Codeforces Round #620 (Div. 2)

5 years ago, # |

← Rev. 2 →

:v Testers in every ranks !

→ Reply

rony160136rr

5 years ago, # |

← Rev. 2 →

-46

→ Reply

SuperJ6

5 years ago, # ^ |

-6

Alright the time will be changed accordingly as cf revolves around your schedule :p (not rlly)

→ Reply

shubhamcr7

5 years ago, # |

Editorials ?????

→ Reply

5 years ago, # ^ |

+123

Here you go.

→ Reply

shubhamcr7

5 years ago, # ^ |

+21

Ohh!!! I am sorry man, I was looking for Round 619 Div2 editorials and ended up posting my comment here. I just realized this.

→ Reply

AM_I_Learning

5 years ago, # |

-8

Hope B will be clear this time. Past 2 contests Test Cases for B were very hard.

→ Reply

5 years ago, # ^ |

-6

Allways read B before submitting A ;)

→ Reply

INeedAPaper

5 years ago, # |

OwO，well...Korea round is my nightmare,hope pass C

→ Reply

SnowGoose

5 years ago, # |

As one of the lower ranked testers, I would highly recommend participation in this round. These are good problems, there is something for every skill level here!

→ Reply

_T_Bag_

5 years ago, # |

I wish positive rating for every participant :)

→ Reply

M.Ibrahim_

5 years ago, # ^ |

We all wish that but this is not possible.

→ Reply

M.Ibrahim_

5 years ago, # ^ |

The total rating decreases after every contest, otherwise, ratings will become higher and higher because new users will join Codeforces with the initial 1500 rating points.

→ Reply

_overrated_

5 years ago, # |

-26

Round with subtask

→ Reply

leejseo

5 years ago, # |

+13

팬이에요! :fan:

→ Reply

Hyperbolic

5 years ago, # ^ |

Is Codeforces started to support the Korean language? Cool.

→ Reply

2022OVO

5 years ago, # |

Can someone tell me what does one of the problems has 2 subtasks means? thanks in advance.

→ Reply

Artful-Dodger

5 years ago, # ^ |

+13

I'm not quite sure, but it may be an easy version and a hard version of the same problem. Like C1 & C2 for example.

→ Reply

79brue

5 years ago, # ^ |

+20

One of the problems has an easy version and a hard version. You'll know which problem has subtasks after the contest starts.

→ Reply

pls_be_short

5 years ago, # |

-10

I hope the problem statements are short and clear.

→ Reply

Mujahidul_Islam_

5 years ago, # |

+30

Want to be Pupil today.Wish me good luck

→ Reply

MagicSpark

5 years ago, # |

← Rev. 2 →

I wonder what is jh05013's real rating.

Because when he became blue.I can only solve div.2B problems

Maybe he have a real rating of Grandmaster or higher :>

→ Reply

libraqz

5 years ago, # |

I found a really interesting thing that jh05013 and I have both the lowest rating -43 and after that, our ratings both increase 3.

→ Reply

blobugh

5 years ago, # |

← Rev. 2 →

+11

How to solve E?

Edited : Thank you to all who replied. Because I got TLE on pretest 14 during the contest, I'll try again using the faster LCA algorithm.

→ Reply

rkm0959

5 years ago, # ^ |

Hint: The added edge is there for parity reasons.

→ Reply

blobugh

5 years ago, # ^ |

I know that, but I failed to optimize the running time.

→ Reply

5 years ago, # ^ |

Do I need to know "LCA"?

→ Reply

_overrated_

5 years ago, # ^ |

Yes.

→ Reply

5 years ago, # ^ |

+11

In fact,I have never solved any problem about "LCA".I need to learn more!

→ Reply

aryanc403

5 years ago, # ^ |

That was the reason I got when I said E is a standard problem.

300iq : "We thought it is not so standard for div2 participants."

→ Reply

memento

5 years ago, # ^ |

-8

wai- how are you even cm?

