AmirrzwM's blog

By AmirrzwM, 2 years ago, In English

Hello Codeforces!

On Oct/15/2022 17:35 (Moscow time) we will host Codeforces Global Round 23.

It is the fifth round of a 2022 series of Codeforces Global Rounds. The rounds are open and rated for everybody.

The prizes for this round:

  • 30 best participants get a t-shirt.
  • 20 t-shirts are randomly distributed among those with ranks between 31 and 500, inclusive.

The prizes for the 6-round series in 2022:

  • In each round top-100 participants get points according to the table.
  • The final result for each participant is equal to the sum of points he gets in the four rounds he placed the highest.
  • The best 20 participants over all series get sweatshirts and place certificates.

Thanks to XTX, which in 2022 supported the global rounds initiative!

The problems were written and prepared by AmirrzwM, MohammadParsaElahimanesh and napstablook, AquaMoon, Cirno_9baka, mejiamejia, ChthollyNotaSeniorious, SSerxhs, TomiokapEace, SomethingNew, pakhomovee, rembocoder, SirShokoladina.

We would also like to thank:

Round Information:

  • Duration: 2 hours and 15 minutes
  • Number of problems: 7 problems and 1 sub task
  • Score distribution: 500-1000-1000-1500-(2000+1750)-2500-3500
  • There is an interactive problem, so please see the guide of interactive problems if you are not familiar with it.

We have tried our best to write clear problem statements and make strong pretests and we are looking forward to your participation!

UPD: Editorial

Announcement of Codeforces Global Round 23
  • Vote: I like it
  • +269
  • Vote: I do not like it

| Write comment?
»
2 years ago, # |
  Vote: I like it +11 Vote: I do not like it

It clashes with Google Kickstart, can this round be postponed pls? Thanks!

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +27 Vote: I do not like it

    If you solve GK in 2h30min you still have 5min to rest before this contest.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it -9 Vote: I do not like it

      I think, for every contest we should try to solve problems till end of the contest. Sometimes, last problem gets accepted in last minute of the contest.

      If the contest gets postponed, it will be good for everyone.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +48 Vote: I do not like it

    Score Distribution?

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +10 Vote: I do not like it

      Who is here after Kickstart ?

      • »
        »
        »
        »
        2 years ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        Not after; Along with kickstarter! I came at 8:05, solved howmany ever were possible, and went back to kickstarter. Glad to share I cracked a problem in kickstarter in this small time period.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +10 Vote: I do not like it

      It's not bolded (or math mode, I guess?) like it usually is, but it's in the third bullet point of the round information (end of post):

      500-1000-1000-1500-(2000+1750)-2500-3500

      E has the subtask, apparently

»
2 years ago, # |
Rev. 3   Vote: I like it +1 Vote: I do not like it

Hello. Is there any chance that the timing will be adjusted according to this request and this request? In fact, there was much more than just these blogs as I have seen.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Gotta emphasize the message again.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it -20 Vote: I do not like it

      Well, is that indeed hard to postpone the contest by half an hour? I genuinely can't see any problems doing that, rather you will get a few hundred more participants that will be able to finish Google in time and then enjoy this event too. I believe there're a lot of people like me who wish to participate in both contest, yet try their best until the last minute on KickStart.

      • »
        »
        »
        »
        2 years ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        No, you will lose participants from China who wants to go to bed half an hour earlier.

        • »
          »
          »
          »
          »
          2 years ago, # ^ |
          Rev. 4   Vote: I like it -14 Vote: I do not like it

          Yeah, that somewhat does make sense, but going at 9.30 or 10,00 PM is I feel something more negligible than missing either a CF round or not going till the end on Google.
          ed. Actually the region thing is both ended: US guys will for example get 30 mins more of sleep if needed and probs more will join, so can't really judge like that.

»
2 years ago, # |
  Vote: I like it +6 Vote: I do not like it

Hope to become IM!!!

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    completely failed after hacking myself(C)

»
2 years ago, # |
  Vote: I like it +51 Vote: I do not like it

as author of 4 problems I wish you experience great contest ...

»
2 years ago, # |
  Vote: I like it +1 Vote: I do not like it

as a. . . Welcome to our round!

»
2 years ago, # |
Rev. 4   Vote: I like it -77 Vote: I do not like it

Dear authors, Can you please postpone or prepone this round a little bit as the time is significantly coinciding with Leetcode Biweekly contest 89. i.e. Codeforces contest will start just 5 minute after the beginning of the Leetcode biweekly contest 89. Please authors, Can you reschedule this round a little bit . It will be hard for many of us to choose one contest between them. Thank You.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +192 Vote: I do not like it

    It isn't hard to choose, just do the global round and forget about Leetcode.

    It's simple, Codeforces > Leetcode

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it -11 Vote: I do not like it

      what about google kickstart? it's from 12:00 to 15:00 UTC and Global round starts fron 14:30 UTC

»
2 years ago, # |
  Vote: I like it +8 Vote: I do not like it

Global round again Wow! Hoping for a nice round again!

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

I guess you forgot to mention t-shirts for top 30 and 20 random participants who will make into top 500.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Interactive Problem ftw!

»
2 years ago, # |
  Vote: I like it +14 Vote: I do not like it

Can it be postponed or preponed as it's coinciding with google kickstart and we need min. 30 min. Interval break as both of the contests are brain storming & wanna enjoy both the contest!!!

»
2 years ago, # |
  Vote: I like it +14 Vote: I do not like it

Hello! Should I participate in this round? Are the problems good?

