ABC165E
https://gyazo.com/433122aabc060c44d595f84db3aff948
Thoughts.
The point is, the difference can't be the same.
No K and N-K either.
I don't know how many assignments would be OK.
M must be less than N/2
Oh, it says > $ M \times 2 + 1 \le N
https://gyazo.com/72626eebb3eb1c72e900d7e5f7afa66a
Now if N is odd, the rest are even, so they're all guaranteed to be at different intervals.
This doesn't work when N is even, the rest of the numbers are also odd and covered.
Looks good if you shift one half.
If M is even, the length taken in the M/2nd and the remaining length match, so we spread 1 there.
After that, they will all be even numbers, so they won't be covered.
I'm sure there's a bug that could cause a discrepancy of 1 or so at the boundary, but it's better to implement it and squash it with a small test case than to think about it in my head.
Official Explanation
Style that creates both odd-length and even-length clumps
The length, not the evenness, prevents overlap.
---
This page is auto-translated from /nishio/ABC165E using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I'm very happy to spread my thought to non-Japanese readers.