→ Reply

wisest

5 years ago, # ^ |

I know parity reasons. but I don't know distance between two node is long than k or not without time limit exceeded.

→ Reply

Jakube

5 years ago, # ^ |

+10

With LCA you can find the distance between two nodes in a tree in $$$O(1)$$$ or $$$O(\log n)$$$ time (depending on which LCA algorithm you use).

→ Reply

5 years ago, # ^ |

I think you need learn "LCA".

→ Reply

wisest

5 years ago, # ^ |

Thank you, I catch a keyword. I'll study LCA algorithm.

→ Reply

5 years ago, # ^ |

I didn't solve this problem as well,because I didn't know how to know the distance between any two nodes quickly.QwQ

→ Reply

danirainy

5 years ago, # ^ |

+12

only three possible ways l = a -> x -> y -> b l = a -> y -> x -> b l = a -> b

if l<=k and l%2==k%2 then we have a solution

→ Reply

Vasiljko

5 years ago, # ^ |

Did the same but didn't work :(

→ Reply

5 years ago, # ^ |

← Rev. 2 →

I do not get it how to find the shortest possible path after adding x,y, we need that to check against k. Somebody explain?

Edit: Or is it expected to do a BFS to find it? That would be to slow.

→ Reply

gupta_samarth

5 years ago, # ^ |

You need to make use of precomputing depths, and LCA ( Least Common Ancestor ) of two nodes. Read more here.

→ Reply

5 years ago, # ^ |

← Rev. 2 →

The LCA does not help if the path after adding the new edge x,y is shorter than the one through the LCA.

Edit: After reading the other answer below I understand. We simply use LCA again to find path length from a to x and y etc. Thanks for explanation.

→ Reply

gupta_samarth

5 years ago, # ^ |

Read my comment below. After adding the edge the shortest path will have to be one of the three alternatives. ( It can be proven ).

→ Reply

Be_dos

5 years ago, # ^ |

The path between two vertices in the query may either go through the added edge or not. Analyze these cases individually, finding path lengths in O(1) using whatever fast LCA algorithm you prefer. Then if the difference between the length of the path obtained in some of these ways and the desired length if divisible by 2, answer is yes, otherwise no.

→ Reply

gupta_samarth

5 years ago, # ^ |

← Rev. 2 →

+14

Root the tree at an arbitrary vertex. Precompute depths of each node to give distances between two nodes in $$$O(log(N))$$$ time using LCA.

When adding a new edge $$$(x,y)$$$, there are now three possible paths ( without going back and forth ), which are

original path between $$$(a,b)$$$
$$$(a,x), (x,y), (y,b)$$$
$$$(a,y), (y,x), (x,b)$$$

Now, if any of these paths can reach $$$b$$$ from $$$a$$$ in less than or equal to $$$k$$$ steps and has same parity as $$$k$$$, then answer is YES.

→ Reply

gratus907

5 years ago, # ^ |

First we do not consider added edge. We can 'hang out' instead of actually making meaningful move as much as we need — which is, if we can arrive $$$b$$$ at $$$t$$$, we can hang out some even number of steps loafing back and forth.

Then we consider new edge. Since new edge form a cycle, it opens new possibility — by loafing around in cycle. If cycle length is $$$l$$$, we can arrive at

$$$p + a\times l + 2b$$$

for any arbitrary nonnegative integer $a, b$, where $$$p$$$ is shortest time we can arrive at $$$b$$$ using the added edge. (If we are not using the new edge, we can't hang out on the cycle).

All we need is careful casework (there aren't many) and computing distance in trees using LCA or some other fast methods.

→ Reply

utsav_deep

5 years ago, # |

Am I the only one who solved A with binary search instead of much more obvious 1 liner solution?