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Sure

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +11 Vote: I do not like it

      How would you know?! You didn't even test >:((

      Te tin minte.

      • »
        »
        »
        »
        2 years ago, # ^ |
        Rev. 2   Vote: I like it +33 Vote: I do not like it

        Well, this was partly a joke (nevertheless, to improve, you should participate in rounds as much as possible, so the answer to the first question might be correct unconditionally).

        I can answer more seriously: recently Um_nik has stated the real opinion in the comment section, and despite me upvoting the comment (I liked the interruption of continuous "the great contest ever" comments chain), I've later agreed with dario's point which stated Um_nik's comment improper. Indeed, it's one more step to reveal in the comment section The Problem Statements Mystery.

        That is, I would not like any involved person answering you evenly. One of the solutions is to claim "good contest", but to do this, you don't even need to be a tester. So I've done it.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +6 Vote: I do not like it

    My dear friend, how do you want me to answer you. . .

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +12 Vote: I do not like it

      Probabilistically it would be better if you didn't answer at all

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Hope, I will become pupil after this contest :).

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +1 Vote: I do not like it

    just enjoy and solve problems, do not thinking about too much on rating changes, i believe u will have a better enjoyment ^-^

»
2 years ago, # |
  Vote: I like it +5 Vote: I do not like it

Nice to see AquaMoon in problem-setter's list

»
2 years ago, # |
Rev. 2   Vote: I like it +15 Vote: I do not like it

Hello authors, Can you please postpone the contest by half an hour as Google kickstart will end at 20:30 IST. We will be very very grateful to you.

»
2 years ago, # |
Rev. 2   Vote: I like it +12 Vote: I do not like it

It will be good if this round can be postponed by half an hour, since google kick start happens 150 minutes earlier and last for 3 hours. (sorry the time cannot be changed, then I will do my best to AK the kick start within 2 hours and half)

»
2 years ago, # |
  Vote: I like it +2 Vote: I do not like it

Just a small clarification needed, will there be an update in the contest timings as it clashes with the GOOGLE Kickstart Round G which is supposed to last till 15:00 UTC?

»
2 years ago, # |
  Vote: I like it -40 Vote: I do not like it

Hmm, as I see, no one wrote a comment about clash with Google Kickstart??!!! So I have to write!

Please please please please, postpone this round!!!!!

»
2 years ago, # |
Rev. 2   Vote: I like it 0 Vote: I do not like it

Interactive problem 1
my solution: 176222751
Please help me. what's wrong in my code ?

  • »
    »
    2 years ago, # ^ |
    Rev. 2   Vote: I like it 0 Vote: I do not like it

    your algorithm will take 10^6 queries, and there are only 25 queries allowed. So you need to learn binary search https://mirror.codeforces.com/edu/course/2/lesson/6/1 , or from any other website which you prefer.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      But why wrong answer on test 1. it will take 12 queries. Is there any other reason ?

      • »
        »
        »
        »
        2 years ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        Sorry for the late reply, The problem in your code is that it will not answer correctly if the number to guess is greater than the number you have already set, you have set that to 1 (ans = 1) so every number is greater than 1 so your while loop condition will never be true. So you are just giving 1 as output everytime. So, if you want to go with the algorithm you have, you should set the ans to 1e6+1 (1000001) in this case otherwise (input_number+1) so that it is always less than that and while loop will keep running until you get the right number... Ofcourse in this case you need to do ans-- in the while loop.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

As a tester, I can confirm that the round is well balanced and you'll enjoy the problems.

»
2 years ago, # |
  Vote: I like it +183 Vote: I do not like it
»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

7 problems and 1 sub task. I like it!

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

I'm not familer with global rounds,could someone tell me is it div.1 or div.2 or div.1+div.2 please?

»
2 years ago, # |
  Vote: I like it +1 Vote: I do not like it

Give me some upvote to get out of the negative contribution please ..

»
2 years ago, # |
  Vote: I like it +1 Vote: I do not like it

Hope to become expert after the cotest!

»
2 years ago, # |
  Vote: I like it +3 Vote: I do not like it

»
2 years ago, # |
  Vote: I like it +3 Vote: I do not like it

Hope to back to Master lol

»
2 years ago, # |
  Vote: I like it +17 Vote: I do not like it

Score Distribution?

»
2 years ago, # |
  Vote: I like it -20 Vote: I do not like it

score distribution?

»
2 years ago, # |
  Vote: I like it -14 Vote: I do not like it

why can't i register in the round ?

  • »
    »
    2 years ago, # ^ |
    Rev. 3   Vote: I like it -14 Vote: I do not like it

    https://postimg.cc/xJQHJC6s/4c459b91 really thanks alot my day was ruined i was excited to take part in this contest ( i even think i have registered earlier but maybe i am not remember) ps : its strange now i can register thnx

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      hell, honesly i don't mind you downvoting me but at least explain me what happened

      • »
        »
        »
        »
        2 years ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        You can't register within 5 minutes of the start of the contest. If you miss the normal registration period you have to wait until they start allowing "extra registration" with is usually 10 minutes after the contest has already started

»
2 years ago, # |
  Vote: I like it -72 Vote: I do not like it

Worst global round ever, downvoted.

»
2 years ago, # |
  Vote: I like it +1 Vote: I do not like it

Should beginers try it? What is the level of dificulty?

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    I would say the first 3 problems could all potentially be solved by someone who doesn't have a lot of experience with cp, but "beginner" includes a wide range of very different people.

