Remove duplicates from array, some first and some last

Hi -- I would like to remove duplicate entries from this array, but sometimes keep the first occurance and sometimes the last.

A = [1 0;
     2 0;
     3 1;
     4 2;
     5 3;
     6 3;
     7 3];

I want to identify duplicates in the second column. At the beginning of the array, I would like to keep the last duplicate and at the end, I would like to keep the first, so my output would be:

ans = [2 0;
       3 1;
       4 2;
       5 3];

I am familiar with, e.g.,

[C,ia] = unique(A(:,2),____)

but can't figure out a slick way to get what I want. I can also imagine a brute-force approach that involves looping through the whole array and doing this "manually", but I feel like there should be a much more elegant (read: clever" solution that I am missing.

I have to process many arrays, and there is no obvious pattern to the duplicates -- some arrays have no duplicates, others have many, some at the beginning, some at the end.

Thanks, all.

Matt

EDIT: I think it should be OK to assume that the arrays are sorted on the first column in ascending order, which I think may be helpful.

Answer 1

[the cyclist]

I believe this does what you want:

A = [1 0;
     2 0;
     3 1;
     4 2;
     5 3;
     6 3;
     7 3];

[~, ia_last]  = unique(A(:,2), 'last', 'stable');
[~, ia_first] = unique(A(:,2), 'first','stable');

A_new = A(ia_last(1):ia_first(end),:)

A_new = 4×2

2 0 3 1 4 2 5 3

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

[Moe Szyslak]

Brilliant! Thanks so much for the help.