Copyright (c) 1992 Association of Mizar Users
environ vocabulary ARYTM, FUNCT_1, ARYTM_3, RELAT_1, WELLORD1, RELAT_2, BOOLE, TARSKI, FINSEQ_1, PRE_TOPC, METRIC_1, SETFAM_1, COMPTS_1, SUBSET_1, ARYTM_1, PCOMPS_1, FINSET_1, PCOMPS_2, GROUP_1, FINSEQ_4; notation TARSKI, XBOOLE_0, ZFMISC_1, SUBSET_1, ORDINAL1, NUMBERS, XCMPLX_0, XREAL_0, REAL_1, TOPS_2, FUNCT_1, FUNCT_2, NAT_1, SETFAM_1, STRUCT_0, METRIC_1, COMPTS_1, PRE_TOPC, PCOMPS_1, PREPOWER, WELLORD1, RELAT_1, FINSEQ_1, FINSET_1, RELAT_2, FINSEQ_2, FINSEQ_4; constructors TOPS_2, NAT_1, COMPTS_1, PCOMPS_1, REAL_1, PREPOWER, WELLORD1, WELLORD2, RELAT_2, FINSEQ_4, MEMBERED; clusters SUBSET_1, PRE_TOPC, FINSET_1, RELSET_1, STRUCT_0, XREAL_0, FINSEQ_1, METRIC_1, NEWTON, MEMBERED, ZFMISC_1, ORDINAL2; requirements REAL, NUMERALS, BOOLE, SUBSET, ARITHM; definitions TARSKI, WELLORD1, RELAT_2, SETFAM_1, XBOOLE_0; theorems PCOMPS_1, COMPTS_1, SETFAM_1, TOPS_1, TOPS_2, PRE_TOPC, SUBSET_1, TARSKI, WELLORD2, AXIOMS, ZFMISC_1, NAT_1, REAL_1, REAL_2, PREPOWER, WELLORD1, RELAT_1, RELAT_2, SEQ_4, FUNCT_1, FINSET_1, METRIC_1, FINSEQ_1, CARD_1, FINSEQ_2, FINSEQ_4, TBSP_1, NEWTON, XREAL_0, XBOOLE_0, XBOOLE_1, XCMPLX_1; schemes NAT_1, FUNCT_2, SETFAM_1, PRE_TOPC, RECDEF_1, TREES_2, FRAENKEL, XBOOLE_0; begin :: 1. Selected properties of real numbers reserve r,u for real number; reserve n,k,i for Nat; canceled 2; theorem Th3: r>0 & u > 0 implies ex k be Nat st u/(2 |^ k) <= r proof assume that A1:r>0 and A2: u>0; consider t be real number such that A3: r * t = 1 by A1,AXIOMS:20; reconsider t as Real by XREAL_0:def 1; A4: t > 0 by A1,A3,REAL_2:123; A5: r = 1/t by A3,XCMPLX_1:74 .= (u * 1) / (u * t) by A2,XCMPLX_1:92 .= u/(u * t); consider n such that A6: (u * t) < n by SEQ_4:10; A7: 0 < u * t by A2,A4,REAL_2:122; then A8: u/(u * t) > u/n by A2,A6,REAL_2:200; defpred P[Nat] means $1 <= 2 |^ $1; 2 |^ 0 = 1 by NEWTON:9; then A9: P[0]; A10: for k be Nat st P[k] holds P[k + 1] proof let k be Nat; assume A11: k <= 2 |^ k; A12: 2 |^ (k + 1) = (2 |^ k) * (2 |^ 1) by NEWTON:13 .= (2 |^ k) * 2 by NEWTON:10 .= 2 |^ k + 2 |^ k by XCMPLX_1:11; 2 |^ k >= 1 by PREPOWER:19; then A13: 2 |^ k + 2 |^ k >= 2 |^ k + 1 by REAL_1:55; (2 |^ k) + 1 >= k + 1 by A11,REAL_1:55; hence thesis by A12,A13,AXIOMS:22; end; for k be Nat holds P[k] from Ind(A9,A10); then A14: n <= (2 |^ n); 0 < n by A6,A7,AXIOMS:22; then A15: u/n >= u/(2 |^ n) by A2,A14,REAL_2:201; take n; thus thesis by A5,A8,A15,AXIOMS:22; end; theorem Th4: k>=n & r >= 1 implies r |^ k >= r |^ n proof assume that A1:k>=n and A2: r >= 1; consider m be Nat such that A3: k = n + m by A1,NAT_1:28; A4: r |^ k = r |^ n * r |^ m by A3,NEWTON:13; r |^ n >= 1 by A2,PREPOWER:19; then A5: r |^ n >= 0 by AXIOMS:22; r |^ m >= 1 by A2,PREPOWER:19; hence thesis by A4,A5,REAL_2:146; end; begin :: 2. Certain functions defined on families of sets reserve R for Relation; reserve A for set; theorem Th5: R well_orders A implies R |_2 A well_orders A & A = field (R |_2 A) proof assume A1: R well_orders A; then A2: R is_reflexive_in A by WELLORD1:def 5; A3: field (R |_2 A) c= A by WELLORD1:20; A4: A c= field (R |_2 A) proof let x be set; assume A5: x in A; then A6: [x,x] in [:A,A:] by ZFMISC_1:106; [x,x] in R by A2,A5,RELAT_2:def 1; then [x,x] in R |_2 A by A6,WELLORD1:16; hence x in field (R |_2 A) by RELAT_1:30; end; A7: R |_2 A is_reflexive_in A proof let x be set; assume A8: x in A; then A9: [x,x] in R by A2,RELAT_2:def 1; [x,x] in [:A,A:] by A8,ZFMISC_1:106; hence [x,x] in R |_2 A by A9,WELLORD1:16; end; A10:R |_2 A c= R by WELLORD1:15; A11: R |_2 A is_transitive_in A proof let x,y,z be set; assume that A12:x in A and A13:y in A and A14:z in A and A15:[x,y] in R |_2 A and A16:[y,z] in R |_2 A; R is_transitive_in A by A1,WELLORD1:def 5; then A17:[x,z] in R by A10,A12,A13,A14,A15,A16,RELAT_2:def 8; [x,z] in [:A,A:] by A12,A14,ZFMISC_1:106; hence [x,z] in R |_2 A by A17,WELLORD1:16; end; A18: R |_2 A is_antisymmetric_in A proof let x,y be set; assume that A19:x in A and A20:y in A and A21: [x,y] in R |_2 A and A22:[y,x] in R |_2 A; R is_antisymmetric_in A by A1,WELLORD1:def 5; hence x=y by A10,A19,A20,A21,A22,RELAT_2:def 4; end; A23: R |_2 A is_connected_in A proof let x,y be set; assume A24: x in A & y in A & x <>y; A25: R is_connected_in A by A1,WELLORD1:def 5; now per cases by A24,A25,RELAT_2:def 6; case A26:[x,y] in R; [x,y] in [:A,A:] by A24,ZFMISC_1:106; hence [x,y] in R |_2 A by A26,WELLORD1:16; case A27:[y,x] in R; [y,x] in [:A,A:] by A24,ZFMISC_1:106; hence [y,x] in R |_2 A by A27,WELLORD1:16; end; hence [x,y] in R |_2 A or [y,x] in R |_2 A; end; R |_2 A is_well_founded_in A proof let Y be set; assume that A28:Y c= A and A29: Y <> {}; R is_well_founded_in A by A1,WELLORD1:def 5; then consider a be set such that A30:a in Y and A31:R-Seg(a) misses Y by A28,A29,WELLORD1:def 3; (R |_2 A)-Seg(a) c= R-Seg(a) by WELLORD1:21; then (R |_2 A)-Seg(a) misses Y by A31,XBOOLE_1:63; hence ex a be set st a in Y & (R |_2 A)-Seg(a) misses Y by A30; end; hence thesis by A3,A4,A7,A11,A18,A23,WELLORD1:def 5,XBOOLE_0:def 10; end; scheme MinSet{A()->set,R()->Relation,P[set]}: ex X be set st X in A() & P[X] & for Y be set st Y in A() & P[Y] holds [X,Y] in R() provided A1: R() well_orders A() and A2: ex X be set st X in A() & P[X] proof defpred Q[set] means P[$1]; consider Y be set such that A3: for x be set holds x in Y iff x in A() & Q[x] from Separation; A4: ex x be set st x in Y proof consider x be set such that A5: x in A() & P[x] by A2; take x; thus x in Y by A3,A5; end; for x be set holds x in Y implies x in A() by A3; then Y c= A() by TARSKI:def 3; then consider X be set such that A6: X in Y and A7: for Z be set st Z in Y holds [X,Z] in R() by A1,A4,WELLORD1:9; A8: X in A() by A3,A6; A9: P[X] by A3,A6; for M be set st M in A() & P[M] holds [X,M] in R() proof let M be set; assume M in A() & P[M]; then M in Y by A3; hence thesis by A7; end; hence thesis by A8,A9; end; definition let FX be set, R be Relation, B be Element of FX; func PartUnion(B,R) equals :Def1: union (R-Seg(B)); coherence; end; definition let FX be set, R be Relation; func DisjointFam(FX,R) means A in it iff ex B be Element of FX st B in FX & A = B\PartUnion(B,R); existence proof defpred P[set] means ex B be Element of FX st B in FX & $1 = B\PartUnion(B,R); consider X being set such that A1: for x being set holds x in X iff x in bool union FX & P[x] from Separation; reconsider X as set; take X; let A; thus A in X implies ex B be Element of FX st B in FX & A = B\PartUnion(B,R) by A1; assume