    Anyways, problems don't hurt, just go ahead and face them.. If you try and find them to be too tough for your level of experience just move on and go back to them another day or perhaps read the editorial once it's available.

»
2 years ago, # |
Rev. 2   Vote: I like it -72 Vote: I do not like it

.

»
2 years ago, # |
Rev. 2   Vote: I like it +1 Vote: I do not like it

Damn in D I just can't find an efficient way to sort each vertex paths and then sort these paths in the parent vertex paths because we can't go in the same visited vertex and once all the vertices of the of the same parent vertex are visited we consider these vertices unvisited (not their children).. I think I know the main idea but can't find a way to code it.

Correct me if I'm wrong

  • »
    »
    2 years ago, # ^ |
    Rev. 2   Vote: I like it 0 Vote: I do not like it

    .

  • »
    »
    2 years ago, # ^ |
    Rev. 2   Vote: I like it 0 Vote: I do not like it

    Well, sorting the paths is easy. The real issue in that approach is 1) the best possible path relies on which paths can be used, requiring 2^|# leaves| states, and 2) it is not clear to me that greedily choosing the best possible path is optimal.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      It should be greedily because once we find all the possible paths we will repeat the same order so we could just sort these values in the array and depending on k%(number of leaves) we will know at which value we should stop

»
2 years ago, # |
Rev. 2   Vote: I like it +4 Vote: I do not like it

D was good problem, yet couldn't solve. How to choose the leftover k%(num_of_leafs) leaves, without breaking the rules(conditions) ?

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Let assume dp(u, k) return best result at u with k and k + 1 paths. Then for each node, we consider x = (k % num_child) best child. And the next best child is the answer for k + 1.176362632

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +4 Vote: I do not like it

    You only will use a child maximum once, so you only push to the parent the best child you didn't used.

»
2 years ago, # |
  Vote: I like it -34 Vote: I do not like it

D is easier, than B

»
2 years ago, # |
Rev. 2   Vote: I like it 0 Vote: I do not like it

E2 when exactly three candidates remain killed me :(

»
2 years ago, # |
  Vote: I like it +9 Vote: I do not like it

Please say that D was DP on trees

»
2 years ago, # |
  Vote: I like it +3 Vote: I do not like it

Is Problem E an IMO problem?

  • »
    »
    2 years ago, # ^ |
    Rev. 2   Vote: I like it +24 Vote: I do not like it

    It's essentially a special case of IMO2012 problem 3, except that this problem has an upper bound on the number of questions. But knowing this didn't help me anything, as in the IMO problem you just have to find a solution for the generalization to only one truthful answer in (k+1) questions (and in the end you can query 2^k + 1 numbers), but then your (mathematical) solution doesn't focus on minimizing the number of questions.

»
2 years ago, # |
  Vote: I like it +3 Vote: I do not like it

Great round! I love it <3

»
2 years ago, # |
Rev. 2   Vote: I like it +1 Vote: I do not like it

F is similar to a problem which I have made in our Online Judge (But this time I didn't solve F TAT).

And can someone help me find out the mistakes in this code ? Thanks a lot.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +18 Vote: I do not like it

    The mistake in your code is that the random numbers that it uses are of poor quality. Specifically, the lowest bit of the numbers returned by rand() has a low period, and adding the previous number and taking the result modulo N (when N is even) doesn't change that. In fact, it is not difficult to make a single test that will break any periodic RNG with a period less than 300000.

    In 176407464, I modified your code to replace rand with mt19937, and it passed.

»
2 years ago, # |
  Vote: I like it +69 Vote: I do not like it

I feel problem E is similar (same?) with BOI 2022 communication, correct me if I'm wrong.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

How to solve E? I was thinking about sending two patterns 1100 and 1010 twice (So e.g. if n=8 the two patterns arrays are [1,2,3,4] and [1,2,5,6]). Then in any case we can reduce the search space by half.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +8 Vote: I do not like it

    For E1, eliminating 1/4 for every 2 queries is fine. I used two intersecting subsets.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +24 Vote: I do not like it

      Though solving n=3 is the hardest part of the task...

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +3 Vote: I do not like it

      Wow, I feel stupid. I had two intersecting subsets as queries drawn on my paper, with various case distinctions about truth and lie etc. but I disregarded elements not in the subsets...

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

losing cyan cuz misread C's statement

i read in each operation it increases 1 to the suffix not i

»
2 years ago, # |
  Vote: I like it +59 Vote: I do not like it

Does anybody else misread C and tried to solve problem when you can increase suffixes just by 1?

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

I feel like I got the correct idea for D but bad implementation led me to a TLE. Is there a more effecient way to pick the optimal paths % child_count results among all children for every parent than simply sorting? Or maybe the intended complexity is actually O(n) and I am missing a detail?

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    I used multiset, but with a memorization (I used vector<map<int, int64_t>> to record).

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +1 Vote: I do not like it

    you can calculate Try(child, k) and Try(child, k+1) at one. This is my submission: 176362632

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      Ahh I thought about returning a pair but did not have enough time to think it through. Yours looks great. Thanks!

»
2 years ago, # |
  Vote: I like it +20 Vote: I do not like it
  • »
    »
    2 years ago, # ^ |
    Rev. 3   Vote: I like it 0 Vote: I do not like it

    My idea was that we prepare two sets, $$$J$$$ and $$$T$$$ (for joking and truth). The idea is that if the latest query was honest, then the target is in $$$T$$$; otherwise, the target is in $$$J$$$. We can initialize them by querying half of the search space, where the half that's consistent with the answer goes to $$$T$$$ and the other half goes to $$$J$$$.

