Problem 1103. Right Triangle Side Lengths (Inspired by Project Euler Problem 39)

Well I don't know why I can't pass the test
could anyone help

function c = right_triangle_sides(p)
c = {};
k=1;
for i=1:ceil(p/3)
for j=i:ceil((p-i)/2)
if i^2+j^2==(p-i-j)^2
c(k)=mat2cell([i j p-i-i],1);
k=k+1;
end
end
end
end

To creator of this group,
Please check the test case 3 in the Problem of calculating side length from area of triangle. One ';' is missing which is not allowing the case to pass for correct solution.

This problem really highlights problems with the Cody scoring system. The score using a double for loop to brute force the lengths beats out a single for loop using equations. I tested my code in Matlab vs double loops and it's faster than most double loops I tested by several orders of magnitude for larger numbers (p>1000). There is also some sort of caching going on where my code gets orders of magnitude faster again if ran a few times vs most double for loops.

There is one exception with a double loop that is really fast by Hung Hoang. They use the for loops in a way that doesn't make it O^2.