annotate src/GeneralRec.v @ 434:92f386c33e94

Finish pass taking advantage of new coqdoc features
author Adam Chlipala <adam@chlipala.net>
date Thu, 26 Jul 2012 16:38:44 -0400
parents 5f25705a10ea
children 8077352044b2
rev   line source
adam@380 1 (* Copyright (c) 2006, 2011-2012, Adam Chlipala
adam@350 2 *
adam@350 3 * This work is licensed under a
adam@350 4 * Creative Commons Attribution-Noncommercial-No Derivative Works 3.0
adam@350 5 * Unported License.
adam@350 6 * The license text is available at:
adam@350 7 * http://creativecommons.org/licenses/by-nc-nd/3.0/
adam@350 8 *)
adam@350 9
adam@350 10 (* begin hide *)
adam@351 11 Require Import Arith List.
adam@350 12
adam@351 13 Require Import CpdtTactics Coinductive.
adam@350 14
adam@350 15 Set Implicit Arguments.
adam@350 16 (* end hide *)
adam@350 17
adam@350 18
adam@350 19 (** %\chapter{General Recursion}% *)
adam@350 20
adam@353 21 (** Termination of all programs is a crucial property of Gallina. Non-terminating programs introduce logical inconsistency, where any theorem can be proved with an infinite loop. Coq uses a small set of conservative, syntactic criteria to check termination of all recursive definitions. These criteria are insufficient to support the natural encodings of a variety of important programming idioms. Further, since Coq makes it so convenient to encode mathematics computationally, with functional programs, we may find ourselves wanting to employ more complicated recursion in mathematical definitions.
adam@351 22
adam@424 23 What exactly are the conservative criteria that we run up against? For _recursive_ definitions, recursive calls are only allowed on _syntactic subterms_ of the original primary argument, a restriction known as%\index{primitive recursion}% _primitive recursion_. In fact, Coq's handling of reflexive inductive types (those defined in terms of functions returning the same type) gives a bit more flexibility than in traditional primitive recursion, but the term is still applied commonly. In Chapter 5, we saw how _co-recursive_ definitions are checked against a syntactic guardedness condition that guarantees productivity.
adam@351 24
adam@353 25 Many natural recursion patterns satisfy neither condition. For instance, there is our simple running example in this chapter, merge sort. We will study three different approaches to more flexible recursion, and the latter two of the approaches will even support definitions that may fail to terminate on certain inputs, without any up-front characterization of which inputs those may be.
adam@351 26
adam@404 27 Before proceeding, it is important to note that the problem here is not as fundamental as it may appear. The final example of Chapter 5 demonstrated what is called a%\index{deep embedding}% _deep embedding_ of the syntax and semantics of a programming language. That is, we gave a mathematical definition of a language of programs and their meanings. This language clearly admitted non-termination, and we could think of writing all our sophisticated recursive functions with such explicit syntax types. However, in doing so, we forfeit our chance to take advantage of Coq's very good built-in support for reasoning about Gallina programs. We would rather use a%\index{shallow embedding}% _shallow embedding_, where we model informal constructs by encoding them as normal Gallina programs. Each of the three techniques of this chapter follows that style. *)
adam@351 28
adam@351 29
adam@351 30 (** * Well-Founded Recursion *)
adam@351 31
adam@404 32 (** The essence of terminating recursion is that there are no infinite chains of nested recursive calls. This intuition is commonly mapped to the mathematical idea of a%\index{well-founded relation}% _well-founded relation_, and the associated standard technique in Coq is%\index{well-founded recursion}% _well-founded recursion_. The syntactic-subterm relation that Coq applies by default is well-founded, but many cases demand alternate well-founded relations. To demonstrate, let us see where we get stuck on attempting a standard merge sort implementation. *)
adam@351 33
adam@351 34 Section mergeSort.
adam@351 35 Variable A : Type.
adam@351 36 Variable le : A -> A -> bool.
adam@424 37 (** We have a set equipped with some "less-than-or-equal-to" test. *)
adam@351 38
adam@351 39 (** A standard function inserts an element into a sorted list, preserving sortedness. *)
adam@351 40
adam@351 41 Fixpoint insert (x : A) (ls : list A) : list A :=
adam@351 42 match ls with
adam@351 43 | nil => x :: nil
adam@351 44 | h :: ls' =>
adam@351 45 if le x h
adam@351 46 then x :: ls
adam@351 47 else h :: insert x ls'
adam@351 48 end.
adam@351 49
adam@351 50 (** We will also need a function to merge two sorted lists. (We use a less efficient implementation than usual, because the more efficient implementation already forces us to think about well-founded recursion, while here we are only interested in setting up the example of merge sort.) *)
adam@351 51
adam@351 52 Fixpoint merge (ls1 ls2 : list A) : list A :=
adam@351 53 match ls1 with
adam@351 54 | nil => ls2
adam@351 55 | h :: ls' => insert h (merge ls' ls2)
adam@351 56 end.
adam@351 57
adam@351 58 (** The last helper function for classic merge sort is the one that follows, to partition a list arbitrarily into two pieces of approximately equal length. *)
adam@351 59
adam@351 60 Fixpoint partition (ls : list A) : list A * list A :=
adam@351 61 match ls with
adam@351 62 | nil => (nil, nil)
adam@351 63 | h :: nil => (h :: nil, nil)
adam@351 64 | h1 :: h2 :: ls' =>
adam@351 65 let (ls1, ls2) := partition ls' in
adam@351 66 (h1 :: ls1, h2 :: ls2)
adam@351 67 end.
adam@351 68
adam@424 69 (** Now, let us try to write the final sorting function, using a natural number "[<=]" test [leb] from the standard library.
adam@351 70 [[
adam@351 71 Fixpoint mergeSort (ls : list A) : list A :=
adam@351 72 if leb (length ls) 2
adam@351 73 then ls
adam@351 74 else let lss := partition ls in
adam@351 75 merge (mergeSort (fst lss)) (mergeSort (snd lss)).
adam@351 76 ]]
adam@351 77
adam@351 78 <<
adam@351 79 Recursive call to mergeSort has principal argument equal to
adam@351 80 "fst (partition ls)" instead of a subterm of "ls".
adam@351 81 >>
adam@351 82
adam@351 83 The definition is rejected for not following the simple primitive recursion criterion. In particular, it is not apparent that recursive calls to [mergeSort] are syntactic subterms of the original argument [ls]; indeed, they are not, yet we know this is a well-founded recursive definition.
adam@351 84
adam@351 85 To produce an acceptable definition, we need to choose a well-founded relation and prove that [mergeSort] respects it. A good starting point is an examination of how well-foundedness is formalized in the Coq standard library. *)
adam@351 86
adam@351 87 Print well_founded.
adam@351 88 (** %\vspace{-.15in}% [[
adam@351 89 well_founded =
adam@351 90 fun (A : Type) (R : A -> A -> Prop) => forall a : A, Acc R a
adam@351 91 ]]
adam@351 92
adam@404 93 The bulk of the definitional work devolves to the%\index{accessibility relation}\index{Gallina terms!Acc}% _accessibility_ relation [Acc], whose definition we may also examine. *)
adam@351 94
adam@424 95 (* begin hide *)
adam@424 96 Definition Acc_intro' := Acc_intro.
adam@424 97 (* end hide *)
adam@424 98
adam@351 99 Print Acc.
