expression matching function, guaranteeing that it always decides the string property that we expect it to decide.
adamc@93 779
adam@425 780 Before defining the syntax of expressions, it is helpful to define an inductive type capturing the meaning of the Kleene star. That is, a string [s] matches regular expression [star e] if and only if [s] can be decomposed into a sequence of substrings that all match [e]. We use Coq's string support, which comes through a combination of the [String] library and some parsing notations built into Coq. Operators like [++] and functions like [length] that we know from lists are defined again for strings. Notation scopes help us control which versions we want to use in particular contexts.%\index{Vernacular commands!Open Scope}% *)
adamc@93 781
adamc@86 782 Require Import Ascii String.
adamc@86 783 Open Scope string_scope.
adamc@86 784
adamc@91 785 Section star.
adamc@91 786 Variable P : string -> Prop.
adamc@91 787
adamc@91 788 Inductive star : string -> Prop :=
adamc@91 789 | Empty : star ""
adamc@91 790 | Iter : forall s1 s2,
adamc@91 791 P s1
adamc@91 792 -> star s2
adamc@91 793 -> star (s1 ++ s2).
adamc@91 794 End star.
adamc@91 795
adam@283 796 (** Now we can make our first attempt at defining a [regexp] type that is indexed by predicates on strings. Here is a reasonable-looking definition that is restricted to constant characters and concatenation. We use the constructor [String], which is the analogue of list cons for the type [string], where [""] is like list nil.
adamc@93 797 [[
adamc@93 798 Inductive regexp : (string -> Prop) -> Set :=
adamc@93 799 | Char : forall ch : ascii,
adamc@93 800 regexp (fun s => s = String ch "")
adamc@93 801 | Concat : forall (P1 P2 : string -> Prop) (r1 : regexp P1) (r2 : regexp P2),
adamc@93 802 regexp (fun s => exists s1, exists s2, s = s1 ++ s2 /\ P1 s1 /\ P2 s2).
adamc@93 803 ]]
adamc@93 804
adam@338 805 <<
adam@338 806 User error: Large non-propositional inductive types must be in Type
adam@338 807 >>
adam@338 808
adam@338 809 What is a %\index{large inductive types}%large inductive type? In Coq, it is an inductive type that has a constructor which quantifies over some type of type [Type]. We have not worked with [Type] very much to this point. Every term of CIC has a type, including [Set] and [Prop], which are assigned type [Type]. The type [string -> Prop] from the failed definition also has type [Type].
adamc@93 810
adamc@93 811 It turns out that allowing large inductive types in [Set] leads to contradictions when combined with certain kinds of classical logic reasoning. Thus, by default, such types are ruled out. There is a simple fix for our [regexp] definition, which is to place our new type in [Type]. While fixing the problem, we also expand the list of constructors to cover the remaining regular expression operators. *)
adamc@93 812
adamc@89 813 Inductive regexp : (string -> Prop) -> Type :=
adamc@86 814 | Char : forall ch : ascii,
adamc@86 815 regexp (fun s => s = String ch "")
adamc@86 816 | Concat : forall P1 P2 (r1 : regexp P1) (r2 : regexp P2),
adamc@87 817 regexp (fun s => exists s1, exists s2, s = s1 ++ s2 /\ P1 s1 /\ P2 s2)
adamc@87 818 | Or : forall P1 P2 (r1 : regexp P1) (r2 : regexp P2),
adamc@91 819 regexp (fun s => P1 s \/ P2 s)
adamc@91 820 | Star : forall P (r : regexp P),
adamc@91 821 regexp (star P).
adamc@86 822
adam@425 823 (** Many theorems about strings are useful for implementing a certified regexp matcher, and few of them are in the [String] library. The book source includes statements, proofs, and hint commands for a handful of such omitted theorems. Since they are orthogonal to our use of dependent types, we hide them in the rendered versions of this book. *)
adamc@93 824
adamc@93 825 (* begin hide *)
adamc@86 826 Open Scope specif_scope.
adamc@86 827
adamc@86 828 Lemma length_emp : length "" <= 0.
adamc@86 829 crush.
