nipkow@10213
|
1 |
(* Title: HOL/Product_Type.thy
|
nipkow@10213
|
2 |
Author: Lawrence C Paulson, Cambridge University Computer Laboratory
|
nipkow@10213
|
3 |
Copyright 1992 University of Cambridge
|
wenzelm@11777
|
4 |
*)
|
nipkow@10213
|
5 |
|
wenzelm@11838
|
6 |
header {* Cartesian products *}
|
nipkow@10213
|
7 |
|
nipkow@15131
|
8 |
theory Product_Type
|
haftmann@33958
|
9 |
imports Typedef Inductive Fun
|
haftmann@24699
|
10 |
uses
|
haftmann@24699
|
11 |
("Tools/split_rule.ML")
|
haftmann@37364
|
12 |
("Tools/inductive_codegen.ML")
|
haftmann@31723
|
13 |
("Tools/inductive_set.ML")
|
nipkow@15131
|
14 |
begin
|
nipkow@10213
|
15 |
|
haftmann@24699
|
16 |
subsection {* @{typ bool} is a datatype *}
|
haftmann@24699
|
17 |
|
haftmann@27104
|
18 |
rep_datatype True False by (auto intro: bool_induct)
|
haftmann@24699
|
19 |
|
haftmann@24699
|
20 |
declare case_split [cases type: bool]
|
haftmann@24699
|
21 |
-- "prefer plain propositional version"
|
haftmann@24699
|
22 |
|
haftmann@28346
|
23 |
lemma
|
haftmann@39086
|
24 |
shows [code]: "HOL.equal False P \<longleftrightarrow> \<not> P"
|
haftmann@39086
|
25 |
and [code]: "HOL.equal True P \<longleftrightarrow> P"
|
haftmann@39086
|
26 |
and [code]: "HOL.equal P False \<longleftrightarrow> \<not> P"
|
haftmann@39086
|
27 |
and [code]: "HOL.equal P True \<longleftrightarrow> P"
|
haftmann@39086
|
28 |
and [code nbe]: "HOL.equal P P \<longleftrightarrow> True"
|
haftmann@39086
|
29 |
by (simp_all add: equal)
|
haftmann@25534
|
30 |
|
haftmann@39086
|
31 |
code_const "HOL.equal \<Colon> bool \<Rightarrow> bool \<Rightarrow> bool"
|
haftmann@25534
|
32 |
(Haskell infixl 4 "==")
|
haftmann@25534
|
33 |
|
haftmann@39086
|
34 |
code_instance bool :: equal
|
haftmann@25534
|
35 |
(Haskell -)
|
haftmann@24699
|
36 |
|
haftmann@26358
|
37 |
|
haftmann@37160
|
38 |
subsection {* The @{text unit} type *}
|
nipkow@10213
|
39 |
|
wenzelm@11838
|
40 |
typedef unit = "{True}"
|
wenzelm@11838
|
41 |
proof
|
haftmann@20588
|
42 |
show "True : ?unit" ..
|
wenzelm@11838
|
43 |
qed
|
wenzelm@11838
|
44 |
|
haftmann@24699
|
45 |
definition
|
wenzelm@11838
|
46 |
Unity :: unit ("'(')")
|
haftmann@24699
|
47 |
where
|
haftmann@24699
|
48 |
"() = Abs_unit True"
|
wenzelm@11838
|
49 |
|
blanchet@35828
|
50 |
lemma unit_eq [no_atp]: "u = ()"
|
wenzelm@11838
|
51 |
by (induct u) (simp add: unit_def Unity_def)
|
wenzelm@11838
|
52 |
|
wenzelm@11838
|
53 |
text {*
|
wenzelm@11838
|
54 |
Simplification procedure for @{thm [source] unit_eq}. Cannot use
|
wenzelm@11838
|
55 |
this rule directly --- it loops!
|
wenzelm@11838
|
56 |
*}
|
wenzelm@11838
|
57 |
|
wenzelm@26480
|
58 |
ML {*
|
wenzelm@13462
|
59 |
val unit_eq_proc =
|
haftmann@24699
|
60 |
let val unit_meta_eq = mk_meta_eq @{thm unit_eq} in
|
wenzelm@38963
|
61 |
Simplifier.simproc_global @{theory} "unit_eq" ["x::unit"]
|
skalberg@15531
|
62 |
(fn _ => fn _ => fn t => if HOLogic.is_unit t then NONE else SOME unit_meta_eq)
|
wenzelm@13462
|
63 |
end;
|
wenzelm@11838
|
64 |
|
wenzelm@11838
|
65 |
Addsimprocs [unit_eq_proc];
|
wenzelm@11838
|
66 |
*}
|
wenzelm@11838
|
67 |
|
haftmann@27104
|
68 |
rep_datatype "()" by simp
|
haftmann@24699
|
69 |
|
wenzelm@11838
|
70 |
lemma unit_all_eq1: "(!!x::unit. PROP P x) == PROP P ()"
|
wenzelm@11838
|
71 |
by simp
|
wenzelm@11838
|
72 |
|
wenzelm@11838
|
73 |
lemma unit_all_eq2: "(!!x::unit. PROP P) == PROP P"
|
wenzelm@11838
|
74 |
by (rule triv_forall_equality)
|
wenzelm@11838
|
75 |
|
wenzelm@11838
|
76 |
text {*
|
wenzelm@11838
|
77 |
This rewrite counters the effect of @{text unit_eq_proc} on @{term
|
wenzelm@11838
|
78 |
[source] "%u::unit. f u"}, replacing it by @{term [source]
|
wenzelm@11838
|
79 |
f} rather than by @{term [source] "%u. f ()"}.
|
wenzelm@11838
|
80 |
*}
|
wenzelm@11838
|
81 |
|
blanchet@35828
|
82 |
lemma unit_abs_eta_conv [simp,no_atp]: "(%u::unit. f ()) = f"
|
wenzelm@11838
|
83 |
by (rule ext) simp
|
wenzelm@11838
|
84 |
|
haftmann@30924
|
85 |
instantiation unit :: default
|
haftmann@30924
|
86 |
begin
|
haftmann@30924
|
87 |
|
haftmann@30924
|
88 |
definition "default = ()"
|
haftmann@30924
|
89 |
|
haftmann@30924
|
90 |
instance ..
|
haftmann@30924
|
91 |
|
haftmann@30924
|
92 |
end
|
wenzelm@11838
|
93 |
|
haftmann@28562
|
94 |
lemma [code]:
|
haftmann@39086
|
95 |
"HOL.equal (u\<Colon>unit) v \<longleftrightarrow> True" unfolding equal unit_eq [of u] unit_eq [of v] by rule+
|
haftmann@26358
|
96 |
|
haftmann@26358
|
97 |
code_type unit
|
haftmann@26358
|
98 |
(SML "unit")
|
haftmann@26358
|
99 |
(OCaml "unit")
|
haftmann@26358
|
100 |
(Haskell "()")
|
haftmann@34886
|
101 |
(Scala "Unit")
|
haftmann@26358
|
102 |
|
haftmann@37160
|
103 |
code_const Unity
|
haftmann@37160
|
104 |
(SML "()")
|
haftmann@37160
|
105 |
(OCaml "()")
|
haftmann@37160
|
106 |
(Haskell "()")
|
haftmann@37160
|
107 |
(Scala "()")
|
haftmann@37160
|
108 |
|
haftmann@39086
|
109 |
code_instance unit :: equal
|
haftmann@26358
|
110 |
(Haskell -)
|
haftmann@26358
|
111 |
|
haftmann@39086
|
112 |
code_const "HOL.equal \<Colon> unit \<Rightarrow> unit \<Rightarrow> bool"
|
haftmann@26358
|
113 |
(Haskell infixl 4 "==")
|
haftmann@26358
|
114 |
|
haftmann@26358
|
115 |
code_reserved SML
|
haftmann@26358
|
116 |
unit
|
haftmann@26358
|
117 |
|
haftmann@26358
|
118 |
code_reserved OCaml
|
haftmann@26358
|
119 |
unit
|
haftmann@26358
|
120 |
|
haftmann@34886
|
121 |
code_reserved Scala
|
haftmann@34886
|
122 |
Unit
|
haftmann@34886
|
123 |
|
haftmann@26358
|
124 |
|
haftmann@37160
|
125 |
subsection {* The product type *}
|
nipkow@10213
|
126 |
|
haftmann@37160
|
127 |
subsubsection {* Type definition *}
|
nipkow@10213
|
128 |
|
haftmann@37160
|
129 |
definition Pair_Rep :: "'a \<Rightarrow> 'b \<Rightarrow> 'a \<Rightarrow> 'b \<Rightarrow> bool" where
|
haftmann@26358
|
130 |
"Pair_Rep a b = (\<lambda>x y. x = a \<and> y = b)"
|
nipkow@10213
|
131 |
|
haftmann@37678
|
132 |
typedef ('a, 'b) prod (infixr "*" 20)
|
haftmann@37364
|
133 |
= "{f. \<exists>a b. f = Pair_Rep (a\<Colon>'a) (b\<Colon>'b)}"
|
oheimb@11025
|
134 |
proof
|
haftmann@37364
|
135 |
fix a b show "Pair_Rep a b \<in> ?prod"
|
haftmann@26358
|
136 |
by rule+
|
oheimb@11025
|
137 |
qed
|
nipkow@10213
|
138 |
|
wenzelm@35537
|
139 |
type_notation (xsymbols)
|
haftmann@37678
|
140 |
"prod" ("(_ \<times>/ _)" [21, 20] 20)
|
wenzelm@35537
|
141 |
type_notation (HTML output)
|
haftmann@37678
|
142 |
"prod" ("(_ \<times>/ _)" [21, 20] 20)
|
nipkow@10213
|
143 |
|
haftmann@37364
|
144 |
definition Pair :: "'a \<Rightarrow> 'b \<Rightarrow> 'a \<times> 'b" where
|
haftmann@37364
|
145 |
"Pair a b = Abs_prod (Pair_Rep a b)"
|
haftmann@37160
|
146 |
|
haftmann@37678
|
147 |
rep_datatype Pair proof -
|
haftmann@37160
|
148 |
fix P :: "'a \<times> 'b \<Rightarrow> bool" and p
|
haftmann@37160
|
149 |
assume "\<And>a b. P (Pair a b)"
|
haftmann@37364
|
150 |
then show "P p" by (cases p) (auto simp add: prod_def Pair_def Pair_Rep_def)
|
haftmann@37160
|
151 |
next
|
haftmann@37160
|
152 |
fix a c :: 'a and b d :: 'b
|
haftmann@37160
|
153 |
have "Pair_Rep a b = Pair_Rep c d \<longleftrightarrow> a = c \<and> b = d"
|
haftmann@37160
|
154 |
by (auto simp add: Pair_Rep_def expand_fun_eq)
|
haftmann@37364
|
155 |
moreover have "Pair_Rep a b \<in> prod" and "Pair_Rep c d \<in> prod"
|
haftmann@37364
|
156 |
by (auto simp add: prod_def)
|
haftmann@37160
|
157 |
ultimately show "Pair a b = Pair c d \<longleftrightarrow> a = c \<and> b = d"
|
haftmann@37364
|
158 |
by (simp add: Pair_def Abs_prod_inject)
|
haftmann@37160
|
159 |
qed
|
haftmann@37160
|
160 |
|
blanchet@37695
|
161 |
declare prod.simps(2) [nitpick_simp del]
|
blanchet@37695
|
162 |
|
haftmann@37386
|
163 |
declare weak_case_cong [cong del]
|
haftmann@37386
|
164 |
|
haftmann@37160
|
165 |
|
haftmann@37160
|
166 |
subsubsection {* Tuple syntax *}
|
haftmann@37160
|
167 |
|
haftmann@37591
|
168 |
abbreviation (input) split :: "('a \<Rightarrow> 'b \<Rightarrow> 'c) \<Rightarrow> 'a \<times> 'b \<Rightarrow> 'c" where
|
haftmann@37591
|
169 |
"split \<equiv> prod_case"
|
wenzelm@19535
|
170 |
|
wenzelm@11777
|
171 |
text {*
|
wenzelm@11777
|
172 |
Patterns -- extends pre-defined type @{typ pttrn} used in
|
wenzelm@11777
|
173 |
abstractions.