adam@351 100 (** %\vspace{-.15in}% [[
adam@351 101 Inductive Acc (A : Type) (R : A -> A -> Prop) (x : A) : Prop :=
adam@351 102 Acc_intro : (forall y : A, R y x -> Acc R y) -> Acc R x
adam@351 103 ]]
adam@351 104
adam@424 105 In prose, an element [x] is accessible for a relation [R] if every element "less than" [x] according to [R] is also accessible. Since [Acc] is defined inductively, we know that any accessibility proof involves a finite chain of invocations, in a certain sense that we can make formal. Building on Chapter 5's examples, let us define a co-inductive relation that is closer to the usual informal notion of "absence of infinite decreasing chains." *)
adam@351 106
adam@351 107 CoInductive isChain A (R : A -> A -> Prop) : stream A -> Prop :=
adam@351 108 | ChainCons : forall x y s, isChain R (Cons y s)
adam@351 109 -> R y x
adam@351 110 -> isChain R (Cons x (Cons y s)).
adam@351 111
adam@351 112 (** We can now prove that any accessible element cannot be the beginning of any infinite decreasing chain. *)
adam@351 113
adam@351 114 (* begin thide *)
adam@351 115 Lemma noChains' : forall A (R : A -> A -> Prop) x, Acc R x
adam@351 116 -> forall s, ~isChain R (Cons x s).
adam@351 117 induction 1; crush;
adam@351 118 match goal with
adam@351 119 | [ H : isChain _ _ |- _ ] => inversion H; eauto
adam@351 120 end.
adam@351 121 Qed.
adam@351 122
adam@351 123 (** From here, the absence of infinite decreasing chains in well-founded sets is immediate. *)
adam@351 124
adam@351 125 Theorem noChains : forall A (R : A -> A -> Prop), well_founded R
adam@351 126 -> forall s, ~isChain R s.
adam@351 127 destruct s; apply noChains'; auto.
adam@351 128 Qed.
adam@351 129 (* end thide *)
adam@351 130
adam@351 131 (** Absence of infinite decreasing chains implies absence of infinitely nested recursive calls, for any recursive definition that respects the well-founded relation. The [Fix] combinator from the standard library formalizes that intuition: *)
adam@351 132
adam@351 133 Check Fix.
adam@351 134 (** %\vspace{-.15in}%[[
adam@351 135 Fix
adam@351 136 : forall (A : Type) (R : A -> A -> Prop),
adam@351 137 well_founded R ->
adam@351 138 forall P : A -> Type,
adam@351 139 (forall x : A, (forall y : A, R y x -> P y) -> P x) ->
adam@351 140 forall x : A, P x
adam@351 141 ]]
adam@351 142
adam@351 143 A call to %\index{Gallina terms!Fix}%[Fix] must present a relation [R] and a proof of its well-foundedness. The next argument, [P], is the possibly dependent range type of the function we build; the domain [A] of [R] is the function's domain. The following argument has this type:
adam@351 144 [[
adam@351 145 forall x : A, (forall y : A, R y x -> P y) -> P x
adam@351 146 ]]
adam@351 147
adam@424 148 This is an encoding of the function body. The input [x] stands for the function argument, and the next input stands for the function we are defining. Recursive calls are encoded as calls to the second argument, whose type tells us it expects a value [y] and a proof that [y] is "less than" [x], according to [R]. In this way, we enforce the well-foundedness restriction on recursive calls.
adam@351 149
adam@353 150 The rest of [Fix]'s type tells us that it returns a function of exactly the type we expect, so we are now ready to use it to implement [mergeSort]. Careful readers may have noticed that [Fix] has a dependent type of the sort we met in the previous chapter.
adam@351 151
adam@351 152 Before writing [mergeSort], we need to settle on a well-founded relation. The right one for this example is based on lengths of lists. *)
adam@351 153
adam@351 154 Definition lengthOrder (ls1 ls2 : list A) :=
adam@351 155 length ls1 < length ls2.
adam@351 156
adam@353 157 (** We must prove that the relation is truly well-founded. To save some space in the rest of this chapter, we skip right to nice, automated proof scripts, though we postpone introducing the principles behind such scripts to Part III of the book. Curious readers may still replace semicolons with periods and newlines to step through these scripts interactively. *)
adam@351 158
adam@351 159 Hint Constructors Acc.
adam@351 160
adam@351 161 Lemma lengthOrder_wf' : forall len, forall ls, length ls <= len -> Acc lengthOrder ls.
adam@351 162 unfold lengthOrder; induction len; crush.
adam@351 163 Defined.
adam@351 164
adam@351 165 Theorem lengthOrder_wf : well_founded lengthOrder.
adam@351 166 red; intro; eapply lengthOrder_wf'; eauto.
adam@351 167 Defined.
adam@351 168
adam@398 169 (** Notice that we end these proofs with %\index{Vernacular commands!Defined}%[Defined], not [Qed]. Recall that [Defined] marks the theorems as %\emph{transparent}%, so that the details of their proofs may be used during program execution. Why could such details possibly matter for computation? It turns out that [Fix] satisfies the primitive recursion restriction by declaring itself as _recursive in the structure of [Acc] proofs_. This is possible because [Acc] proofs follow a predictable inductive structure. We must do work, as in the last theorem's proof, to establish that all elements of a type belong to [Acc], but the automatic unwinding of those proofs during recursion is straightforward. If we ended the proof with [Qed], the proof details would be hidden from computation, in which case the unwinding process would get stuck.
adam@351 170
adam@351 171 To justify our two recursive [mergeSort] calls, we will also need to prove that [partition] respects the [lengthOrder] relation. These proofs, too, must be kept transparent, to avoid stuckness of [Fix] evaluation. *)
adam@351 172
adam@351 173 Lemma partition_wf : forall len ls, 2 <= length ls <= len
adam@351 174 -> let (ls1, ls2) := partition ls in
adam@351 175 lengthOrder ls1 ls /\ lengthOrder ls2 ls.
adam@351 176 unfold lengthOrder; induction len; crush; do 2 (destruct ls; crush);
adam@351 177 destruct (le_lt_dec 2 (length ls));
adam@351 178 repeat (match goal with
adam@351 179 | [ _ : length ?E < 2 |- _ ] => destruct E
adam@351 180 | [ _ : S (length ?E) < 2 |- _ ] => destruct E
adam@351 181 | [ IH : _ |- context[partition ?L] ] =>
adam@351 182 specialize (IH L); destruct (partition L); destruct IH
adam@351 183 end; crush).
adam@351 184 Defined.
adam@351 185
adam@351 186 Ltac partition := intros ls ?; intros; generalize (@partition_wf (length ls) ls);
adam@351 187 destruct (partition ls); destruct 1; crush.
adam@351 188
adam@351 189 Lemma partition_wf1 : forall ls, 2 <= length ls
adam@351 190 -> lengthOrder (fst (partition ls)) ls.
adam@351 191 partition.
adam@351 192 Defined.
adam@351 193
adam@351 194 Lemma partition_wf2 : forall ls, 2 <= length ls
adam@351 195 -> lengthOrder (snd (partition ls)) ls.
adam@351 196 partition.
adam@351 197 Defined.
adam@351 198
adam@351 199 Hint Resolve partition_wf1 partition_wf2.
adam@351 200
adam@353 201 (** To write the function definition itself, we use the %\index{tactics!refine}%[refine] tactic as a convenient way to write a program that needs to manipulate proofs, without writing out those proofs manually. We also use a replacement [le_lt_dec] for [leb] that has a more interesting dependent type. *)
adam@351 202
adam@351 203 Definition mergeSort : list A -> list A.
adam@351 204 (* begin thide *)
adam@351 205 refine (Fix lengthOrder_wf (fun _ => list A)
adam@351 206 (fun (ls : list A)
adam@351 207 (mergeSort : forall ls' : list A, lengthOrder ls' ls -> list A) =>
adam@351 208 if le_lt_dec 2 (length ls)
adam@351 209 then let lss := partition ls in
adam@351 210 merge (mergeSort (fst lss) _) (mergeSort (snd lss) _)
adam@351 211 else ls)); subst lss; eauto.