A2:ex B be Element of FX st B in FX & A = B\PartUnion(B,R); then consider B be Element of FX such that A3: B in FX & A = B\PartUnion(B,R); B c= union FX by A3,ZFMISC_1:92; then A c= union FX by A3,XBOOLE_1:109; hence A in X by A1,A2; end; uniqueness proof defpred P[set] means ex B be Element of FX st B in FX & $1 = B\PartUnion(B,R); thus for X1,X2 being set st (for x be set holds x in X1 iff P[x]) & (for x be set holds x in X2 iff P[x]) holds X1 = X2 from SetEq; end; end; definition let X be set, n be Nat, f be Function of NAT,bool X; func PartUnionNat(n,f) equals union (f.:(Seg(n)\{n})); coherence; end; begin :: 3. Paracompactness of metrizable spaces reserve PT for non empty TopSpace; reserve PM for MetrSpace; reserve FX,GX,HX for Subset-Family of PT; reserve Y,V,W for Subset of PT; theorem Th6:PT is_T3 implies for FX st FX is_a_cover_of PT & FX is open ex HX st HX is open & HX is_a_cover_of PT & for V st V in HX ex W st W in FX & Cl V c= W proof assume A1: PT is_T3; let FX; assume that A2:FX is_a_cover_of PT and A3:FX is open; defpred P[set] means ex V1 being Subset of PT st V1 = $1 & V1 is open & ex W st W in FX & Cl V1 c= W; consider HX such that A4: for V being Subset of PT holds V in HX iff P[V] from SubFamExS; take HX; for V being Subset of PT st V in HX holds V is open proof let V be Subset of PT; assume V in HX; then consider V1 being Subset of PT such that A5: V1 = V & V1 is open & ex W st W in FX & Cl V1 c= W by A4; thus thesis by A5; end; hence HX is open by TOPS_2:def 1; thus HX is_a_cover_of PT proof [#](PT) = union HX proof thus [#](PT) c= union HX proof let x be set; assume x in [#](PT); then reconsider x as Point of PT; consider V being Subset of PT such that A6: x in V & V in FX by A2,PCOMPS_1:5; reconsider V as Subset of PT; now per cases; suppose A7:V`<> {}; A8: not x in V` by A6,SUBSET_1:54; V is open by A3,A6,TOPS_2:def 1; then V` is closed by TOPS_1:30; then consider X,Y being Subset of PT such that A9: X is open & Y is open & x in X & V` c= Y & X misses Y by A1,A7,A8,COMPTS_1:def 5; A10: Y` c= V by A9,SUBSET_1:36; A11: X c= Y` by A9,SUBSET_1:43; Y` is closed by A9,TOPS_1:30; then Cl X c= Y` by A11,TOPS_1:31; then Cl X c= V by A10,XBOOLE_1:1; then X in HX by A4,A6,A9; hence x in union HX by A9,TARSKI:def 4; suppose A12: V` = {}; consider X being Subset of PT such that A13: X=[#](PT); A14: V = ({}(PT))` by A12 .= [#](PT) by PRE_TOPC:27; A15: X is open by A13,TOPS_1:53; A16: x in X by A13,PRE_TOPC:13; Cl X = V by A13,A14,TOPS_1:27; then X in HX by A4,A6,A15; hence x in union HX by A16,TARSKI:def 4; end; hence thesis; end; thus union HX c= [#](PT) proof let x be set; assume x in union HX; then consider V be set such that A17: x in V & V in HX by TARSKI:def 4; reconsider x as Point of PT by A17; x in [#](PT) by PRE_TOPC:13; hence thesis; end; end; hence thesis by PRE_TOPC:def 8; end; let V be Subset of PT; assume V in HX; then ex V1 being Subset of PT st V1 = V & V1 is open & ex W st W in FX & Cl V1 c= W by A4; hence ex W st W in FX & Cl V c= W; end; theorem for PT,FX st PT is_T2 & PT is paracompact & FX is_a_cover_of PT & FX is open ex GX st GX is open & GX is_a_cover_of PT & clf GX is_finer_than FX & GX is locally_finite proof let PT,FX; assume that A1: PT is_T2 and A2: PT is paracompact and A3: FX is_a_cover_of PT and A4: FX is open; PT is_T3 by A1,A2,PCOMPS_1:27; then consider HX such that A5: HX is open & HX is_a_cover_of PT & for V st V in HX ex W st W in FX & Cl V c= W by A3,A4,Th6; consider GX such that A6:GX is open & GX is_a_cover_of PT & GX is_finer_than HX & GX is locally_finite by A2,A5,PCOMPS_1:def 4; A7: for V st V in GX ex W st W in FX & (Cl V) c= W proof let V; assume V in GX; then consider X being set such that A8: X in HX & V c= X by A6,SETFAM_1:def 2; reconsider X as Subset of PT by A8; consider W such that A9:W in FX & Cl X c= W by A5,A8; take W; Cl V c= Cl X by A8,PRE_TOPC:49; hence thesis by A9,XBOOLE_1:1; end; clf GX is_finer_than FX proof let X be set; assume A10:X in clf GX; then reconsider X as Subset of PT; consider V such that A11: X = Cl V & V in GX by A10,PCOMPS_1:def 3; consider W such that A12: W in FX & (Cl V) c= W by A7,A11; take W; thus thesis by A11,A12; end; hence thesis by A6; end; theorem Th8: for f being Function of [:the carrier of PT,the carrier of PT:],REAL st f is_metric_of ( the carrier of PT) holds PM = SpaceMetr(the carrier of PT,f) implies the carrier of PM = the carrier of PT proof let f be Function of [:the carrier of PT,the carrier of PT:],REAL; assume A1: f is_metric_of the carrier of PT; assume PM = SpaceMetr(the carrier of PT,f); then PM = MetrStruct(#the carrier of PT,f#) by A1,PCOMPS_1:def 8; hence thesis; end; canceled 2; theorem for f being Function of [:the carrier of PT,the carrier of PT:],REAL st f is_metric_of ( the carrier of PT) holds PM = SpaceMetr(the carrier of PT,f) implies (FX is Subset-Family of PT iff FX is Subset-Family of PM) by Th8; reserve Mn for Relation; reserve n,k,l,q,p,q1 for Nat; definition let PM be non empty set; let g be Function of NAT,(bool bool PM)*; let n; redefine func g.n -> FinSequence of bool bool PM; coherence proof g.n is Element of (bool bool PM)*; hence g.n is FinSequence of bool bool PM; end; end; XXX1 { PM() -> non empty set, UB() -> Element of bool bool PM(), F(set,set) -> Subset of PM(), P[set], Q[set,set,set,set]} : ex f being Function of NAT, bool bool PM() st f.0 = UB() & for n holds f.(n+1) = {union { F(c,n) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds Q[c,V,n,fq] } where V is Element of bool PM() : P[V]} proof defpred T[Nat,FinSequence of bool bool PM(),set] means $3 = $2^<*{union { F(c,$1) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= $1 & fq = $2/.(q+1) holds Q[c,V,$1,fq] } where V is Element of bool PM() : P[V]}*>; A1: for n being Element of NAT for x being Element of (bool bool PM())* ex y being Element of (bool bool PM())* st T[n,x,y] proof let n; let x be Element of (bool bool PM())*; set T = { union { F(c,n) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= n & fq = x/.(q+1) holds Q[c,V,n,fq]} where V is Element of bool PM() : P[V]}; T in bool bool PM() proof now let X be set; assume X in T; then consider V be Element of bool PM() such that A2: X=union { F(c,n) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= n & fq = x/.(q+1) holds Q[c,V,n,fq] } and P[V]; now let a be set; assume a in X; then consider P be set such that A3: a in P and A4: P in { F(c,n) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= n & fq = x/.