|
wenzelm@11777
|
174 |
*}
|
nipkow@10213
|
175 |
|
nipkow@10213
|
176 |
nonterminals
|
nipkow@10213
|
177 |
tuple_args patterns
|
nipkow@10213
|
178 |
|
nipkow@10213
|
179 |
syntax
|
nipkow@10213
|
180 |
"_tuple" :: "'a => tuple_args => 'a * 'b" ("(1'(_,/ _'))")
|
nipkow@10213
|
181 |
"_tuple_arg" :: "'a => tuple_args" ("_")
|
nipkow@10213
|
182 |
"_tuple_args" :: "'a => tuple_args => tuple_args" ("_,/ _")
|
oheimb@11025
|
183 |
"_pattern" :: "[pttrn, patterns] => pttrn" ("'(_,/ _')")
|
oheimb@11025
|
184 |
"" :: "pttrn => patterns" ("_")
|
oheimb@11025
|
185 |
"_patterns" :: "[pttrn, patterns] => patterns" ("_,/ _")
|
nipkow@10213
|
186 |
|
nipkow@10213
|
187 |
translations
|
wenzelm@35118
|
188 |
"(x, y)" == "CONST Pair x y"
|
nipkow@10213
|
189 |
"_tuple x (_tuple_args y z)" == "_tuple x (_tuple_arg (_tuple y z))"
|
haftmann@37591
|
190 |
"%(x, y, zs). b" == "CONST prod_case (%x (y, zs). b)"
|
haftmann@37591
|
191 |
"%(x, y). b" == "CONST prod_case (%x y. b)"
|
wenzelm@35118
|
192 |
"_abs (CONST Pair x y) t" => "%(x, y). t"
|
haftmann@37160
|
193 |
-- {* The last rule accommodates tuples in `case C ... (x,y) ... => ...'
|
haftmann@37160
|
194 |
The (x,y) is parsed as `Pair x y' because it is logic, not pttrn *}
|
nipkow@10213
|
195 |
|
wenzelm@35118
|
196 |
(*reconstruct pattern from (nested) splits, avoiding eta-contraction of body;
|
wenzelm@35118
|
197 |
works best with enclosing "let", if "let" does not avoid eta-contraction*)
|
schirmer@14359
|
198 |
print_translation {*
|
wenzelm@35118
|
199 |
let
|
wenzelm@35118
|
200 |
fun split_tr' [Abs (x, T, t as (Abs abs))] =
|
wenzelm@35118
|
201 |
(* split (%x y. t) => %(x,y) t *)
|
wenzelm@35118
|
202 |
let
|
wenzelm@35118
|
203 |
val (y, t') = atomic_abs_tr' abs;
|
wenzelm@35118
|
204 |
val (x', t'') = atomic_abs_tr' (x, T, t');
|
wenzelm@35118
|
205 |
in
|
wenzelm@35118
|
206 |
Syntax.const @{syntax_const "_abs"} $
|
wenzelm@35118
|
207 |
(Syntax.const @{syntax_const "_pattern"} $ x' $ y) $ t''
|
wenzelm@35118
|
208 |
end
|
haftmann@37591
|
209 |
| split_tr' [Abs (x, T, (s as Const (@{const_syntax prod_case}, _) $ t))] =
|
wenzelm@35118
|
210 |
(* split (%x. (split (%y z. t))) => %(x,y,z). t *)
|
wenzelm@35118
|
211 |
let
|
wenzelm@35118
|
212 |
val Const (@{syntax_const "_abs"}, _) $
|
wenzelm@35118
|
213 |
(Const (@{syntax_const "_pattern"}, _) $ y $ z) $ t' = split_tr' [t];
|
wenzelm@35118
|
214 |
val (x', t'') = atomic_abs_tr' (x, T, t');
|
wenzelm@35118
|
215 |
in
|
wenzelm@35118
|
216 |
Syntax.const @{syntax_const "_abs"} $
|
wenzelm@35118
|
217 |
(Syntax.const @{syntax_const "_pattern"} $ x' $
|
wenzelm@35118
|
218 |
(Syntax.const @{syntax_const "_patterns"} $ y $ z)) $ t''
|
wenzelm@35118
|
219 |
end
|
haftmann@37591
|
220 |
| split_tr' [Const (@{const_syntax prod_case}, _) $ t] =
|
wenzelm@35118
|
221 |
(* split (split (%x y z. t)) => %((x, y), z). t *)
|
wenzelm@35118
|
222 |
split_tr' [(split_tr' [t])] (* inner split_tr' creates next pattern *)
|
wenzelm@35118
|
223 |
| split_tr' [Const (@{syntax_const "_abs"}, _) $ x_y $ Abs abs] =
|
wenzelm@35118
|
224 |
(* split (%pttrn z. t) => %(pttrn,z). t *)
|
wenzelm@35118
|
225 |
let val (z, t) = atomic_abs_tr' abs in
|
wenzelm@35118
|
226 |
Syntax.const @{syntax_const "_abs"} $
|
wenzelm@35118
|
227 |
(Syntax.const @{syntax_const "_pattern"} $ x_y $ z) $ t
|
wenzelm@35118
|
228 |
end
|
wenzelm@35118
|
229 |
| split_tr' _ = raise Match;
|
haftmann@37591
|
230 |
in [(@{const_syntax prod_case}, split_tr')] end
|
schirmer@14359
|
231 |
*}
|
schirmer@14359
|
232 |
|
schirmer@15422
|
233 |
(* print "split f" as "\<lambda>(x,y). f x y" and "split (\<lambda>x. f x)" as "\<lambda>(x,y). f x y" *)
|
schirmer@15422
|
234 |
typed_print_translation {*
|
schirmer@15422
|
235 |
let
|
wenzelm@35118
|
236 |
fun split_guess_names_tr' _ T [Abs (x, _, Abs _)] = raise Match
|
wenzelm@35118
|
237 |
| split_guess_names_tr' _ T [Abs (x, xT, t)] =
|
schirmer@15422
|
238 |
(case (head_of t) of
|
haftmann@37591
|
239 |
Const (@{const_syntax prod_case}, _) => raise Match
|
wenzelm@35118
|
240 |
| _ =>
|
wenzelm@35118
|
241 |
let
|
wenzelm@35118
|
242 |
val (_ :: yT :: _) = binder_types (domain_type T) handle Bind => raise Match;
|
wenzelm@35118
|
243 |
val (y, t') = atomic_abs_tr' ("y", yT, incr_boundvars 1 t $ Bound 0);
|
wenzelm@35118
|
244 |
val (x', t'') = atomic_abs_tr' (x, xT, t');
|
wenzelm@35118
|
245 |
in
|
wenzelm@35118
|
246 |
Syntax.const @{syntax_const "_abs"} $
|
wenzelm@35118
|
247 |
(Syntax.const @{syntax_const "_pattern"} $ x' $ y) $ t''
|
wenzelm@35118
|
248 |
end)
|
schirmer@15422
|
249 |
| split_guess_names_tr' _ T [t] =
|
wenzelm@35118
|
250 |
(case head_of t of
|
haftmann@37591
|
251 |
Const (@{const_syntax prod_case}, _) => raise Match
|
wenzelm@35118
|
252 |
| _ =>
|
wenzelm@35118
|
253 |
let
|
wenzelm@35118
|
254 |
val (xT :: yT :: _) = binder_types (domain_type T) handle Bind => raise Match;
|
wenzelm@35118
|
255 |
val (y, t') = atomic_abs_tr' ("y", yT, incr_boundvars 2 t $ Bound 1 $ Bound 0);
|
wenzelm@35118
|
256 |
val (x', t'') = atomic_abs_tr' ("x", xT, t');
|
wenzelm@35118
|
257 |
in
|
wenzelm@35118
|
258 |
Syntax.const @{syntax_const "_abs"} $
|
wenzelm@35118
|
259 |
(Syntax.const @{syntax_const "_pattern"} $ x' $ y) $ t''
|
wenzelm@35118
|
260 |
end)
|
schirmer@15422
|
261 |
| split_guess_names_tr' _ _ _ = raise Match;
|
haftmann@37591
|
262 |
in [(@{const_syntax prod_case}, split_guess_names_tr')] end
|
schirmer@15422
|
263 |
*}
|
schirmer@15422
|
264 |
|
nipkow@10213
|
265 |
|
haftmann@37160
|
266 |
subsubsection {* Code generator setup *}
|
haftmann@37160
|
267 |
|
haftmann@37678
|
268 |
code_type prod
|
haftmann@37160
|
269 |
(SML infix 2 "*")
|
haftmann@37160
|
270 |
(OCaml infix 2 "*")
|
haftmann@37160
|
271 |
(Haskell "!((_),/ (_))")
|
haftmann@37160
|
272 |
(Scala "((_),/ (_))")
|
haftmann@37160
|
273 |
|
haftmann@37160
|
274 |
code_const Pair
|
haftmann@37160
|
275 |
(SML "!((_),/ (_))")
|
haftmann@37160
|
276 |
(OCaml "!((_),/ (_))")
|
haftmann@37160
|
277 |
(Haskell "!((_),/ (_))")
|
haftmann@37160
|
278 |
(Scala "!((_),/ (_))")
|
haftmann@37160
|
279 |
|
haftmann@39086
|
280 |
code_instance prod :: equal
|
haftmann@37160
|
281 |
(Haskell -)
|
haftmann@37160
|
282 |
|
haftmann@39086
|
283 |
code_const "HOL.equal \<Colon> 'a \<times> 'b \<Rightarrow> 'a \<times> 'b \<Rightarrow> bool"
|
haftmann@37160
|
284 |
(Haskell infixl 4 "==")
|
haftmann@37160
|
285 |
|
haftmann@37160
|
286 |
types_code
|
haftmann@37678
|
287 |
"prod" ("(_ */ _)")
|
haftmann@37160
|
288 |
attach (term_of) {*
|
haftmann@37678
|
289 |
fun term_of_prod aF aT bF bT (x, y) = HOLogic.pair_const aT bT $ aF x $ bF y;
|
haftmann@37160
|
290 |
*}
|
haftmann@37160
|
291 |
attach (test) {*
|
bulwahn@37808
|
292 |
fun gen_prod aG aT bG bT i =
|
haftmann@37160
|
293 |
let
|
haftmann@37160
|
294 |
val (x, t) = aG i;
|
haftmann@37160
|
295 |
val (y, u) = bG i
|
haftmann@37160
|
296 |
in ((x, y), fn () => HOLogic.pair_const aT bT $ t () $ u ()) end;
|
haftmann@37160
|
297 |
*}
|
haftmann@37160
|
298 |
|
haftmann@37160
|
299 |
consts_code
|
haftmann@37160
|
300 |
"Pair" ("(_,/ _)")
|
haftmann@37160
|
301 |
|
haftmann@37160
|
302 |
setup {*
|
haftmann@37160
|
303 |
let
|
haftmann@37160
|
304 |
|
haftmann@37160
|
305 |
fun strip_abs_split 0 t = ([], t)
|
haftmann@37160
|
306 |
| strip_abs_split i (Abs (s, T, t)) =
|
haftmann@37160
|
307 |
let
|
haftmann@37160
|
308 |
val s' = Codegen.new_name t s;
|
haftmann@37160
|
309 |
val v = Free (s', T)
|
haftmann@37160
|
310 |
in apfst (cons v) (strip_abs_split (i-1) (subst_bound (v, t))) end
|
haftmann@37591
|
311 |
| strip_abs_split i (u as Const (@{const_name prod_case}, _) $ t) =
|
haftmann@37160
|
312 |
(case strip_abs_split (i+1) t of
|
haftmann@37160
|
313 |
(v :: v' :: vs, u) => (HOLogic.mk_prod (v, v') :: vs, u)
|
haftmann@37160
|
314 |
| _ => ([], u))
|
haftmann@37160
|
315 |
| strip_abs_split i t =
|
haftmann@37160
|
316 |
strip_abs_split i (Abs ("x", hd (binder_types (fastype_of t)), t $ Bound 0));
|
haftmann@37160
|
317 |
|
haftmann@37160
|
318 |
