Ok, so the question is kind of awkwardly phrased, but I hope this will clear things up.
I have this sample 2d array.
$array = array(
array(1, 0, 0, 0, 1, 0, 0, 1),
array(0, 0, 1, 1, 1, 1, 0, 1),
array(0, 1, 1, 0, 1, 0, 0, 0),
array(0, 1, 1, 0, 0, 0, 1, 0),
array(1, 0, 0, 0, 1, 1, 1, 1),
array(0, 1, 1, 0, 1, 0, 1, 0),
array(0, 0, 0, 0, 0, 0, 0, 1)
);
When iterated by rows (and terminating each row with \n), and for every row then iterated by column, it will echo something like this: (░░ = 0, ▓▓ = 1)
▓▓░░░░░░▓▓░░░░▓▓
░░░░▓▓▓▓▓▓▓▓░░▓▓
░░▓▓▓▓░░▓▓░░░░░░
░░▓▓▓▓░░░░░░▓▓░░
▓▓░░░░░░▓▓▓▓▓▓▓▓
░░▓▓▓▓░░▓▓░░▓▓░░
░░░░░░░░░░░░░░▓▓
But what I'd like to do is to "analyse" the array and only leave 1 contiguous shape (the one with the most "cells"), in this example, the result would be:
░░░░░░░░▓▓░░░░░░
░░░░▓▓▓▓▓▓▓▓░░░░
░░▓▓▓▓░░▓▓░░░░░░
░░▓▓▓▓░░░░░░░░░░
▓▓░░░░░░░░░░░░░░
░░▓▓▓▓░░░░░░░░░░
░░░░░░░░░░░░░░░░
My initial approach was to:
Assign each ▓▓ cell a unique number (be it completely random, or the current iteration number):
01 02 03 04050607 08 0910 11 1213 14 15 16171819 2021 22 23 24
Iterate through the array many, MANY times: every iteration, each ▓▓ cell assumes the largest unique number among his neighbours. The loop would go on indefinitely until there's no change detected between the current state and the previous state. After the last iteration, the result would be this:
01 21 08 21212121 08 2121 21 2121 24 21 24242424 2121 24 24 24
Now it all comes down to counting the value that occurs the most. Then, iterating once again, to turn all the cells whose value is not the most popular one, to 0, giving me the desired result.
However, I feel it's quite a roundabout and computationally heavy approach for such a simple task and there has to be a better way. Any ideas would be greatly appreciated, cheers!
BONUS POINTS: Divide all the blobs into an array of 2D arrays, ordered by number of cells, so we can do something with the smallest blob, too