→ Reply

5 years ago, # ^ |

If there is the point to which we get after $$$k$$$ steps from both $$$x$$$ and $$$y$$$ is

$$$x + k \cdot a = y - k \cdot b$$$ $$$k \cdot a + k \cdot b = y - x$$$ $$$k\cdot (a+b) = y -x$$$ $$$k = \frac{y-x}{a+b}$$$

So, if $$$(y-x)\%(a+b)=0$$$

then answer is $$$k$$$, in other case the answer is $$$-1$$$.

→ Reply

utsav_deep

5 years ago, # ^ |

I realised this simple solution when I revisited problem A for locking it, after solving up to problem D.

→ Reply

xodeChemist

5 years ago, # ^ |

this is owsam explanation but I used binary_search here

→ Reply

5 years ago, # ^ |

I used binary search too, and needed the editorial to make me think of manipulating the equation :)

→ Reply

5 years ago, # |

How to solve C?

→ Reply

arshad2117

5 years ago, # ^ |

← Rev. 2 →

After each customer comes in, you have an effective range of temperature, let it be (x, y). For the next customer find the intersection of (x — t, y + t) (where t is difference between entry times) and his temperature range. That is your new effective range. Initial range is (m, m). If all the intersections are non-empty you can satisfy, otherwise no.

→ Reply

5 years ago, # ^ |

Thank you, it certainly makes sense to me now

→ Reply

Eddagdeg

5 years ago, # ^ |

I did the same thing but I got WA in pretest 3

→ Reply

arshad2117

5 years ago, # ^ |

I got the same result in the beginning. Apparently, I was breaking out of the for loop without taking all of the input.

→ Reply

5 years ago, # ^ |

yea one need to be careful of that

→ Reply

gratus907

5 years ago, # ^ |

I got 1 WA for same reason. By coincedence it passes the example input QAQ

→ Reply

yaswanth

5 years ago, # ^ |

Group all customers attending at a given time t.
For each group check if there is an overlapping temperature range and store that. For instance if the ranges are [(1, 4), (2, 5)], the overlapping range is (2, 4).
If it's not possible to form an overlapping range at any t, then it's not possible
Assuming each t has a range, check if it's possible to move from range at ti to range at ti+1. Also update the range of temperatures possible at each step.
If possible range at step i is (lo, hi), Then possible range at step i + 1 is (lo - (ti+1 - ti), hi + (ti + 1 - ti))
Just check if that is acceptable for all customers at ti+1 and update lo and hi. (Not sure if this step can be optimised further but it passes the problem limits)

My submission 71152069

→ Reply

5 years ago, # ^ |

← Rev. 2 →

I have a solution that is easier to implement in my opinion.

Loop through all pairs of customers. If it is impossible to reach the second Customer's range from the 1st Customer's range in the time difference between them, then the answer is impossible.

O(N^2) solution.

→ Reply

5 years ago, # ^ |

I think your solution will fail systest

→ Reply

5 years ago, # ^ |

← Rev. 2 →

It already passed 71138487

→ Reply

5 years ago, # ^ |

congratulations on becoming CM. Now I see why your solution passed.

→ Reply

5 years ago, # ^ |

Thanks!

→ Reply

5 years ago, # |

How to solve C and D guys ?

→ Reply

5 years ago, # ^ |

For C

Let's say that we have minimal and maximal temperatures with which we can get to the next consumer, and if the intersection of our range and range (l,r) is not empty, then it's okay, now we just update our range — basically it would be equal to intersection of ours and (l, r).

My submission: 71146715

→ Reply

5 years ago, # ^ |

Thanks, I only know how to solve case (x < l) and (x > h) but not (l <= x <= h) :<

→ Reply

xodeChemist

5 years ago, # ^ |

good explanation thanks Mr.Hakimov

→ Reply

5 years ago, # ^ |

For D

The main idea here is that for minimum we should build our array in such a way that bigger numbers have lower indexes and for maximum bigger numbers should have bigger indexes.

You can chek out my submission: 71159214

→ Reply

5 years ago, # ^ |

Thanks. But can I solve D problem using prefix array & suffix array ?