fun let_codegen thy defs dep thyname brack t gr =
|
haftmann@37160
|
319 |
(case strip_comb t of
|
haftmann@37160
|
320 |
(t1 as Const (@{const_name Let}, _), t2 :: t3 :: ts) =>
|
haftmann@37160
|
321 |
let
|
haftmann@37160
|
322 |
fun dest_let (l as Const (@{const_name Let}, _) $ t $ u) =
|
haftmann@37160
|
323 |
(case strip_abs_split 1 u of
|
haftmann@37160
|
324 |
([p], u') => apfst (cons (p, t)) (dest_let u')
|
haftmann@37160
|
325 |
| _ => ([], l))
|
haftmann@37160
|
326 |
| dest_let t = ([], t);
|
haftmann@37160
|
327 |
fun mk_code (l, r) gr =
|
haftmann@37160
|
328 |
let
|
haftmann@37160
|
329 |
val (pl, gr1) = Codegen.invoke_codegen thy defs dep thyname false l gr;
|
haftmann@37160
|
330 |
val (pr, gr2) = Codegen.invoke_codegen thy defs dep thyname false r gr1;
|
haftmann@37160
|
331 |
in ((pl, pr), gr2) end
|
haftmann@37160
|
332 |
in case dest_let (t1 $ t2 $ t3) of
|
haftmann@37160
|
333 |
([], _) => NONE
|
haftmann@37160
|
334 |
| (ps, u) =>
|
haftmann@37160
|
335 |
let
|
haftmann@37160
|
336 |
val (qs, gr1) = fold_map mk_code ps gr;
|
haftmann@37160
|
337 |
val (pu, gr2) = Codegen.invoke_codegen thy defs dep thyname false u gr1;
|
haftmann@37160
|
338 |
val (pargs, gr3) = fold_map
|
haftmann@37160
|
339 |
(Codegen.invoke_codegen thy defs dep thyname true) ts gr2
|
haftmann@37160
|
340 |
in
|
haftmann@37160
|
341 |
SOME (Codegen.mk_app brack
|
haftmann@37160
|
342 |
(Pretty.blk (0, [Codegen.str "let ", Pretty.blk (0, flat
|
haftmann@37160
|
343 |
(separate [Codegen.str ";", Pretty.brk 1] (map (fn (pl, pr) =>
|
haftmann@37160
|
344 |
[Pretty.block [Codegen.str "val ", pl, Codegen.str " =",
|
haftmann@37160
|
345 |
Pretty.brk 1, pr]]) qs))),
|
haftmann@37160
|
346 |
Pretty.brk 1, Codegen.str "in ", pu,
|
haftmann@37160
|
347 |
Pretty.brk 1, Codegen.str "end"])) pargs, gr3)
|
haftmann@37160
|
348 |
end
|
haftmann@37160
|
349 |
end
|
haftmann@37160
|
350 |
| _ => NONE);
|
haftmann@37160
|
351 |
|
haftmann@37160
|
352 |
fun split_codegen thy defs dep thyname brack t gr = (case strip_comb t of
|
haftmann@37591
|
353 |
(t1 as Const (@{const_name prod_case}, _), t2 :: ts) =>
|
haftmann@37160
|
354 |
let
|
haftmann@37160
|
355 |
val ([p], u) = strip_abs_split 1 (t1 $ t2);
|
haftmann@37160
|
356 |
val (q, gr1) = Codegen.invoke_codegen thy defs dep thyname false p gr;
|
haftmann@37160
|
357 |
val (pu, gr2) = Codegen.invoke_codegen thy defs dep thyname false u gr1;
|
haftmann@37160
|
358 |
val (pargs, gr3) = fold_map
|
haftmann@37160
|
359 |
(Codegen.invoke_codegen thy defs dep thyname true) ts gr2
|
haftmann@37160
|
360 |
in
|
haftmann@37160
|
361 |
SOME (Codegen.mk_app brack
|
haftmann@37160
|
362 |
(Pretty.block [Codegen.str "(fn ", q, Codegen.str " =>",
|
haftmann@37160
|
363 |
Pretty.brk 1, pu, Codegen.str ")"]) pargs, gr2)
|
haftmann@37160
|
364 |
end
|
haftmann@37160
|
365 |
| _ => NONE);
|
haftmann@37160
|
366 |
|
haftmann@37160
|
367 |
in
|
haftmann@37160
|
368 |
|
haftmann@37160
|
369 |
Codegen.add_codegen "let_codegen" let_codegen
|
haftmann@37160
|
370 |
#> Codegen.add_codegen "split_codegen" split_codegen
|
haftmann@37160
|
371 |
|
haftmann@37160
|
372 |
end
|
haftmann@37160
|
373 |
*}
|
haftmann@37160
|
374 |
|
haftmann@37160
|
375 |
|
haftmann@37160
|
376 |
subsubsection {* Fundamental operations and properties *}
|
nipkow@10213
|
377 |
|
haftmann@26358
|
378 |
lemma surj_pair [simp]: "EX x y. p = (x, y)"
|
haftmann@37160
|
379 |
by (cases p) simp
|
wenzelm@11838
|
380 |
|
haftmann@37364
|
381 |
definition fst :: "'a \<times> 'b \<Rightarrow> 'a" where
|
haftmann@37364
|
382 |
"fst p = (case p of (a, b) \<Rightarrow> a)"
|
haftmann@26358
|
383 |
|
haftmann@37364
|
384 |
definition snd :: "'a \<times> 'b \<Rightarrow> 'b" where
|
haftmann@37364
|
385 |
"snd p = (case p of (a, b) \<Rightarrow> b)"
|
oheimb@11025
|
386 |
|
haftmann@22886
|
387 |
lemma fst_conv [simp, code]: "fst (a, b) = a"
|
haftmann@37160
|
388 |
unfolding fst_def by simp
|
wenzelm@11032
|
389 |
|
haftmann@22886
|
390 |
lemma snd_conv [simp, code]: "snd (a, b) = b"
|
haftmann@37160
|
391 |
unfolding snd_def by simp
|
wenzelm@11032
|
392 |
|
haftmann@37160
|
393 |
code_const fst and snd
|
haftmann@37160
|
394 |
(Haskell "fst" and "snd")
|
haftmann@26358
|
395 |
|
blanchet@37695
|
396 |
lemma prod_case_unfold [nitpick_def]: "prod_case = (%c p. c (fst p) (snd p))"
|
haftmann@37160
|
397 |
by (simp add: expand_fun_eq split: prod.split)
|
haftmann@26358
|
398 |
|
wenzelm@11838
|
399 |
lemma fst_eqD: "fst (x, y) = a ==> x = a"
|
wenzelm@11838
|
400 |
by simp
|
oheimb@11025
|
401 |
|
wenzelm@11838
|
402 |
lemma snd_eqD: "snd (x, y) = a ==> y = a"
|
wenzelm@11838
|
403 |
by simp
|
wenzelm@11838
|
404 |
|
haftmann@26358
|
405 |
lemma pair_collapse [simp]: "(fst p, snd p) = p"
|
wenzelm@11838
|
406 |
by (cases p) simp
|
wenzelm@11838
|
407 |
|
haftmann@26358
|
408 |
lemmas surjective_pairing = pair_collapse [symmetric]
|
wenzelm@11838
|
409 |
|
haftmann@37160
|
410 |
lemma Pair_fst_snd_eq: "s = t \<longleftrightarrow> fst s = fst t \<and> snd s = snd t"
|
haftmann@37160
|
411 |
by (cases s, cases t) simp
|
haftmann@37160
|
412 |
|
haftmann@37160
|
413 |
lemma prod_eqI [intro?]: "fst p = fst q \<Longrightarrow> snd p = snd q \<Longrightarrow> p = q"
|
haftmann@37160
|
414 |
by (simp add: Pair_fst_snd_eq)
|
haftmann@37160
|
415 |
|
haftmann@37160
|
416 |
lemma split_conv [simp, code]: "split f (a, b) = f a b"
|
haftmann@37591
|
417 |
by (fact prod.cases)
|
haftmann@37160
|
418 |
|
haftmann@37160
|
419 |
lemma splitI: "f a b \<Longrightarrow> split f (a, b)"
|
haftmann@37160
|
420 |
by (rule split_conv [THEN iffD2])
|
haftmann@37160
|
421 |
|
haftmann@37160
|
422 |
lemma splitD: "split f (a, b) \<Longrightarrow> f a b"
|
haftmann@37160
|
423 |
by (rule split_conv [THEN iffD1])
|
haftmann@37160
|
424 |
|
haftmann@37160
|
425 |
lemma split_Pair [simp]: "(\<lambda>(x, y). (x, y)) = id"
|
haftmann@37591
|
426 |
by (simp add: expand_fun_eq split: prod.split)
|
haftmann@37160
|
427 |
|
haftmann@37160
|
428 |
lemma split_eta: "(\<lambda>(x, y). f (x, y)) = f"
|
haftmann@37160
|
429 |
-- {* Subsumes the old @{text split_Pair} when @{term f} is the identity function. *}
|
haftmann@37591
|
430 |
by (simp add: expand_fun_eq split: prod.split)
|
haftmann@37160
|
431 |
|
haftmann@37160
|
432 |
lemma split_comp: "split (f \<circ> g) x = f (g (fst x)) (snd x)"
|
haftmann@37160
|
433 |
by (cases x) simp
|
haftmann@37160
|
434 |
|
haftmann@37160
|
435 |
lemma split_twice: "split f (split g p) = split (\<lambda>x y. split f (g x y)) p"
|
haftmann@37160
|
436 |
by (cases p) simp
|
haftmann@37160
|
437 |
|
haftmann@37160
|
438 |
lemma The_split: "The (split P) = (THE xy. P (fst xy) (snd xy))"
|
haftmann@37591
|
439 |
by (simp add: prod_case_unfold)
|
haftmann@37160
|
440 |
|
haftmann@37160
|
441 |
lemma split_weak_cong: "p = q \<Longrightarrow> split c p = split c q"
|
haftmann@37160
|
442 |
-- {* Prevents simplification of @{term c}: much faster *}
|
haftmann@37678
|
443 |
by (fact weak_case_cong)
|
haftmann@37160
|
444 |
|
haftmann@37160
|
445 |
lemma cond_split_eta: "(!!x y. f x y = g (x, y)) ==> (%(x, y). f x y) = g"
|
haftmann@37160
|
446 |
by (simp add: split_eta)
|
haftmann@37160
|
447 |
|
wenzelm@11838
|
448 |
lemma split_paired_all: "(!!x. PROP P x) == (!!a b. PROP P (a, b))"
|
wenzelm@11820
|
449 |
proof
|
wenzelm@11820
|
450 |
fix a b
|
wenzelm@11820
|
451 |
assume "!!x. PROP P x"
|
wenzelm@19535
|
452 |
then show "PROP P (a, b)" .
|
wenzelm@11820
|
453 |
next
|
wenzelm@11820
|
454 |
fix x
|
wenzelm@11820
|
455 |
assume "!!a b. PROP P (a, b)"
|
wenzelm@19535
|
456 |
from `PROP P (fst x, snd x)` show "PROP P x" by simp
|
wenzelm@11820
|
457 |
qed
|
wenzelm@11820
|
458 |
|
wenzelm@11838
|
459 |
text {*
|
wenzelm@11838
|
460 |
The rule @{thm [source] split_paired_all} does not work with the
|
wenzelm@11838
|
461 |
Simplifier because it also affects premises in congrence rules,
|
wenzelm@11838
|
462 |
where this can lead to premises of the form @{text "!!a b. ... =
|
wenzelm@11838
|
463 |
?P(a, b)"} which cannot be solved by reflexivity.
