A:B with two matrices instead of scalars

I have two matrices a = [1;2;3;4;5;6] and b = [5;6;7;8;9;10].

I want to compute [1:5;2:6;3:7;4:8;5:9;6:10] as a whole matrix. But want it in a for loop so it is not computed manually for each row.

Currently, I have used the code:

for i = [1:6]

      c = [a(i,1):b(i,1)];
  end

This only does it for the last term (ie 6:10) but not the 1st, 2nd, 3rd, 4th and 5th term.

How can it be done like this?

Answer 1

[Walter Roberson]

c = (a(1):b(1)) + (a - a(1));

Requires R2016b or later, and requires that the difference b-a remains constant.

For earlier releases,

c = bsxfun(@plus, (a(1):b(1)) , a-a(1));

Answer 2

[Image Analyst]

Walter gave the more MATLABy way of doing it. If you really want a for loop way of doing it, here is a partially robust way of doing it.

a = [1;2;3;4;5;6] 
b = [5;6;7;8;9;10]
numColumns = max(b-a) + 1 % Assume a and b are integers
maxRows = min([length(a), length(b)])
c = zeros(maxRows, numColumns)
for row = 1 : maxRows  % Assume a and b are the same length, otherwise get error.
  % Assume a(row) : b(row) is the same number of elements for every row,
  % otherwise you'll get an error.
  c(row, :) = a(row) : b(row);
end
c % Report results to command window.

Again, like Walter's solution, it assumes certain things.