Generating a matrix using other matrixes

Hi everyone I need to use 3 different permutation matrixes to make a new one here's how I have it so far

n=randi([1 10],1,1);

x1=[randperm(n)];

x2=[randperm(n)];

o=[randperm(n)];

cc=randsample(n,2);

c1=min(cc);

c2=max(cc);

so far as you know I have produced 3 matrixes with same dimension and 2 random number which indicate the places between arrays not arrays themself.

now I need to produce Y with same dimension using x1 x2 and o with these constraints applied on Y :

from c1+1 to c2 I need to have arrays from c1+1 to c2 using x1 (exact same permutation)

from c2 to end I need to have arrays from o which are not in Y so far(exact same permutation)

from 1 to c1 I need to have those arrays from x2 that are not used in Y so far(exact same permutation)

to be more clear Y =[ something from x2 , something from x1 , something from o ]

can anyone help me?

[Michael Haderlein]

Come on, that's your 5th question on the basically same problem. You should start writing code by yourself. I didn't follow all of the previous answers, so it might be that I present now techniques which are inferior to answers already posted.

On your question, I think the solution is

Y=nan(size(x1));
Y(c1+1:c2)=x1(c1+1:c2);
temp_o=o(~ismember(o,Y));
Y(c2+1:end)=temp_o(end-length(Y)+c2+1:end);
temp_x2=x2(~ismember(x2,Y));
Y(1:c1)=temp_x2(1:c1);