adam@351 212 Defined.
adam@351 213 (* end thide *)
adam@351 214 End mergeSort.
adam@351 215
adam@351 216 (** The important thing is that it is now easy to evaluate calls to [mergeSort]. *)
adam@351 217
adam@351 218 Eval compute in mergeSort leb (1 :: 2 :: 36 :: 8 :: 19 :: nil).
adam@351 219 (** [= 1 :: 2 :: 8 :: 19 :: 36 :: nil] *)
adam@351 220
adam@351 221 (** Since the subject of this chapter is merely how to define functions with unusual recursion structure, we will not prove any further correctness theorems about [mergeSort]. Instead, we stop at proving that [mergeSort] has the expected computational behavior, for all inputs, not merely the one we just tested. *)
adam@351 222
adam@351 223 (* begin thide *)
adam@351 224 Theorem mergeSort_eq : forall A (le : A -> A -> bool) ls,
adam@351 225 mergeSort le ls = if le_lt_dec 2 (length ls)
adam@351 226 then let lss := partition ls in
adam@351 227 merge le (mergeSort le (fst lss)) (mergeSort le (snd lss))
adam@351 228 else ls.
adam@351 229 intros; apply (Fix_eq (@lengthOrder_wf A) (fun _ => list A)); intros.
adam@351 230
adam@424 231 (** The library theorem [Fix_eq] imposes one more strange subgoal upon us. We must prove that the function body is unable to distinguish between "self" arguments that map equal inputs to equal outputs. One might think this should be true of any Gallina code, but in fact this general%\index{extensionality}% _function extensionality_ property is neither provable nor disprovable within Coq. The type of [Fix_eq] makes clear what we must show manually: *)
adam@351 232
adam@351 233 Check Fix_eq.
adam@351 234 (** %\vspace{-.15in}%[[
adam@351 235 Fix_eq
adam@351 236 : forall (A : Type) (R : A -> A -> Prop) (Rwf : well_founded R)
adam@351 237 (P : A -> Type)
adam@351 238 (F : forall x : A, (forall y : A, R y x -> P y) -> P x),
adam@351 239 (forall (x : A) (f g : forall y : A, R y x -> P y),
adam@351 240 (forall (y : A) (p : R y x), f y p = g y p) -> F x f = F x g) ->
adam@351 241 forall x : A,
adam@351 242 Fix Rwf P F x = F x (fun (y : A) (_ : R y x) => Fix Rwf P F y)
adam@351 243 ]]
adam@351 244
adam@351 245 Most such obligations are dischargable with straightforward proof automation, and this example is no exception. *)
adam@351 246
adam@351 247 match goal with
adam@351 248 | [ |- context[match ?E with left _ => _ | right _ => _ end] ] => destruct E
adam@351 249 end; simpl; f_equal; auto.
adam@351 250 Qed.
adam@351 251 (* end thide *)
adam@351 252
adam@351 253 (** As a final test of our definition's suitability, we can extract to OCaml. *)
adam@351 254
adam@351 255 Extraction mergeSort.
adam@351 256
adam@351 257 (** <<
adam@351 258 let rec mergeSort le x =
adam@351 259 match le_lt_dec (S (S O)) (length x) with
adam@351 260 | Left ->
adam@351 261 let lss = partition x in
adam@351 262 merge le (mergeSort le (fst lss)) (mergeSort le (snd lss))
adam@351 263 | Right -> x
adam@351 264 >>
adam@351 265
adam@353 266 We see almost precisely the same definition we would have written manually in OCaml! It might be a good exercise for the reader to use the commands we saw in the previous chapter to clean up some remaining differences from idiomatic OCaml.
adam@351 267
adam@351 268 One more piece of the full picture is missing. To go on and prove correctness of [mergeSort], we would need more than a way of unfolding its definition. We also need an appropriate induction principle matched to the well-founded relation. Such a principle is available in the standard library, though we will say no more about its details here. *)
adam@351 269
adam@351 270 Check well_founded_induction.
adam@351 271 (** %\vspace{-.15in}%[[
adam@351 272 well_founded_induction
adam@351 273 : forall (A : Type) (R : A -> A -> Prop),
adam@351 274 well_founded R ->
adam@351 275 forall P : A -> Set,
adam@351 276 (forall x : A, (forall y : A, R y x -> P y) -> P x) ->
adam@351 277 forall a : A, P a
adam@351 278 ]]
adam@351 279
adam@351 280 Some more recent Coq features provide more convenient syntax for defining recursive functions. Interested readers can consult the Coq manual about the commands %\index{Function}%[Function] and %\index{Program Fixpoint}%[Program Fixpoint]. *)
adam@352 281
adam@352 282
adam@354 283 (** * A Non-Termination Monad Inspired by Domain Theory *)
adam@352 284
adam@424 285 (** The key insights of %\index{domain theory}%domain theory%~\cite{WinskelDomains}% inspire the next approach to modeling non-termination. Domain theory is based on _information orders_ that relate values representing computation results, according to how much information these values convey. For instance, a simple domain might include values "the program does not terminate" and "the program terminates with the answer 5." The former is considered to be an _approximation_ of the latter, while the latter is _not_ an approximation of "the program terminates with the answer 6." The details of domain theory will not be important in what follows; we merely borrow the notion of an approximation ordering on computation results.
adam@355 286
adam@355 287 Consider this definition of a type of computations. *)
adam@355 288
adam@352 289 Section computation.
adam@352 290 Variable A : Type.
adam@355 291 (** The type [A] describes the result a computation will yield, if it terminates.
adam@355 292
adam@355 293 We give a rich dependent type to computations themselves: *)
adam@352 294
adam@352 295 Definition computation :=
adam@352 296 {f : nat -> option A
adam@352 297 | forall (n : nat) (v : A),
adam@352 298 f n = Some v
adam@352 299 -> forall (n' : nat), n' >= n
adam@352 300 -> f n' = Some v}.
adam@352 301
adam@398 302 (** A computation is fundamentally a function [f] from an _approximation level_ [n] to an optional result. Intuitively, higher [n] values enable termination in more cases than lower values. A call to [f] may return [None] to indicate that [n] was not high enough to run the computation to completion; higher [n] values may yield [Some]. Further, the proof obligation within the sigma type asserts that [f] is _monotone_ in an appropriate sense: when some [n] is sufficient to produce termination, so are all higher [n] values, and they all yield the same program result [v].
adam@355 303
adam@355 304 It is easy to define a relation characterizing when a computation runs to a particular result at a particular approximation level. *)
adam@355 305
adam@352 306 Definition runTo (m : computation) (n : nat) (v : A) :=
adam@352 307 proj1_sig m n = Some v.
adam@352 308
adam@355 309 (** On top of [runTo], we also define [run], which is the most abstract notion of when a computation runs to a value. *)
adam@355 310
adam@352 311 Definition run (m : computation) (v : A) :=
adam@352 312 exists n, runTo m n v.
adam@352 313 End computation.
adam@352 314
adam@355 315 (** The book source code contains at this point some tactics, lemma proofs, and hint commands, to be used in proving facts about computations. Since their details are orthogonal to the message of this chapter, I have omitted them in the rendered version. *)
adam@355 316 (* begin hide *)
adam@355 317
adam@352 318 Hint Unfold runTo.