    For each query afterwards, we can split $$$J$$$ into two halves: $$$J_1$$$ and $$$J_2$$$, and also split $$$T$$$ into two halves: $$$T_1$$$ and $$$T_2$$$. Then we query $$$J_1 \cup T_1$$$.

    • If YES: Then $$$J' = T_2$$$ and $$$T' = J_1 \cup T_1$$$. We eliminate $$$J_2$$$ since consecutive answers cannot both be jokes.
    • If NO: Then $$$J' = T_1$$$ and $$$T' = J_2 \cup T_2$$$. We eliminate $$$J_1$$$ since consecutive answers cannot both be jokes.

    Let $$$J'$$$ and $$$T'$$$ be the new $$$J$$$ and $$$T$$$, and repeat. This way, we reduce the search space ($$$J$$$ and $$$T$$$ combined) by 75% in each query.

    EDIT: Upon testing, I realized this doesn't work, since only half of $$$J$$$ is removed this way. The sizes of $$$J'$$$ and $$$T'$$$ are not necessarily the same, so half of $$$J_1$$$ is not necessarily 25% of $$$J \cup T$$$.

  • »
    »
    2 years ago, # ^ |
    Rev. 2   Vote: I like it -8 Vote: I do not like it

    Divide number set $$$S$$$ into $$$a, b, c,$$$ and $$$d$$$,then query1:$$$a∪b$$$,query2:$$$b∪c$$$. Trust the results of two queries,we get a new number set $$$S'=a∪b∪c/a∪b∪d/...$$$(size of $$$3/4|S|$$$).

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

I am curious about the max rating fall on codeforces. It is a black weekend for me.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

please someone tell me, what the wrong with my code for problem D 176366211

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Any corner case for problem D?

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

How to solve C?

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    +1

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      +1

      • »
        »
        »
        »
        2 years ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        You can start from a[1], you want all the numbers after it to be bigger than a[1] so you can just add a[1] to the suffix [2:n]. then go to a[2] and add a[2] to the suffix [3:n] ans so on. Notice that for example adding a[1] to the suffix [2:n] would not affects on the differences between the numbers in the segment [2:n] so it would not affects how you have to deal with them.

»
2 years ago, # |
  Vote: I like it +25 Vote: I do not like it

G is unsuitable for a codeforces contest.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +13 Vote: I do not like it

    Why?

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +21 Vote: I do not like it

      My idea during the contest is to use a segment tree to simulate augmentations in max cost flow. It is simple but requires tedious coding and effort to fit the memory limit. I thought this was the intended solution, but it seems that the intended solution is a much cleaner one, so I take my previous word back.

      • »
        »
        »
        »
        2 years ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        My solution works for any matroid. It is interesting that for this particular matroid there is another solution. Can you please explain how to solve it using flows? Does it work for arbitrary number of colors? What’s the complexity?

        • »
          »
          »
          »
          »
          2 years ago, # ^ |
          Rev. 3   Vote: I like it +23 Vote: I do not like it

          Overview of my solution 176404703:

          Flow Graph.

          Nodes:

          1. source node, sink node
          2. one node per task
          3. one node per topic

          Edges:

          1. Sort the tasks in increasing order of day, and add directed edges from each task to the one after it with cost $$$0$$$ and capacity $$$\infty$$$.
          2. For each day $$$x\in {1,2,\dots,a+b+c}$$$, add one edge from the source to the leftmost task with $$$d_i\ge x$$$ with cost $$$0$$$ and capacity $$$1$$$.
          3. For each task $$$i$$$, add an edge from it to $$$type_i$$$ with cost $$$r_i$$$ and capacity $$$1$$$.
          4. For each topic, an edge to the sink with cost $$$0$$$ and capacity equal to $$$a$$$, $$$b$$$, or $$$c$$$.

          We want to find a maximum cost flow that saturates all edges of type 2. We will do this by adding an augmenting path for each day in decreasing order of day. All augmenting paths start from the source, pass through one or more topic nodes, and then the sink.

          Efficiently adding augmenting paths.

          It suffices to construct a data structure that maintains the following subpaths and their distances:

          • For each topic, the maximum-cost path from the endpoint of the day $$$x$$$ type-2 edge to the topic node using only forward edges of type 1 and a forward edge of type 3.
          • For each pair of topic nodes $$$l$$$ and $$$r$$$, the maximum-cost path from $$$l$$$ to $$$r$$$ using a reverse edge of type 3, backward or forward edges of type 1, and a forward edge of type 3.

          The maximum-cost augmenting path is the union of at most $$$k$$$ of these subpaths, plus a forward edge of type 4. Our data structure should also support updates of the form: Add or remove a range of backward edges of type 1.

          In general, we can use a lazy segment tree to maintain the subpaths. The segment tree should support updates of the form: Add or remove some range of backward edges of type 1. The segment tree uses $$$O(nk^2)$$$ memory and takes $$$O(k^3\log n)$$$ time to process each day $$$x$$$, if there are $$$k$$$ topics in total.

          • »
            »
            »
            »
            »
            »
            2 years ago, # ^ |
              Vote: I like it 0 Vote: I do not like it

            Is there some reason why you process days in decreasing order? Seems like your solution doesn't depend on that.