(q+1) holds Q[c,V,n,fq]} by A2,TARSKI:def 4; consider c be Element of PM() such that A5: P = F(c,n) and for fq be Element of bool bool PM(),q st q <= n & fq = x/.(q+1) holds Q[c,V,n,fq] by A4; thus a in PM() by A3,A5; end; then X c= PM() by TARSKI:def 3; hence X in bool PM(); end; then T c= bool PM() by TARSKI:def 3; hence thesis; end; then reconsider T as Element of bool bool PM(); reconsider T1 = <*T*> as FinSequence of bool bool PM(); consider y be FinSequence of bool bool PM() such that A6: y = x^T1; reconsider y as Element of (bool bool PM())* by FINSEQ_1:def 11; take y; thus thesis by A6; end; reconsider A = <*UB()*> as Element of (bool bool PM())* by FINSEQ_1:def 11; consider g be Function of NAT,(bool bool PM())* such that A7: g.0=A and A8: for n be Element of NAT holds T[n,g.n,g.(n+1)] from RecExD(A1); deffunc G(Nat) = (g.$1)/.len(g.$1); consider f be Function of NAT,bool bool PM() such that A9: for n holds f.n=G(n) from LambdaD; take f; len <*UB()*> = 1 by FINSEQ_1:56; hence f.0 = <*UB()*>/.1 by A7,A9 .= UB() by FINSEQ_4:25; defpred R[Nat] means len(g.$1) = $1 + 1; A10: R[0] by A7,FINSEQ_1:56; A11: for k st R[k] holds R[k+1] proof let k such that A12: len(g.k) = k+1; len(g.(k+1))=len((g.k)^<*{ union { F(c,k) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= k & fq = (g.k)/.(q+1) holds Q[c,V,k,fq] } where V is Element of bool PM() : P[V]}*>) by A8 .= len(g.k)+1 by FINSEQ_2:19; hence thesis by A12; end; A13: for n holds R[n] from Ind(A10,A11); defpred T[Nat] means for q st q <= $1 holds f.q = (g.$1)/.(q+1); A14: T[0] proof let q; assume q <= 0; then A15: q = 0 by NAT_1:18; thus f.q = (g.q)/.len(g.q) by A9 .= (g.0)/.(q+1) by A7,A15,FINSEQ_1:56; end; A16: for k st T[k] holds T[k+1] proof let k; assume A17: for q st q <= k holds f.q = (g.k)/.(q+1); let q; assume A18: q <= k+1; now per cases by A18,REAL_1:def 5; suppose A19: q = k+1; thus f.q=(g.q)/.len(g.q) by A9 .= (g.(k+1))/.(q+1) by A13,A19; suppose A20: q < k+1; then A21: q+1 <= k+1 by NAT_1:38; A22: q <= k by A20,NAT_1:38; q+1>=1 by NAT_1:29; then A23: q+1 in Seg(k+1) by A21,FINSEQ_1:3; k+1+1>=k+1 by NAT_1:29; then Seg(k+1) c= Seg(k+1+1) by FINSEQ_1:7; then q+1 in Seg(k+1+1) by A23; then q+1 in Seg(len(g.(k+1))) by A13; then A24: q+1 in dom(g.(k+1)) by FINSEQ_1:def 3; q+1 in Seg len(g.k) by A13,A23; then A25: q+1 in dom (g.k) by FINSEQ_1:def 3; thus (g.(k+1))/.(q+1) = (g.(k+1)).(q+1) by A24,FINSEQ_4:def 4 .= ((g.k)^<*{ union { F(c,k) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= k & fq = (g.k)/.(q+1) holds Q[c,V,k,fq]} where V is Element of bool PM() : P[V]}*>).(q+1) by A8 .= (g.k).(q+1) by A25,FINSEQ_1:def 7 .= (g.k)/.(q+1) by A25,FINSEQ_4:def 4 .= f.q by A17,A22; end; hence thesis; end; A26: for n holds T[n] from Ind(A14,A16); let n; defpred P2[set] means P[$1]; deffunc F2(set) = union { F(c,n) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds Q[c,$1,n,fq]}; deffunc G2(set) = union { F(c,n) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= n & fq = (g.n)/.(q+1) holds Q[c,$1,n,fq]}; set NF = { F2(V) where V is Element of bool PM(): P2[V] }; A27: for V be Element of bool PM() st P2[V] holds F2(V) = G2(V) proof let V be Element of bool PM() such that P[V]; deffunc F1(set) = F($1,n); defpred P1[set] means for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds Q[$1,V,n,fq]; defpred P2[set] means for fq be Element of bool bool PM(),q st q <= n & fq = (g.n)/.(q+1) holds Q[$1,V,n,fq]; A28:for c be Element of PM() holds P1[c] iff P2[c] proof let c be Element of PM(); thus (for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds Q[c,V,n,fq]) implies for fq be Element of bool bool PM(),q st q <= n & fq = (g.n)/.(q+1) holds Q[c,V,n,fq] proof assume A29: for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds Q[c,V,n,fq]; let fq be Element of bool bool PM(),q; assume A30: q <= n & fq = (g.n)/.(q+1); then fq = f.q by A26; hence thesis by A29,A30; end; assume A31: for fq be Element of bool bool PM(),q st q <= n & fq = (g.n)/.(q+1) holds Q[c,V,n,fq]; let fq be Element of bool bool PM(),q; assume A32: q <= n & fq = f.q; then f.q = (g.n)/.(q+1) by A26; hence Q[c,V,n,fq] by A31,A32; end; { F1(c) where c is Element of PM(): P1[c] } = { F1(c) where c is Element of PM(): P2[c] } from Fraenkel6'(A28); hence thesis; end; A33: NF = { G2(V) where V is Element of bool PM() : P2[V] } from FraenkelF'R(A27); then A34: len(g.(n+1))=len((g.n)^<*NF*>) by A8 .= len(g.n)+1 by FINSEQ_2:19; A35: len(g.n)+1 in dom (g.(n+1)) proof len(g.(n+1))= n+1+1 by A13; then A36: dom(g.(n+1)) = Seg(n+1+1) by FINSEQ_1:def 3; len(g.n)+1=n+1+1 by A13; hence thesis by A36,FINSEQ_1:6; end; thus f.(n+1) = (g.(n+1))/.(len(g.n)+1) by A9,A34 .= g.(n+1).(len(g.n)+1) by A35,FINSEQ_4:def 4 .= ((g.n)^<*NF*>).(len(g.n)+1) by A8,A33 .= NF by FINSEQ_1:59; end; XXX { PM() -> non empty set, UB() -> Element of bool bool PM(), F(set,set) -> Subset of PM(), P[set], Q[set,set,set]} : ex f being Function of NAT, bool bool PM() st f.0 = UB() & for n holds f.(n+1) = { union { F(c,n) where c is Element of PM(): Q[c,V,n] & not c in union{union(f.q): q <= n } } where V is Element of bool PM() : P[V]} proof defpred P1[set] means P[$1]; defpred Q1[set,set,set,set] means Q[$1,$2,$3] & not $1 in union $4; deffunc F1(set,set) = F($1,$2); consider f being Function of NAT, bool bool PM() such that A1: f.0 = UB() and A2: for n holds f.(n+1) = { union { F1(c,n) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds Q1[c,V,n,fq] } where V is Element of bool PM() : P1[V]} from XXX1; take f; thus f.0 = UB() by A1; let n; defpred P2[set] means P[$1]; deffunc F2(set) = union { F(c,n) where c is Element of PM(): for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds Q[c,$1,n] & not c in union(fq) }; deffunc G2(set) = union { F(c,n) where c is Element of PM(): Q[c,$1,n] & not c in union{union(f.q): q <= n } }; set fxxx1 = { F2(V) where V is Element of bool PM() : P2[V] }; set fxxx = { G2(V) where V is Element of bool PM() : P2[V] }; A3:now let V be Element of bool PM(); assume P2[V]; deffunc F1(set) = F($1,n); defpred P1[set] means for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds Q[$1,V,n] & not $1 in union(fq); defpred Q1[set] means Q[$1,V,n] & not $1 in union{union(f.q1): q1 <= n }; A4:now let c be Element of PM(); A5: ( for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds not c in union(fq)) iff not c in union{union(f.q): q <= n } proof thus (for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds not c in union(fq)) implies not c in union{union(f.q): q <= n } proof assume