→ Reply

5 years ago, # ^ |

That's a good question, but I don't think that there is a necessity in them :)

→ Reply

5 years ago, # ^ |

If I use prefix-suffix array, what should I approach

→ Reply

jatinag22

5 years ago, # |

+16

how to solve D?

→ Reply

5 years ago, # ^ |

For Min: starting from the first '<', start at n and go down but if there are <<<< or something sort that subarray then for the rest just go down by 1 each time so say for the 2nd sample, my answer would be 5 4 3 7 2 1 6

Max: for max it's the same thing but all the numbers after a < are going up so 5 4 3 6 2 1 7

→ Reply

5 years ago, # |

Thanks for the contest, have a good day problem setters <3

→ Reply

rahulpkvij

5 years ago, # |

In Problem B of DIV2. the problem statement mentioned that the string can be "REORDERED" and not "REVERSED" but my code was failing the first testcase here : https://codeforces.net/contest/1304/submission/71145055

→ Reply

Jakube

5 years ago, # ^ |

No, you can't reorder a string. You can reorder the list of strings.

→ Reply

rahulpkvij

5 years ago, # ^ |

Ohh Okay. But this got me very confused and thus wasted a lot of my time.

→ Reply

_asmah98

5 years ago, # ^ |

strings can be reordered not a particular string !

→ Reply

Vovkaez

5 years ago, # ^ |

I can't see your code, but "Strings can be reordered" means order of concatenation can be changed, not the strings themselves

→ Reply

5 years ago, # ^ |

The list of strings can be reordered, not the strings themselfes.

→ Reply

yessirski

5 years ago, # ^ |

My AC solution outputs the test case: 2 2 aa aa -> 2 aa. When I think it should be aaaa, 4. Can somebody explain me why !?

→ Reply

5 years ago, # ^ |

It is an invalid input because all strings must be distinct.

→ Reply

5 years ago, # |

← Rev. 2 →

Can someone please tell me why this code for C is giving rte. The only place where an exception can be thrown is the assert in which case the test cases are wrong. https://codeforces.net/contest/1304/submission/71155599

→ Reply

Kognition

5 years ago, # ^ |

Perhaps it has to do with your usage of the word "time"? I think it's a reserved keyword that can still do arithmetic with integers. You're right, the code itself doesn't seem like it can segfault as far as I can see.

→ Reply

5 years ago, # ^ |

I had submitted the same code without the assert line. That gave wa.

→ Reply

Kognition

5 years ago, # ^ |

That doesn't necessarily imply much; undefined behavior is still within your set of possibilities. In fact, I'm much more inclined to think that you have UB somewhere; this would probably cause a segfault sometimes and a WA at others.

→ Reply

5 years ago, # ^ |

The thing is that I have been getting RTE only when I have included assert in the code. I have submitted the same code without assert thrice (introducing one useless variable each time so that it allows me to submit) and got WA in all those.

→ Reply

5 years ago, # ^ |

I've gotten several questions regarding this.

The problem is that you're "returning" the function while you haven't gotten the full test case yet. Some people used 'break' inside a loop. These codes generally got incorrect on pretest 3.

→ Reply

5 years ago, # ^ |

Yes that's the mistake. Thanks for pointing out and sorry for wasting your time.

→ Reply

thelonewolf

5 years ago, # ^ |

I got similar Runtime error in Problem B today. It was resolved by removing #define int long long. I couldn't figure out Why it happened today, I haven't faced any problem due to this statement earlier.

My Submission link: Submission

→ Reply

rupav

5 years ago, # ^ |

You are returning from function, "solveTestCase", before taking all inputs. Instead return only at the end of the function.

→ Reply

5 years ago, # ^ |

Okay yes. That is the error. Thanks for pointing out!

→ Reply

TheLethalCode

5 years ago, # |

+70

Every time........!