adamc@86 830 Qed.
adamc@86 831
adamc@86 832 Lemma append_emp : forall s, s = "" ++ s.
adamc@86 833 crush.
adamc@86 834 Qed.
adamc@86 835
adamc@86 836 Ltac substring :=
adamc@86 837 crush;
adamc@86 838 repeat match goal with
adamc@86 839 | [ |- context[match ?N with O => _ | S _ => _ end] ] => destruct N; crush
adamc@86 840 end.
adamc@86 841
adamc@86 842 Lemma substring_le : forall s n m,
adamc@86 843 length (substring n m s) <= m.
adamc@86 844 induction s; substring.
adamc@86 845 Qed.
adamc@86 846
adamc@86 847 Lemma substring_all : forall s,
adamc@86 848 substring 0 (length s) s = s.
adamc@86 849 induction s; substring.
adamc@86 850 Qed.
adamc@86 851
adamc@86 852 Lemma substring_none : forall s n,
adamc@93 853 substring n 0 s = "".
adamc@86 854 induction s; substring.
adamc@86 855 Qed.
adamc@86 856
adam@375 857 Hint Rewrite substring_all substring_none.
adamc@86 858
adamc@86 859 Lemma substring_split : forall s m,
adamc@86 860 substring 0 m s ++ substring m (length s - m) s = s.
adamc@86 861 induction s; substring.
adamc@86 862 Qed.
adamc@86 863
adamc@86 864 Lemma length_app1 : forall s1 s2,
adamc@86 865 length s1 <= length (s1 ++ s2).
adamc@86 866 induction s1; crush.
adamc@86 867 Qed.
adamc@86 868
adamc@86 869 Hint Resolve length_emp append_emp substring_le substring_split length_app1.
adamc@86 870
adamc@86 871 Lemma substring_app_fst : forall s2 s1 n,
adamc@86 872 length s1 = n
adamc@86 873 -> substring 0 n (s1 ++ s2) = s1.
adamc@86 874 induction s1; crush.
adamc@86 875 Qed.
adamc@86 876
adamc@86 877 Lemma substring_app_snd : forall s2 s1 n,
adamc@86 878 length s1 = n
adamc@86 879 -> substring n (length (s1 ++ s2) - n) (s1 ++ s2) = s2.
adam@375 880 Hint Rewrite <- minus_n_O.
adamc@86 881
adamc@86 882 induction s1; crush.
adamc@86 883 Qed.
adamc@86 884
adam@375 885 Hint Rewrite substring_app_fst substring_app_snd using solve [trivial].
adamc@93 886 (* end hide *)
adamc@93 887
adamc@93 888 (** A few auxiliary functions help us in our final matcher definition. The function [split] will be used to implement the regexp concatenation case. *)
adamc@86 889
adamc@86 890 Section split.
adamc@86 891 Variables P1 P2 : string -> Prop.
adamc@214 892 Variable P1_dec : forall s, {P1 s} + {~ P1 s}.
adamc@214 893 Variable P2_dec : forall s, {P2 s} + {~ P2 s}.
adamc@93 894 (** We require a choice of two arbitrary string predicates and functions for deciding them. *)
adamc@86 895
adamc@86 896 Variable s : string.
adamc@93 897 (** Our computation will take place relative to a single fixed string, so it is easiest to make it a [Variable], rather than an explicit argument to our functions. *)
adamc@93 898
adam@338 899 (** The function [split'] is the workhorse behind [split]. It searches through the possible ways of splitting [s] into two pieces, checking the two predicates against each such pair. The execution of [split'] progresses right-to-left, from splitting all of [s] into the first piece to splitting all of [s] into the second piece. It takes an extra argument, [n], which specifies how far along we are in this search process. *)
adamc@86 900
adam@297 901 Definition split' : forall n : nat, n <= length s
adamc@86 902 -> {exists s1, exists s2, length s1 <= n /\ s1 ++ s2 = s /\ P1 s1 /\ P2 s2}
adamc@214 903 + {forall s1 s2, length s1 <= n -> s1 ++ s2 = s -> ~ P1 s1 \/ ~ P2 s2}.