adam@352 319
adam@352 320 Ltac run' := unfold run, runTo in *; try red; crush;
adam@352 321 repeat (match goal with
adam@352 322 | [ _ : proj1_sig ?E _ = _ |- _ ] =>
adam@352 323 match goal with
adam@352 324 | [ x : _ |- _ ] =>
adam@352 325 match x with
adam@352 326 | E => destruct E
adam@352 327 end
adam@352 328 end
adam@352 329 | [ |- context[match ?M with exist _ _ => _ end] ] => let Heq := fresh "Heq" in
adam@352 330 case_eq M; intros ? ? Heq; try rewrite Heq in *; try subst
adam@352 331 | [ _ : context[match ?M with exist _ _ => _ end] |- _ ] => let Heq := fresh "Heq" in
adam@352 332 case_eq M; intros ? ? Heq; try rewrite Heq in *; subst
adam@352 333 | [ H : forall n v, ?E n = Some v -> _,
adam@352 334 _ : context[match ?E ?N with Some _ => _ | None => _ end] |- _ ] =>
adam@426 335 specialize (H N); destruct (E N); try rewrite (H _ (eq_refl _)) by auto; try discriminate
adam@352 336 | [ H : forall n v, ?E n = Some v -> _, H' : ?E _ = Some _ |- _ ] => rewrite (H _ _ H') by auto
adam@352 337 end; simpl in *); eauto 7.
adam@352 338
adam@352 339 Ltac run := run'; repeat (match goal with
adam@352 340 | [ H : forall n v, ?E n = Some v -> _
adam@352 341 |- context[match ?E ?N with Some _ => _ | None => _ end] ] =>
adam@426 342 specialize (H N); destruct (E N); try rewrite (H _ (eq_refl _)) by auto; try discriminate
adam@352 343 end; run').
adam@352 344
adam@352 345 Lemma ex_irrelevant : forall P : Prop, P -> exists n : nat, P.
adam@352 346 exists 0; auto.
adam@352 347 Qed.
adam@352 348
adam@352 349 Hint Resolve ex_irrelevant.
adam@352 350
adam@352 351 Require Import Max.
adam@352 352
adam@380 353 Theorem max_spec_le : forall n m, n <= m /\ max n m = m \/ m <= n /\ max n m = n.
adam@380 354 induction n; destruct m; simpl; intuition;
adam@380 355 specialize (IHn m); intuition.
adam@380 356 Qed.
adam@380 357
adam@352 358 Ltac max := intros n m; generalize (max_spec_le n m); crush.
adam@352 359
adam@352 360 Lemma max_1 : forall n m, max n m >= n.
adam@352 361 max.
adam@352 362 Qed.
adam@352 363
adam@352 364 Lemma max_2 : forall n m, max n m >= m.
adam@352 365 max.
adam@352 366 Qed.
adam@352 367
adam@352 368 Hint Resolve max_1 max_2.
adam@352 369
adam@352 370 Lemma ge_refl : forall n, n >= n.
adam@352 371 crush.
adam@352 372 Qed.
adam@352 373
adam@352 374 Hint Resolve ge_refl.
adam@352 375
adam@352 376 Hint Extern 1 => match goal with
adam@352 377 | [ H : _ = exist _ _ _ |- _ ] => rewrite H
adam@352 378 end.
adam@355 379 (* end hide *)
adam@355 380 (** remove printing exists *)
adam@355 381
adam@357 382 (** Now, as a simple first example of a computation, we can define [Bottom], which corresponds to an infinite loop. For any approximation level, it fails to terminate (returns [None]). Note the use of [abstract] to create a new opaque lemma for the proof found by the [run] tactic. In contrast to the previous section, opaque proofs are fine here, since the proof components of computations do not influence evaluation behavior. *)
adam@352 383
adam@352 384 Section Bottom.
adam@352 385 Variable A : Type.
adam@352 386
adam@352 387 Definition Bottom : computation A.
adam@352 388 exists (fun _ : nat => @None A); abstract run.
adam@352 389 Defined.
adam@352 390
adam@352 391 Theorem run_Bottom : forall v, ~run Bottom v.
adam@352 392 run.
adam@352 393 Qed.
adam@352 394 End Bottom.
adam@352 395
adam@355 396 (** A slightly more complicated example is [Return], which gives the same terminating answer at every approximation level. *)
adam@355 397
adam@352 398 Section Return.
adam@352 399 Variable A : Type.
adam@352 400 Variable v : A.
adam@352 401
adam@352 402 Definition Return : computation A.
adam@352 403 intros; exists (fun _ : nat => Some v); abstract run.
adam@352 404 Defined.
adam@352 405
adam@352 406 Theorem run_Return : run Return v.
adam@352 407 run.
adam@352 408 Qed.
adam@352 409 End Return.
adam@352 410
adam@356 411 (** The name [Return] was meant to be suggestive of the standard operations of %\index{monad}%monads%~\cite{Monads}%. The other standard operation is [Bind], which lets us run one computation and, if it terminates, pass its result off to another computation. *)
adam@352 412
adam@352 413 Section Bind.
adam@352 414 Variables A B : Type.
adam@352 415 Variable m1 : computation A.
adam@352 416 Variable m2 : A -> computation B.
adam@352 417
adam@352 418 Definition Bind : computation B.
adam@352 419 exists (fun n =>
adam@357 420 let (f1, _) := m1 in
adam@352 421 match f1 n with
adam@352 422 | None => None
adam@352 423 | Some v =>
adam@357 424 let (f2, _) := m2 v in
adam@352 425 f2 n
adam@352 426 end); abstract run.
adam@352 427 Defined.
adam@352 428
adam@352 429 Theorem run_Bind : forall (v1 : A) (v2 : B),
adam@352 430 run m1 v1
adam@352 431 -> run (m2 v1) v2
adam@352 432 -> run Bind v2.
adam@352 433 run; match goal with
adam@352 434 | [ x : nat, y : nat |- _ ] => exists (max x y)
adam@352 435 end; run.
adam@352 436 Qed.
adam@352 437 End Bind.
adam@352 438
adam@355 439 (** A simple notation lets us write [Bind] calls the way they appear in Haskell. *)
adam@352 440
adam@352 441 Notation "x <- m1 ; m2" :=
adam@352 442 (Bind m1 (fun x => m2)) (right associativity, at level 70).
adam@352 443
adam@424 444 (** We can verify that we have indeed defined a monad, by proving the standard monad laws. Part of the exercise is choosing an appropriate notion of equality between computations. We use "equality at all approximation levels." *)
adam@355 445
adam@352 446 Definition meq A (m1 m2 : computation A) := forall n, proj1_sig m1 n = proj1_sig m2 n.
adam@352 447
adam@352 448 Theorem left_identity : forall A B (a : A) (f : A -> computation B),
adam@352 449 meq (Bind (Return a) f) (f a).
adam@352 450 run.
adam@352 451 Qed.
adam@352 452
adam@352 453 Theorem right_identity : forall A (m : computation A),
adam@352 454 meq (Bind m (@Return _)) m.
adam@352 455 run.
adam@352 456 Qed.
adam@352 457
adam@357 458 Theorem associativity : forall A B C (m : computation A)
adam@357 459 (f : A -> computation B) (g : B -> computation C),
adam@352 460 meq (Bind (Bind m f) g) (Bind m (fun x => Bind (f x) g)).
adam@352 461 run.
adam@352 462 Qed.
adam@352 463
adam@398 464 (** Now we come to the piece most directly inspired by domain theory. We want to support general recursive function definitions, but domain theory tells us that not all definitions are reasonable; some fail to be _continuous_ and thus represent unrealizable computations. To formalize an analogous notion of continuity for our non-termination monad, we write down the approximation relation on computation results that we have had in mind all along. *)
adam@352 465
adam@352 466 Section lattice.
