Easy columns merge question

Hi all, I have an easy merge question.Let's say I have 2 columns that look like this: A = [1 NaN 3 NaN 5 6 ...] and B = [NaN 2 NaN 4 5 6 ...]. Now, I want to merge them so I get C = [1 2 3 4 5 6 ...]. Any ideas? Thanks all!

Answer 1

[Azzi Abdelmalek]

A = [1 NaN 3 NaN 5 6] 
B= [NaN 2 NaN 4 5 6]
C=unique([A B])
C(isnan(C))=[]

[012786534]

Thank you for your great answer but what if I need to keep the NaNs when there are no values and keep the numbers in order? Like say A = [1 NaN NaN 4 5 NaN], B = [1 2 NaN NaN 5 NaN] so that C = [1 2 NaN 4 5 NaN]? Thanks!!!

[Azzi Abdelmalek]

Your original question is different. Please edit your question

[Azzi Abdelmalek]

A = [1 NaN NaN 4 5 NaN], 
B = [1 2 NaN NaN 5 NaN] 
i1=isnan(A)
i2=~isnan(B)
C=A
C(i1&i2)=B(i1&i2)

Answer 2

[Image Analyst]

Try this:

C = A; % Initialize
% Replace any nan's with B
C(isnan(C)) = B(isnan(C))

[Image Analyst]

Will you ever have A and B have different numbers in the same row? If so, which value does C take on: the value from A or the value from B?

[012786534]

Yes, the numbers may be the same. If so, C take the value of A

[Image Analyst]

Then my code works.

It works even if you meant "Yes, the numbers may be different. If so, C take the value of A"

If you wanted B numbers in preference to A, then initialize to B instead of A.

A = [1 NaN NaN 4 5 NaN]
B = [1 2 NaN NaN 5 NaN]
C = A; % Initialize
% Replace any nan's with B
C(isnan(C)) = B(isnan(C))

C =
     1     2   NaN     4     5   NaN