adamc@86 904 refine (fix F (n : nat) : n <= length s
adamc@86 905 -> {exists s1, exists s2, length s1 <= n /\ s1 ++ s2 = s /\ P1 s1 /\ P2 s2}
adamc@214 906 + {forall s1 s2, length s1 <= n -> s1 ++ s2 = s -> ~ P1 s1 \/ ~ P2 s2} :=
adamc@214 907 match n with
adamc@86 908 | O => fun _ => Reduce (P1_dec "" && P2_dec s)
adamc@93 909 | S n' => fun _ => (P1_dec (substring 0 (S n') s)
adamc@93 910 && P2_dec (substring (S n') (length s - S n') s))
adamc@86 911 || F n' _
adamc@86 912 end); clear F; crush; eauto 7;
adamc@86 913 match goal with
adamc@86 914 | [ _ : length ?S <= 0 |- _ ] => destruct S
adam@338 915 | [ _ : length ?S' <= S ?N |- _ ] => destruct (eq_nat_dec (length S') (S N))
adamc@86 916 end; crush.
adamc@86 917 Defined.
adamc@86 918
adam@338 919 (** There is one subtle point in the [split'] code that is worth mentioning. The main body of the function is a [match] on [n]. In the case where [n] is known to be [S n'], we write [S n'] in several places where we might be tempted to write [n]. However, without further work to craft proper [match] annotations, the type-checker does not use the equality between [n] and [S n']. Thus, it is common to see patterns repeated in [match] case bodies in dependently typed Coq code. We can at least use a [let] expression to avoid copying the pattern more than once, replacing the first case body with:
adamc@93 920 [[
adamc@93 921 | S n' => fun _ => let n := S n' in
adamc@93 922 (P1_dec (substring 0 n s)
adamc@93 923 && P2_dec (substring n (length s - n) s))
adamc@93 924 || F n' _
adamc@214 925
adamc@93 926 ]]
adamc@93 927
adam@338 928 The [split] function itself is trivial to implement in terms of [split']. We just ask [split'] to begin its search with [n = length s]. *)
adamc@93 929
adamc@86 930 Definition split : {exists s1, exists s2, s = s1 ++ s2 /\ P1 s1 /\ P2 s2}
adamc@214 931 + {forall s1 s2, s = s1 ++ s2 -> ~ P1 s1 \/ ~ P2 s2}.
adamc@86 932 refine (Reduce (split' (n := length s) _)); crush; eauto.
adamc@86 933 Defined.
adamc@86 934 End split.
adamc@86 935
adamc@86 936 Implicit Arguments split [P1 P2].
adamc@86 937
adamc@93 938 (* begin hide *)
adamc@91 939 Lemma app_empty_end : forall s, s ++ "" = s.
adamc@91 940 induction s; crush.
adamc@91 941 Qed.
adamc@91 942
adam@375 943 Hint Rewrite app_empty_end.
adamc@91 944
adamc@91 945 Lemma substring_self : forall s n,
adamc@91 946 n <= 0
adamc@91 947 -> substring n (length s - n) s = s.
adamc@91 948 induction s; substring.
adamc@91 949 Qed.
adamc@91 950
adamc@91 951 Lemma substring_empty : forall s n m,
adamc@91 952 m <= 0
adamc@91 953 -> substring n m s = "".
adamc@91 954 induction s; substring.
adamc@91 955 Qed.
adamc@91 956
adam@375 957 Hint Rewrite substring_self substring_empty using omega.
adamc@91 958
adamc@91 959 Lemma substring_split' : forall s n m,
adamc@91 960 substring n m s ++ substring (n + m) (length s - (n + m)) s
adamc@91 961 = substring n (length s - n) s.
adam@375 962 Hint Rewrite substring_split.
adamc@91 963
adamc@91 964 induction s; substring.
adamc@91 965 Qed.
adamc@91 966
adamc@91 967 Lemma substring_stack : forall s n2 m1 m2,
adamc@91 968 m1 <= m2
adamc@91 969 -> substring 0 m1 (substring n2 m2 s)
adamc@91 970 = substring n2 m1 s.
