(* This file is a part of IsarMathLib - a library of formalized mathematics for Isabelle/Isar. Copyright (C) 2008 Slawomir Kolodynski This program is free software; Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. The name of the author may not be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. *) header{*\isaheader{Enumeration\_ZF.thy}*} theory Enumeration_ZF imports NatOrder_ZF FiniteSeq_ZF FinOrd_ZF begin text{*Suppose $r$ is a linear order on a set $A$ that has $n$ elements, where $n \in\mathbb{N}$ . In the @{text "FinOrd_ZF"} theory we prove a theorem stating that there is a unique order isomorphism between $n = \{0,1,..,n-1\}$ (with natural order) and $A$. Another way of stating that is that there is a unique way of counting the elements of $A$ in the order increasing according to relation $r$. Yet another way of stating the same thing is that there is a unique sorted list of elements of $A$. We will call this list the @{text "Enumeration"} of $A$. *} section{*Enumerations: definition and notation*} text{*In this section we introduce the notion of enumeration and define a proof context (a ''locale'' in Isabelle terms) that sets up the notation for writing about enumarations.*} text{*We define enumeration as the only order isomorphism beween a set $A$ and the number of its elements. We are using the formula $\bigcup \{x\} = x$ to extract the only element from a singleton. @{text "Le"} is the (natural) order on natural numbers, defined is @{text "Nat_ZF"} theory in the standard Isabelle library.*} definition "Enumeration(A,r) ≡ ⋃ ord_iso(|A|,Le,A,r)" text{*To set up the notation we define a locale @{text "enums"}. In this locale we will assume that $r$ is a linear order on some set $X$. In most applications this set will be just the set of natural numbers. Standard Isabelle uses $\leq $ to denote the "less or equal" relation on natural numbers. We will use the @{text "\<lsq>"} symbol to denote the relation $r$. Those two symbols usually look the same in the presentation, but they are different in the source.To shorten the notation the enumeration @{text "Enumeration(A,r)"} will be denoted as $\sigma (A)$. Similarly as in the @{text "Semigroup"} theory we will write $a\hookleftarrow x$ for the result of appending an element $x$ to the finite sequence (list) $a$. Finally, $a\sqcup b$ will denote the concatenation of the lists $a$ and $b$.*} locale enums = fixes X r assumes linord: "IsLinOrder(X,r)" fixes ler (infix "\<lsq>" 70) defines ler_def[simp]: "x \<lsq> y ≡ ⟨x,y⟩ ∈ r" fixes σ defines σ_def [simp]: "σ(A) ≡ Enumeration(A,r)" fixes append (infix "↩" 72) defines append_def[simp]: "a ↩ x ≡ Append(a,x)" fixes concat (infixl "⊔" 69) defines concat_def[simp]: "a ⊔ b ≡ Concat(a,b)" section{*Properties of enumerations*} text{*In this section we prove basic facts about enumerations. *} text{*A special case of the existence and uniqueess of the order isomorphism for finite sets when the first set is a natural number.*} lemma (in enums) ord_iso_nat_fin: assumes "A ∈ FinPow(X)" and "n ∈ nat" and "A ≈ n" shows "∃!f. f ∈ ord_iso(n,Le,A,r)" using assms NatOrder_ZF_1_L2 linord nat_finpow_nat fin_ord_iso_ex_uniq by simp text{*An enumeration is an order isomorhism, a bijection, and a list.