            Another thing is that theoretically there might be an issue of overlapping backward subpaths, but it can be fixed manually by removing overlapping part of the path, right?

            • »
              »
              »
              »
              »
              »
              »
              2 years ago, # ^ |
              Rev. 2   Vote: I like it +13 Vote: I do not like it

              It makes my implementation simpler (but it's not necessary).

              I believe that if you take the maximum-cost path that passes through the minimum number of topic nodes, then it won't use the same edge twice. This should work because the graph can never contain any positive-cost cycles.

»
2 years ago, # |
  Vote: I like it +1 Vote: I do not like it

why this code got WA. T^T? I want to know WA testcast.. 176378204

»
2 years ago, # |
  Vote: I like it +11 Vote: I do not like it

Garbage pretests on C, lots of hacks and wrong submissions passed.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    It's not about garbage pretests but strong programming debug testing skills IMO

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +1 Vote: I do not like it

      Yeah yeah yeah next time replace the pretests with the samples. That can train your debugging skills so well!

»
2 years ago, # |
  Vote: I like it +1 Vote: I do not like it

Can anyone explain the approach for problem D?

»
2 years ago, # |
  Vote: I like it +90 Vote: I do not like it

Only I think F much easier than E1 and E1 much easier than E2?

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +62 Vote: I do not like it

    Maybe because I am not good at math and interactive.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Anyone explain C ( I thought you can increase by one at each operation therefore I find next greater element for each from last and then made this current element equal to that and push this index min(diff,k) times but it was not that simple)

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +3 Vote: I do not like it

    Each operation on index i only affects the relation between i and i+1, therefore each operation should be dedicated to some consecutive pair of items in which the first one is bigger than the second. Let's create a differences array in which we will store a[i] - a[i + 1] for each 0 <= i < n - 1 that satisfies a[i] > a[i + 1]. Sort this array in an ascending order, and these are your "missions". Now obviously we rather use the earlier operations to satisfy the small missions, so iterate over the missions array and subtract the value of the current operation (starting from the first operation) until you complete it.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      What's wrong if i just print 1,2,....,n? Can you tell me some counter test?

      • »
        »
        »
        »
        2 years ago, # ^ |
        Rev. 2   Vote: I like it 0 Vote: I do not like it
        Testcase
        • »
          »
          »
          »
          »
          2 years ago, # ^ |
            Vote: I like it +1 Vote: I do not like it

          What would be wrong if I print index for every i such that a[i] + index = n+1?

          So for example for array : 1 3 4 2

          Why this answer is wrong? : 4 2 1 3.

          After all the operations , array would be: 5 9 11 12.

          So inversions is zero here. Did I understood question wrong? Jury is giving WA at this TC.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

My Approach to E1 was the following: You ask for set A from [0 to 2 * (remain_size/ 3)] and set B from [remain_size/ 3 to remain_size]. if the responses are different, it means that the number is in the intersection between A and B, if the responses are the same, it means that the number is not in the intersection between A and B. Could someone please tell me what is wrong with this reasoning?

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +5 Vote: I do not like it

    if the responses are different, it means that the number is in the intersection between A and B

    Not necessarily. The number can be in the intersection, with YES being the honest response, and NO being a joke.

    if the responses are the same, it means that the number is not in the intersection between A and B

    Also not necessarily. WLOG suppose the number is in A but not B. Answering YES to both is valid, since the second answer can be a joke. Answering NO to both is also valid, since the first answer can be a joke.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    It sounds like you are counting joking responses as lies (ie valid but inverted) whereas they should just be ignored entirely. YES YES doesn't eliminate anything because it means it's either in A or B without saying anything about the intersection.

»
2 years ago, # |
Rev. 2   Vote: I like it +10 Vote: I do not like it

I think A is not that trivial. You might hesitate if you think a little bit more.

I guess many just reduce to k then use operation 2 once, but that won't work in the corner case "0 1 0" where k=2.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    You don't need 1st operation at all. Just apply 2nd until array become [1] or [0].

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      Suppose n=20 and k = 15. After one operation-2 the array's length is 6 where you cannot apply operation-2 anymore.

      • »
        »
        »
        »
        2 years ago, # ^ |
          Vote: I like it +1 Vote: I do not like it

        Heck, then I misread more than one problem in this contest :^(

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Can't you just do this - 1. If there is atleast one "1", answer is YES 2. Otherwise answer is NO

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      Yes that's still correct, but the actual operations to achieve [1] might need case analysis.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      Yeah, but we are trying to find out why this works.

      • »
        »
        »
        »
        2 years ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        If n > 3 then you can choose any 1 and there will always be two adjacent values not including that value which can be used to reduce n with operation 1 until n == k.

        Then you just have to check that n, k in (2,2), (3,2) and (3,3) are all possible which is easy to do (can all be done without ever needing operation 1).

        • »
          »
          »
          »
          »
          2 years ago, # ^ |
            Vote: I like it 0 Vote: I do not like it

          Yeah, this one also works, thanks for sharing!

  • »
    »
    2 years ago, # ^ |
    Rev. 2   Vote: I like it +3 Vote: I do not like it

    Actually, your approach is correct with single distinction: you should reduce to 2*k-1 and apply 2nd operation twice (like in this corner case you mentioned).

    • »
      »
      »
      2 years ago, # ^ |
      Rev. 2   Vote: I like it 0 Vote: I do not like it

      And if n < 2*k-1 — reduce to k as you supposed initially.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      Yes we can just use operation2 when k=2. k>=3 is the general case.

