summaryrefslogtreecommitdiff
path: root/theories/Init/Tactics.v
blob: 23d9d10eb42ba3b3273992bd1b0182d2b92d18b1 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
(************************************************************************)
(*  v      *   The Coq Proof Assistant  /  The Coq Development Team     *)
(* <O___,, *   INRIA - CNRS - LIX - LRI - PPS - Copyright 1999-2012     *)
(*   \VV/  **************************************************************)
(*    //   *      This file is distributed under the terms of the       *)
(*         *       GNU Lesser General Public License Version 2.1        *)
(************************************************************************)

Require Import Notations.
Require Import Logic.
Require Import Specif.

(** * Useful tactics *)

(** Ex falso quodlibet : a tactic for proving False instead of the current goal.
    This is just a nicer name for tactics such as [elimtype False]
    and other [cut False]. *)

Ltac exfalso := elimtype False.

(** A tactic for proof by contradiction. With contradict H,
    -   H:~A |-  B    gives       |-  A
    -   H:~A |- ~B    gives  H: B |-  A
    -   H: A |-  B    gives       |- ~A
    -   H: A |- ~B    gives  H: B |- ~A
    -   H:False leads to a resolved subgoal.
   Moreover, negations may be in unfolded forms,
   and A or B may live in Type *)

Ltac contradict H :=
  let save tac H := let x:=fresh in intro x; tac H; rename x into H
  in
  let negpos H := case H; clear H
  in
  let negneg H := save negpos H
  in
  let pospos H :=
    let A := type of H in (exfalso; revert H; try fold (~A))
  in
  let posneg H := save pospos H
  in
  let neg H := match goal with
   | |- (~_) => negneg H
   | |- (_->False) => negneg H
   | |- _ => negpos H
  end in
  let pos H := match goal with
   | |- (~_) => posneg H
   | |- (_->False) => posneg H
   | |- _ => pospos H
  end in
  match type of H with
   | (~_) => neg H
   | (_->False) => neg H
   | _ => (elim H;fail) || pos H
  end.

(* Transforming a negative goal [ H:~A |- ~B ] into a positive one [ B |- A ]*)

Ltac swap H :=
  idtac "swap is OBSOLETE: use contradict instead.";
  intro; apply H; clear H.

(* To contradict an hypothesis without copying its type. *)

Ltac absurd_hyp H :=
  idtac "absurd_hyp is OBSOLETE: use contradict instead.";
  let T := type of H in
  absurd T.

(* A useful complement to contradict. Here H:A while G allows to conclude ~A *)

Ltac false_hyp H G :=
  let T := type of H in absurd T; [ apply G | assumption ].

(* A case with no loss of information. *)

Ltac case_eq x := generalize (eq_refl x); pattern x at -1; case x.

(* use either discriminate or injection on a hypothesis *)

Ltac destr_eq H := discriminate H || (try (injection H; clear H; intro H)).

(* Similar variants of destruct *)

Tactic Notation "destruct_with_eqn" constr(x) :=
  destruct x eqn:?.
Tactic Notation "destruct_with_eqn" ident(n) :=
  try intros until n; destruct n eqn:?.
Tactic Notation "destruct_with_eqn" ":" ident(H) constr(x) :=
  destruct x eqn:H.
Tactic Notation "destruct_with_eqn" ":" ident(H) ident(n) :=
  try intros until n; destruct n eqn:H.

(** Break every hypothesis of a certain type *)

Ltac destruct_all t :=
 match goal with
  | x : t |- _ => destruct x; destruct_all t
  | _ => idtac
 end.

(* Rewriting in all hypothesis several times everywhere *)

Tactic Notation "rewrite_all" constr(eq) := repeat rewrite eq in *.
Tactic Notation "rewrite_all" "<-" constr(eq) := repeat rewrite <- eq in *.

(** Tactics for applying equivalences.

The following code provides tactics "apply -> t", "apply <- t",
"apply -> t in H" and "apply <- t in H". Here t is a term whose type
consists of nested dependent and nondependent products with an
equivalence A <-> B as the conclusion. The tactics with "->" in their
names apply A -> B while those with "<-" in the name apply B -> A. *)

(* The idea of the tactics is to first provide a term in the context
whose type is the implication (in one of the directions), and then
apply it. The first idea is to produce a statement "forall ..., A ->
B" (call this type T) and then do "assert (H : T)" for a fresh H.
Thus, T can be proved from the original equivalence and then used to
perform the application. However, currently in Ltac it is difficult
to produce such T from the original formula.