adam@352 467 Variable A : Type.
adam@352 468
adam@352 469 Definition leq (x y : option A) :=
adam@352 470 forall v, x = Some v -> y = Some v.
adam@352 471 End lattice.
adam@352 472
adam@355 473 (** We now have the tools we need to define a new [Fix] combinator that, unlike the one we saw in the prior section, does not require a termination proof, and in fact admits recursive definition of functions that fail to terminate on some or all inputs. *)
adam@352 474
adam@352 475 Section Fix.
adam@355 476 (** First, we have the function domain and range types. *)
adam@355 477
adam@352 478 Variables A B : Type.
adam@355 479
adam@355 480 (** Next comes the function body, which is written as though it can be parameterized over itself, for recursive calls. *)
adam@355 481
adam@352 482 Variable f : (A -> computation B) -> (A -> computation B).
adam@352 483
adam@355 484 (** Finally, we impose an obligation to prove that the body [f] is continuous. That is, when [f] terminates according to one recursive version of itself, it also terminates with the same result at the same approximation level when passed a recursive version that refines the original, according to [leq]. *)
adam@355 485
adam@352 486 Hypothesis f_continuous : forall n v v1 x,
adam@352 487 runTo (f v1 x) n v
adam@352 488 -> forall (v2 : A -> computation B),
adam@352 489 (forall x, leq (proj1_sig (v1 x) n) (proj1_sig (v2 x) n))
adam@352 490 -> runTo (f v2 x) n v.
adam@352 491
adam@355 492 (** The computational part of the [Fix] combinator is easy to define. At approximation level 0, we diverge; at higher levels, we run the body with a functional argument drawn from the next lower level. *)
adam@355 493
adam@352 494 Fixpoint Fix' (n : nat) (x : A) : computation B :=
adam@352 495 match n with
adam@352 496 | O => Bottom _
adam@352 497 | S n' => f (Fix' n') x
adam@352 498 end.
adam@352 499
adam@355 500 (** Now it is straightforward to package [Fix'] as a computation combinator [Fix]. *)
adam@355 501
adam@352 502 Hint Extern 1 (_ >= _) => omega.
adam@352 503 Hint Unfold leq.
adam@352 504
adam@352 505 Lemma Fix'_ok : forall steps n x v, proj1_sig (Fix' n x) steps = Some v
adam@352 506 -> forall n', n' >= n
adam@352 507 -> proj1_sig (Fix' n' x) steps = Some v.
adam@352 508 unfold runTo in *; induction n; crush;
adam@352 509 match goal with
adam@352 510 | [ H : _ >= _ |- _ ] => inversion H; crush; eauto
adam@352 511 end.
adam@352 512 Qed.
adam@352 513
adam@352 514 Hint Resolve Fix'_ok.
adam@352 515
adam@352 516 Hint Extern 1 (proj1_sig _ _ = _) => simpl;
adam@352 517 match goal with
adam@352 518 | [ |- proj1_sig ?E _ = _ ] => eapply (proj2_sig E)
adam@352 519 end.
adam@352 520
adam@352 521 Definition Fix : A -> computation B.
adam@352 522 intro x; exists (fun n => proj1_sig (Fix' n x) n); abstract run.
adam@352 523 Defined.
adam@352 524
adam@355 525 (** Finally, we can prove that [Fix] obeys the expected computation rule. *)
adam@352 526
adam@352 527 Theorem run_Fix : forall x v,
adam@352 528 run (f Fix x) v
adam@352 529 -> run (Fix x) v.
adam@352 530 run; match goal with
adam@352 531 | [ n : nat |- _ ] => exists (S n); eauto
adam@352 532 end.
adam@352 533 Qed.
adam@352 534 End Fix.
adam@352 535
adam@355 536 (* begin hide *)
adam@352 537 Lemma leq_Some : forall A (x y : A), leq (Some x) (Some y)
adam@352 538 -> x = y.
adam@426 539 intros ? ? ? H; generalize (H _ (eq_refl _)); crush.
adam@352 540 Qed.
adam@352 541
adam@352 542 Lemma leq_None : forall A (x y : A), leq (Some x) None
adam@352 543 -> False.
adam@426 544 intros ? ? ? H; generalize (H _ (eq_refl _)); crush.
adam@352 545 Qed.
adam@352 546
adam@355 547 Ltac mergeSort' := run;
adam@355 548 repeat (match goal with
adam@355 549 | [ |- context[match ?E with O => _ | S _ => _ end] ] => destruct E
adam@355 550 end; run);
adam@355 551 repeat match goal with
adam@355 552 | [ H : forall x, leq (proj1_sig (?f x) _) (proj1_sig (?g x) _) |- _ ] =>
adam@355 553 match goal with
adam@355 554 | [ H1 : f ?arg = _, H2 : g ?arg = _ |- _ ] =>
adam@355 555 generalize (H arg); rewrite H1; rewrite H2; clear H1 H2; simpl; intro
adam@355 556 end
adam@355 557 end; run; repeat match goal with
adam@355 558 | [ H : _ |- _ ] => (apply leq_None in H; tauto) || (apply leq_Some in H; subst)
adam@355 559 end; auto.
adam@355 560 (* end hide *)
adam@355 561
adam@355 562 (** After all that work, it is now fairly painless to define a version of [mergeSort] that requires no proof of termination. We appeal to a program-specific tactic whose definition is hidden here but present in the book source. *)
adam@355 563
adam@352 564 Definition mergeSort' : forall A, (A -> A -> bool) -> list A -> computation (list A).
adam@352 565 refine (fun A le => Fix
adam@352 566 (fun (mergeSort : list A -> computation (list A))
adam@352 567 (ls : list A) =>
adam@352 568 if le_lt_dec 2 (length ls)
adam@352 569 then let lss := partition ls in
adam@352 570 ls1 <- mergeSort (fst lss);
adam@352 571 ls2 <- mergeSort (snd lss);
adam@352 572 Return (merge le ls1 ls2)
adam@355 573 else Return ls) _); abstract mergeSort'.
adam@352 574 Defined.
adam@352 575
adam@424 576 (** Furthermore, "running" [mergeSort'] on concrete inputs is as easy as choosing a sufficiently high approximation level and letting Coq's computation rules do the rest. Contrast this with the proof work that goes into deriving an evaluation fact for a deeply embedded language, with one explicit proof rule application per execution step. *)
adam@352 577
adam@352 578 Lemma test_mergeSort' : run (mergeSort' leb (1 :: 2 :: 36 :: 8 :: 19 :: nil))
adam@352 579 (1 :: 2 :: 8 :: 19 :: 36 :: nil).
adam@352 580 exists 4; reflexivity.
adam@352 581 Qed.
adam@352 582
adam@355 583 (** There is another benefit of our new [Fix] compared with one we used in the previous section: we can now write recursive functions that sometimes fail to terminate, without losing easy reasoning principles for the terminating cases. Consider this simple example, which appeals to another tactic whose definition we elide here. *)
adam@355 584
adam@355 585 (* begin hide *)
adam@355 586 Ltac looper := unfold leq in *; run;
adam@355 587 repeat match goal with
adam@355 588 | [ x : unit |- _ ] => destruct x
adam@355 589 | [ x : bool |- _ ] => destruct x
adam@355 590 end; auto.
adam@355 591 (* end hide *)
adam@355 592
adam@352 593 Definition looper : bool -> computation unit.
adam@352 594 refine (Fix (fun looper (b : bool) =>
adam@355 595 if b then Return tt else looper b) _); abstract looper.
adam@352 596 Defined.
adam@352 597
adam@352 598 Lemma test_looper : run (looper true) tt.
adam@352 599 exists 1; reflexivity.