adamc@91 971 induction s; substring.
adamc@91 972 Qed.
adamc@91 973
adamc@91 974 Ltac substring' :=
adamc@91 975 crush;
adamc@91 976 repeat match goal with
adamc@91 977 | [ |- context[match ?N with O => _ | S _ => _ end] ] => case_eq N; crush
adamc@91 978 end.
adamc@91 979
adamc@91 980 Lemma substring_stack' : forall s n1 n2 m1 m2,
adamc@91 981 n1 + m1 <= m2
adamc@91 982 -> substring n1 m1 (substring n2 m2 s)
adamc@91 983 = substring (n1 + n2) m1 s.
adamc@91 984 induction s; substring';
adamc@91 985 match goal with
adamc@91 986 | [ |- substring ?N1 _ _ = substring ?N2 _ _ ] =>
adamc@91 987 replace N1 with N2; crush
adamc@91 988 end.
adamc@91 989 Qed.
adamc@91 990
adamc@91 991 Lemma substring_suffix : forall s n,
adamc@91 992 n <= length s
adamc@91 993 -> length (substring n (length s - n) s) = length s - n.
adamc@91 994 induction s; substring.
adamc@91 995 Qed.
adamc@91 996
adamc@91 997 Lemma substring_suffix_emp' : forall s n m,
adamc@91 998 substring n (S m) s = ""
adamc@91 999 -> n >= length s.
adamc@91 1000 induction s; crush;
adamc@91 1001 match goal with
adamc@91 1002 | [ |- ?N >= _ ] => destruct N; crush
adamc@91 1003 end;
adamc@91 1004 match goal with
adamc@91 1005 [ |- S ?N >= S ?E ] => assert (N >= E); [ eauto | omega ]
adamc@91 1006 end.
adamc@91 1007 Qed.
adamc@91 1008
adamc@91 1009 Lemma substring_suffix_emp : forall s n m,
adamc@92 1010 substring n m s = ""
adamc@92 1011 -> m > 0
adamc@91 1012 -> n >= length s.
adam@335 1013 destruct m as [ | m]; [crush | intros; apply substring_suffix_emp' with m; assumption].
adamc@91 1014 Qed.
adamc@91 1015
adamc@91 1016 Hint Rewrite substring_stack substring_stack' substring_suffix
adam@375 1017 using omega.
adamc@91 1018
adamc@91 1019 Lemma minus_minus : forall n m1 m2,
adamc@91 1020 m1 + m2 <= n
adamc@91 1021 -> n - m1 - m2 = n - (m1 + m2).
adamc@91 1022 intros; omega.
adamc@91 1023 Qed.
adamc@91 1024
adamc@91 1025 Lemma plus_n_Sm' : forall n m : nat, S (n + m) = m + S n.
adamc@91 1026 intros; omega.
adamc@91 1027 Qed.
adamc@91 1028
adam@375 1029 Hint Rewrite minus_minus using omega.
adamc@93 1030 (* end hide *)
adamc@93 1031
adamc@93 1032 (** One more helper function will come in handy: [dec_star], for implementing another linear search through ways of splitting a string, this time for implementing the Kleene star. *)
adamc@91 1033
adamc@91 1034 Section dec_star.
adamc@91 1035 Variable P : string -> Prop.
adamc@214 1036 Variable P_dec : forall s, {P s} + {~ P s}.
adamc@91 1037
adam@338 1038 (** Some new lemmas and hints about the [star] type family are useful. We omit them here; they are included in the book source at this point. *)
adamc@93 1039
adamc@93 1040 (* begin hide *)
adamc@91 1041 Hint Constructors star.
adamc@91 1042
adamc@91 1043 Lemma star_empty : forall s,
adamc@91 1044 length s = 0
adamc@91 1045 -> star P s.
adamc@91 1046 destruct s; crush.
adamc@91 1047 Qed.
adamc@91 1048
adamc@91 1049 Lemma star_singleton : forall s, P s -> star P s.
adamc@91 1050 intros; rewrite <- (app_empty_end s); auto.
adamc@91 1051 Qed.