      • »
        »
        »
        »
        2 years ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        Yeah, I noticed the flaw :)
        Corrected for the general case.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    In my opinion that's the best way a problem A/B can be. Still challenging, so you have to think to come up with a solution and beginners can develop their problem solving skills further than just improving implementation skills, but with a simple solution, so any beginner can solve it.

»
2 years ago, # |
  Vote: I like it +4 Vote: I do not like it

Memory limit exceeded on C, wtf ?

»
2 years ago, # |
  Vote: I like it +8 Vote: I do not like it

How the fuck does taking the difference array and not sorting it pass pretests in C?

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Wrong Ans on test case 46, what? :( 176324763

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

nice contest but little sad that couldn't finish coding D in time , GRs should be given 2:30 hrs I think atleast

»
2 years ago, # |
  Vote: I like it +6 Vote: I do not like it

Is F solvable with Mo?

  • »
    »
    2 years ago, # ^ |
      Vote: I like it -10 Vote: I do not like it

    Maybe hard because of the variety of $$$k$$$

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +10 Vote: I do not like it

    It has updates, so that would be O(N^5/3)

  • »
    »
    2 years ago, # ^ |
    Rev. 4   Vote: I like it +3 Vote: I do not like it

    I coded Mo with updates in $$$O(n^\frac{5}{3})$$$ but I can't make it pass. Maybe it's doable with tweaking the constants enough. The general idea is the following.

    Let's maintain $$$\mathit{cnt}[x]$$$ — the number of occurrences of a value $$$x$$$ and $$$\mathit{cnt}_2[i]$$$ — the amount of values that have $$$\mathit{cnt}[x] = i$$$. That is trivial with Mo with updates.

    How to obtain the answer for the query $$$k$$$ from that? Well, we want to take gcd of all $$$i$$$ such that $$$\mathit{cnt}_2[i]$$$ is non-zero. Then check if it's divisible by $$$k$$$. Obviously, there are $$$O(\sqrt{n})$$$ non-zero values.

    How to use that fact? Well, let's split the values into the ones that occur at least $$$P$$$ times in the entire array at least once throughout the updates. There will be about $$$\frac{n + m}{P}$$$ of them. To find them, maintain a set of values sorted by their $$$\mathit{cnt}$$$ and mark the top ones as big after every update. Must be $$$O(m \cdot \frac{n + m}{P})$$$ in total.

    So, we can check the values of $$$i$$$ from $$$0$$$ to $$$P - 1$$$ naively and then $$$i = \mathit{cnt}[x]$$$ for all big numbers $$$x$$$. That sums up to a $$$O(P + \frac{n + m}{P})$$$ query. So the optimal $$$P$$$ is around $$$\sqrt{n + m}$$$. Since taking consecutive gcds doesn't make your complexity multiply by log, it's still just sqrt.

    So there is a Mo with updates part that is pretty slow complexity-wise and all the other parts that don't exceed sqrt but have a large constant I guess.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      Thank you, got it
      It's good idea to find answer dividing cnt on small and large values, I came up with similar idea, but couldn't implement
      Yeah, my question was more likely "is there some implementation which fits TL" rather than "is it solvable"

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Why does this pass lmao. I have not seen a solution like mine and I'm going to assume I somehow BSed through the problem.

https://mirror.codeforces.com/contest/1746/submission/176328481

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +3 Vote: I do not like it

    It's correct. The number of position where given and sorted differ (which will always be even) is exactly twice the number of swaps needed. If answer is X, first X mismatches would have a[i] = 1 and b[i] = 0, and second X mismatches would have a[i] = 0 and b[i] = 1. These lead to X swaps.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +9 Vote: I do not like it

      Thanks for the explanation. I had that train of thought and decided to go with it, and I wasn't sure if it was true for all cases and if there was a corner case that I was missing. Thanks again.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Because it's correct solution.

»
2 years ago, # |
  Vote: I like it +9 Vote: I do not like it

Weak prestests for C

»
2 years ago, # |
  Vote: I like it +1 Vote: I do not like it

Too many authors make a round tough.

»
2 years ago, # |
  Vote: I like it +12 Vote: I do not like it

Why so many FSTs on C? I guess many small cases were included in pretest.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +26 Vote: I do not like it

    Actually, most fsts were from hacks with really small n (like 6 or 8). They just didnt put all permutations of small size on either pretests or systests.

»
2 years ago, # |
  Vote: I like it +2 Vote: I do not like it

what is the test 47 on problem C ? A lot of people are fallen down in test 47

»
2 years ago, # |
  Vote: I like it +16 Vote: I do not like it

I want to know how the answer of problem E changes.

My solution is: ask $$$82$$$ queries, each query contains $$$\dfrac{n}{2}$$$ numbers, randomly choose from $$$1$$$ to $$$n$$$.

A number $$$y$$$ can be the answer if and only if for every $$$i$$$, it satisfis query $$$i$$$ or query $$$i+1$$$. That is, if the answer is fixed, the possibility of recognize a wrong answer is $$$1-0.75^{81}$$$, It will get correct answer of a single test case with a possibility of $$$\gt 0.99999$$$. But the solution doesn't work on test $$$13$$$ for many times.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +16 Vote: I do not like it

    The grader is adaptive

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +8 Vote: I do not like it

      But I want to know how it works.

      • »
        »
        »
        »
        2 years ago, # ^ |
          Vote: I like it +16 Vote: I do not like it

        The grader probably has two sets: the possible x values if the last answer was a lie and if it wasn't.