|
wenzelm@11838
|
464 |
*}
|
wenzelm@11838
|
465 |
|
haftmann@26358
|
466 |
lemmas split_tupled_all = split_paired_all unit_all_eq2
|
haftmann@26358
|
467 |
|
wenzelm@26480
|
468 |
ML {*
|
wenzelm@11838
|
469 |
(* replace parameters of product type by individual component parameters *)
|
wenzelm@11838
|
470 |
val safe_full_simp_tac = generic_simp_tac true (true, false, false);
|
wenzelm@11838
|
471 |
local (* filtering with exists_paired_all is an essential optimization *)
|
wenzelm@16121
|
472 |
fun exists_paired_all (Const ("all", _) $ Abs (_, T, t)) =
|
wenzelm@11838
|
473 |
can HOLogic.dest_prodT T orelse exists_paired_all t
|
wenzelm@11838
|
474 |
| exists_paired_all (t $ u) = exists_paired_all t orelse exists_paired_all u
|
wenzelm@11838
|
475 |
| exists_paired_all (Abs (_, _, t)) = exists_paired_all t
|
wenzelm@11838
|
476 |
| exists_paired_all _ = false;
|
wenzelm@11838
|
477 |
val ss = HOL_basic_ss
|
wenzelm@26340
|
478 |
addsimps [@{thm split_paired_all}, @{thm unit_all_eq2}, @{thm unit_abs_eta_conv}]
|
wenzelm@11838
|
479 |
addsimprocs [unit_eq_proc];
|
wenzelm@11838
|
480 |
in
|
wenzelm@11838
|
481 |
val split_all_tac = SUBGOAL (fn (t, i) =>
|
wenzelm@11838
|
482 |
if exists_paired_all t then safe_full_simp_tac ss i else no_tac);
|
wenzelm@11838
|
483 |
val unsafe_split_all_tac = SUBGOAL (fn (t, i) =>
|
wenzelm@11838
|
484 |
if exists_paired_all t then full_simp_tac ss i else no_tac);
|
wenzelm@11838
|
485 |
fun split_all th =
|
wenzelm@26340
|
486 |
if exists_paired_all (Thm.prop_of th) then full_simplify ss th else th;
|
wenzelm@11838
|
487 |
end;
|
wenzelm@26340
|
488 |
*}
|
wenzelm@11838
|
489 |
|
wenzelm@26340
|
490 |
declaration {* fn _ =>
|
wenzelm@26340
|
491 |
Classical.map_cs (fn cs => cs addSbefore ("split_all_tac", split_all_tac))
|
wenzelm@16121
|
492 |
*}
|
wenzelm@11838
|
493 |
|
wenzelm@11838
|
494 |
lemma split_paired_All [simp]: "(ALL x. P x) = (ALL a b. P (a, b))"
|
wenzelm@11838
|
495 |
-- {* @{text "[iff]"} is not a good idea because it makes @{text blast} loop *}
|
wenzelm@11838
|
496 |
by fast
|
wenzelm@11838
|
497 |
|
haftmann@26358
|
498 |
lemma split_paired_Ex [simp]: "(EX x. P x) = (EX a b. P (a, b))"
|
haftmann@26358
|
499 |
by fast
|
haftmann@26358
|
500 |
|
wenzelm@11838
|
501 |
lemma split_paired_The: "(THE x. P x) = (THE (a, b). P (a, b))"
|
wenzelm@11838
|
502 |
-- {* Can't be added to simpset: loops! *}
|
haftmann@26358
|
503 |
by (simp add: split_eta)
|
wenzelm@11838
|
504 |
|
wenzelm@11838
|
505 |
text {*
|
wenzelm@11838
|
506 |
Simplification procedure for @{thm [source] cond_split_eta}. Using
|
wenzelm@11838
|
507 |
@{thm [source] split_eta} as a rewrite rule is not general enough,
|
wenzelm@11838
|
508 |
and using @{thm [source] cond_split_eta} directly would render some
|
wenzelm@11838
|
509 |
existing proofs very inefficient; similarly for @{text
|
haftmann@26358
|
510 |
split_beta}.
|
haftmann@26358
|
511 |
*}
|
wenzelm@11838
|
512 |
|
wenzelm@26480
|
513 |
ML {*
|
wenzelm@11838
|
514 |
local
|
wenzelm@35364
|
515 |
val cond_split_eta_ss = HOL_basic_ss addsimps @{thms cond_split_eta};
|
wenzelm@35364
|
516 |
fun Pair_pat k 0 (Bound m) = (m = k)
|
wenzelm@35364
|
517 |
| Pair_pat k i (Const (@{const_name Pair}, _) $ Bound m $ t) =
|
wenzelm@35364
|
518 |
i > 0 andalso m = k + i andalso Pair_pat k (i - 1) t
|
wenzelm@35364
|
519 |
| Pair_pat _ _ _ = false;
|
wenzelm@35364
|
520 |
fun no_args k i (Abs (_, _, t)) = no_args (k + 1) i t
|
wenzelm@35364
|
521 |
| no_args k i (t $ u) = no_args k i t andalso no_args k i u
|
wenzelm@35364
|
522 |
| no_args k i (Bound m) = m < k orelse m > k + i
|
wenzelm@35364
|
523 |
| no_args _ _ _ = true;
|
wenzelm@35364
|
524 |
fun split_pat tp i (Abs (_, _, t)) = if tp 0 i t then SOME (i, t) else NONE
|
haftmann@37591
|
525 |
| split_pat tp i (Const (@{const_name prod_case}, _) $ Abs (_, _, t)) = split_pat tp (i + 1) t
|
wenzelm@35364
|
526 |
| split_pat tp i _ = NONE;
|
wenzelm@20044
|
527 |
fun metaeq ss lhs rhs = mk_meta_eq (Goal.prove (Simplifier.the_context ss) [] []
|
wenzelm@35364
|
528 |
(HOLogic.mk_Trueprop (HOLogic.mk_eq (lhs, rhs)))
|
wenzelm@18328
|
529 |
(K (simp_tac (Simplifier.inherit_context ss cond_split_eta_ss) 1)));
|
wenzelm@11838
|
530 |
|
wenzelm@35364
|
531 |
fun beta_term_pat k i (Abs (_, _, t)) = beta_term_pat (k + 1) i t
|
wenzelm@35364
|
532 |
| beta_term_pat k i (t $ u) =
|
wenzelm@35364
|
533 |
Pair_pat k i (t $ u) orelse (beta_term_pat k i t andalso beta_term_pat k i u)
|
wenzelm@35364
|
534 |
| beta_term_pat k i t = no_args k i t;
|
wenzelm@35364
|
535 |
fun eta_term_pat k i (f $ arg) = no_args k i f andalso Pair_pat k i arg
|
wenzelm@35364
|
536 |
| eta_term_pat _ _ _ = false;
|
wenzelm@11838
|
537 |
fun subst arg k i (Abs (x, T, t)) = Abs (x, T, subst arg (k+1) i t)
|
wenzelm@35364
|
538 |
| subst arg k i (t $ u) =
|
wenzelm@35364
|
539 |
if Pair_pat k i (t $ u) then incr_boundvars k arg
|
wenzelm@35364
|
540 |
else (subst arg k i t $ subst arg k i u)
|
wenzelm@35364
|
541 |
| subst arg k i t = t;
|
haftmann@37591
|
542 |
fun beta_proc ss (s as Const (@{const_name prod_case}, _) $ Abs (_, _, t) $ arg) =
|
wenzelm@11838
|
543 |
(case split_pat beta_term_pat 1 t of
|
wenzelm@35364
|
544 |
SOME (i, f) => SOME (metaeq ss s (subst arg 0 i f))
|
skalberg@15531
|
545 |
| NONE => NONE)
|
wenzelm@35364
|
546 |
| beta_proc _ _ = NONE;
|
haftmann@37591
|
547 |
fun eta_proc ss (s as Const (@{const_name prod_case}, _) $ Abs (_, _, t)) =
|
wenzelm@11838
|
548 |
(case split_pat eta_term_pat 1 t of
|
wenzelm@35364
|
549 |
SOME (_, ft) => SOME (metaeq ss s (let val (f $ arg) = ft in f end))
|
skalberg@15531
|
550 |
| NONE => NONE)
|
wenzelm@35364
|
551 |
| eta_proc _ _ = NONE;
|
wenzelm@11838
|
552 |
in
|
wenzelm@38963
|
553 |
val split_beta_proc = Simplifier.simproc_global @{theory} "split_beta" ["split f z"] (K beta_proc);
|
wenzelm@38963
|
554 |
val split_eta_proc = Simplifier.simproc_global @{theory} "split_eta" ["split f"] (K eta_proc);
|
wenzelm@11838
|
555 |
end;
|
wenzelm@11838
|
556 |
|
wenzelm@11838
|
557 |
Addsimprocs [split_beta_proc, split_eta_proc];
|
wenzelm@11838
|
558 |
*}
|
wenzelm@11838
|
559 |
|
berghofe@26798
|
560 |
lemma split_beta [mono]: "(%(x, y). P x y) z = P (fst z) (snd z)"
|
wenzelm@11838
|
561 |
by (subst surjective_pairing, rule split_conv)
|
wenzelm@11838
|
562 |
|
blanchet@35828
|
563 |
lemma split_split [no_atp]: "R(split c p) = (ALL x y. p = (x, y) --> R(c x y))"
|
wenzelm@11838
|
564 |
-- {* For use with @{text split} and the Simplifier. *}
|
paulson@15481
|
565 |
by (insert surj_pair [of p], clarify, simp)
|
wenzelm@11838
|
566 |
|
wenzelm@11838
|
567 |
text {*
|
wenzelm@11838
|
568 |
@{thm [source] split_split} could be declared as @{text "[split]"}
|
wenzelm@11838
|
569 |
done after the Splitter has been speeded up significantly;
|
wenzelm@11838
|
570 |
precompute the constants involved and don't do anything unless the
|
wenzelm@11838
|
571 |
current goal contains one of those constants.
|
wenzelm@11838
|
572 |
*}
|
wenzelm@11838
|
573 |
|
blanchet@35828
|
574 |
lemma split_split_asm [no_atp]: "R (split c p) = (~(EX x y. p = (x, y) & (~R (c x y))))"
|
paulson@14208
|
575 |
by (subst split_split, simp)
|
wenzelm@11838
|
576 |
|
wenzelm@11838
|
577 |
text {*
|
wenzelm@11838
|
578 |
\medskip @{term split} used as a logical connective or set former.
|
wenzelm@11838
|
579 |
|
wenzelm@11838
|
580 |
\medskip These rules are for use with @{text blast}; could instead
|
wenzelm@11838
|
581 |
call @{text simp} using @{thm [source] split} as rewrite. *}
|
wenzelm@11838
|
582 |
|
wenzelm@11838
|
583 |
lemma splitI2: "!!p. [| !!a b. p = (a, b) ==> c a b |] ==> split c p"
|
wenzelm@11838
|
584 |
apply (simp only: split_tupled_all)
|
wenzelm@11838
|
585 |
apply (simp (no_asm_simp))
|
wenzelm@11838
|
586 |
done
|
wenzelm@11838
|
587 |
|
wenzelm@11838
|
588 |
lemma splitI2': "!!p. [| !!a b. (a, b) = p ==> c a b x |] ==> split c p x"
|
wenzelm@11838
|
589 |
apply (simp only: split_tupled_all)
|
wenzelm@11838
|
590 |
apply (simp (no_asm_simp))
|
wenzelm@11838
|
591 |
done
|
wenzelm@11838
|
592 |
|
wenzelm@11838
|
593 |
lemma splitE: "split c p ==> (!!x y. p = (x, y) ==> c x y ==> Q) ==> Q"
|
haftmann@37591
|
594 |
by (induct p) auto
|
wenzelm@11838
|
595 |
|
wenzelm@11838
|
596 |
lemma splitE': "split c p z ==> (!!x y. p = (x, y) ==> c x y z ==> Q) ==> Q"
|
haftmann@37591
|
597 |
by (induct p) auto
|
wenzelm@11838
|
598 |
|
wenzelm@11838
|
599 |
lemma splitE2:
|
wenzelm@11838
|
600 |
"[| Q (split P z); !!x y. [|z = (x, y); Q (P x y)|] ==> R |] ==> R"
|
wenzelm@11838
|
601 |
proof -
|
wenzelm@11838
|
602 |
assume q: "Q (split P z)"
|
wenzelm@11838
|
603 |
assume r: "!!x y. [|z = (x, y); Q (P x y)|] ==> R"
|
wenzelm@11838
|
604 |
show R
|
wenzelm@11838
|
605 |
apply (rule r surjective_pairing)+
|
wenzelm@11838
|
606 |
apply (rule split_beta [THEN subst], rule q)
|
wenzelm@11838
|
607 |
done
|
wenzelm@11838
|
608 |
qed
|
wenzelm@11838
|
609 |
|
wenzelm@11838
|
610 |
lemma splitD': "split R (a,b) c ==> R a b c"
|
wenzelm@11838
|
611 |
by simp
|
wenzelm@11838
|
612 |
|
wenzelm@11838
|
613 |
lemma mem_splitI: "z: c a b ==> z: split c (a, b)"
|
wenzelm@11838
|
614 |
by simp
|
wenzelm@11838
|
615 |
|
wenzelm@11838
|
616 |
lemma mem_splitI2: "!!p. [| !!a b. p = (a, b) ==> z: c a b |] ==> z: split c p"
|
paulson@14208
|
617 |
by (simp only: split_tupled_all, simp)
|
wenzelm@11838
|
618 |
|
wenzelm@18372
|
619 |
lemma mem_splitE:
|
haftmann@37160
|
620 |
assumes major: "z \<in> split c p"
|
haftmann@37160
|
621 |
and cases: "\<And>x y. p = (x, y) \<Longrightarrow> z \<in> c x y \<Longrightarrow> Q"
|
wenzelm@18372
|
622 |
shows Q
|
haftmann@37591
|
623 |
by (rule major [unfolded prod_case_unfold] cases surjective_pairing)+
|
wenzelm@11838
|
624 |
|
wenzelm@11838
|
625 |
declare mem_splitI2 [intro!] mem_splitI [intro!] splitI2' [intro!] splitI2 [intro!] splitI [intro!]