adam@352 600 Qed.
adam@354 601
adam@355 602 (** As before, proving outputs for specific inputs is as easy as demonstrating a high enough approximation level.
adam@355 603
adam@424 604 There are other theorems that are important to prove about combinators like [Return], [Bind], and [Fix]. In general, for a computation [c], we sometimes have a hypothesis proving [run c v] for some [v], and we want to perform inversion to deduce what [v] must be. Each combinator should ideally have a theorem of that kind, for [c] built directly from that combinator. We have omitted such theorems here, but they are not hard to prove. In general, the domain theory-inspired approach avoids the type-theoretic "gotchas" that tend to show up in approaches that try to mix normal Coq computation with explicit syntax types. The next section of this chapter demonstrates two alternate approaches of that sort. *)
adam@355 605
adam@354 606
adam@354 607 (** * Co-Inductive Non-Termination Monads *)
adam@354 608
adam@356 609 (** There are two key downsides to both of the previous approaches: both require unusual syntax based on explicit calls to fixpoint combinators, and both generate immediate proof obligations about the bodies of recursive definitions. In Chapter 5, we have already seen how co-inductive types support recursive definitions that exhibit certain well-behaved varieties of non-termination. It turns out that we can leverage that co-induction support for encoding of general recursive definitions, by adding layers of co-inductive syntax. In effect, we mix elements of shallow and deep embeddings.
adam@356 610
adam@356 611 Our first example of this kind, proposed by Capretta%~\cite{Capretta}%, defines a silly-looking type of thunks; that is, computations that may be forced to yield results, if they terminate. *)
adam@356 612
adam@354 613 CoInductive thunk (A : Type) : Type :=
adam@354 614 | Answer : A -> thunk A
adam@354 615 | Think : thunk A -> thunk A.
adam@354 616
adam@356 617 (** A computation is either an immediate [Answer] or another computation wrapped inside [Think]. Since [thunk] is co-inductive, every [thunk] type is inhabited by an infinite nesting of [Think]s, standing for non-termination. Terminating results are [Answer] wrapped inside some finite number of [Think]s.
adam@356 618
adam@424 619 Why bother to write such a strange definition? The definition of [thunk] is motivated by the ability it gives us to define a "bind" operation, similar to the one we defined in the previous section. *)
adam@356 620
adam@356 621 CoFixpoint TBind A B (m1 : thunk A) (m2 : A -> thunk B) : thunk B :=
adam@354 622 match m1 with
adam@354 623 | Answer x => m2 x
adam@354 624 | Think m1' => Think (TBind m1' m2)
adam@354 625 end.
adam@354 626
adam@356 627 (** Note that the definition would violate the co-recursion guardedness restriction if we left out the seemingly superfluous [Think] on the righthand side of the second [match] branch.
adam@356 628
adam@356 629 We can prove that [Answer] and [TBind] form a monad for [thunk]. The proof is omitted here but present in the book source. As usual for this sort of proof, a key element is choosing an appropriate notion of equality for [thunk]s. *)
adam@356 630
adam@356 631 (* begin hide *)
adam@354 632 CoInductive thunk_eq A : thunk A -> thunk A -> Prop :=
adam@354 633 | EqAnswer : forall x, thunk_eq (Answer x) (Answer x)
adam@354 634 | EqThinkL : forall m1 m2, thunk_eq m1 m2 -> thunk_eq (Think m1) m2
adam@354 635 | EqThinkR : forall m1 m2, thunk_eq m1 m2 -> thunk_eq m1 (Think m2).
adam@354 636
adam@354 637 Section thunk_eq_coind.
adam@354 638 Variable A : Type.
adam@354 639 Variable P : thunk A -> thunk A -> Prop.
adam@354 640
adam@354 641 Hypothesis H : forall m1 m2, P m1 m2
adam@354 642 -> match m1, m2 with
adam@354 643 | Answer x1, Answer x2 => x1 = x2
adam@354 644 | Think m1', Think m2' => P m1' m2'
adam@354 645 | Think m1', _ => P m1' m2
adam@354 646 | _, Think m2' => P m1 m2'
adam@354 647 end.
adam@354 648
adam@354 649 Theorem thunk_eq_coind : forall m1 m2, P m1 m2 -> thunk_eq m1 m2.
adam@354 650 cofix; intros;
adam@354 651 match goal with
adam@354 652 | [ H' : P _ _ |- _ ] => specialize (H H'); clear H'
adam@354 653 end; destruct m1; destruct m2; subst; repeat constructor; auto.
adam@354 654 Qed.
adam@354 655 End thunk_eq_coind.
adam@356 656 (* end hide *)
adam@356 657
adam@356 658 (** In the proofs to follow, we will need a function similar to one we saw in Chapter 5, to pull apart and reassemble a [thunk] in a way that provokes reduction of co-recursive calls. *)
adam@354 659
adam@354 660 Definition frob A (m : thunk A) : thunk A :=
adam@354 661 match m with
adam@354 662 | Answer x => Answer x
adam@354 663 | Think m' => Think m'
adam@354 664 end.
adam@354 665
adam@354 666 Theorem frob_eq : forall A (m : thunk A), frob m = m.
adam@354 667 destruct m; reflexivity.
adam@354 668 Qed.
adam@354 669
adam@356 670 (* begin hide *)
adam@354 671 Theorem thunk_eq_frob : forall A (m1 m2 : thunk A),
adam@354 672 thunk_eq (frob m1) (frob m2)
adam@354 673 -> thunk_eq m1 m2.
adam@354 674 intros; repeat rewrite frob_eq in *; auto.
adam@354 675 Qed.
adam@354 676
adam@354 677 Ltac findDestr := match goal with
adam@354 678 | [ |- context[match ?E with Answer _ => _ | Think _ => _ end] ] =>
adam@354 679 match E with
adam@354 680 | context[match _ with Answer _ => _ | Think _ => _ end] => fail 1
adam@354 681 | _ => destruct E
adam@354 682 end
adam@354 683 end.
adam@354 684
adam@354 685 Theorem thunk_eq_refl : forall A (m : thunk A), thunk_eq m m.
adam@354 686 intros; apply (thunk_eq_coind (fun m1 m2 => m1 = m2)); crush; findDestr; reflexivity.
adam@354 687 Qed.
adam@354 688
adam@354 689 Hint Resolve thunk_eq_refl.
adam@354 690
adam@354 691 Theorem tleft_identity : forall A B (a : A) (f : A -> thunk B),
adam@354 692 thunk_eq (TBind (Answer a) f) (f a).
adam@354 693 intros; apply thunk_eq_frob; crush.
adam@354 694 Qed.
adam@354 695
adam@354 696 Theorem tright_identity : forall A (m : thunk A),
adam@354 697 thunk_eq (TBind m (@Answer _)) m.
adam@354 698 intros; apply (thunk_eq_coind (fun m1 m2 => m1 = TBind m2 (@Answer _))); crush;
adam@354 699 findDestr; reflexivity.
adam@354 700 Qed.
adam@354 701
adam@354 702 Lemma TBind_Answer : forall (A B : Type) (v : A) (m2 : A -> thunk B),
adam@354 703 TBind (Answer v) m2 = m2 v.
adam@354 704 intros; rewrite <- (frob_eq (TBind (Answer v) m2));
adam@354 705 simpl; findDestr; reflexivity.
adam@354 706 Qed.
adam@354 707
adam@375 708 Hint Rewrite TBind_Answer.
adam@354 709
adam@355 710 (** printing exists $\exists$ *)
adam@355 711
adam@354 712 Theorem tassociativity : forall A B C (m : thunk A) (f : A -> thunk B) (g : B -> thunk C),
adam@354 713 thunk_eq (TBind (TBind m f) g) (TBind m (fun x => TBind (f x) g)).