adamc@91 1052
adamc@91 1053 Lemma star_app : forall s n m,
adamc@91 1054 P (substring n m s)
adamc@91 1055 -> star P (substring (n + m) (length s - (n + m)) s)
adamc@91 1056 -> star P (substring n (length s - n) s).
adamc@91 1057 induction n; substring;
adamc@91 1058 match goal with
adamc@91 1059 | [ H : P (substring ?N ?M ?S) |- _ ] =>
adamc@91 1060 solve [ rewrite <- (substring_split S M); auto
adamc@91 1061 | rewrite <- (substring_split' S N M); auto ]
adamc@91 1062 end.
adamc@91 1063 Qed.
adamc@91 1064
adamc@91 1065 Hint Resolve star_empty star_singleton star_app.
adamc@91 1066
adamc@91 1067 Variable s : string.
adamc@91 1068
adamc@91 1069 Lemma star_inv : forall s,
adamc@91 1070 star P s
adamc@91 1071 -> s = ""
adamc@91 1072 \/ exists i, i < length s
adamc@91 1073 /\ P (substring 0 (S i) s)
adamc@91 1074 /\ star P (substring (S i) (length s - S i) s).
adamc@91 1075 Hint Extern 1 (exists i : nat, _) =>
adamc@91 1076 match goal with
adamc@91 1077 | [ H : P (String _ ?S) |- _ ] => exists (length S); crush
adamc@91 1078 end.
adamc@91 1079
adamc@91 1080 induction 1; [
adamc@91 1081 crush
adamc@91 1082 | match goal with
adamc@91 1083 | [ _ : P ?S |- _ ] => destruct S; crush
adamc@91 1084 end
adamc@91 1085 ].
adamc@91 1086 Qed.
adamc@91 1087
adamc@91 1088 Lemma star_substring_inv : forall n,
adamc@91 1089 n <= length s
adamc@91 1090 -> star P (substring n (length s - n) s)
adamc@91 1091 -> substring n (length s - n) s = ""
adamc@91 1092 \/ exists l, l < length s - n
adamc@91 1093 /\ P (substring n (S l) s)
adamc@91 1094 /\ star P (substring (n + S l) (length s - (n + S l)) s).
adam@375 1095 Hint Rewrite plus_n_Sm'.
adamc@91 1096
adamc@91 1097 intros;
adamc@91 1098 match goal with
adamc@91 1099 | [ H : star _ _ |- _ ] => generalize (star_inv H); do 3 crush; eauto
adamc@91 1100 end.
adamc@91 1101 Qed.
adamc@93 1102 (* end hide *)
adamc@93 1103
adamc@93 1104 (** The function [dec_star''] implements a single iteration of the star. That is, it tries to find a string prefix matching [P], and it calls a parameter function on the remainder of the string. *)
adamc@91 1105
adamc@91 1106 Section dec_star''.
adamc@91 1107 Variable n : nat.
adamc@93 1108 (** [n] is the length of the prefix of [s] that we have already processed. *)
adamc@91 1109
adamc@91 1110 Variable P' : string -> Prop.
adamc@91 1111 Variable P'_dec : forall n' : nat, n' > n
adamc@91 1112 -> {P' (substring n' (length s - n') s)}
adamc@214 1113 + {~ P' (substring n' (length s - n') s)}.
adamc@93 1114 (** When we use [dec_star''], we will instantiate [P'_dec] with a function for continuing the search for more instances of [P] in [s]. *)
adamc@93 1115
adamc@93 1116 (** Now we come to [dec_star''] itself. It takes as an input a natural [l] that records how much of the string has been searched so far, as we did for [split']. The return type expresses that [dec_star''] is looking for an index into [s] that splits [s] into a nonempty prefix and a suffix, such that the prefix satisfies [P] and the suffix satisfies [P']. *)
adamc@91 1117
adam@297 1118 Definition dec_star'' : forall l : nat,
adam@297 1119 {exists l', S l' <= l
adamc@91 1120 /\ P (substring n (S l') s) /\ P' (substring (n + S l') (length s - (n + S l')) s)}
adamc@91 1121 + {forall l', S l' <= l
adamc@214 1122 -> ~ P (substring n (S l') s)
adamc@214 1123 \/ ~ P' (substring (n + S l') (length s - (n + S l')) s)}.