→ Reply

Lucky_Glass

5 years ago, # ^ |

I feel exactly the same. :(

→ Reply

e.m.blagov

5 years ago, # |

Thanks for the round, hope pretests were good enough.... I wish everyone to increase their rating ;)

→ Reply

gyh20

5 years ago, # |

Oh,just five minutes after the contest over my solution passed all the examples.:( The code for the round are a little bit long. I think a longer time will be better.:)

→ Reply

dvshah

5 years ago, # |

+16

Why does codeforces skip main test and don't take the best submission? I missed my chance of becoming Master today.

→ Reply

kingcobrabt

5 years ago, # |

I don't understand what is wrong in my submission, the answer is printed but still it shows run time error. Here is my submission 71154644

→ Reply

cool__dude__69

5 years ago, # ^ |

if(second.size()==0) It will give run time error .

→ Reply

kingcobrabt

5 years ago, # ^ |

Could you explain why?

→ Reply

cool__dude__69

5 years ago, # ^ |

Because second.size does not returns an int value therefore second.size() -1 will not be -1.

→ Reply

shamdin

5 years ago, # |

here i come #808080

→ Reply

tumaryui

5 years ago, # |

really sad, failed B on test 13.

→ Reply

Lucky_Glass

5 years ago, # |

It was such a good contest for the beginners. As the problems were not so hard as before :) But I think it should be harder, especially the Problem F.

→ Reply

YakovPolykovsky

5 years ago, # ^ |

If contest was 4 hours instead 2 I was solve all :-(

→ Reply

ACdreams

5 years ago, # |

My color is going to change. I love these questions!!!

→ Reply

code: https://codeforces.net/contest/1304/submission/71165457

5 years ago, # |

Thank you for strong pretests!

→ Reply

fnjnwmlclla

5 years ago, # |

Question B : Why there is run time error coming in https://codeforces.net/contest/1304/submission/71161367 this code can anyone tell me?

→ Reply

coding_panda

5 years ago, # ^ |

because of this loop for(i=v.size()-1;i>=0;i--). v.size() does not return int value.

→ Reply

akshaykumar99

5 years ago, # |

https://codeforces.net/contest/1304/submission/71145443 : what's wrong with this code

→ Reply

Vingying0

5 years ago, # ^ |

The problem is multi-case. which means you can't puts("NO") then simply return, or as the input continues, in fact your program will read the wrong data.

→ Reply

akshaykumar99

5 years ago, # ^ |

oh! thanks...

→ Reply

doppler_effect

5 years ago, # |

← Rev. 4 →

What is the mistake in this code, I am getting wa on prestest 8

→ Reply

5 years ago, # ^ |

If you erase elements in a set while iterating through it, especially your current pointer, you might end up skipping a few elements. This is shown when you run your code on the following case:

4 2
ab
cd
ba
dc

A different approach is to scrap the set entirely and instead use an array of strings, with boolean markers that say if you've used that string already (example, with the mk array).

→ Reply

doppler_effect

5 years ago, # ^ |

Thanks a lot man!

But why does it skips a few elements?

→ Reply

5 years ago, # ^ |

I didn't realize that you called a vector s, but it functions the same as a set in this context. My bad.

Imagine your vector like this:

ab <- (pointer)
cd
ba
dc

Now you find ba to be a match with ab, so you erase them both. Where does the pointer go? It can only advance to cd:

cd <-
dc

Now you increment the pointer, and it ends up at:

cd
dc <-

Since you only check ahead of your current iterator for matches, you don't find anything else and terminate. There's probably a topic on StackOverflow addressing this issue, but the basic premise is to only increment when you don't erase.

→ Reply

master_noobie

5 years ago, # |

I believe test cases for D are a bit weak?I changed my code for D a bit during contest as it had 1 mistake, but after the contest, it still passed....

→ Reply

5 years ago, # |

71166353 works on pretest 4 on my machine, and on ideone link but failed during contest and also on custom invocation. MikeMirzayanov djm03178

→ Reply

5 years ago, # ^ |

71177225 unordered_map -> map => WA -> AC

→ Reply

5 years ago, # ^ |

← Rev. 2 →

the same code gives correct output on my machine and another ide and on codeforces if used with c++17 but fails with c++14, i want to understand why is that the case, exactly what is not working in this case.