adam@354 714 intros; apply (thunk_eq_coind (fun m1 m2 => (exists m,
adam@354 715 m1 = TBind (TBind m f) g
adam@354 716 /\ m2 = TBind m (fun x => TBind (f x) g))
adam@354 717 \/ m1 = m2)); crush; eauto; repeat (findDestr; crush; eauto).
adam@354 718 Qed.
adam@356 719 (* end hide *)
adam@356 720
adam@356 721 (** As a simple example, here is how we might define a tail-recursive factorial function. *)
adam@354 722
adam@354 723 CoFixpoint fact (n acc : nat) : thunk nat :=
adam@354 724 match n with
adam@354 725 | O => Answer acc
adam@354 726 | S n' => Think (fact n' (S n' * acc))
adam@354 727 end.
adam@354 728
adam@356 729 (** To test our definition, we need an evaluation relation that characterizes results of evaluating [thunk]s. *)
adam@356 730
adam@354 731 Inductive eval A : thunk A -> A -> Prop :=
adam@354 732 | EvalAnswer : forall x, eval (Answer x) x
adam@354 733 | EvalThink : forall m x, eval m x -> eval (Think m) x.
adam@354 734
adam@375 735 Hint Rewrite frob_eq.
adam@354 736
adam@354 737 Lemma eval_frob : forall A (c : thunk A) x,
adam@354 738 eval (frob c) x
adam@354 739 -> eval c x.
adam@354 740 crush.
adam@354 741 Qed.
adam@354 742
adam@354 743 Theorem eval_fact : eval (fact 5 1) 120.
adam@354 744 repeat (apply eval_frob; simpl; constructor).
adam@354 745 Qed.
adam@354 746
adam@356 747 (** We need to apply constructors of [eval] explicitly, but the process is easy to automate completely for concrete input programs.
adam@356 748
adam@357 749 Now consider another very similar definition, this time of a Fibonacci number funtion. *)
adam@357 750
adam@357 751 Notation "x <- m1 ; m2" :=
adam@357 752 (TBind m1 (fun x => m2)) (right associativity, at level 70).
adam@357 753
adam@404 754 (* begin hide *)
adam@424 755 Definition fib := pred.
adam@404 756 (* end hide *)
adam@404 757
adam@357 758 (** %\vspace{-.15in}%[[
adam@354 759 CoFixpoint fib (n : nat) : thunk nat :=
adam@354 760 match n with
adam@354 761 | 0 => Answer 1
adam@354 762 | 1 => Answer 1
adam@357 763 | _ => n1 <- fib (pred n);
adam@357 764 n2 <- fib (pred (pred n));
adam@357 765 Answer (n1 + n2)
adam@354 766 end.
adam@354 767 ]]
adam@354 768
adam@356 769 Coq complains that the guardedness condition is violated. The two recursive calls are immediate arguments to [TBind], but [TBind] is not a constructor of [thunk]. Rather, it is a defined function. This example shows a very serious limitation of [thunk] for traditional functional programming: it is not, in general, possible to make recursive calls and then make further recursive calls, depending on the first call's result. The [fact] example succeeded because it was already tail recursive, meaning no further computation is needed after a recursive call.
adam@356 770
adam@356 771 %\medskip%
adam@356 772
adam@424 773 I know no easy fix for this problem of [thunk], but we can define an alternate co-inductive monad that avoids the problem, based on a proposal by Megacz%~\cite{Megacz}%. We ran into trouble because [TBind] was not a constructor of [thunk], so let us define a new type family where "bind" is a constructor. *)
adam@354 774
adam@354 775 CoInductive comp (A : Type) : Type :=
adam@354 776 | Ret : A -> comp A
adam@354 777 | Bnd : forall B, comp B -> (B -> comp A) -> comp A.
adam@354 778
adam@404 779 (** This example shows off Coq's support for%\index{recursively non-uniform parameters}% _recursively non-uniform parameters_, as in the case of the parameter [A] declared above, where each constructor's type ends in [comp A], but there is a recursive use of [comp] with a different parameter [B]. Beside that technical wrinkle, we see the simplest possible definition of a monad, via a type whose two constructors are precisely the monad operators.
adam@356 780
adam@356 781 It is easy to define the semantics of terminating [comp] computations. *)
adam@356 782
adam@354 783 Inductive exec A : comp A -> A -> Prop :=
adam@354 784 | ExecRet : forall x, exec (Ret x) x
adam@354 785 | ExecBnd : forall B (c : comp B) (f : B -> comp A) x1 x2, exec (A := B) c x1
adam@354 786 -> exec (f x1) x2
adam@354 787 -> exec (Bnd c f) x2.
adam@354 788
adam@356 789 (** We can also prove that [Ret] and [Bnd] form a monad according to a notion of [comp] equality based on [exec], but we omit details here; they are in the book source at this point. *)
adam@356 790
adam@356 791 (* begin hide *)
adam@354 792 Hint Constructors exec.
adam@354 793
adam@354 794 Definition comp_eq A (c1 c2 : comp A) := forall r, exec c1 r <-> exec c2 r.
adam@354 795
adam@354 796 Ltac inverter := repeat match goal with
adam@354 797 | [ H : exec _ _ |- _ ] => inversion H; []; crush
adam@354 798 end.
adam@354 799
adam@354 800 Theorem cleft_identity : forall A B (a : A) (f : A -> comp B),
adam@354 801 comp_eq (Bnd (Ret a) f) (f a).
adam@354 802 red; crush; inverter; eauto.
adam@354 803 Qed.
adam@354 804
adam@354 805 Theorem cright_identity : forall A (m : comp A),
adam@354 806 comp_eq (Bnd m (@Ret _)) m.
adam@354 807 red; crush; inverter; eauto.
adam@354 808 Qed.
adam@354 809
adam@354 810 Lemma cassociativity1 : forall A B C (f : A -> comp B) (g : B -> comp C) r c,
adam@354 811 exec c r
adam@354 812 -> forall m, c = Bnd (Bnd m f) g
adam@354 813 -> exec (Bnd m (fun x => Bnd (f x) g)) r.
adam@354 814 induction 1; crush.
adam@354 815 match goal with
adam@354 816 | [ H : Bnd _ _ = Bnd _ _ |- _ ] => injection H; clear H; intros; try subst
adam@354 817 end.
adam@354 818 move H3 after A.
adam@354 819 generalize dependent B0.
adam@354 820 do 2 intro.
adam@354 821 subst.
adam@354 822 crush.
adam@354 823 inversion H; clear H; crush.
adam@354 824 eauto.
adam@354 825 Qed.
adam@354 826
adam@354 827 Lemma cassociativity2 : forall A B C (f : A -> comp B) (g : B -> comp C) r c,
adam@354 828 exec c r
adam@354 829 -> forall m, c = Bnd m (fun x => Bnd (f x) g)
adam@354 830 -> exec (Bnd (Bnd m f) g) r.
adam@354 831 induction 1; crush.
adam@354 832 match goal with
adam@354 833 | [ H : Bnd _ _ = Bnd _ _ |- _ ] => injection H; clear H; intros; try subst
adam@354 834 end.
adam@354 835 move H3 after B.
adam@354 836 generalize dependent B0.
adam@354 837 do 2 intro.
adam@354 838 subst.
adam@354 839 crush.
adam@354 840 inversion H0; clear H0; crush.
adam@354 841 eauto.
adam@354 842 Qed.
adam@354 843
adam@354 844 Hint Resolve cassociativity1 cassociativity2.