adamc@91 1124 refine (fix F (l : nat) : {exists l', S l' <= l
adamc@91 1125 /\ P (substring n (S l') s) /\ P' (substring (n + S l') (length s - (n + S l')) s)}
adamc@91 1126 + {forall l', S l' <= l
adamc@214 1127 -> ~ P (substring n (S l') s)
adamc@214 1128 \/ ~ P' (substring (n + S l') (length s - (n + S l')) s)} :=
adam@380 1129 match l return {exists l', S l' <= l
adam@380 1130 /\ P (substring n (S l') s) /\ P' (substring (n + S l') (length s - (n + S l')) s)}
adam@380 1131 + {forall l', S l' <= l
adam@380 1132 -> ~ P (substring n (S l') s)
adam@380 1133 \/ ~ P' (substring (n + S l') (length s - (n + S l')) s)} with
adamc@91 1134 | O => _
adamc@91 1135 | S l' =>
adamc@91 1136 (P_dec (substring n (S l') s) && P'_dec (n' := n + S l') _)
adamc@91 1137 || F l'
adamc@91 1138 end); clear F; crush; eauto 7;
adamc@91 1139 match goal with
adamc@91 1140 | [ H : ?X <= S ?Y |- _ ] => destruct (eq_nat_dec X (S Y)); crush
adamc@91 1141 end.
adamc@91 1142 Defined.
adamc@91 1143 End dec_star''.
adamc@91 1144
adamc@93 1145 (* begin hide *)
adamc@92 1146 Lemma star_length_contra : forall n,
adamc@92 1147 length s > n
adamc@92 1148 -> n >= length s
adamc@92 1149 -> False.
adamc@92 1150 crush.
adamc@92 1151 Qed.
adamc@92 1152
adamc@92 1153 Lemma star_length_flip : forall n n',
adamc@92 1154 length s - n <= S n'
adamc@92 1155 -> length s > n
adamc@92 1156 -> length s - n > 0.
adamc@92 1157 crush.
adamc@92 1158 Qed.
adamc@92 1159
adamc@92 1160 Hint Resolve star_length_contra star_length_flip substring_suffix_emp.
adamc@93 1161 (* end hide *)
adamc@92 1162
adamc@93 1163 (** The work of [dec_star''] is nested inside another linear search by [dec_star'], which provides the final functionality we need, but for arbitrary suffixes of [s], rather than just for [s] overall. *)
adamc@93 1164
adam@297 1165 Definition dec_star' : forall n n' : nat, length s - n' <= n
adamc@91 1166 -> {star P (substring n' (length s - n') s)}
adamc@214 1167 + {~ star P (substring n' (length s - n') s)}.
adamc@214 1168 refine (fix F (n n' : nat) : length s - n' <= n
adamc@91 1169 -> {star P (substring n' (length s - n') s)}
adamc@214 1170 + {~ star P (substring n' (length s - n') s)} :=
adamc@214 1171 match n with
adamc@91 1172 | O => fun _ => Yes
adamc@91 1173 | S n'' => fun _ =>
adamc@91 1174 le_gt_dec (length s) n'
adam@338 1175 || dec_star'' (n := n') (star P)
adam@338 1176 (fun n0 _ => Reduce (F n'' n0 _)) (length s - n')
adamc@92 1177 end); clear F; crush; eauto;
adamc@92 1178 match goal with
adamc@92 1179 | [ H : star _ _ |- _ ] => apply star_substring_inv in H; crush; eauto
adamc@92 1180 end;
adamc@92 1181 match goal with
adamc@92 1182 | [ H1 : _ < _ - _, H2 : forall l' : nat, _ <= _ - _ -> _ |- _ ] =>
adamc@92 1183 generalize (H2 _ (lt_le_S _ _ H1)); tauto
adamc@92 1184 end.
adamc@91 1185 Defined.