A6: for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds not c in union(fq); assume c in union{union(f.q): q <= n}; then consider C be set such that A7: c in C and A8: C in {union(f.q): q <= n}by TARSKI:def 4; consider q be Nat such that A9: C = union(f.q) and A10: q <= n by A8; thus contradiction by A6,A7,A9,A10; end; assume A11: not c in union{union(f.q): q <= n }; let fq be Element of bool bool PM(),q; assume that A12: q <= n and A13: fq = f.q; assume A14: c in union(fq); union(fq) in {union(f.p): p <= n} by A12,A13; hence contradiction by A11,A14,TARSKI:def 4; end; thus P1[c] iff Q1[c] proof hereby assume A15:( for fq be Element of bool bool PM(),q st q <= n & fq = f.q holds Q[c,V,n] & not c in union(fq)); consider q such that A16: q <= n; consider fq be Element of bool bool PM() such that A17: fq = f.q; thus Q[c,V,n] by A15,A16,A17; thus not c in union{union(f.p): p <= n } by A5,A15; end; assume Q[c,V,n] & not c in union{union(f.q): q <= n }; hence thesis by A5; end; end; { F1(c) where c is Element of PM(): P1[c] } = { F1(c) where c is Element of PM(): Q1[c] } from Fraenkel6'(A4); hence F2(V) = G2(V); end; fxxx1 = fxxx from FraenkelF'R(A3); hence thesis by A2; end; theorem :: Stone Theorem - general case Th12: PT is metrizable implies for FX being Subset-Family of PT st FX is_a_cover_of PT & FX is open ex GX being Subset-Family of PT st GX is open & GX is_a_cover_of PT & GX is_finer_than FX & GX is locally_finite proof assume PT is metrizable; then consider metr being Function of [:the carrier of PT,the carrier of PT:],REAL such that A1: metr is_metric_of (the carrier of PT) and A2: Family_open_set( SpaceMetr (the carrier of PT,metr) ) = the topology of PT by PCOMPS_1:def 9; consider PM being non empty MetrSpace such that A3: PM = SpaceMetr(the carrier of PT,metr); let FX; assume that A4:FX is_a_cover_of PT and A5: FX is open; consider R such that A6: R well_orders FX by WELLORD2:26; consider Mn such that A7: Mn = R |_2 FX; A8: Mn well_orders FX by A6,A7,Th5; set UB = {union {Ball(c,1/2) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V} where V is Subset of PM: V in FX}; UB is Element of bool bool the carrier of PM proof reconsider UB as set; UB c= bool the carrier of PM proof let x be set; assume x in UB; then consider V be Subset of PM such that A9: x = union {Ball(c,1/2) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V} and V in FX; x c= the carrier of PM proof let y be set; assume y in x; then consider W be set such that A10: y in W and A11: W in {Ball(c,1/2) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V} by A9,TARSKI:def 4; consider c be Element of PM such that A12: W = Ball(c,1/2) and c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V by A11; thus y in the carrier of PM by A10,A12; end; hence thesis; end; hence thesis; end; then reconsider UB as Element of bool bool the carrier of PM; deffunc F1(Element of PM,Nat) = Ball($1,1/(2 |^($2+1))); defpred Q1[Element of PM, Element of bool the carrier of PM,Nat] means $1 in $2\PartUnion($2,Mn) & Ball($1,3/(2 |^ ($3+1))) c= $2; defpred P1[set] means $1 in FX; consider f being Function of NAT, bool bool the carrier of PM such that A13: f.0 = UB and A14: for n holds f.(n+1) = {union { F1(c,n) where c is Element of PM: Q1[c,V,n] & not c in union{union (f.q): q <= n } } where V is Element of bool the carrier of PM: P1[V]} from XXX; defpred P2[set] means ex n st $1 in f.n; consider GX being Subset-Family of PM such that A15: for X being Subset of PM holds X in GX iff P2[X] from SubFamEx; GX is Subset-Family of PT by A1,A3,Th8; then reconsider GX as Subset-Family of PT; take GX; thus A16: GX is open proof for X being Subset of PT st X in GX holds X is open proof let X be Subset of PT; assume A17: X in GX; then reconsider X as Subset of PM; consider n such that A18: X in f.n by A15,A17; now per cases by NAT_1:19; suppose A19: n=0; thus X in the topology of PT proof consider V be Subset of PM such that A20: X = union {Ball(c,1/2) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V} and V in FX by A13,A18,A19; set NF = {Ball(c,1/2) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V}; NF c= bool the carrier of PM proof let a be set; assume a in NF; then consider c be Element of PM such that A21: a = Ball(c,1/2) and c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V; thus a in bool the carrier of PM by A21; end; then reconsider NF as Subset-Family of PM by SETFAM_1:def 7; NF c= Family_open_set(PM) proof let a be set; assume a in NF; then consider c be Element of PM such that A22: a = Ball(c,1/2) and c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V; thus a in Family_open_set(PM) by A22,PCOMPS_1:33; end; hence thesis by A2,A3,A20,PCOMPS_1:36; end; suppose n>0; then consider k such that A23: n = k + 1 by NAT_1:22; thus X in the topology of PT proof X in {union { Ball(c,1/(2 |^ (k+1))) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/(2 |^ (k+1))) c= V & not c in union{union(f.q) where q is Nat: q <= k } } where V is Element of bool the carrier of PM: V in FX} by A14,A18,A23; then consider V be Element of bool the carrier of PM such that A24: X = union { Ball(c,1/(2 |^ (k+1))) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/(2 |^ (k+1))) c= V & not c in union{union(f.q) where q is Nat: q <= k } } and V in FX; reconsider NF = { Ball(c,1/(2 |^ (k+1))) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/(2 |^ (k+1))) c= V & not c in union { union(f.q) where q is Nat: q <= k } } as set; NF c= bool the carrier of PM proof let a be set; assume a in NF; then consider c be Element of PM such that A25: a = Ball(c,1/(2 |^ (k+1))) and c in V\PartUnion(V,Mn) & Ball(c,3/(2 |^ (k+1))) c= V & not c in union { union(f.l): l <= k}; thus a in bool the carrier of PM by A25; end; then reconsider NF as Subset-Family of PM by SETFAM_1:def 7; NF c= Family_open_set(PM) proof let a be set; assume a in NF; then consider c be Element of PM such that A26: a = Ball(c,1/(2 |^ (k+1))) and c in V\PartUnion(V,Mn) & Ball(c,3/(2 |^ (k+1))) c= V & not c in union{union(f.l): l <= k}; thus a in Family_open_set(PM) by A26,PCOMPS_1:33; end; hence thesis by A2,A3,A24,PCOMPS_1:36; end; end; hence thesis by PRE_TOPC:def 5; end; hence thesis by TOPS_2:def 1; end; thus A27: GX is_a_cover_of PT proof A28: [#](PT) c= union GX proof let x be set; defpred P1[set] means x in $1; assume A29: x in [#](PT); then ex G be Subset of PT st x in G & G in FX by A4,PCOMPS_1:5; then A30: ex G be set st G in FX & P1[G]; consider X be set such that A31: X