Find values in array a based on unique values in array b as well as based on the range of values in b

Hello Everyone,

I have two long arrays that look like this:

a = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17]
b = [10,14,3,21,21,21,21,3,14,14,14,2,75,13,13,13,100]

I need to construct a cell ( c) that will have values in array a that correspond to unique values in b. However, when b has a value that repeats (e.g. 21 above) then the entry in c should be the smallest value and the largest value in a for that value in b. That is, the result should be as follows:

c = {1,2,3,[4,7],8,[9,11],12,13,[14,16],17}

My code is too complicated and it also does not work in all cases. It looks like this:

[~,uniq_fir] = unique(b,'first');
[~,uniq_las] = unique(b,'last');
uniq_share = intersect(uniq_fir,uniq_las);
uniq_fir_dif = setdiff(uniq_fir,uniq_las);
uniq_las_dif = setdiff(uniq_las,uniq_fir);
uniq1 = a(uniq_share);
uniq2 = sort(a(uniq_fir_dif));
uniq3 = sort(a(uniq_las_dif));
c(uniq_share) = num2cell(uniq1);
empt_cell_temp = find(cellfun(@isempty,c));
for i = 1:length(uniq_fir_dif)
   c{empt_cell_temp(i)} = [uniq2(i),uniq3(i)];
end

Please can anybody help me with this problem? Much obliged!

Answer 1

[Stephen23]

A = [ 1, 2, 3, 4, 5, 6, 7,8, 9,10,11,12,13,14,15,16, 17];
B = [10,14, 3,21,21,21,21,3,14,14,14, 2,75,13,13,13,100];
X = find([true,diff(B)~=0,true]);
fun = @(b,e)A(unique([b,e]));
out = arrayfun(fun,X(1:end-1),X(2:end)-1,'uni',0);

and the output:

>> out{:}
ans =  1
ans =  2
ans =  3
ans =
   4   7
ans =  8
ans =
    9   11
ans =  12
ans =  13
ans =
   14   16
ans =  17

[djr]

Thank you so much. So elegant and works perfectly. You saw my messy code that did not work by the way :) Thanks!

Answer 2

[Andrei Bobrov]

out = accumarray(cumsum([true;diff(b(:))] ~= 0),a(:),[],@(x){unique([x(1),x(end)])})