|
wenzelm@11838
|
626 |
declare mem_splitE [elim!] splitE' [elim!] splitE [elim!]
|
wenzelm@11838
|
627 |
|
wenzelm@26340
|
628 |
ML {*
|
wenzelm@11838
|
629 |
local (* filtering with exists_p_split is an essential optimization *)
|
haftmann@37591
|
630 |
fun exists_p_split (Const (@{const_name prod_case},_) $ _ $ (Const (@{const_name Pair},_)$_$_)) = true
|
wenzelm@11838
|
631 |
| exists_p_split (t $ u) = exists_p_split t orelse exists_p_split u
|
wenzelm@11838
|
632 |
| exists_p_split (Abs (_, _, t)) = exists_p_split t
|
wenzelm@11838
|
633 |
| exists_p_split _ = false;
|
wenzelm@35364
|
634 |
val ss = HOL_basic_ss addsimps @{thms split_conv};
|
wenzelm@11838
|
635 |
in
|
wenzelm@11838
|
636 |
val split_conv_tac = SUBGOAL (fn (t, i) =>
|
wenzelm@11838
|
637 |
if exists_p_split t then safe_full_simp_tac ss i else no_tac);
|
wenzelm@11838
|
638 |
end;
|
wenzelm@26340
|
639 |
*}
|
wenzelm@26340
|
640 |
|
wenzelm@11838
|
641 |
(* This prevents applications of splitE for already splitted arguments leading
|
wenzelm@11838
|
642 |
to quite time-consuming computations (in particular for nested tuples) *)
|
wenzelm@26340
|
643 |
declaration {* fn _ =>
|
wenzelm@26340
|
644 |
Classical.map_cs (fn cs => cs addSbefore ("split_conv_tac", split_conv_tac))
|
wenzelm@16121
|
645 |
*}
|
wenzelm@11838
|
646 |
|
blanchet@35828
|
647 |
lemma split_eta_SetCompr [simp,no_atp]: "(%u. EX x y. u = (x, y) & P (x, y)) = P"
|
wenzelm@18372
|
648 |
by (rule ext) fast
|
wenzelm@11838
|
649 |
|
blanchet@35828
|
650 |
lemma split_eta_SetCompr2 [simp,no_atp]: "(%u. EX x y. u = (x, y) & P x y) = split P"
|
wenzelm@18372
|
651 |
by (rule ext) fast
|
wenzelm@11838
|
652 |
|
wenzelm@11838
|
653 |
lemma split_part [simp]: "(%(a,b). P & Q a b) = (%ab. P & split Q ab)"
|
wenzelm@11838
|
654 |
-- {* Allows simplifications of nested splits in case of independent predicates. *}
|
wenzelm@18372
|
655 |
by (rule ext) blast
|
wenzelm@11838
|
656 |
|
nipkow@14337
|
657 |
(* Do NOT make this a simp rule as it
|
nipkow@14337
|
658 |
a) only helps in special situations
|
nipkow@14337
|
659 |
b) can lead to nontermination in the presence of split_def
|
nipkow@14337
|
660 |
*)
|
nipkow@14337
|
661 |
lemma split_comp_eq:
|
paulson@20415
|
662 |
fixes f :: "'a => 'b => 'c" and g :: "'d => 'a"
|
paulson@20415
|
663 |
shows "(%u. f (g (fst u)) (snd u)) = (split (%x. f (g x)))"
|
wenzelm@18372
|
664 |
by (rule ext) auto
|
oheimb@14101
|
665 |
|
haftmann@26358
|
666 |
lemma pair_imageI [intro]: "(a, b) : A ==> f a b : (%(a, b). f a b) ` A"
|
haftmann@26358
|
667 |
apply (rule_tac x = "(a, b)" in image_eqI)
|
haftmann@26358
|
668 |
apply auto
|
haftmann@26358
|
669 |
done
|
haftmann@26358
|
670 |
|
wenzelm@11838
|
671 |
lemma The_split_eq [simp]: "(THE (x',y'). x = x' & y = y') = (x, y)"
|
wenzelm@11838
|
672 |
by blast
|
wenzelm@11838
|
673 |
|
wenzelm@11838
|
674 |
(*
|
wenzelm@11838
|
675 |
the following would be slightly more general,
|
wenzelm@11838
|
676 |
but cannot be used as rewrite rule:
|
wenzelm@11838
|
677 |
### Cannot add premise as rewrite rule because it contains (type) unknowns:
|
wenzelm@11838
|
678 |
### ?y = .x
|
wenzelm@11838
|
679 |
Goal "[| P y; !!x. P x ==> x = y |] ==> (@(x',y). x = x' & P y) = (x,y)"
|
paulson@14208
|
680 |
by (rtac some_equality 1)
|
paulson@14208
|
681 |
by ( Simp_tac 1)
|
paulson@14208
|
682 |
by (split_all_tac 1)
|
paulson@14208
|
683 |
by (Asm_full_simp_tac 1)
|
wenzelm@11838
|
684 |
qed "The_split_eq";
|
wenzelm@11838
|
685 |
*)
|
wenzelm@11838
|
686 |
|
haftmann@26358
|
687 |
text {*
|
haftmann@26358
|
688 |
Setup of internal @{text split_rule}.
|
haftmann@26358
|
689 |
*}
|
haftmann@26358
|
690 |
|
haftmann@26358
|
691 |
lemmas prod_caseI = prod.cases [THEN iffD2, standard]
|
haftmann@26358
|
692 |
|
haftmann@26358
|
693 |
lemma prod_caseI2: "!!p. [| !!a b. p = (a, b) ==> c a b |] ==> prod_case c p"
|
haftmann@37678
|
694 |
by (fact splitI2)
|
wenzelm@11838
|
695 |
|
haftmann@26358
|
696 |
lemma prod_caseI2': "!!p. [| !!a b. (a, b) = p ==> c a b x |] ==> prod_case c p x"
|
haftmann@37678
|
697 |
by (fact splitI2')
|
haftmann@26358
|
698 |
|
haftmann@26358
|
699 |
lemma prod_caseE: "prod_case c p ==> (!!x y. p = (x, y) ==> c x y ==> Q) ==> Q"
|
haftmann@37678
|
700 |
by (fact splitE)
|
haftmann@26358
|
701 |
|
haftmann@26358
|
702 |
lemma prod_caseE': "prod_case c p z ==> (!!x y. p = (x, y) ==> c x y z ==> Q) ==> Q"
|
haftmann@37678
|
703 |
by (fact splitE')
|
haftmann@26358
|
704 |
|
haftmann@37678
|
705 |
declare prod_caseI [intro!]
|
haftmann@26358
|
706 |
|
haftmann@26358
|
707 |
lemma prod_case_beta:
|
haftmann@26358
|
708 |
"prod_case f p = f (fst p) (snd p)"
|
haftmann@37591
|
709 |
by (fact split_beta)
|
haftmann@26358
|
710 |
|
haftmann@26358
|
711 |
lemma prod_cases3 [cases type]:
|
haftmann@26358
|
712 |
obtains (fields) a b c where "y = (a, b, c)"
|
haftmann@26358
|
713 |
by (cases y, case_tac b) blast
|
haftmann@26358
|
714 |
|
haftmann@26358
|
715 |
lemma prod_induct3 [case_names fields, induct type]:
|
haftmann@26358
|
716 |
"(!!a b c. P (a, b, c)) ==> P x"
|
haftmann@26358
|
717 |
by (cases x) blast
|
haftmann@26358
|
718 |
|
haftmann@26358
|
719 |
lemma prod_cases4 [cases type]:
|
haftmann@26358
|
720 |
obtains (fields) a b c d where "y = (a, b, c, d)"
|
haftmann@26358
|
721 |
by (cases y, case_tac c) blast
|
haftmann@26358
|
722 |
|
haftmann@26358
|
723 |
lemma prod_induct4 [case_names fields, induct type]:
|
haftmann@26358
|
724 |
"(!!a b c d. P (a, b, c, d)) ==> P x"
|
haftmann@26358
|
725 |
by (cases x) blast
|
haftmann@26358
|
726 |
|
haftmann@26358
|
727 |
lemma prod_cases5 [cases type]:
|
haftmann@26358
|
728 |
obtains (fields) a b c d e where "y = (a, b, c, d, e)"
|
haftmann@26358
|
729 |
by (cases y, case_tac d) blast
|
haftmann@26358
|
730 |
|
haftmann@26358
|
731 |
lemma prod_induct5 [case_names fields, induct type]:
|
haftmann@26358
|
732 |
"(!!a b c d e. P (a, b, c, d, e)) ==> P x"
|
haftmann@26358
|
733 |
by (cases x) blast
|
haftmann@26358
|
734 |
|
haftmann@26358
|
735 |
lemma prod_cases6 [cases type]:
|
haftmann@26358
|
736 |
obtains (fields) a b c d e f where "y = (a, b, c, d, e, f)"
|
haftmann@26358
|
737 |
by (cases y, case_tac e) blast
|
haftmann@26358
|
738 |
|
haftmann@26358
|
739 |
lemma prod_induct6 [case_names fields, induct type]:
|
haftmann@26358
|
740 |
"(!!a b c d e f. P (a, b, c, d, e, f)) ==> P x"
|
haftmann@26358
|
741 |
by (cases x) blast
|
haftmann@26358
|
742 |
|
haftmann@26358
|
743 |
lemma prod_cases7 [cases type]:
|
haftmann@26358
|
744 |
obtains (fields) a b c d e f g where "y = (a, b, c, d, e, f, g)"
|
haftmann@26358
|
745 |
by (cases y, case_tac f) blast
|
haftmann@26358
|
746 |
|
haftmann@26358
|
747 |
lemma prod_induct7 [case_names fields, induct type]:
|
haftmann@26358
|
748 |
"(!!a b c d e f g. P (a, b, c, d, e, f, g)) ==> P x"
|
haftmann@26358
|
749 |
by (cases x) blast
|
haftmann@26358
|
750 |
|
haftmann@37160
|
751 |
lemma split_def:
|
haftmann@37160
|
752 |
"split = (\<lambda>c p. c (fst p) (snd p))"
|
haftmann@37591
|
753 |
by (fact prod_case_unfold)
|
haftmann@37160
|
754 |
|
haftmann@37160
|
755 |
definition internal_split :: "('a \<Rightarrow> 'b \<Rightarrow> 'c) \<Rightarrow> 'a \<times> 'b \<Rightarrow> 'c" where
|
haftmann@37160
|
756 |
"internal_split == split"
|
haftmann@37160
|
757 |
|
haftmann@37160
|
758 |
lemma internal_split_conv: "internal_split c (a, b) = c a b"
|
haftmann@37160
|
759 |
by (simp only: internal_split_def split_conv)
|
haftmann@37160
|
760 |
|
haftmann@37160
|
761 |
use "Tools/split_rule.ML"
|
haftmann@37160
|
762 |
setup Split_Rule.setup
|
haftmann@37160
|
763 |
|
haftmann@37160
|
764 |
hide_const internal_split
|
haftmann@37160
|
765 |
|
haftmann@26358
|
766 |
|
haftmann@26358
|
767 |
subsubsection {* Derived operations *}
|
wenzelm@11838
|
768 |
|
haftmann@37362
|
769 |
definition curry :: "('a \<times> 'b \<Rightarrow> 'c) \<Rightarrow> 'a \<Rightarrow> 'b \<Rightarrow> 'c" where
|
haftmann@37362
|
770 |
"curry = (\<lambda>c x y. c (x, y))"
|
haftmann@37160
|
771 |
|
haftmann@37160
|
772 |
lemma curry_conv [simp, code]: "curry f a b = f (a, b)"
|
haftmann@37160
|
773 |
by (simp add: curry_def)
|
haftmann@37160
|
774 |
|
haftmann@37160
|
775 |
lemma curryI [intro!]: "f (a, b) \<Longrightarrow> curry f a b"
|
haftmann@37160
|
776 |
by (simp add: curry_def)
|
haftmann@37160
|
777 |
|
haftmann@37160
|
778 |
lemma curryD [dest!]: "curry f a b \<Longrightarrow> f (a, b)"
|
haftmann@37160
|
779 |
by (simp add: curry_def)
|
haftmann@37160
|
780 |
|
haftmann@37160
|
781 |
lemma curryE: "curry f a b \<Longrightarrow> (f (a, b) \<Longrightarrow> Q) \<Longrightarrow> Q"
|
haftmann@37160
|
782 |
by (simp add: curry_def)
|
haftmann@37160
|
783 |
|
haftmann@37160
|
784 |
lemma curry_split [simp]: "curry (split f) = f"
|
haftmann@37160
|
785 |
by (simp add: curry_def split_def)
|
haftmann@37160
|
786 |
|
haftmann@37160
|
787 |
lemma split_curry [simp]: "split (curry f) = f"
|
haftmann@37160
|
788 |
by (simp add: curry_def split_def)
|
haftmann@37160
|
789 |
|
wenzelm@11838
|
790 |
text {*
|
haftmann@26358
|
791 |
The composition-uncurry combinator.