in FX and A32: P1[X] and A33: for Y be set st Y in FX & P1[Y] holds [X,Y] in Mn from MinSet(A8,A30); reconsider X as Subset of PT by A31; X is open by A5,A31,TOPS_2:def 1; then A34: X in Family_open_set(PM) by A2,A3,PRE_TOPC:def 5; reconsider X as Subset of PM by A1,A3,Th8; reconsider x'=x as Element of PM by A1,A3,A29,Th8; consider r be Real such that A35: r>0 and A36: Ball(x',r) c= X by A32,A34,PCOMPS_1:def 5; defpred P2[Nat] means 3/(2 |^ $1) <= r; A37: ex k be Nat st P2[k] by A35,Th3; consider k such that A38: P2[k] and for l st P2[l] holds k <= l from Min(A37); A39: 2 |^ k > 0 by PREPOWER:13; 2 |^ (k+1) = 2 |^ k * 2 by NEWTON:11; then 2 |^ (k+1) >= 2 |^ k by A39,REAL_2:146; then 3/2 |^ (k+1) <= 3/2 |^ k by A39,REAL_2:201; then A40: 3/2 |^ (k+1) <= r by A38,AXIOMS:22; assume A41: not x in union GX; A42: for V be set,n st V in f.n holds not x in V proof let V be set; let n; assume V in f.n; then V in GX by A15; hence thesis by A41,TARSKI:def 4; end; A43: for n holds not x in union (f.n) proof let n; assume x in union (f.n); then consider V be set such that A44: x in V & V in f.n by TARSKI:def 4; thus contradiction by A42,A44; end; now set W = union{ Ball(y,1/(2 |^ (k+1))) where y is Element of PM: y in X\PartUnion(X,Mn) & Ball(y,3/(2 |^ (k+1))) c= X & not y in union{ union(f.q) where q is Nat: q <= k} }; A45: x in W proof consider A be Subset of PM such that A46: A = Ball(x',1/(2 |^ (k+1))); 0 < 2 |^ (k+1) by PREPOWER:13; then 0 < 1/(2 |^ (k+1)) by REAL_2:127; then A47: x in A by A46,TBSP_1:16; not x' in PartUnion(X,Mn) proof assume x' in PartUnion(X,Mn); then x' in union (Mn-Seg(X)) by Def1; then consider M be set such that A48: x' in M and A49: M in Mn-Seg(X) by TARSKI:def 4; reconsider FX as set; A50: M <> X & [M,X] in Mn by A49,WELLORD1:def 1; then M in field Mn by RELAT_1:30; then A51: M in FX by A6,A7,Th5; then A52: [X,M] in Mn by A33,A48; Mn is_antisymmetric_in FX by A8,WELLORD1:def 5; hence contradiction by A31,A50,A51,A52,RELAT_2:def 4 ; end; then A53: x' in X\PartUnion(X,Mn) by A32,XBOOLE_0:def 4; Ball(x',3/(2 |^ (k+1))) c= Ball(x',r) by A40,PCOMPS_1:1; then A54: Ball(x',3/(2 |^ (k+1))) c= X by A36,XBOOLE_1:1; not x' in union { union(f.q) where q is Nat: q <= k} proof assume x' in union { union(f.q) where q is Nat: q <= k}; then consider D be set such that A55: x' in D and A56: D in { union(f.q) where q is Nat: q <= k} by TARSKI:def 4; consider q be Nat such that A57: D = union (f.q) and q <= k by A56; thus contradiction by A43,A55,A57; end; then A in { Ball(y,1/(2 |^ (k+1))) where y is Element of PM: y in X\PartUnion(X,Mn) & Ball(y,3/(2 |^ (k+1))) c= X & not y in union { union(f.q) where q is Nat: q <= k}} by A46,A53,A54; hence thesis by A47,TARSKI:def 4; end; reconsider W as set; W in {union{ Ball(y,1/(2 |^ (k+1))) where y is Element of PM: y in V\PartUnion(V,Mn) & Ball(y,3/(2 |^ (k+1))) c= V & not y in union { union(f.q) where q is Nat: q <= k}} where V is Element of bool the carrier of PM: V in FX} by A31; then W in f.(k+1) by A14; hence ex W be set,l be Nat st W in f.l & x in W by A45; end; hence contradiction by A42; end; union GX c= [#](PT) proof let x be set; assume x in union GX; then consider X be set such that A58: x in X and A59: X in GX by TARSKI:def 4; reconsider x'=x as Point of PT by A58,A59; x' in the carrier of PT; hence x in [#](PT) by PRE_TOPC:12; end; then [#](PT) = union GX by A28,XBOOLE_0:def 10; hence thesis by PRE_TOPC:def 8; end; thus GX is_finer_than FX proof for X be set st X in GX ex Y be set st Y in FX & X c= Y proof let X be set; assume A60: X in GX; then reconsider X as Subset of PM; consider n such that A61: X in f.n by A15,A60; now per cases by NAT_1:19; suppose A62: n=0; thus ex Y be set st Y in FX & X c= Y proof consider V be Subset of PM such that A63: X = union {Ball(c,1/2) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V} and A64: V in FX by A13,A61,A62; set NF = {Ball(c,1/2) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V}; NF c= bool the carrier of PM proof let a be set; assume a in NF; then consider c be Element of PM such that A65: a = Ball(c,1/2) and c in V\PartUnion(V,Mn) & Ball(c,3/2) c= V; thus a in bool the carrier of PM by A65; end; then reconsider NF as Subset-Family of PM by SETFAM_1:def 7; A66: for W be set st W in NF holds W c= V proof let W be set; assume W in NF; then consider c be Element of PM such that A67: W = Ball(c,1/2) & c in V\PartUnion(V,Mn) and A68: Ball(c,3/2) c= V; Ball(c,1/2) c= Ball(c,3/2) by PCOMPS_1:1; hence thesis by A67,A68,XBOOLE_1:1; end; reconsider V as set; take Y = V; thus Y in FX by A64; thus X c= Y by A63,A66,ZFMISC_1:94; end; suppose n>0; then consider k such that A69: n = k + 1 by NAT_1:22; thus ex Y be set st Y in FX & X c= Y proof X in {union { Ball(c,1/(2 |^ (k+1))) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/(2 |^ (k+1))) c= V & not c in union { union(f.q) where q is Nat: q <= k}} where V is Element of bool the carrier of PM: V in FX} by A14,A61,A69; then consider V be Element of bool the carrier of PM such that A70: X = union { Ball(c,1/(2 |^ (k+1))) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/(2 |^ (k+1))) c= V & not c in union{union(f.q) where q is Nat: q <= k}} and A71: V in FX; reconsider NF = { Ball(c,1/(2 |^ (k+1))) where c is Element of PM: c in V\PartUnion(V,Mn) & Ball(c,3/(2 |^ (k+1))) c= V & not c in union{union(f.q) where q is Nat: q <= k}} as set; NF c= bool the carrier of PM proof let a be set; assume a in NF; then consider c be Element of PM such that A72: a = Ball(c,1/(2 |^ (k+1))) and c in V\PartUnion(V,Mn) & Ball(c,3/(2 |^ (k+1))) c= V & not c in union { union(f.q) where q is Nat: q <= k}; thus a in bool the carrier of PM by A72; end; then reconsider NF as Subset-Family of PM by SETFAM_1:def 7; A73: for W be set st W in NF holds W c= V proof let W be set; assume W in NF; then consider c be Element of PM such that A74: W = Ball(c,1/(2 |^ (k+1))) & c in V\PartUnion(V,Mn) and A75: Ball(c,3/(2 |^ (k+1))) c= V & not c in union { union(f.q) where q is Nat: q <= k}; 0 < 2 |^ (k+1) by PREPOWER:13; then 1/(2 |^ (k+1)) <= 3/(2 |^ (k+1)) by REAL_1:73; then Ball(c,1/(2 |^ (k+1))) c= Ball(c,3/(2 |^ (k+1))) by PCOMPS_1:1; hence thesis by A74,A75,XBOOLE_1:1; end; reconsider V as set; take Y = V; thus Y in FX by A71; thus X c= Y by A70,A73,ZFMISC_1:94; end; end; hence thesis; end; hence thesis by SETFAM_1:def 2; end; thus GX is locally_finite proof for x be Point of PT ex W be Subset of PT st x in W & W is open & { V : V in GX & V meets W } is finite proof let x be Point of PT; consider X be Subset of PT such that A76: x in X and A77: X in GX by A27,PCOMPS_1:5; reconsider X as Subset of PT; defpred P3[set] means X in f.