Therefore, we first pose the original equivalence as H. If the type of
H is a dependent product, we create an existential variable and apply
H to this variable. If the type of H has the form C -> D, then we do a
cut on C. Once we eliminate all products, we split (i.e., destruct)
the conjunction into two parts and apply the relevant one. *)

Ltac find_equiv H :=
let T := type of H in
lazymatch T with
| ?A -> ?B =>
  let H1 := fresh in
  let H2 := fresh in
  cut A;
  [intro H1; pose proof (H H1) as H2; clear H H1;
   rename H2 into H; find_equiv H |
   clear H]
| forall x : ?t, _ =>
  let a := fresh "a" with
      H1 := fresh "H" in
    evar (a : t); pose proof (H a) as H1; unfold a in H1;
    clear a; clear H; rename H1 into H; find_equiv H
| ?A <-> ?B => idtac
| _ => fail "The given statement does not seem to end with an equivalence."
end.

Ltac bapply lemma todo :=
let H := fresh in
  pose proof lemma as H;
  find_equiv H; [todo H; clear H | .. ].

Tactic Notation "apply" "->" constr(lemma) :=
bapply lemma ltac:(fun H => destruct H as [H _]; apply H).

Tactic Notation "apply" "<-" constr(lemma) :=
bapply lemma ltac:(fun H => destruct H as [_ H]; apply H).

Tactic Notation "apply" "->" constr(lemma) "in" hyp(J) :=
bapply lemma ltac:(fun H => destruct H as [H _]; apply H in J).

Tactic Notation "apply" "<-" constr(lemma) "in" hyp(J) :=
bapply lemma ltac:(fun H => destruct H as [_ H]; apply H in J).

(** An experimental tactic simpler than auto that is useful for ending
    proofs "in one step" *)

Ltac easy :=
  let rec use_hyp H :=
    match type of H with
    | _ /\ _ => exact H || destruct_hyp H
    | _ => try solve [inversion H]
    end
  with do_intro := let H := fresh in intro H; use_hyp H
  with destruct_hyp H := case H; clear H; do_intro; do_intro in
  let rec use_hyps :=
    match goal with
    | H : _ /\ _ |- _  => exact H || (destruct_hyp H; use_hyps)
    | H : _ |- _ => solve [inversion H]
    | _ => idtac
    end in
  let rec do_atom :=
    solve [reflexivity | symmetry; trivial] ||
    contradiction ||
    (split; do_atom)
  with do_ccl := trivial with eq_true; repeat do_intro; do_atom in
  (use_hyps; do_ccl) || fail "Cannot solve this goal".

Tactic Notation "now" tactic(t) := t; easy.

(** Slightly more than [easy]*)

Ltac easy' := repeat split; simpl; easy || now destruct 1.

(** A tactic to document or check what is proved at some point of a script *)

Ltac now_show c := change c.

(** Support for rewriting decidability statements *)

Set Implicit Arguments.

Lemma decide_left : forall (C:Prop) (decide:{C}+{~C}),
  C -> forall P:{C}+{~C}->Prop, (forall H:C, P (left _ H)) -> P decide.
Proof.
intros; destruct decide. apply H0. contradiction.
Qed.

Lemma decide_right : forall (C:Prop) (decide:{C}+{~C}),
  ~C -> forall P:{C}+{~C}->Prop, (forall H:~C, P (right _ H)) -> P decide.
Proof.
intros; destruct decide. contradiction. apply H0.
Qed.

Tactic Notation "decide" constr(lemma) "with" constr(H) :=
  let try_to_merge_hyps H :=
     try (clear H; intro H) ||
     (let H' := fresh H "bis" in intro H'; try clear H') ||
     (let H' := fresh in intro H'; try clear H') in
  match type of H with
  | ~ ?C => apply (decide_right lemma H); try_to_merge_hyps H
  | ?C -> False => apply (decide_right lemma H); try_to_merge_hyps H
  | _ => apply (decide_left lemma H); try_to_merge_hyps H
  end.

(** Clear an hypothesis and its dependencies *)

Tactic Notation "clear" "dependent" hyp(h) :=
 let rec depclear h :=
  clear h ||
  match goal with
   | H : context [ h ] |- _ => depclear H; depclear h
  end ||
  fail "hypothesis to clear is used in the conclusion (maybe indirectly)"
 in depclear h.

(** Revert an hypothesis and its dependencies :
    this is actually generalize dependent... *)

Tactic Notation "revert" "dependent" hyp(h) :=
 generalize dependent h.