|
haftmann@26358
|
792 |
*}
|
haftmann@26358
|
793 |
|
haftmann@37750
|
794 |
notation fcomp (infixl "\<circ>>" 60)
|
haftmann@26588
|
795 |
|
haftmann@37750
|
796 |
definition scomp :: "('a \<Rightarrow> 'b \<times> 'c) \<Rightarrow> ('b \<Rightarrow> 'c \<Rightarrow> 'd) \<Rightarrow> 'a \<Rightarrow> 'd" (infixl "\<circ>\<rightarrow>" 60) where
|
haftmann@37750
|
797 |
"f \<circ>\<rightarrow> g = (\<lambda>x. prod_case g (f x))"
|
haftmann@26358
|
798 |
|
haftmann@37678
|
799 |
lemma scomp_unfold: "scomp = (\<lambda>f g x. g (fst (f x)) (snd (f x)))"
|
haftmann@37750
|
800 |
by (simp add: expand_fun_eq scomp_def prod_case_unfold)
|
haftmann@37678
|
801 |
|
haftmann@37750
|
802 |
lemma scomp_apply [simp]: "(f \<circ>\<rightarrow> g) x = prod_case g (f x)"
|
haftmann@37750
|
803 |
by (simp add: scomp_unfold prod_case_unfold)
|
haftmann@26358
|
804 |
|
haftmann@37750
|
805 |
lemma Pair_scomp: "Pair x \<circ>\<rightarrow> f = f x"
|
haftmann@26588
|
806 |
by (simp add: expand_fun_eq scomp_apply)
|
haftmann@26358
|
807 |
|
haftmann@37750
|
808 |
lemma scomp_Pair: "x \<circ>\<rightarrow> Pair = x"
|
haftmann@26588
|
809 |
by (simp add: expand_fun_eq scomp_apply)
|
haftmann@26358
|
810 |
|
haftmann@37750
|
811 |
lemma scomp_scomp: "(f \<circ>\<rightarrow> g) \<circ>\<rightarrow> h = f \<circ>\<rightarrow> (\<lambda>x. g x \<circ>\<rightarrow> h)"
|
haftmann@37678
|
812 |
by (simp add: expand_fun_eq scomp_unfold)
|
haftmann@26358
|
813 |
|
haftmann@37750
|
814 |
lemma scomp_fcomp: "(f \<circ>\<rightarrow> g) \<circ>> h = f \<circ>\<rightarrow> (\<lambda>x. g x \<circ>> h)"
|
haftmann@37678
|
815 |
by (simp add: expand_fun_eq scomp_unfold fcomp_def)
|
haftmann@26358
|
816 |
|
haftmann@37750
|
817 |
lemma fcomp_scomp: "(f \<circ>> g) \<circ>\<rightarrow> h = f \<circ>> (g \<circ>\<rightarrow> h)"
|
haftmann@37678
|
818 |
by (simp add: expand_fun_eq scomp_unfold fcomp_apply)
|
haftmann@26358
|
819 |
|
haftmann@31202
|
820 |
code_const scomp
|
haftmann@31202
|
821 |
(Eval infixl 3 "#->")
|
haftmann@31202
|
822 |
|
haftmann@37750
|
823 |
no_notation fcomp (infixl "\<circ>>" 60)
|
haftmann@37750
|
824 |
no_notation scomp (infixl "\<circ>\<rightarrow>" 60)
|
haftmann@26358
|
825 |
|
haftmann@26358
|
826 |
text {*
|
haftmann@26358
|
827 |
@{term prod_fun} --- action of the product functor upon
|
krauss@36664
|
828 |
functions.
|
wenzelm@11838
|
829 |
*}
|
wenzelm@11838
|
830 |
|
haftmann@26358
|
831 |
definition prod_fun :: "('a \<Rightarrow> 'c) \<Rightarrow> ('b \<Rightarrow> 'd) \<Rightarrow> 'a \<times> 'b \<Rightarrow> 'c \<times> 'd" where
|
haftmann@37765
|
832 |
"prod_fun f g = (\<lambda>(x, y). (f x, g y))"
|
haftmann@26358
|
833 |
|
haftmann@28562
|
834 |
lemma prod_fun [simp, code]: "prod_fun f g (a, b) = (f a, g b)"
|
wenzelm@11838
|
835 |
by (simp add: prod_fun_def)
|
wenzelm@11838
|
836 |
|
nipkow@37277
|
837 |
lemma fst_prod_fun[simp]: "fst (prod_fun f g x) = f (fst x)"
|
nipkow@37277
|
838 |
by (cases x, auto)
|
nipkow@37277
|
839 |
|
nipkow@37277
|
840 |
lemma snd_prod_fun[simp]: "snd (prod_fun f g x) = g (snd x)"
|
nipkow@37277
|
841 |
by (cases x, auto)
|
nipkow@37277
|
842 |
|
nipkow@37277
|
843 |
lemma fst_comp_prod_fun[simp]: "fst \<circ> prod_fun f g = f \<circ> fst"
|
nipkow@37277
|
844 |
by (rule ext) auto
|
nipkow@37277
|
845 |
|
nipkow@37277
|
846 |
lemma snd_comp_prod_fun[simp]: "snd \<circ> prod_fun f g = g \<circ> snd"
|
nipkow@37277
|
847 |
by (rule ext) auto
|
nipkow@37277
|
848 |
|
nipkow@37277
|
849 |
|
nipkow@37277
|
850 |
lemma prod_fun_compose:
|
nipkow@37277
|
851 |
"prod_fun (f1 o f2) (g1 o g2) = (prod_fun f1 g1 o prod_fun f2 g2)"
|
nipkow@37277
|
852 |
by (rule ext) auto
|
wenzelm@11838
|
853 |
|
wenzelm@11838
|
854 |
lemma prod_fun_ident [simp]: "prod_fun (%x. x) (%y. y) = (%z. z)"
|
haftmann@26358
|
855 |
by (rule ext) auto
|
wenzelm@11838
|
856 |
|
wenzelm@11838
|
857 |
lemma prod_fun_imageI [intro]: "(a, b) : r ==> (f a, g b) : prod_fun f g ` r"
|
wenzelm@11838
|
858 |
apply (rule image_eqI)
|
paulson@14208
|
859 |
apply (rule prod_fun [symmetric], assumption)
|
wenzelm@11838
|
860 |
done
|
wenzelm@11838
|
861 |
|
wenzelm@11838
|
862 |
lemma prod_fun_imageE [elim!]:
|
wenzelm@18372
|
863 |
assumes major: "c: (prod_fun f g)`r"
|
wenzelm@18372
|
864 |
and cases: "!!x y. [| c=(f(x),g(y)); (x,y):r |] ==> P"
|
wenzelm@18372
|
865 |
shows P
|
wenzelm@18372
|
866 |
apply (rule major [THEN imageE])
|
haftmann@37160
|
867 |
apply (case_tac x)
|
wenzelm@18372
|
868 |
apply (rule cases)
|
wenzelm@18372
|
869 |
apply (blast intro: prod_fun)
|
wenzelm@18372
|
870 |
apply blast
|
wenzelm@18372
|
871 |
done
|
wenzelm@11838
|
872 |
|
nipkow@37277
|
873 |
|
haftmann@37160
|
874 |
definition apfst :: "('a \<Rightarrow> 'c) \<Rightarrow> 'a \<times> 'b \<Rightarrow> 'c \<times> 'b" where
|
haftmann@37160
|
875 |
"apfst f = prod_fun f id"
|
wenzelm@11838
|
876 |
|
haftmann@37160
|
877 |
definition apsnd :: "('b \<Rightarrow> 'c) \<Rightarrow> 'a \<times> 'b \<Rightarrow> 'a \<times> 'c" where
|
haftmann@37160
|
878 |
"apsnd f = prod_fun id f"
|
oheimb@14101
|
879 |
|
haftmann@26358
|
880 |
lemma apfst_conv [simp, code]:
|
haftmann@26358
|
881 |
"apfst f (x, y) = (f x, y)"
|
haftmann@26358
|
882 |
by (simp add: apfst_def)
|
oheimb@14101
|
883 |
|
hoelzl@33638
|
884 |
lemma apsnd_conv [simp, code]:
|
haftmann@26358
|
885 |
"apsnd f (x, y) = (x, f y)"
|
haftmann@26358
|
886 |
by (simp add: apsnd_def)
|
haftmann@26358
|
887 |
|
haftmann@33585
|
888 |
lemma fst_apfst [simp]:
|
haftmann@33585
|
889 |
"fst (apfst f x) = f (fst x)"
|
haftmann@33585
|
890 |
by (cases x) simp
|
haftmann@33585
|
891 |
|
haftmann@33585
|
892 |
lemma fst_apsnd [simp]:
|
haftmann@33585
|
893 |
"fst (apsnd f x) = fst x"
|
haftmann@33585
|
894 |
by (cases x) simp
|
haftmann@33585
|
895 |
|
haftmann@33585
|
896 |
lemma snd_apfst [simp]:
|
haftmann@33585
|
897 |
"snd (apfst f x) = snd x"
|
haftmann@33585
|
898 |
by (cases x) simp
|
haftmann@33585
|
899 |
|
haftmann@33585
|
900 |
lemma snd_apsnd [simp]:
|
haftmann@33585
|
901 |
"snd (apsnd f x) = f (snd x)"
|
haftmann@33585
|
902 |
by (cases x) simp
|
haftmann@33585
|
903 |
|
haftmann@33585
|
904 |
lemma apfst_compose:
|
haftmann@33585
|
905 |
"apfst f (apfst g x) = apfst (f \<circ> g) x"
|
haftmann@33585
|
906 |
by (cases x) simp
|
haftmann@33585
|
907 |
|
haftmann@33585
|
908 |
lemma apsnd_compose:
|
haftmann@33585
|
909 |
"apsnd f (apsnd g x) = apsnd (f \<circ> g) x"
|
haftmann@33585
|
910 |
by (cases x) simp
|
haftmann@33585
|
911 |
|
haftmann@33585
|
912 |
lemma apfst_apsnd [simp]:
|
haftmann@33585
|
913 |
"apfst f (apsnd g x) = (f (fst x), g (snd x))"
|
haftmann@33585
|
914 |
by (cases x) simp
|
haftmann@33585
|
915 |
|
haftmann@33585
|
916 |
lemma apsnd_apfst [simp]:
|
haftmann@33585
|
917 |
"apsnd f (apfst g x) = (g (fst x), f (snd x))"
|
haftmann@33585
|
918 |
by (cases x) simp
|
haftmann@33585
|
919 |
|
haftmann@33585
|
920 |
lemma apfst_id [simp] :
|
haftmann@33585
|
921 |
"apfst id = id"
|
haftmann@33585
|
922 |
by (simp add: expand_fun_eq)
|
haftmann@33585
|
923 |
|
haftmann@33585
|
924 |
lemma apsnd_id [simp] :
|
haftmann@33585
|
925 |
"apsnd id = id"
|
haftmann@33585
|
926 |
by (simp add: expand_fun_eq)
|
haftmann@33585
|
927 |
|
haftmann@33585
|
928 |
lemma apfst_eq_conv [simp]:
|
haftmann@33585
|
929 |
"apfst f x = apfst g x \<longleftrightarrow> f (fst x) = g (fst x)"
|
haftmann@33585
|
930 |
by (cases x) simp
|
haftmann@33585
|
931 |
|
haftmann@33585
|
932 |
lemma apsnd_eq_conv [simp]:
|
haftmann@33585
|
933 |
"apsnd f x = apsnd g x \<longleftrightarrow> f (snd x) = g (snd x)"
|
haftmann@33585
|
934 |
by (cases x) simp
|
haftmann@33585
|
935 |
|
hoelzl@33638
|
936 |
lemma apsnd_apfst_commute:
|
hoelzl@33638
|
937 |
"apsnd f (apfst g p) = apfst g (apsnd f p)"
|
hoelzl@33638
|
938 |
by simp
|
oheimb@14101
|
939 |
|
wenzelm@11838
|
940 |
text {*
|
haftmann@26358
|
941 |
Disjoint union of a family of sets -- Sigma.