$1; A78: ex n st P3[n] by A15,A77; consider k such that A79: P3[k] and for l st P3[l] holds k <= l from Min(A78); X is open by A16,A77,TOPS_2:def 1; then A80: X in Family_open_set(PM) by A2,A3,PRE_TOPC:def 5; reconsider x'=x as Element of PM by A1,A3,Th8; consider r be Real such that A81: r>0 and A82: Ball(x',r) c= X by A76,A80,PCOMPS_1:def 5; consider m be Nat such that A83: 1/(2 |^ m) <= r by A81,Th3; 2 |^ (m+1+k+1) > 0 by PREPOWER:13; then A84: 1/(2 |^ (m+1+k+1)) > 0 by REAL_2:127; consider W be Subset of PM such that A85: W = Ball(x',1/(2 |^ (m+1+k+1))); A86:x in W by A84,A85,TBSP_1:16; W is Subset of PT by A1,A3,Th8; then reconsider W as Subset of PT; W in the topology of PT by A2,A3,A85,PCOMPS_1:33; then A87:W is open by PRE_TOPC:def 5; { V : V in GX & V meets W } is finite proof set NZ={ V : V in GX & V meets W }; defpred P4[set,set] means $1 in f.$2; A88: for p be set st p in NZ ex n st P4[p,n] proof let p be set; assume p in NZ; then consider V be Subset of PT such that A89: p = V and A90: V in GX and V meets W; consider k be Nat such that A91: V in f.k by A15,A90; thus thesis by A89,A91; end; consider g be Function such that A92: dom g = NZ and A93: for y be set st y in NZ ex n st g.y=n & P4[y,n] & for t be Nat st P4[y,t] holds n <=t from FuncExOfMinNat(A88); A94: rng g c= {i: i < (m+1+k+1)} proof let t be set; assume t in rng g; then consider a be set such that A95: a in dom g and A96: t = g.a by FUNCT_1:def 5; consider n be Nat such that A97: g.a = n and A98:a in f.n and for p be Nat st a in f.p holds n <= p by A92,A93,A95; assume A99: not t in {i: i < (m+1+k+1)}; consider V such that A100:a=V and V in GX and A101: V meets W by A92,A95; reconsider t as Nat by A96,A97; A102:t >= (m+1+k+1) by A99; consider y being set such that A103: y in V & y in W by A101,XBOOLE_0:3; now per cases by NAT_1:19; suppose A104: t=0; m+1+k+1 >= 1 by NAT_1:29; hence contradiction by A102,A104,AXIOMS:22; suppose t>0; then consider l be Nat such that A105: t=l+1 by NAT_1:22; V in {union { Ball(c,1/(2 |^ (l+1))) where c is Element of PM: c in Y\PartUnion(Y,Mn) & Ball(c,3/(2 |^ (l+1))) c= Y & not c in union{union(f.q) where q is Nat: q <= l}} where Y is Element of bool the carrier of PM: Y in FX} by A14,A96,A97,A98,A100,A105; then consider Y be Element of bool the carrier of PM such that A106: V = union { Ball(c,1/(2 |^ (l+1))) where c is Element of PM: c in Y\PartUnion(Y,Mn) & Ball(c,3/(2 |^ (l+1))) c= Y & not c in union{union(f.q) where q is Nat: q <= l}} and Y in FX; reconsider NF = { Ball(c,1/(2 |^ (l+1))) where c is Element of PM: c in Y\PartUnion(Y,Mn) & Ball(c,3/(2 |^ (l+1))) c= Y & not c in union{union(f.q) where q is Nat: q <= l}} as set; consider T be set such that A107: y in T and A108: T in NF by A103,A106,TARSKI:def 4; consider c be Element of PM such that A109: T = Ball(c,1/(2 |^ (l+1))) and A110: c in Y\PartUnion(Y,Mn) & Ball(c,3/(2 |^ (l+1))) c= Y & not c in union{union (f.q) where q is Nat: q <= l} by A108; reconsider y as Element of PM by A103; A111:2 |^ t >= 2 |^ (m+1+k+1) by A102,Th4; A112:2 |^ (m+1+k+1) > 0 by PREPOWER:13; then A113: 1/(2 |^ (m+1+k+1)) >= 1/(2 |^ t) by A111,REAL_2:201; 2 |^ (1+k+1) <> 0 by PREPOWER:12; then A114:1/(2 |^ m) - 1/(2 |^ (m+1+k+1)) = (1*(2 |^ (1+k+1)))/((2 |^ m)*(2 |^ (1+k+1))) - 1/(2 |^ (m+1+k+1)) by XCMPLX_1:92 .= (1*(2 |^ (1+k+1)))/(2 |^ (m+((1+k)+1))) - 1/(2 |^ (m+1+k+1)) by NEWTON:13 .= (1*(2 |^ (1+k+1)))/(2 |^ ((m+(1+k))+1)) - 1/(2 |^ (m+1+k+1)) by XCMPLX_1:1 .= 2 |^ (1+k+1)/(2 |^ (m+1+k+1)) - 1/(2 |^ (m+1+k+1)) by XCMPLX_1:1 .= ((2 |^ (1+k))*2)/(2 |^ (m+1+k+1)) - 1/(2 |^ (m+1+k+1)) by NEWTON:11 .= (((2 |^ (1+k))*2)-1)/(2 |^ (m+1+k+1)) by XCMPLX_1:121; (2 |^ (1+k)) >= 1 by PREPOWER:19; then (2 |^ (1+k))*2>=2 by REAL_2:146; then ((2 |^ (1+k))*2)-1>=2-1 by REAL_1:49; then A115: 1/(2 |^ m) - 1/(2 |^ (m+1+k+1)) >= 1/(2 |^ (m+1+k+1)) by A112,A114, REAL_1:73; A116: for t be Nat st t < l holds not c in union(f.t) proof let t be Nat; assume A117: t < l; assume A118: c in union(f.t); union(f.t) in {union(f.q) where q is Nat: q <= l} by A117; hence contradiction by A110,A118,TARSKI:def 4; end; dist(c,y) < 1/(2 |^ t) by A105,A107,A109,METRIC_1:12; then dist(c,y) < 1/(2 |^ (m+1+k+1)) by A113,AXIOMS:22; then A119: dist(c,y) + dist(y,x') < 1/(2 |^ (m+1+k+1)) + dist(y,x') by REAL_1:53; A120: dist(c,x') >= 1/(2 |^ m) proof assume not dist(c,x') >= 1/(2 |^ m); then dist(x',c) < r by A83,AXIOMS:22; then c in Ball(x',r) by METRIC_1:12; then A121: c in union (f.k) by A79,A82,TARSKI:def 4; A122: k <> l proof assume k=l; then union (f.k) in {union(f.q) where q is Nat: q <= l}; hence contradiction by A110,A121,TARSKI:def 4; end;A123:l >= k+(m+1) by A102,A105,REAL_1:53; k+(m+1)>=k by NAT_1:29; then k <= l by A123,AXIOMS:22; then k < l by A122,REAL_1:def 5; hence contradiction by A116,A121; end; dist(c,y) + dist(y,x') >= dist(c,x') by METRIC_1:4; then dist(c,y) + dist(y,x') >= 1/(2 |^ m) by A120,AXIOMS:22; then 1/(2 |^ (m+1+k+1)) + dist(y,x') > 1/(2 |^ m) by A119,AXIOMS:22; then dist(y,x') >= 1/(2 |^ m) - 1/(2 |^ (m+1+k+1)) by REAL_1:84; then dist(y,x') >= 1/(2 |^ (m+1+k+1)) by A115,AXIOMS:22; hence contradiction by A85,A103,METRIC_1:12; end; hence contradiction; end; g is one-to-one proof for x1,x2 be set st x1 in dom g & x2 in dom g & g.x1 = g.x2 holds x1 = x2 proof let x1,x2 be set; assume that A124: x1 in dom g and A125: x2 in dom g and A126: g.x1 = g.x2; assume A127: x1<>x2; consider n1 be Nat such that A128: g.x1 = n1 and A129: x1 in f.n1 and for t be Nat st x1 in f.t holds n1 <= t by A92,A93,A124; consider n2 be Nat such that A130: g.x2 = n2 and A131: x2 in f.n2 and for t be Nat st x2 in f.t holds n2 <= t by A92,A93,A125; consider V1 be Subset of PT such that A132: x1=V1 and V1 in GX and A133: V1 meets W by A92,A124; consider w1 being set such that A134: w1 in W and A135: w1 in x1 by A132,A133,XBOOLE_0:3; consider V2 be Subset of PT such that A136: x2=V2 and V2 in GX and A137: V2 meets W by A92,A125; consider w2 being set such that A138: w2 in W and A139: w2 in x2 by A136,A137,XBOOLE_0:3; reconsider w1, w2 as Element of PM by A134,A138; now per cases by NAT_1:19; suppose A140: n1=0; then consider M1 be Subset of PM such that A141: x1 = union {Ball(c,1/2) where c is Element of PM: c in M1\PartUnion(M1,Mn) & Ball(c,3/2) c= M1} and A142: M1 in FX by A13,A129; consider k1 be set such that A143: w1 in k1 and