#codechef

Thank god codechef is the only site that doesn't punish failed submissions, because let me fucking tell you, this was a trip.

N Triplets - Man, fuck this problem. It's not hard, it's literally just "more than one distinct factors smaller than it" but it has a nuance that I didn't pick up on: first, not only does this exclude primes, it also excludes squares of primes. Second, I basically shit the bed implementing it for a while. Forgot to use the sqrt trick, forgot that I only need to find two distinct divisors other than 1, etc etc. Poor performance.

No Sequence - A fun and interesting problem! Unfortunately I once again drove myself crazy trying to figure out why I was getting exceptions before realizing that I'm an idiot and was giving myself overflow errors by not boundary checking my modulus. This is the second time this has happened on codechef, which is why I was able to pick up on it. Then I had to remember to pad with leading zeros. Another bad performance, but at least conceptually I was on top of things.

Good Sequence - This one went a little smoother, thank god. Thought about how to do it, carefully navigated my cases, dealt with all the offsets. One failed submission because I made something a > instead of a >=, but I figured it out and my next sub worked. Felt good about that one.

Terrible start, good middle. The last four problems all feel kind of out of my league - I can explain more or less how to do the fourth problem, but I have a few stumbling blocks in my implementation that I don't think I can figure out how to resolve, and I don't really feel like trying for another hour.

The way div 2 tends to go is there are 3 "easy" problems, 2 hard ones, and 1 impossible one. You have to get all 3 easy, everyone does, and whether you get 0, 1, or 2 hard determines whether you lose a little rating, gain a little rating, or gain a lot of rating. Do not attempt the impossible problem.

Pet Store - The easiest one, make a counter, check that your values are even.

Maximizing LCS - I straight up forgot how to do these problems, so I had to spend a bunch of time remembering how to do this. The solution I came up with was pretty clever, so I'd be happy, except I wrote a bunch of bugs and had to fix it three times. Meh.

Flip the Prefix - Did this one wrong initially and had to rewrite it. Still ended up getting it reasonably quickly tho. There was probably a cleaner way to do it but I did some ugly casing.

Smallest Difference - One of the two hard problems, I looked at it and knew I could find a way to do it. It took a while but solved it on the first time. It involved figuring out first that the optimal solution would always force to either the bottom or the right ASAP (whichever is closer) and then finding the best slice of integers that you could do that with.

Expected Sum - Really feels like I should've been able to get this one, but I didn't have enough time left and spent too much time fucking with my algorithm library. Still need to finish learning how to use my tools and all that. I think I understand it conceptually.

Unique Mode - Impossible, but really interesting. Need to read the editorial for it.