»
2 years ago, # |
Rev. 4   Vote: I like it +37 Vote: I do not like it

There are some solutions to E1 can be searched on the internet.

First , Second , Third

Should it be unrated ?

And what happend to F ?

»
2 years ago, # |
  Vote: I like it +13 Vote: I do not like it

Gonna tell my kids that contests used to have strong test cases

»
2 years ago, # |
  Vote: I like it +1 Vote: I do not like it

Really interesting E1-E2 problems!

»
2 years ago, # |
  Vote: I like it +8 Vote: I do not like it

Lmao FSTed both C and D :pepe_cry:

»
2 years ago, # |
  Vote: I like it +3 Vote: I do not like it

Tough round. What is the logic for problem C? I didn't get it, but many others thought they got it but failed system testing.

  • »
    »
    2 years ago, # ^ |
    Rev. 2   Vote: I like it -25 Vote: I do not like it

    .

  • »
    »
    2 years ago, # ^ |
    Rev. 4   Vote: I like it +12 Vote: I do not like it

    If you have a permutation of size n, you can always add to it in a way that all numbers become n+1. For example, for the permutation 3 2 4 1:

    [3, 2, 4, 1] + [2, 3, 1, 4] = [5, 5, 5, 5]

    (this ignores what is added to the suffixes)

    Finally, it is important to notice that the suffix additions won't change the relative order of the numbers in a way that it creates an inversion. The number of inversions will be 0. This way you can construct the answer.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +4 Vote: I do not like it

      Spent over 2 hours on the question and didn't even understand the premise. Thought it was increasing by 1, instead of i.

»
2 years ago, # |
Rev. 2   Vote: I like it -17 Vote: I do not like it

.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

can some1 help me in my B submission getting runtime error in test2 while all cases i can think of are coming right

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    This line here is wrong : if(num0==n || num1==n) cout<<0<<endl;

    look at this test case :

    4
    1111
    

    answer is 3 not 0, another question what is the purpose of these variables : count, ok, temp2, I think they are useless.

»
2 years ago, # |
  Vote: I like it +1 Vote: I do not like it

editorial?

»
2 years ago, # |
  Vote: I like it +4 Vote: I do not like it

I misread ques C I thought that we can increase suffix of a by 1 rather than 'i' T_T

»
2 years ago, # |
  Vote: I like it +134 Vote: I do not like it

Sad Story: I passed pretest of problem F and got WA3 on system test. After system testing, I resubmitted 10 times (with different comments each time) and passed system test each time. What a terrible luck!

These are the accepted submission IDs after system testing: 176386615 176386593 176386555 176386511 176386479 176386450 176386432 176386407 176386386 176386363

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Me too. The same code got AC after ST but failed ST. So sad.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +52 Vote: I do not like it

    Setters should definitely be more careful than this when setting a randomized problem.

    Editorial mentions "chance of failure is around $$$\frac{1}{2^{30}}$$$ so it's safe to submit", but nowhere they take into account the fact that you're answering 300000 queries for 100 test cases, so when you're running your algorithm 30000000 times, it doesn't look so foolproof anymore, does it?

    As soon as I read the problem I realized there was a decent chance someone would fail system tests here, and unfortunately it seems to have happened.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Need testcases for D as the test 2 is quite a big one.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Never waited this eagerly for any editorial.

»
2 years ago, # |
  Vote: I like it +29 Vote: I do not like it

Attention!

Your solution 176348990 for the problem 1746E2 significantly coincides with solutions Tutis/176348990, null_awe/176378653. Such a coincidence is a clear rules violation. Note that unintentional leakage is also a violation. For example, do not use ideone.com with the default settings (public access to your code). If you have conclusive evidence that a coincidence has occurred due to the use of a common source published before the competition, write a comment to post about the round with all the details. More information can be found at http://mirror.codeforces.com/blog/entry/8790. Such violation of the rules may be the reason for blocking your account or other penalties. In case of repeated violations, your account may be blocked.

I copied code from https://oj.uz/submission/592003

»
2 years ago, # |
Rev. 2   Vote: I like it 0 Vote: I do not like it

Can some one please help me why my subission failed for problem C?

In C I print index for every i such that a[i] + index = n+1?

So for example for array : 1 3 4 2

Why this answer is wrong? : 4 2 1 3.

After all the operations , array would be: 5 9 11 12.

So inversions is zero here. Did I misunderstood the question? Jury is giving WA at this TC(Test case 2).

Submission No: 176389069

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    1 3 4 2 -> 1 3 4 3 -> 1 5 6 5 -> 4 8 9 8 -> 4 8 13 12

    as you can see there is one inversion in final array

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Because the array would be: 4 8 13 12

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    If you add 1 to the number in position 4, you are not adding 1 to the 4 in the original array. Following the approach so that all numbers become n+1 (ignoring suffixes) the answer is 3 2 4 1. Here 1 is added to the number in position 3.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

how to solve E1/E2?

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

any hint for today's D? I know the fact that I need I start with 1 and c1 will be k ,then go to its' children and divide their parent's value of C among them in a way that that everyone has C value of parent / child count & (parent / child count) + 1 but in which order should I distribute I am confused about that, any hint ?

  • »
    »
    2 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Since larger C is always better, sort by the difference of two possible target function values of children.