→ Reply

5 years ago, # ^ |

have you figured it out?

→ Reply

5 years ago, # ^ |

no, changing the order of input strings makes the test case pass in custom invocation on cf, when i tried to debug in cf custom invocation it was weird, somehow the entries in my unordered_map were getting corrupted during the execution. AK18

→ Reply

5 years ago, # ^ |

AK18 however the logic is same, unordered_map -> map should make no difference really, to me it seems like deep magic

→ Reply

5 years ago, # ^ |

I just changed unordered_map -> map and used C++14, it's AC. I guess it has something to do with the unordered map which I am not aware of. 71278944

→ Reply

https://en.cppreference.com/w/cpp/container/unordered_map/operator_at https://en.cppreference.com/w/cpp/container/map/operator_at

5 years ago, # ^ |

The problem is int count_r = freq[r];. It causes the unordered_map to insert r if there was no such element. By result, rehashing can occur, and therefore all iterators are invalidated, affecting the loop for(auto x: freq) in a bad way.

In contrast, using map::operator[] doesn't affect iterators at all. Check these two documents:

→ Reply

5 years ago, # ^ |

Thanks :)

→ Reply

saurabh_dtu

5 years ago, # |

My solution of B is failing in this test case but somehow it managed to pass system testing —
3 3
tab
bat
tab

→ Reply

5 years ago, # ^ |

All strings are guaranteed to be distinct, so your test case isn't valid.

→ Reply

saurabh_dtu

5 years ago, # ^ |

sorry

→ Reply

shamdin

5 years ago, # |

where i am wrong? https://codeforces.net/contest/1304/submission/71174871 problem C

i am iterating in reverse, finding the largest intersection possible so that both current and previous customer are satisfied with temperature

→ Reply

kikoawesome

5 years ago, # |

← Rev. 2 →

Not sure if I should be posting this here, but:

I just got a message saying that my solution for problem E coincides with that of some other participants. This is likely because I used the CP Algorithms implementation for LCA which you can find here: https://cp-algorithms.com/graph/lca.html

Although I added a few additional methods in my source code, a big chunk of my solution was for the LCA structure. So I guess that could've set off the response. I hope this clears things up.

If you need any more information, please let me know. MikeMirzayanov

→ Reply

SrbhVj_321

5 years ago, # |

← Rev. 2 →

I am not sure whether I should comment here . but i just got a warning about coinciding my code with another user.Since I am new here , I dint know about the codes copying from ideone . I was using IDEONE in default setting. My code is https://codeforces.net/contest/1304/submission/71137281. And the guy who copied my code is https://codeforces.net/profile/8bit_thug I request you not to do this. this kills contest spirit. And yeah Since He submmitted code after 3 minutes when I submitted hence it is cleared that he copied my code. Moreover all his submissions are skipped. Do I need to provide more information? MikeMirzayanov

→ Reply

smartraider

5 years ago, # |

My solutions for this contest were skipped most likely because I used the implementation of LCA from here which is probably where some other participants picked the code up from.

djm03178 can you look into this?

→ Reply

5 years ago, # ^ |

Sorry, there's nothing that can be done from my side.

→ Reply

smartraider

5 years ago, # ^ |

Alright. MikeMirzayanov

→ Reply

Elisey_Muxin

5 years ago, # |

Мое решение 71159863 по задаче 1304E совпадает с решением другого пользователя, так как видимо мы использовали один и тот же источник, а именно https://e-maxx.ru/algo/lca.

→ Reply

Pragyan_c2

5 years ago, # |

← Rev. 2 →

I got a message from User system regarding code similarity about the E problem (71164831). I have used the code for lowest common ancestor from this site (same code published in the site mentioned by other affected users) which was last updated on 15th June 2018 (here) which is conclusive evidence of this being published well before the contest.