adamc@91 1186
adam@380 1187 (** Finally, we have [dec_star], defined by straightforward reduction from [dec_star']. *)
adamc@93 1188
adamc@214 1189 Definition dec_star : {star P s} + {~ star P s}.
adam@380 1190 refine (Reduce (dec_star' (n := length s) 0 _)); crush.
adamc@91 1191 Defined.
adamc@91 1192 End dec_star.
adamc@91 1193
adamc@93 1194 (* begin hide *)
adamc@86 1195 Lemma app_cong : forall x1 y1 x2 y2,
adamc@86 1196 x1 = x2
adamc@86 1197 -> y1 = y2
adamc@86 1198 -> x1 ++ y1 = x2 ++ y2.
adamc@86 1199 congruence.
adamc@86 1200 Qed.
adamc@86 1201
adamc@86 1202 Hint Resolve app_cong.
adamc@93 1203 (* end hide *)
adamc@93 1204
adamc@93 1205 (** With these helper functions completed, the implementation of our [matches] function is refreshingly straightforward. We only need one small piece of specific tactic work beyond what [crush] does for us. *)
adamc@86 1206
adam@297 1207 Definition matches : forall P (r : regexp P) s, {P s} + {~ P s}.
adamc@214 1208 refine (fix F P (r : regexp P) s : {P s} + {~ P s} :=
adamc@86 1209 match r with
adamc@86 1210 | Char ch => string_dec s (String ch "")
adamc@86 1211 | Concat _ _ r1 r2 => Reduce (split (F _ r1) (F _ r2) s)
adamc@87 1212 | Or _ _ r1 r2 => F _ r1 s || F _ r2 s
adamc@91 1213 | Star _ r => dec_star _ _ _
adamc@86 1214 end); crush;
adamc@86 1215 match goal with
adam@426 1216 | [ H : _ |- _ ] => generalize (H _ _ (eq_refl _))
adamc@93 1217 end; tauto.
adamc@86 1218 Defined.
adamc@86 1219
adam@283 1220 (** It is interesting to pause briefly to consider alternate implementations of [matches]. Dependent types give us much latitude in how specific correctness properties may be encoded with types. For instance, we could have made [regexp] a non-indexed inductive type, along the lines of what is possible in traditional ML and Haskell. We could then have implemented a recursive function to map [regexp]s to their intended meanings, much as we have done with types and programs in other examples. That style is compatible with the [refine]-based approach that we have used here, and it might be an interesting exercise to redo the code from this subsection in that alternate style or some further encoding of the reader's choice. The main advantage of indexed inductive types is that they generally lead to the smallest amount of code. *)
adam@283 1221
adamc@93 1222 (* begin hide *)
adamc@86 1223 Example hi := Concat (Char "h"%char) (Char "i"%char).
adam@380 1224 Eval hnf in matches hi "hi".
adam@380 1225 Eval hnf in matches hi "bye".
adamc@87 1226
adamc@87 1227 Example a_b := Or (Char "a"%char) (Char "b"%char).
adam@380 1228 Eval hnf in matches a_b "".
adam@380 1229 Eval hnf in matches a_b "a".
adam@380 1230 Eval hnf in matches a_b "aa".
adam@380 1231 Eval hnf in matches a_b "b".
adam@283 1232 (* end hide *)
adam@283 1233
adam@405 1234 (** Many regular expression matching problems are easy to test. The reader may run each of the following queries to verify that it gives the correct answer. We use evaluation strategy %\index{tactics!hnf}%[hnf] to reduce each term to%\index{head-normal form}% _head-normal form_, where the datatype constructor used to build its value is known. *)
adamc@91 1235
adamc@91 1236 Example a_star := Star (Char "a"%char).
adam@380 1237 Eval hnf in matches a_star "".
adam@380 1238 Eval hnf in matches a_star "a".
adam@380 1239 Eval hnf in matches a_star "b".
adam@380 1240 Eval hnf in matches a_star "aa".
adam@283 1241
adam@283 1242 (** Evaluation inside Coq does not scale very well, so it is easy to build other tests that run for hours or more. Such cases are better suited to execution with the extracted OCaml code. *)