Summary updata~

The first is the evil problem D. Finally in from the aura of the shine of a, the price is helpless 35 wrong.

Own two points have been compared to engage in. I am clear on this point, took the code from Ye, to tell the truth for a long time, did not understand.

I have tried all kinds of dog blood methods, a lower_bound, a upper_bound to do, or not

The ... Finally, finally understand. In fact, every time the two points to find strictly less than the number of numbers there (now want to

Come from ye also this idea ah ... ), and then change the boundary to ... 囧

followed by a variety of uncomfortable, first problem G inexplicably compiled the wrong time, resulting in no 1A (very engaged

, G is the binary pressure state, I am very skilled, and the code check for a long time. The dog has gone to the day)

And then again tonight's problem C was data pit for a night, has been WA in 19 (actually also 1a...! Now want to

Come on, it's cool.

KMP the way I suddenly realized. Haha, did not wait to guide me from the master. 囧

Well, not much to say, to write M (listen to Liao said is a water problem < Liao means code only 56 lines >).

---------------------------------------------------------------------

First, it was written last night, M ... I also think more, p^ (r-l+1) also want to every quick power, direct tle on TEST 6

After a while suspected that their algorithm has a problem ... Of course, eventually I still believe my algorithm, first p^ (r-l+1) preprocessing, and then hand over a

I wrote this evening, the idea is clear ... I saw the hint hint to use double hash ... is also not believe data can get stuck to me. Forced WA to a pitch,

Afterwards the double hash was also successful a dropped L.

At this point, there are two questions, together I suspect I game search, as for another tot STRING, haven't seen ...

Summary updata~