A144: k1 in {Ball(c,1/2) where c is Element of PM: c in M1\PartUnion(M1,Mn) & Ball(c,3/2) c= M1} by A135,A141,TARSKI:def 4; consider c1 be Element of PM such that A145: k1 = Ball(c1,1/2) and A146: c1 in M1\PartUnion(M1,Mn) and A147: Ball(c1,3/2) c= M1 by A144; consider M2 be Subset of PM such that A148: x2 = union {Ball(c,1/2) where c is Element of PM: c in M2\PartUnion(M2,Mn) & Ball(c,3/2) c= M2} and A149: M2 in FX by A13,A126,A128,A130,A131,A140; consider k2 be set such that A150: w2 in k2 and A151: k2 in {Ball(c,1/2) where c is Element of PM: c in M2\PartUnion(M2,Mn) & Ball(c,3/2) c= M2} by A139,A148,TARSKI:def 4; consider c2 be Element of PM such that A152: k2 = Ball(c2,1/2) and A153: c2 in M2\PartUnion(M2,Mn) and A154: Ball(c2,3/2) c= M2 by A151; A155: dist(x',c2) <= dist(x',w2) + dist(w2,c2) by METRIC_1:4; dist(c1,x') <= dist(c1,w1) + dist(w1,x') by METRIC_1:4; then A156: dist(c1,x') + dist(x',c2) <= (dist(c1,w1) + dist(w1,x')) + (dist(x',w2) + dist(w2,c2)) by A155,REAL_1:55; dist(c1,c2) <= dist(c1,x') + dist(x',c2) by METRIC_1:4; then A157: dist(c1,c2) <= (dist(c1,w1) + dist(w1,x')) + (dist(x',w2) + dist(w2,c2)) by A156,AXIOMS:22; A158: dist(c1,w1) < 1/2 by A143,A145,METRIC_1:12; A159: dist(x',w1) < 1/(2 |^ (m+1+k+1)) by A85,A134,METRIC_1:12; A160: dist(x',w2) < 1/(2 |^ (m+1+k+1)) by A85,A138,METRIC_1:12; A161: dist(c2,w2) < 1/2 by A150,A152,METRIC_1:12; dist(c1,c2) <= dist(c1,w1) + (dist(w1,x') + (dist(x',w2) + dist(w2,c2))) by A157,XCMPLX_1:1; then dist(c1,c2) - (dist(w1,x') + (dist(x',w2) + dist(w2,c2))) <= dist(c1,w1) by REAL_1:86; then dist(c1,c2) - (dist(w1,x') + (dist(x',w2) + dist(w2,c2))) < 1/2 by A158,AXIOMS:22; then dist(c1,c2) < 1/2 + (dist(w1,x') + (dist(x',w2) + dist(w2,c2))) by REAL_1:84; then dist(c1,c2) < dist(w1,x') + (1/2 + (dist(x',w2) + dist(w2,c2))) by XCMPLX_1:1; then dist(c1,c2) - (1/2 + (dist(x',w2) + dist(w2,c2))) < dist(w1,x') by REAL_1:84; then dist(c1,c2) - (1/2 + (dist(x',w2) + dist(w2,c2))) < 1/(2 |^ (m+1+k+1)) by A159,AXIOMS:22; then dist(c1,c2) < 1/(2 |^ (m+1+k+1)) + (1/2 + (dist(x',w2) + dist(w2,c2))) by REAL_1:84; then dist(c1,c2) < (dist(x',w2) + dist(w2,c2)) + (1/(2 |^ (m+1+k+1)) + 1/2) by XCMPLX_1:1; then dist(c1,c2) < dist(x',w2) + (dist(w2,c2) + (1/(2 |^ (m+1+k+1)) + 1/2)) by XCMPLX_1:1; then dist(c1,c2) - (dist(w2,c2) + (1/(2 |^ (m+1+k+1)) + 1/2)) < dist(x',w2) by REAL_1:84; then dist(c1,c2) - (dist(w2,c2) + (1/(2 |^ (m+1+k+1)) + 1/2)) < 1/(2 |^ (m+1+k+1)) by A160,AXIOMS:22; then dist(c1,c2) < 1/(2 |^ (m+1+k+1)) + (dist(w2,c2) + (1/(2 |^ (m+1+k+1)) + 1/2)) by REAL_1:84; then dist(c1,c2) < dist(w2,c2) + (1/(2 |^ (m+1+k+1)) + (1/(2 |^ (m+1+k+1)) + 1/2)) by XCMPLX_1:1; then dist(c1,c2) - (1/(2 |^ (m+1+k+1)) + (1/(2 |^ (m+1+k+1)) + 1/2)) < dist(w2,c2) by REAL_1:84; then dist(c1,c2) - (1/(2 |^ (m+1+k+1)) + (1/(2 |^ (m+1+k+1)) + 1/2)) < 1/2 by A161,AXIOMS:22; then dist(c1,c2) < 1/2 + (1/(2 |^ (m+1+k+1)) + (1/(2 |^ (m+1+k+1)) + 1/2)) by REAL_1:84; then A162: dist(c1,c2) < (1/2 + 1/(2 |^ (m+1+k+1))) + (1/(2 |^ (m+1+k+1)) + 1/2) by XCMPLX_1:1; A163: (1/2 + 1/(2 |^ (m+1+k+1))) + (1/(2 |^ (m+1+k+1)) + 1/2) = 1/2 + ((1/2 + 1/(2 |^ (m+1+k+1))) + 1/(2 |^ (m+1+k+1))) by XCMPLX_1:1 .= 1/2 + (1/2 + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (m+1+k+1)))) by XCMPLX_1:1 .= (1/2 + 1/2) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (m+1+k+1))) by XCMPLX_1:1 .= (1+1)/2 + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (m+1+k+1))) .= 2/2 + 2/(2 |^ (m+1+k+1)) by XCMPLX_1:63; m+k+1 >= 1 by NAT_1:29; then A164: m+1+k >= 1 by XCMPLX_1:1; A165: 2/(2 |^ (m+1+k+1)) =(1*2)/(2 |^ (m+1+k)*2) by NEWTON:11 .= 1/(2 |^ (m+1+k)) by XCMPLX_1:92; 2 |^ (m+1+k) >= 2 |^ 1 by A164,Th4; then 2 |^ (m+1+k) >= 2 by NEWTON:10; then A166: 1/(2 |^ (m+1+k)) <= 1/2 by REAL_2:201; (1/2 + 1/(2 |^ (m+1+k+1))) + (1/(2 |^ (m+1+k+1)) + 1/2) - 2/2 = 1/(2 |^ (m+1+k)) by A163,A165,XCMPLX_1:26; then (1/2 + 1/(2 |^ (m+1+k+1))) + (1/(2 |^ (m+1+k+1)) + 1/2) <= 2/2 + 1/2 by A166,REAL_1:86; then A167: dist(c1,c2) < 3/2 by A162,AXIOMS:22; then A168: c2 in Ball(c1,3/2) by METRIC_1:12; A169: c1 in Ball(c2,3/2) by A167,METRIC_1:12; A170: Mn is_connected_in FX by A8,WELLORD1:def 5; A171: M1 <> M2 by A127,A141,A148; now per cases by A142,A149,A170,A171,RELAT_2:def 6; suppose [M1,M2] in Mn; then M1 in Mn-Seg(M2) by A171,WELLORD1:def 1; then c2 in union(Mn-Seg(M2)) by A147,A168,TARSKI:def 4; then c2 in PartUnion(M2,Mn) by Def1; hence contradiction by A153,XBOOLE_0:def 4; suppose [M2,M1] in Mn; then M2 in Mn-Seg(M1) by A171,WELLORD1:def 1; then c1 in union(Mn-Seg(M1)) by A154,A169,TARSKI:def 4; then c1 in PartUnion(M1,Mn) by Def1; hence contradiction by A146,XBOOLE_0:def 4; end; hence contradiction; suppose n1>0; then consider l be Nat such that A172: n1 = l+1 by NAT_1:22; x1 in {union {Ball(c,1/(2 |^ (l+1))) where c is Element of PM: c in M1\PartUnion(M1,Mn) & Ball(c,3/(2 |^ (l+1))) c= M1 & not c in union { union(f.q) where q is Nat: q <= l}} where M1 is Element of bool the carrier of PM: M1 in FX} by A14,A129,A172; then consider M1 be Element of bool the carrier of PM such that A173: x1 = union {Ball(c,1/(2 |^ (l+1))) where c is Element of PM: c in M1\PartUnion(M1,Mn) & Ball(c,3/(2 |^ (l+1))) c= M1 & not c in union { union(f.q) where q is Nat: q <= l}} and A174: M1 in FX; reconsider NF = {Ball(c,1/(2 |^ (l+1))) where c is Element of PM: c in M1\PartUnion(M1,Mn) & Ball(c,3/(2 |^ (l+1))) c= M1 & not c in union { union(f.q) where q is Nat: q <= l}} as set; consider k1 be set such that A175: w1 in k1 and A176: k1 in NF by A135,A173,TARSKI:def 4; consider c1 be Element of PM such that A177: k1 = Ball(c1,1/(2 |^ (l+1))) and A178: c1 in M1\PartUnion(M1,Mn) and A179: Ball(c1,3/(2 |^ (l+1))) c= M1 and not c1 in union { union(f.q) where q is Nat: q <= l} by A176; x2 in {union {Ball(c,1/(2 |^ (l+1))) where c is Element of PM: c in M2\PartUnion(M2,Mn) & Ball(c,3/(2 |^ (l+1))) c= M2 & not c in union { union(f.q) where q is Nat: q <= l}} where M2 is Element of bool the carrier of PM: M2 in FX} by A14,A126,A128,A130,A131,A172; then consider M2 be Element of bool the carrier of PM such that A180: x2 = union {Ball(c,1/(2 |^ (l+1))) where c is Element of PM: c in M2\PartUnion(M2,Mn) & Ball(c,3/(2 |^ (l+1))) c= M2 & not c in union { union(f.q) where q is Nat: q <= l}} and A181: M2 in FX; reconsider NF = {Ball(c,1/(2 |^ (l+1))) where c is Element of PM: c in M2\PartUnion(M2,Mn) & Ball(c,3/(2 |^ (l+1))) c= M2 & not c in union { union(f.q) where q is Nat: q <= l}} as set; consider k2 be set such that A182: w2 in k2 