|
wenzelm@11838
|
942 |
*}
|
wenzelm@11838
|
943 |
|
haftmann@26358
|
944 |
definition Sigma :: "['a set, 'a => 'b set] => ('a \<times> 'b) set" where
|
haftmann@26358
|
945 |
Sigma_def: "Sigma A B == UN x:A. UN y:B x. {Pair x y}"
|
haftmann@26358
|
946 |
|
haftmann@26358
|
947 |
abbreviation
|
haftmann@26358
|
948 |
Times :: "['a set, 'b set] => ('a * 'b) set"
|
haftmann@26358
|
949 |
(infixr "<*>" 80) where
|
haftmann@26358
|
950 |
"A <*> B == Sigma A (%_. B)"
|
haftmann@26358
|
951 |
|
haftmann@26358
|
952 |
notation (xsymbols)
|
haftmann@26358
|
953 |
Times (infixr "\<times>" 80)
|
haftmann@26358
|
954 |
|
haftmann@26358
|
955 |
notation (HTML output)
|
haftmann@26358
|
956 |
Times (infixr "\<times>" 80)
|
haftmann@26358
|
957 |
|
haftmann@26358
|
958 |
syntax
|
wenzelm@35118
|
959 |
"_Sigma" :: "[pttrn, 'a set, 'b set] => ('a * 'b) set" ("(3SIGMA _:_./ _)" [0, 0, 10] 10)
|
haftmann@26358
|
960 |
translations
|
wenzelm@35118
|
961 |
"SIGMA x:A. B" == "CONST Sigma A (%x. B)"
|
haftmann@26358
|
962 |
|
wenzelm@11838
|
963 |
lemma SigmaI [intro!]: "[| a:A; b:B(a) |] ==> (a,b) : Sigma A B"
|
wenzelm@11838
|
964 |
by (unfold Sigma_def) blast
|
wenzelm@11838
|
965 |
|
paulson@14952
|
966 |
lemma SigmaE [elim!]:
|
wenzelm@11838
|
967 |
"[| c: Sigma A B;
|
wenzelm@11838
|
968 |
!!x y.[| x:A; y:B(x); c=(x,y) |] ==> P
|
wenzelm@11838
|
969 |
|] ==> P"
|
wenzelm@11838
|
970 |
-- {* The general elimination rule. *}
|
wenzelm@11838
|
971 |
by (unfold Sigma_def) blast
|
wenzelm@11838
|
972 |
|
wenzelm@11838
|
973 |
text {*
|
schirmer@15422
|
974 |
Elimination of @{term "(a, b) : A \<times> B"} -- introduces no
|
wenzelm@11838
|
975 |
eigenvariables.
|
wenzelm@11838
|
976 |
*}
|
wenzelm@11838
|
977 |
|
wenzelm@11838
|
978 |
lemma SigmaD1: "(a, b) : Sigma A B ==> a : A"
|
wenzelm@18372
|
979 |
by blast
|
wenzelm@11838
|
980 |
|
wenzelm@11838
|
981 |
lemma SigmaD2: "(a, b) : Sigma A B ==> b : B a"
|
wenzelm@18372
|
982 |
by blast
|
wenzelm@11838
|
983 |
|
wenzelm@11838
|
984 |
lemma SigmaE2:
|
wenzelm@11838
|
985 |
"[| (a, b) : Sigma A B;
|
wenzelm@11838
|
986 |
[| a:A; b:B(a) |] ==> P
|
wenzelm@11838
|
987 |
|] ==> P"
|
paulson@14952
|
988 |
by blast
|
wenzelm@11838
|
989 |
|
paulson@14952
|
990 |
lemma Sigma_cong:
|
schirmer@15422
|
991 |
"\<lbrakk>A = B; !!x. x \<in> B \<Longrightarrow> C x = D x\<rbrakk>
|
schirmer@15422
|
992 |
\<Longrightarrow> (SIGMA x: A. C x) = (SIGMA x: B. D x)"
|
wenzelm@18372
|
993 |
by auto
|
wenzelm@11838
|
994 |
|
wenzelm@11838
|
995 |
lemma Sigma_mono: "[| A <= C; !!x. x:A ==> B x <= D x |] ==> Sigma A B <= Sigma C D"
|
wenzelm@11838
|
996 |
by blast
|
wenzelm@11838
|
997 |
|
wenzelm@11838
|
998 |
lemma Sigma_empty1 [simp]: "Sigma {} B = {}"
|
wenzelm@11838
|
999 |
by blast
|
wenzelm@11838
|
1000 |
|
wenzelm@11838
|
1001 |
lemma Sigma_empty2 [simp]: "A <*> {} = {}"
|
wenzelm@11838
|
1002 |
by blast
|
wenzelm@11838
|
1003 |
|
wenzelm@11838
|
1004 |
lemma UNIV_Times_UNIV [simp]: "UNIV <*> UNIV = UNIV"
|
wenzelm@11838
|
1005 |
by auto
|
wenzelm@11838
|
1006 |
|
wenzelm@11838
|
1007 |
lemma Compl_Times_UNIV1 [simp]: "- (UNIV <*> A) = UNIV <*> (-A)"
|
wenzelm@11838
|
1008 |
by auto
|
wenzelm@11838
|
1009 |
|
wenzelm@11838
|
1010 |
lemma Compl_Times_UNIV2 [simp]: "- (A <*> UNIV) = (-A) <*> UNIV"
|
wenzelm@11838
|
1011 |
by auto
|
wenzelm@11838
|
1012 |
|
wenzelm@11838
|
1013 |
lemma mem_Sigma_iff [iff]: "((a,b): Sigma A B) = (a:A & b:B(a))"
|
wenzelm@11838
|
1014 |
by blast
|
wenzelm@11838
|
1015 |
|
wenzelm@11838
|
1016 |
lemma Times_subset_cancel2: "x:C ==> (A <*> C <= B <*> C) = (A <= B)"
|
wenzelm@11838
|
1017 |
by blast
|
wenzelm@11838
|
1018 |
|
wenzelm@11838
|
1019 |
lemma Times_eq_cancel2: "x:C ==> (A <*> C = B <*> C) = (A = B)"
|
wenzelm@11838
|
1020 |
by (blast elim: equalityE)
|
wenzelm@11838
|
1021 |
|
wenzelm@11838
|
1022 |
lemma SetCompr_Sigma_eq:
|
wenzelm@11838
|
1023 |
"Collect (split (%x y. P x & Q x y)) = (SIGMA x:Collect P. Collect (Q x))"
|
wenzelm@11838
|
1024 |
by blast
|
wenzelm@11838
|
1025 |
|
wenzelm@11838
|
1026 |
lemma Collect_split [simp]: "{(a,b). P a & Q b} = Collect P <*> Collect Q"
|
wenzelm@11838
|
1027 |
by blast
|
wenzelm@11838
|
1028 |
|
wenzelm@11838
|
1029 |
lemma UN_Times_distrib:
|
wenzelm@11838
|
1030 |
"(UN (a,b):(A <*> B). E a <*> F b) = (UNION A E) <*> (UNION B F)"
|
wenzelm@11838
|
1031 |
-- {* Suggested by Pierre Chartier *}
|
wenzelm@11838
|
1032 |
by blast
|
wenzelm@11838
|
1033 |
|
blanchet@35828
|
1034 |
lemma split_paired_Ball_Sigma [simp,no_atp]:
|
wenzelm@11838
|
1035 |
"(ALL z: Sigma A B. P z) = (ALL x:A. ALL y: B x. P(x,y))"
|
wenzelm@11838
|
1036 |
by blast
|
wenzelm@11838
|
1037 |
|
blanchet@35828
|
1038 |
lemma split_paired_Bex_Sigma [simp,no_atp]:
|
wenzelm@11838
|
1039 |
"(EX z: Sigma A B. P z) = (EX x:A. EX y: B x. P(x,y))"
|
wenzelm@11838
|
1040 |
by blast
|
wenzelm@11838
|
1041 |
|
wenzelm@11838
|
1042 |
lemma Sigma_Un_distrib1: "(SIGMA i:I Un J. C(i)) = (SIGMA i:I. C(i)) Un (SIGMA j:J. C(j))"
|
wenzelm@11838
|
1043 |
by blast
|
wenzelm@11838
|
1044 |
|
wenzelm@11838
|
1045 |
lemma Sigma_Un_distrib2: "(SIGMA i:I. A(i) Un B(i)) = (SIGMA i:I. A(i)) Un (SIGMA i:I. B(i))"
|
wenzelm@11838
|
1046 |
by blast
|
wenzelm@11838
|
1047 |
|
wenzelm@11838
|
1048 |
lemma Sigma_Int_distrib1: "(SIGMA i:I Int J. C(i)) = (SIGMA i:I. C(i)) Int (SIGMA j:J. C(j))"
|
wenzelm@11838
|
1049 |
by blast
|
wenzelm@11838
|
1050 |
|
wenzelm@11838
|
1051 |
lemma Sigma_Int_distrib2: "(SIGMA i:I. A(i) Int B(i)) = (SIGMA i:I. A(i)) Int (SIGMA i:I. B(i))"
|
wenzelm@11838
|
1052 |
by blast
|
wenzelm@11838
|
1053 |
|
wenzelm@11838
|
1054 |
lemma Sigma_Diff_distrib1: "(SIGMA i:I - J. C(i)) = (SIGMA i:I. C(i)) - (SIGMA j:J. C(j))"
|
wenzelm@11838
|
1055 |
by blast
|
wenzelm@11838
|
1056 |
|
wenzelm@11838
|
1057 |
lemma Sigma_Diff_distrib2: "(SIGMA i:I. A(i) - B(i)) = (SIGMA i:I. A(i)) - (SIGMA i:I. B(i))"
|
wenzelm@11838
|
1058 |
by blast
|
wenzelm@11838
|
1059 |
|
wenzelm@11838
|
1060 |
lemma Sigma_Union: "Sigma (Union X) B = (UN A:X. Sigma A B)"
|
wenzelm@11838
|
1061 |
by blast
|
wenzelm@11838
|
1062 |
|
wenzelm@11838
|
1063 |
text {*
|
wenzelm@11838
|
1064 |
Non-dependent versions are needed to avoid the need for higher-order
|
wenzelm@11838
|
1065 |
matching, especially when the rules are re-oriented.
