let G1 be finite real-weighted WGraph; :: thesis: for n being Nat
for G2 being inducedSubgraph of G1,((PRIM:CompSeq G1) . n) `1 ,((PRIM:CompSeq G1) . n) `2 holds G2 is Tree-like

set PCS = PRIM:CompSeq G1;
defpred S1[ Nat] means for G2 being inducedSubgraph of G1,((PRIM:CompSeq G1) . $1) `1 ,((PRIM:CompSeq G1) . $1) `2 holds G2 is Tree-like ;
set G0 = (PRIM:CompSeq G1) . 0;
set src = choose (the_Vertices_of G1);
now
let n be Nat; :: thesis: ( S1[n] implies S1[n + 1] )
set Gn = (PRIM:CompSeq G1) . n;
set Gn1 = (PRIM:CompSeq G1) . (n + 1);
set Next = PRIM:NextBestEdges ((PRIM:CompSeq G1) . n);
set e = choose (PRIM:NextBestEdges ((PRIM:CompSeq G1) . n));
set G3 = the inducedSubgraph of G1,((PRIM:CompSeq G1) . n) `1 ,((PRIM:CompSeq G1) . n) `2 ;
A1: (PRIM:CompSeq G1) . (n + 1) = PRIM:Step ((PRIM:CompSeq G1) . n) by Def17;
A2: ((PRIM:CompSeq G1) . n) `2 c= G1 .edgesBetween (((PRIM:CompSeq G1) . n) `1) by Th30;
((PRIM:CompSeq G1) . n) `1 is non empty Subset of (the_Vertices_of G1) by Th30;
then A3: ( the_Vertices_of the inducedSubgraph of G1,((PRIM:CompSeq G1) . n) `1 ,((PRIM:CompSeq G1) . n) `2 = ((PRIM:CompSeq G1) . n) `1 & the_Edges_of the inducedSubgraph of G1,((PRIM:CompSeq G1) . n) `1 ,((PRIM:CompSeq G1) . n) `2 = ((PRIM:CompSeq G1) . n) `2 ) by A2, GLIB_000:def 37;
assume A4: S1[n] ; :: thesis: S1[n + 1]
then A5: the inducedSubgraph of G1,((PRIM:CompSeq G1) . n) `1 ,((PRIM:CompSeq G1) . n) `2 is Tree-like ;
now
A6: the inducedSubgraph of G1,((PRIM:CompSeq G1) . n) `1 ,((PRIM:CompSeq G1) . n) `2 .order() = ( the inducedSubgraph of G1,((PRIM:CompSeq G1) . n) `1 ,((PRIM:CompSeq G1) . n) `2 .size()) + 1 by A5, GLIB_002:46;
let G2 be inducedSubgraph of G1,((PRIM:CompSeq G1) . (n + 1)) `1 ,((PRIM:CompSeq G1) . (n + 1)) `2 ; :: thesis: G2 is Tree-like
A7: ((PRIM:CompSeq G1) . (n + 1)) `2 c= G1 .edgesBetween (((PRIM:CompSeq G1) . (n + 1)) `1) by Th30;
((PRIM:CompSeq G1) . (n + 1)) `1 is non empty Subset of (the_Vertices_of G1) by Th30;
then A8: the_Vertices_of G2 = ((PRIM:CompSeq G1) . (n + 1)) `1 by A7, GLIB_000:def 37;
now
per cases ( PRIM:NextBestEdges ((PRIM:CompSeq G1) . n) = {} or PRIM:NextBestEdges ((PRIM:CompSeq G1) . n) <> {} ) ;
suppose A9: PRIM:NextBestEdges ((PRIM:CompSeq G1) . n) <> {} ; :: thesis: G2 is Tree-like
set GnV = ((PRIM:CompSeq G1) . n) `1 ;
set GnVg = (the_Vertices_of G1) \ (((PRIM:CompSeq G1) . n) `1);
A10: choose (PRIM:NextBestEdges ((PRIM:CompSeq G1) . n)) SJoins ((PRIM:CompSeq G1) . n) `1 ,(the_Vertices_of G1) \ (((PRIM:CompSeq G1) . n) `1),G1 by A9, Def13;
consider v being Vertex of G1 such that
A14: not v in ((PRIM:CompSeq G1) . n) `1 and
A15: (PRIM:CompSeq G1) . (n + 1) = [((((PRIM:CompSeq G1) . n) `1) \/ {v}),((((PRIM:CompSeq G1) . n) `2) \/ {(choose (PRIM:NextBestEdges ((PRIM:CompSeq G1) . n)))})] by A1, A9, Th28;
A16: card (((PRIM:CompSeq G1) . (n + 1)) `1) = card ((((PRIM:CompSeq G1) . n) `1) \/ {v}) by A15, MCART_1:7
.= (card (((PRIM:CompSeq G1) . n) `1)) + 1 by A14, CARD_2:41 ;
card (((PRIM:CompSeq G1) . (n + 1)) `2) = card ((((PRIM:CompSeq G1) . n) `2) \/ {(choose (PRIM:NextBestEdges ((PRIM:CompSeq G1) . n)))}) by A15, MCART_1:7
.= (card (((PRIM:CompSeq G1) . n) `2)) + 1 by A11, CARD_2:41 ;
then G2 .order() = (G2 .size()) + 1 by A3, A7, A8, A6, A16, GLIB_000:def 37;
hence G2 is Tree-like by GLIB_002:47; :: thesis: verum
end;
end;
end;
hence G2 is Tree-like ; :: thesis: verum
end;
hence S1[n + 1] ; :: thesis: verum
end;
then A17: for n being Nat st S1[n] holds
S1[n + 1] ;
(PRIM:CompSeq G1) . 0 = PRIM:Init G1 by Def17;
then ( ((PRIM:CompSeq G1) . 0) `1 = {(choose (the_Vertices_of G1))} & ((PRIM:CompSeq G1) . 0) `2 = {} ) by MCART_1:7;
then A18: S1[ 0 ] ;
for n being Nat holds S1[n] from NAT_1:sch 2(A18, A17);
hence for n being Nat
for G2 being inducedSubgraph of G1,((PRIM:CompSeq G1) . n) `1 ,((PRIM:CompSeq G1) . n) `2 holds G2 is Tree-like ; :: thesis: verum