adam@354 845
adam@354 846 Theorem cassociativity : forall A B C (m : comp A) (f : A -> comp B) (g : B -> comp C),
adam@354 847 comp_eq (Bnd (Bnd m f) g) (Bnd m (fun x => Bnd (f x) g)).
adam@354 848 red; crush; eauto.
adam@354 849 Qed.
adam@356 850 (* end hide *)
adam@356 851
adam@424 852 (** Not only can we define the Fibonacci function with the new monad, but even our running example of merge sort becomes definable. By shadowing our previous notation for "bind,", we can write almost exactly the same code as in our previous [mergeSort'] definition, but with less syntactic clutter. *)
adam@356 853
adam@356 854 Notation "x <- m1 ; m2" := (Bnd m1 (fun x => m2)).
adam@354 855
adam@354 856 CoFixpoint mergeSort'' A (le : A -> A -> bool) (ls : list A) : comp (list A) :=
adam@354 857 if le_lt_dec 2 (length ls)
adam@354 858 then let lss := partition ls in
adam@356 859 ls1 <- mergeSort'' le (fst lss);
adam@356 860 ls2 <- mergeSort'' le (snd lss);
adam@356 861 Ret (merge le ls1 ls2)
adam@354 862 else Ret ls.
adam@354 863
adam@356 864 (** To execute this function, we go through the usual exercise of writing a function to catalyze evaluation of co-recursive calls. *)
adam@356 865
adam@354 866 Definition frob' A (c : comp A) :=
adam@354 867 match c with
adam@354 868 | Ret x => Ret x
adam@354 869 | Bnd _ c' f => Bnd c' f
adam@354 870 end.
adam@354 871
adam@354 872 Lemma exec_frob : forall A (c : comp A) x,
adam@354 873 exec (frob' c) x
adam@354 874 -> exec c x.
adam@356 875 destruct c; crush.
adam@354 876 Qed.
adam@354 877
adam@356 878 (** Now the same sort of proof script that we applied for testing [thunk]s will get the job done. *)
adam@356 879
adam@354 880 Lemma test_mergeSort'' : exec (mergeSort'' leb (1 :: 2 :: 36 :: 8 :: 19 :: nil))
adam@354 881 (1 :: 2 :: 8 :: 19 :: 36 :: nil).
adam@354 882 repeat (apply exec_frob; simpl; econstructor).
adam@354 883 Qed.
adam@354 884
adam@356 885 (** Have we finally reached the ideal solution for encoding general recursive definitions, with minimal hassle in syntax and proof obligations? Unfortunately, we have not, as [comp] has a serious expressivity weakness. Consider the following definition of a curried addition function: *)
adam@356 886
adam@354 887 Definition curriedAdd (n : nat) := Ret (fun m : nat => Ret (n + m)).
adam@354 888
adam@356 889 (** This definition works fine, but we run into trouble when we try to apply it in a trivial way.
adam@356 890 [[
adam@356 891 Definition testCurriedAdd := Bnd (curriedAdd 2) (fun f => f 3).
adam@354 892 ]]
adam@354 893
adam@354 894 <<
adam@354 895 Error: Universe inconsistency.
adam@354 896 >>
adam@356 897
adam@356 898 The problem has to do with rules for inductive definitions that we still study in more detail in Chapter 12. Briefly, recall that the type of the constructor [Bnd] quantifies over a type [B]. To make [testCurriedAdd] work, we would need to instantiate [B] as [nat -> comp nat]. However, Coq enforces a %\emph{predicativity restriction}% that (roughly) no quantifier in an inductive or co-inductive type's definition may ever be instantiated with a term that contains the type being defined. Chapter 12 presents the exact mechanism by which this restriction is enforced, but for now our conclusion is that [comp] is fatally flawed as a way of encoding interesting higher-order functional programs that use general recursion. *)
adam@354 899
adam@354 900
adam@357 901 (** * Comparing the Alternatives *)
adam@354 902
adam@356 903 (** We have seen four different approaches to encoding general recursive definitions in Coq. Among them there is no clear champion that dominates the others in every important way. Instead, we close the chapter by comparing the techniques along a number of dimensions. Every technique allows recursive definitions with terminaton arguments that go beyond Coq's built-in termination checking, so we must turn to subtler points to highlight differences.
adam@356 904
adam@356 905 One useful property is automatic integration with normal Coq programming. That is, we would like the type of a function to be the same, whether or not that function is defined using an interesting recursion pattern. Only the first of the four techniques, well-founded recursion, meets this criterion. It is also the only one of the four to meet the related criterion that evaluation of function calls can take place entirely inside Coq's built-in computation machinery. The monad inspired by domain theory occupies some middle ground in this dimension, since generally standard computation is enough to evaluate a term once a high enough approximation level is provided.
adam@356 906
adam@356 907 Another useful property is that a function and its termination argument may be developed separately. We may even want to define functions that fail to terminate on some or all inputs. The well-founded recursion technique does not have this property, but the other three do.
adam@356 908
adam@356 909 One minor plus is the ability to write recursive definitions in natural syntax, rather than with calls to higher-order combinators. This downside of the first two techniques is actually rather easy to get around using Coq's notation mechanism, though we leave the details as an exercise for the reader.
adam@356 910
adam@356 911 The first two techniques impose proof obligations that are more basic than terminaton arguments, where well-founded recursion requires a proof of extensionality and domain-theoretic recursion requires a proof of continuity. A function may not be defined, and thus may not be computed with, until these obligations are proved. The co-inductive techniques avoid this problem, as recursive definitions may be made without any proof obligations.
adam@356 912
adam@356 913 We can also consider support for common idioms in functional programming. For instance, the [thunk] monad effectively only supports recursion that is tail recursion, while the others allow arbitrary recursion schemes.
adam@356 914
adam@356 915 On the other hand, the [comp] monad does not support the effective mixing of higher-order functions and general recursion, while all the other techniques do. For instance, we can finish the failed [curriedAdd] example in the domain-theoretic monad. *)
adam@356 916
adam@354 917 Definition curriedAdd' (n : nat) := Return (fun m : nat => Return (n + m)).
adam@354 918
adam@356 919 Definition testCurriedAdd := Bind (curriedAdd' 2) (fun f => f 3).
adam@356 920
adam@357 921 (** The same techniques also apply to more interesting higher-order functions like list map, and, as in all four techniques, we can mix primitive and general recursion, preferring the former when possible to avoid proof obligations. *)
adam@354 922
adam@354 923 Fixpoint map A B (f : A -> computation B) (ls : list A) : computation (list B) :=
adam@354 924 match ls with
adam@354 925 | nil => Return nil
adam@354 926 | x :: ls' => Bind (f x) (fun x' =>
adam@354 927 Bind (map f ls') (fun ls'' =>
adam@354 928 Return (x' :: ls'')))
adam@354 929 end.
adam@354 930
adam@355 931 (** remove printing exists *)
adam@356 932 Theorem test_map : run (map (fun x => Return (S x)) (1 :: 2 :: 3 :: nil))
adam@356 933 (2 :: 3 :: 4 :: nil).
adam@354 934 exists 1; reflexivity.
adam@354 935 Qed.
adam@356 936
adam@356 937 (** One further disadvantage of [comp] is that we cannot prove an inversion lemma for executions of [Bind] without appealing to an %\emph{axiom}%, a logical complication that we discuss at more length in Chapter 12. The other three techniques allow proof of all the important theorems within the normal logic of Coq.
adam@356 938
adam@357 939 Perhaps one theme of our comparison is that one must trade off between, on one hand, functional programming expressiveness and compatibility with normal Coq types and computation; and, on the other hand, the level of proof obligations one is willing to handle at function definition time. *)