|
wenzelm@11838
|
1066 |
*}
|
wenzelm@11838
|
1067 |
|
wenzelm@11838
|
1068 |
lemma Times_Un_distrib1: "(A Un B) <*> C = (A <*> C) Un (B <*> C)"
|
nipkow@28719
|
1069 |
by blast
|
wenzelm@11838
|
1070 |
|
wenzelm@11838
|
1071 |
lemma Times_Int_distrib1: "(A Int B) <*> C = (A <*> C) Int (B <*> C)"
|
nipkow@28719
|
1072 |
by blast
|
wenzelm@11838
|
1073 |
|
wenzelm@11838
|
1074 |
lemma Times_Diff_distrib1: "(A - B) <*> C = (A <*> C) - (B <*> C)"
|
nipkow@28719
|
1075 |
by blast
|
wenzelm@11838
|
1076 |
|
hoelzl@36610
|
1077 |
lemma Times_empty[simp]: "A \<times> B = {} \<longleftrightarrow> A = {} \<or> B = {}"
|
hoelzl@36610
|
1078 |
by auto
|
hoelzl@36610
|
1079 |
|
hoelzl@36610
|
1080 |
lemma fst_image_times[simp]: "fst ` (A \<times> B) = (if B = {} then {} else A)"
|
hoelzl@36610
|
1081 |
by (auto intro!: image_eqI)
|
hoelzl@36610
|
1082 |
|
hoelzl@36610
|
1083 |
lemma snd_image_times[simp]: "snd ` (A \<times> B) = (if A = {} then {} else B)"
|
hoelzl@36610
|
1084 |
by (auto intro!: image_eqI)
|
hoelzl@36610
|
1085 |
|
nipkow@28719
|
1086 |
lemma insert_times_insert[simp]:
|
nipkow@28719
|
1087 |
"insert a A \<times> insert b B =
|
nipkow@28719
|
1088 |
insert (a,b) (A \<times> insert b B \<union> insert a A \<times> B)"
|
nipkow@28719
|
1089 |
by blast
|
wenzelm@11838
|
1090 |
|
paulson@33271
|
1091 |
lemma vimage_Times: "f -` (A \<times> B) = ((fst \<circ> f) -` A) \<inter> ((snd \<circ> f) -` B)"
|
haftmann@37160
|
1092 |
by (auto, case_tac "f x", auto)
|
paulson@33271
|
1093 |
|
nipkow@37277
|
1094 |
text{* The following @{const prod_fun} lemmas are due to Joachim Breitner: *}
|
nipkow@37277
|
1095 |
|
nipkow@37277
|
1096 |
lemma prod_fun_inj_on:
|
nipkow@37277
|
1097 |
assumes "inj_on f A" and "inj_on g B"
|
nipkow@37277
|
1098 |
shows "inj_on (prod_fun f g) (A \<times> B)"
|
nipkow@37277
|
1099 |
proof (rule inj_onI)
|
nipkow@37277
|
1100 |
fix x :: "'a \<times> 'c" and y :: "'a \<times> 'c"
|
nipkow@37277
|
1101 |
assume "x \<in> A \<times> B" hence "fst x \<in> A" and "snd x \<in> B" by auto
|
nipkow@37277
|
1102 |
assume "y \<in> A \<times> B" hence "fst y \<in> A" and "snd y \<in> B" by auto
|
nipkow@37277
|
1103 |
assume "prod_fun f g x = prod_fun f g y"
|
nipkow@37277
|
1104 |
hence "fst (prod_fun f g x) = fst (prod_fun f g y)" by (auto)
|
nipkow@37277
|
1105 |
hence "f (fst x) = f (fst y)" by (cases x,cases y,auto)
|
nipkow@37277
|
1106 |
with `inj_on f A` and `fst x \<in> A` and `fst y \<in> A`
|
nipkow@37277
|
1107 |
have "fst x = fst y" by (auto dest:dest:inj_onD)
|
nipkow@37277
|
1108 |
moreover from `prod_fun f g x = prod_fun f g y`
|
nipkow@37277
|
1109 |
have "snd (prod_fun f g x) = snd (prod_fun f g y)" by (auto)
|
nipkow@37277
|
1110 |
hence "g (snd x) = g (snd y)" by (cases x,cases y,auto)
|
nipkow@37277
|
1111 |
with `inj_on g B` and `snd x \<in> B` and `snd y \<in> B`
|
nipkow@37277
|
1112 |
have "snd x = snd y" by (auto dest:dest:inj_onD)
|
nipkow@37277
|
1113 |
ultimately show "x = y" by(rule prod_eqI)
|
nipkow@37277
|
1114 |
qed
|
nipkow@37277
|
1115 |
|
nipkow@37277
|
1116 |
lemma prod_fun_surj:
|
nipkow@37277
|
1117 |
assumes "surj f" and "surj g"
|
nipkow@37277
|
1118 |
shows "surj (prod_fun f g)"
|
nipkow@37277
|
1119 |
unfolding surj_def
|
nipkow@37277
|
1120 |
proof
|
nipkow@37277
|
1121 |
fix y :: "'b \<times> 'd"
|
nipkow@37277
|
1122 |
from `surj f` obtain a where "fst y = f a" by (auto elim:surjE)
|
nipkow@37277
|
1123 |
moreover
|
nipkow@37277
|
1124 |
from `surj g` obtain b where "snd y = g b" by (auto elim:surjE)
|
nipkow@37277
|
1125 |
ultimately have "(fst y, snd y) = prod_fun f g (a,b)" by auto
|
nipkow@37277
|
1126 |
thus "\<exists>x. y = prod_fun f g x" by auto
|
nipkow@37277
|
1127 |
qed
|
nipkow@37277
|
1128 |
|
nipkow@37277
|
1129 |
lemma prod_fun_surj_on:
|
nipkow@37277
|
1130 |
assumes "f ` A = A'" and "g ` B = B'"
|
nipkow@37277
|
1131 |
shows "prod_fun f g ` (A \<times> B) = A' \<times> B'"
|
nipkow@37277
|
1132 |
unfolding image_def
|
nipkow@37277
|
1133 |
proof(rule set_ext,rule iffI)
|
nipkow@37277
|
1134 |
fix x :: "'a \<times> 'c"
|
nipkow@37277
|
1135 |
assume "x \<in> {y\<Colon>'a \<times> 'c. \<exists>x\<Colon>'b \<times> 'd\<in>A \<times> B. y = prod_fun f g x}"
|
nipkow@37277
|
1136 |
then obtain y where "y \<in> A \<times> B" and "x = prod_fun f g y" by blast
|
nipkow@37277
|
1137 |
from `image f A = A'` and `y \<in> A \<times> B` have "f (fst y) \<in> A'" by auto
|
nipkow@37277
|
1138 |
moreover from `image g B = B'` and `y \<in> A \<times> B` have "g (snd y) \<in> B'" by auto
|
nipkow@37277
|
1139 |
ultimately have "(f (fst y), g (snd y)) \<in> (A' \<times> B')" by auto
|
nipkow@37277
|
1140 |
with `x = prod_fun f g y` show "x \<in> A' \<times> B'" by (cases y, auto)
|
nipkow@37277
|
1141 |
next
|
nipkow@37277
|
1142 |
fix x :: "'a \<times> 'c"
|
nipkow@37277
|
1143 |
assume "x \<in> A' \<times> B'" hence "fst x \<in> A'" and "snd x \<in> B'" by auto
|
nipkow@37277
|
1144 |
from `image f A = A'` and `fst x \<in> A'` have "fst x \<in> image f A" by auto
|
nipkow@37277
|
1145 |
then obtain a where "a \<in> A" and "fst x = f a" by (rule imageE)
|
nipkow@37277
|
1146 |
moreover from `image g B = B'` and `snd x \<in> B'`
|
nipkow@37277
|
1147 |
obtain b where "b \<in> B" and "snd x = g b" by auto
|
nipkow@37277
|
1148 |
ultimately have "(fst x, snd x) = prod_fun f g (a,b)" by auto
|
nipkow@37277
|
1149 |
moreover from `a \<in> A` and `b \<in> B` have "(a , b) \<in> A \<times> B" by auto
|
nipkow@37277
|
1150 |
ultimately have "\<exists>y \<in> A \<times> B. x = prod_fun f g y" by auto
|
nipkow@37277
|
1151 |
thus "x \<in> {x. \<exists>y \<in> A \<times> B. x = prod_fun f g y}" by auto
|
nipkow@37277
|
1152 |
qed
|
nipkow@37277
|
1153 |
|
haftmann@35822
|
1154 |
lemma swap_inj_on:
|
hoelzl@36610
|
1155 |
"inj_on (\<lambda>(i, j). (j, i)) A"
|
hoelzl@36610
|
1156 |
by (auto intro!: inj_onI)
|
haftmann@35822
|
1157 |
|
haftmann@35822
|
1158 |
lemma swap_product:
|
haftmann@35822
|
1159 |
"(%(i, j). (j, i)) ` (A \<times> B) = B \<times> A"
|
haftmann@35822
|
1160 |
by (simp add: split_def image_def) blast
|
haftmann@35822
|
1161 |
|
hoelzl@36610
|
1162 |
lemma image_split_eq_Sigma:
|
hoelzl@36610
|
1163 |
"(\<lambda>x. (f x, g x)) ` A = Sigma (f ` A) (\<lambda>x. g ` (f -` {x} \<inter> A))"
|
hoelzl@36610
|
1164 |
proof (safe intro!: imageI vimageI)
|
hoelzl@36610
|
1165 |
fix a b assume *: "a \<in> A" "b \<in> A" and eq: "f a = f b"
|
hoelzl@36610
|
1166 |
show "(f b, g a) \<in> (\<lambda>x. (f x, g x)) ` A"
|
hoelzl@36610
|
1167 |
using * eq[symmetric] by auto
|
hoelzl@36610
|
1168 |
qed simp_all
|
haftmann@35822
|
1169 |
|
haftmann@21908
|
1170 |
|
haftmann@37160
|
1171 |
subsection {* Inductively defined sets *}
|
berghofe@15394
|
1172 |
|
haftmann@37364
|
1173 |
use "Tools/inductive_codegen.ML"
|
haftmann@37364
|
1174 |
setup Inductive_Codegen.setup
|
haftmann@37364
|
1175 |
|
haftmann@31723
|
1176 |
use "Tools/inductive_set.ML"
|
haftmann@31723
|
1177 |
setup Inductive_Set.setup
|
haftmann@24699
|
1178 |
|
haftmann@37160
|
1179 |
|
haftmann@37160
|
1180 |
subsection {* Legacy theorem bindings and duplicates *}
|
haftmann@37160
|
1181 |
|
haftmann@37160
|
1182 |
lemma PairE:
|
haftmann@37160
|
1183 |
obtains x y where "p = (x, y)"
|
haftmann@37160
|
1184 |
by (fact prod.exhaust)
|
haftmann@37160
|
1185 |
|
haftmann@37160
|
1186 |
lemma Pair_inject:
|
haftmann@37160
|
1187 |
assumes "(a, b) = (a', b')"
|
haftmann@37160
|
1188 |
and "a = a' ==> b = b' ==> R"
|
haftmann@37160
|
1189 |
shows R
|
haftmann@37160
|
1190 |
using assms by simp
|
haftmann@37160
|
1191 |
|
haftmann@37160
|
1192 |
lemmas Pair_eq = prod.inject
|
haftmann@37160
|
1193 |
|
haftmann@37160
|
1194 |
lemmas split = split_conv -- {* for backwards compatibility *}
|
haftmann@37160
|
1195 |
|
nipkow@10213
|
1196 |
end
|