»
2 years ago, # |
  Vote: I like it +6 Vote: I do not like it

can someone explain the solution for problem C

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +4 Vote: I do not like it

    You can start from a[1], you want all the numbers after it to be bigger than a[1] so you can just add a[1] to the suffix [2:n]. then go to a[2] and add a[2] to the suffix [3:n] ans so on. Notice that for example adding a[1] to the suffix [2:n] would not affects on the differences between the numbers in the segment [2:n] so it would not affects how you have to deal with them.

»
2 years ago, # |
  Vote: I like it +3 Vote: I do not like it

I think google kickstart is the common source from where people copied but i want to know proper reason why my code is not accepted for problem A and B and there may be chance that some concept matched with the other but whole code is not matched.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +4 Vote: I do not like it

    You can start from a[1], you want all the numbers after it to be bigger than a[1] so you can just add a[1] to the suffix [2:n]. then go to a[2] and add a[2] to the suffix [3:n] ans so on. Notice that for example adding a[1] to the suffix [2:n] would not affects on the differences between the numbers in the segment [2:n] so it would not affects how you have to deal with them.

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      What I written the code my code is correct because during the contest it is showing accepted but status is skipped after end of contest so i want proper reason why my solution of A and B is skipped.

»
2 years ago, # |
  Vote: I like it +3 Vote: I do not like it

Anyone knows how to solve interactive problems on c#? I got Idleness limit exceeded despite flushing console Submission:176402733

»
2 years ago, # |
  Vote: I like it -22 Vote: I do not like it

Bad pretest. Got FST and dropped to candidate master.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +54 Vote: I do not like it

    Have you tried proving your solutions instead of guessing?

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it -26 Vote: I do not like it

      I didn't, but I realized the mistake I made in the code just after I saw "failed system test". So I think stronger pretest can avoid such kind of sad stories.

      I think what you said have reason. But it can't be an excuse for bad pretests.

»
2 years ago, # |
Rev. 2   Vote: I like it -35 Vote: I do not like it

I'm ikun, Henry Xu please shut up

»
2 years ago, # |
  Vote: I like it +11 Vote: I do not like it

FST for D because the inf I set is too small. LOL.

»
2 years ago, # |
  Vote: I like it 0 Vote: I do not like it

too weak pretest for problem c

»
2 years ago, # |
  Vote: I like it +39 Vote: I do not like it

There is a problem,if someone in the same room of me, and he also solve some problems,then some of his friends want to cheat, he don not wanna take any risk,so he might take picture of my code and send to his friends. Or there is no way for my solution D are similar to 2 people who submit a long time after, that might be a really serious problem

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +24 Vote: I do not like it

    I can't find another way how that 2 people get my solution for the problem D. I do not send mysolution to anyone ever

    • »
      »
      »
      2 years ago, # ^ |
        Vote: I like it +13 Vote: I do not like it

      However no one in your room locked D. It's impossible.

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +21 Vote: I do not like it

    What if this actually happened? What should we do to prove that we're innocent?

  • »
    »
    2 years ago, # ^ |
      Vote: I like it +20 Vote: I do not like it

    It could be a really serious problem,and someone might have done this before.

»
2 years ago, # |
Rev. 2   Vote: I like it -14 Vote: I do not like it

I m surprised that few people didn't thought about two pointer in B. Got ac with two pointer.

»
2 years ago, # |
  Vote: I like it +44 Vote: I do not like it

When/where will the drawing for the t-shirts be announced ? For once I managed to do top 500

»
2 years ago, # |
  Vote: I like it -18 Vote: I do not like it

ao tao dau duma

»
2 years ago, # |
  Vote: I like it +54 Vote: I do not like it

Congratulations to tshirts winners! In a few weeks you will be contacted via private messages with instructions to receive your prize.

As usual, we used the following two scripts for generating random winners, seed is the score of the winner.

get_tshirts.py
randgen.cpp
List place Contest Rank Name
1 1746 1 tourist
2 1746 2 Um_nik
3 1746 3 AoLiGei
4 1746 4 duality
5 1746 5 zh0ukangyang
6 1746 6 jiangly
7 1746 7 DearMargaret
8 1746 8 353cerega
9 1746 9 hanbyeol_
10 1746 10 Sana
11 1746 11 ecnerwala
12 1746 12 WYZFL
13 1746 13 QuietBeautifulThoughts
14 1746 14 ksun48
15 1746 15 Ormlis
16 1746 16 Maksim1744
17 1746 17 Melusine
18 1746 18 noimi
19 1746 19 Benq
20 1746 20 mango_lassi
21 1746 21 3.141592653
22 1746 22 hos.lyric
23 1746 23 Chenyu_Qiu
24 1746 24 hitonanode
25 1746 25 heno239
26 1746 26 244mhq
27 1746 27 maspy
28 1746 28 turmax
29 1746 29 blackbori
30 1746 30 tokusakurai
34 1746 34 S2speed
46 1746 46 Thienu
52 1746 52 He_Ren
68 1746 68 conqueror_of_tourist
95 1746 95 chenxiaoyan
96 1746 96 Isonan
113 1746 113 BalintR
153 1746 152 ztcakioi
159 1746 159 RGB_ICPC1
166 1746 166 zmj2008AKIOI
196 1746 196 frokaikan
217 1746 217 mindino
229 1746 227 lingfunny
269 1746 269 toooooosimple
307 1746 307 PubabaOnO
337 1746 337 IanDeHaan
338 1746 337 mban259
342 1746 340 ToniB
499 1746 499 codelegend
500 1746 499 rniya