*} lemma (in enums) enum_props: assumes "A ∈ FinPow(X)" shows "σ(A) ∈ ord_iso(|A|,Le, A,r)" "σ(A) ∈ bij(|A|,A)" "σ(A) : |A| → A" proof - from assms have "IsLinOrder(nat,Le)" and "|A| ∈ FinPow(nat)" and "A ≈ |A|" using NatOrder_ZF_1_L2 card_fin_is_nat nat_finpow_nat by auto with assms show "σ(A) ∈ ord_iso(|A|,Le, A,r)" using linord fin_ord_iso_ex_uniq sigleton_extract Enumeration_def by simp then show "σ(A) ∈ bij(|A|,A)" and "σ(A) : |A| → A" using ord_iso_def bij_def surj_def by auto qed text{*A corollary from @{text "enum_props"}. Could have been attached as another assertion, but this slows down verification of some other proofs. *} lemma (in enums) enum_fun: assumes "A ∈ FinPow(X)" shows "σ(A) : |A| → X" proof - from assms have "σ(A) : |A| → A" and "A⊆X" using enum_props FinPow_def by auto then show "σ(A) : |A| → X" by (rule func1_1_L1B) qed text{*If a list is an order isomorphism then it must be the enumeration. *} lemma (in enums) ord_iso_enum: assumes A1: "A ∈ FinPow(X)" and A2: "n ∈ nat" and A3: "f ∈ ord_iso(n,Le,A,r)" shows "f = σ(A)" proof - from A3 have "n ≈ A" using ord_iso_def eqpoll_def by auto then have "A ≈ n" by (rule eqpoll_sym) with A1 A2 have "∃!f. f ∈ ord_iso(n,Le,A,r)" using ord_iso_nat_fin by simp with assms `A ≈ n` show "f = σ(A)" using enum_props card_card by blast qed text{*What is the enumeration of the empty set?*} lemma (in enums) empty_enum: shows "σ(0) = 0" proof - have "0 ∈ FinPow(X)" and "0 ∈ nat" and "0 ∈ ord_iso(0,Le,0,r)" using empty_in_finpow empty_ord_iso_empty by auto then show "σ(0) = 0" using ord_iso_enum by blast qed text{*Adding a new maximum to a set appends it to the enumeration.*} lemma (in enums) enum_append: assumes A1: "A ∈ FinPow(X)" and A2: "b ∈ X-A" and A3: "∀a∈A. a\<lsq>b" shows " σ(A ∪ {b}) = σ(A)↩ b" proof - let ?f = "σ(A) ∪ {⟨|A|,b⟩}" from A1 have "|A| ∈ nat" using card_fin_is_nat by simp from A1 A2 have "A ∪ {b} ∈ FinPow(X)" using singleton_in_finpow union_finpow by simp moreover from this have "|A ∪ {b}| ∈ nat" using card_fin_is_nat by simp moreover have "?f ∈ ord_iso(|A ∪ {b}| , Le, A ∪ {b} ,r)" proof - from A1 A2 have "σ(A) ∈ ord_iso(|A|,Le, A,r)" and "|A| ∉ |A|" and "b ∉ A" using enum_props mem_not_refl by auto moreover from `|A| ∈ nat` have "∀k ∈ |A|. ⟨k, |A|⟩ ∈ Le" using elem_nat_is_nat by blast moreover from A3 have "∀a∈A. ⟨a,b⟩ ∈ r" by simp moreover have "antisym(Le)" and "antisym(r)" using linord NatOrder_ZF_1_L2 IsLinOrder_def by auto moreover from A2 `|A| ∈ nat` have "⟨|A|,|A|⟩ ∈ Le" and "⟨b,b⟩ ∈ r" using linord NatOrder_ZF_1_L2 IsLinOrder_def total_is_refl refl_def by auto hence "⟨|A|,|A|⟩ ∈ Le ⟷ ⟨b,b⟩ ∈ r" by simp ultimately have "?f ∈ ord_iso(|A| ∪ {|A|} , Le, A ∪ {b} ,r)" by (rule ord_iso_extend) with A1 A2 show "?f ∈ ord_iso(|A ∪ {b}| , Le, A ∪ {b} ,r)" using card_fin_add_one by simp qed ultimately have "?f = σ(A ∪ {b})" using ord_iso_enum by simp moreover have "σ(A)↩ b = ?f" proof - have "σ(A)↩ b = σ(A) ∪ {⟨domain(σ(A)),b⟩}" using Append_def by simp moreover from A1 have "domain(σ(A)) = |A|" using enum_props func1_1_L1 by blast ultimately show "σ(A)↩ b = ?f" by simp qed ultimately show "σ(A ∪ {b}) = σ(A)↩ b" by simp qed text{*What is the enumeration of a singleton?*} lemma (in enums) enum_singleton: assumes A1: "x∈X" shows "σ({x}): 1 → X" and "σ({x})`(0) = x" proof - from A1 have "0 ∈ FinPow(X)" and "x ∈ (X - 0)" and "∀a∈0. a\<lsq>x" using empty_in_finpow by auto then have "σ(0 ∪ {x}) = σ(0)↩ x" by (rule enum_append) with A1 show "σ({x}): 1 → X" and "σ({x})`(0) = x" using empty_enum empty_append1 by auto qed end