and A183: k2 in NF by A139,A180,TARSKI:def 4; consider c2 be Element of PM such that A184: k2 = Ball(c2,1/(2 |^ (l+1))) and A185: c2 in M2\PartUnion(M2,Mn) and A186: Ball(c2,3/(2 |^ (l+1))) c= M2 and not c2 in union { union(f.q) where q is Nat: q <= l} by A183; A187: dist(x',c2) <= dist(x',w2) + dist(w2,c2) by METRIC_1:4; dist(c1,x') <= dist(c1,w1) + dist(w1,x') by METRIC_1:4; then A188: dist(c1,x') + dist(x',c2) <= (dist(c1,w1) + dist(w1,x')) + (dist(x',w2) + dist(w2,c2)) by A187,REAL_1:55; dist(c1,c2) <= dist(c1,x') + dist(x',c2) by METRIC_1:4; then A189: dist(c1,c2) <= (dist(c1,w1) + dist(w1,x')) + (dist(x',w2) + dist(w2,c2)) by A188,AXIOMS:22; A190: dist(c1,w1) < 1/(2 |^ (l+1)) by A175,A177,METRIC_1:12; A191: dist(x',w1) < 1/(2 |^ (m+1+k+1)) by A85,A134,METRIC_1:12; A192: dist(x',w2) < 1/(2 |^ (m+1+k+1)) by A85,A138,METRIC_1:12; A193: dist(c2,w2) < 1/(2 |^ (l+1)) by A182,A184,METRIC_1:12; dist(c1,c2) <= dist(c1,w1) + (dist(w1,x') + (dist(x',w2) + dist(w2,c2))) by A189,XCMPLX_1:1; then dist(c1,c2) - (dist(w1,x') + (dist(x',w2) + dist(w2,c2))) <= dist(c1,w1) by REAL_1:86; then dist(c1,c2) - (dist(w1,x') + (dist(x',w2) + dist(w2,c2))) < 1/(2 |^ (l+1)) by A190,AXIOMS:22; then dist(c1,c2) < 1/(2 |^ (l+1)) + (dist(w1,x') + (dist(x',w2) + dist(w2,c2))) by REAL_1:84; then dist(c1,c2) < dist(w1,x') + (1/(2 |^ (l+1)) + (dist(x',w2) + dist(w2,c2))) by XCMPLX_1:1; then dist(c1,c2) - (1/(2 |^ (l+1)) + (dist(x',w2) + dist(w2,c2))) < dist(w1,x') by REAL_1:84; then dist(c1,c2) - (1/(2 |^ (l+1)) + (dist(x',w2) + dist(w2,c2))) < 1/(2 |^ (m+1+k+1)) by A191,AXIOMS:22; then dist(c1,c2) < 1/(2 |^ (m+1+k+1)) + (1/(2 |^ (l+1)) + (dist(x',w2) + dist(w2,c2))) by REAL_1:84; then dist(c1,c2) < (dist(x',w2) + dist(w2,c2)) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1))) by XCMPLX_1:1; then dist(c1,c2) < dist(x',w2) + (dist(w2,c2) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1)))) by XCMPLX_1:1; then dist(c1,c2) - (dist(w2,c2) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1)))) < dist(x',w2) by REAL_1:84; then dist(c1,c2) - (dist(w2,c2) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1)))) < 1/(2 |^ (m+1+k+1)) by A192,AXIOMS:22; then dist(c1,c2) < 1/(2 |^ (m+1+k+1)) + (dist(w2,c2) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1)))) by REAL_1:84; then dist(c1,c2) < dist(w2,c2) + (1/(2 |^ (m+1+k+1)) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1)))) by XCMPLX_1:1; then dist(c1,c2) - (1/(2 |^ (m+1+k+1)) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1)))) < dist(w2,c2) by REAL_1:84; then dist(c1,c2) - (1/(2 |^ (m+1+k+1)) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1)))) < 1/(2 |^ (l+1)) by A193,AXIOMS:22; then dist(c1,c2) < 1/(2 |^ (l+1)) + (1/(2 |^ (m+1+k+1)) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1)))) by REAL_1:84; then A194: dist(c1,c2) < (1/(2 |^ (l+1)) + 1/(2 |^ (m+1+k+1))) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1))) by XCMPLX_1:1; A195: (1/(2 |^ (l+1)) + 1/(2 |^ (m+1+k+1))) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1))) = 1/(2 |^ (l+1)) + ((1/(2 |^ (l+1)) + 1/(2 |^ (m+1+k+1))) + 1/(2 |^ (m+1+k+1))) by XCMPLX_1:1 .= 1/(2 |^ (l+1)) + (1/(2 |^ (l+1)) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (m+1+k+1)))) by XCMPLX_1:1 .= (1/(2 |^ (l+1)) + 1/(2 |^ (l+1))) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (m+1+k+1))) by XCMPLX_1:1 .= (1+1)/(2 |^ (l+1)) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (m+1+k+1))) by XCMPLX_1:63 .= 2/(2 |^ (l+1)) + 2/(2 |^ (m+1+k+1)) by XCMPLX_1:63; n1 in rng g by A124,A128,FUNCT_1:def 5; then n1 in {i: i<m+1+k+1} by A94; then consider i such that A196:n1=i & i < m+1+k+1; consider h be Nat such that A197: m+1+k+1 = (l+1) + h by A172,A196,NAT_1:28; h <> 0 by A172,A196,A197; then consider i be Nat such that A198: h = i + 1 by NAT_1:22; A199: 2 |^ (l+1) > 0 by PREPOWER:13; A200: 2/(2 |^ (m+1+k+1)) = 2/(2 |^ (((l+1)+i)+1)) by A197,A198,XCMPLX_1:1 .= (1*2)/(2 |^ ((l+1)+i)*2) by NEWTON:11 .= 1/(2 |^ ((l+1)+i)) by XCMPLX_1:92; (l+1)+i >= l+1 by NAT_1:29; then 2 |^ ((l+1)+i) >= 2 |^ (l+1) by Th4; then A201:1/(2 |^ ((l+1)+i)) <= 1/(2 |^ (l+1)) by A199,REAL_2:201; (1/(2 |^ (l+1)) + 1/(2 |^ (m+1+k+1))) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1))) - 2/(2 |^ (l+1)) = 1/(2 |^ ((l+1)+i)) by A195,A200,XCMPLX_1:26; then (1/(2 |^ (l+1)) + 1/(2 |^ (m+1+k+1))) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1))) <= 2/(2 |^ (l+1)) + 1/(2 |^ (l+1)) by A201,REAL_1:86; then (1/(2 |^ (l+1)) + 1/(2 |^ (m+1+k+1))) + (1/(2 |^ (m+1+k+1)) + 1/(2 |^ (l+1))) <= (2+1)/(2 |^ (l+1)) by XCMPLX_1:63; then A202: dist(c1,c2) < 3/(2 |^ (l+1)) by A194,AXIOMS:22; then A203: c2 in Ball(c1,3/(2 |^ (l+1))) by METRIC_1:12; A204: c1 in Ball(c2,3/(2 |^ (l+1))) by A202,METRIC_1:12; A205: Mn is_connected_in FX by A8,WELLORD1:def 5; A206: M1 <> M2 by A127,A173,A180; now per cases by A174,A181,A205,A206,RELAT_2:def 6; suppose [M1,M2] in Mn; then M1 in Mn-Seg(M2) by A206,WELLORD1:def 1; then c2 in union(Mn-Seg(M2)) by A179,A203,TARSKI:def 4; then c2 in PartUnion(M2,Mn) by Def1; hence contradiction by A185,XBOOLE_0:def 4; suppose [M2,M1] in Mn; then M2 in Mn-Seg(M1) by A206,WELLORD1:def 1; then c1 in union(Mn-Seg(M1)) by A186,A204,TARSKI:def 4; then c1 in PartUnion(M1,Mn) by Def1; hence contradiction by A178,XBOOLE_0:def 4; end; hence contradiction; end; hence contradiction; end; hence thesis by FUNCT_1:def 8; end; then A207: NZ,rng g are_equipotent by A92,WELLORD2:def 4; {i: i < m+1+k+1 } is finite proof {i: i < m+1+k+1 } c= {0} \/ Seg(m+1+k+1) proof let x be set; assume x in {i: i < m+1+k+1}; then consider i be Nat such that A208: x = i and A209: i < (m+1+k+1); reconsider x1=x as Nat by A208; now per cases by NAT_1:19; suppose x1 = 0; hence x1 in {0} or x1 in Seg(m+1+k+1) by TARSKI:def 1; suppose x1 > 0; then consider l be Nat such that A210: x1 = l +1 by NAT_1:22; x1 >= 1 by A210,NAT_1:29; hence x1 in {0} or x1 in Seg(m+1+k+1) by A208,A209,FINSEQ_1:3; end; hence x in {0} \/ Seg(m+1+k+1) by XBOOLE_0:def 2; end; hence thesis by FINSET_1:13; end; then rng g is finite by A94,FINSET_1:13; hence thesis by A207,CARD_1:68; end; hence thesis by A86,A87; end; hence thesis by PCOMPS_1:def 2; end; end; theorem :: Stone Theorem PT is metrizable implies PT is paracompact proof assume PT is metrizable; then for FX being Subset-Family of PT st FX is_a_cover_of PT & FX is open ex GX being Subset-Family of PT st GX is open & GX is_a_cover_of PT & GX is_finer_than FX & GX is locally_finite by Th12; hence thesis by PCOMPS_1:def 4; end;