→ Reply

hanga97

5 years ago, # |

I hope this is the right place to put this, otherwise, please tell me where I should write this. I received the following message:

Attention!

Your solution 71159069 for the problem 1304E significantly coincides with solutions ksun48/71134293, hoke_t/71152328, Venia/71156816, hanga97/71159069. 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://codeforces.net/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.

For the solution of problem E, I used the RMQ (range minimum query) and LCA (lowest common ancestor) implementations from https://github.com/kth-competitive-programming/kactl/

Specifically from https://github.com/kth-competitive-programming/kactl/blob/master/content/data-structures/RMQ.h and https://github.com/kth-competitive-programming/kactl/blob/master/content/graph/LCA.h

This code was online long before the start of the round, so I hope, this issue can be resolved.

Thanks for the round btw. I enjoyed it.

→ Reply

meetpr8

5 years ago, # |

Can someone answer why the compiler of Codefoces gives run time error if i have not type casted size of a vector whereas most of the other compilers don't give that error. In this contest only i got stuck in problem B due to this error.

→ Reply

gregor13

5 years ago, # ^ |

Me too

→ Reply

Littleboy123

5 years ago, # ^ |

the .size() function returns an unsigned integer, so when you try to do.size() — 1 when the size is 0, it overflows and causes runtime error.

→ Reply

meetpr8

5 years ago, # ^ |

Got that...but when i did this for(long long i=a.size()-1;i>=0;i--), it gave me run time error but for this for(int i=a.size()-1;i>=0;i--) it did not. Can you tell me why this happened? And thanks in advance :)

→ Reply

ethicalrushi23

5 years ago, # |

← Rev. 2 →

Can someone please explain how I am getting a different output for E on my local compiler than the codeforces judge.71178404

I am the getting the expected output on my machine. I believe the logic is correct. https://ideone.com/LfR13i Expected output with same input

→ Reply

gregor13

5 years ago, # |

Nice problemset!

→ Reply

animal-liberty

5 years ago, # |

Hello MikeMirzayanov, I have received a message about being disqualified from the contest, because of plagiarism. I have used the LCA calculation from cp-algorithm. Here is a link to the web archive, as evidence, that the code has been published before the contest. I have not used any online IDE. Kindly look into this, Thanks a lot!

→ Reply

5 years ago, # ^ |

← Rev. 2 →

Hi MikeMirzayanov, I have received the same email, I have also used the LCA part from cp-algorithmm, same source as animal-liberty and also quite a few others as mentioned in the email. Link to source.

→ Reply

Febrin

5 years ago, # |

Hey,
My solutions were skipped because one of them coincided with solutions of other participants.
It was 100% caused because of this implementation of LCA, which I use for a long time.
Regarding rules, one can use publicly available implementations posted before the round.
MikeMirzayanov can you do something about this?

→ Reply

wucstdio

5 years ago, # |

← Rev. 2 →

+29

Me in this contest:

Firstly, solve problem C, and then B-A-D-E-F.

After solving all the problems, I looked at the standings, and found rank 2 is only 61 points lower than me, and he has locked all his problems and started hacking.

So I locked my ABCDE and started hacking too.

When I opened a solution to problem B, I was surprised to find there was a very silly BUG. A code with such silly BUG wouldn't pass the pretests.

So I took a look at the statement, only to find that I had misread the problem. I thought I can reverse any strings as well as reorder them.

1000 points! If I had not made such a silly mistake, I would probably get on the 1st place, but now I couldn't resubmit because I had locked my problem. So I poured out my sadness to jiangly, who had solved all the problems too and become rank 3 at that time.

Here is the chat history:

Yes, I misread the problem again. I forgot that all the strings are distinct.

So from my story we can conclude that it's very important to read the statement carefully, but if you misread the statement, you still have chance to pass the problem.:)

→ Reply

dvshah

5 years ago, # ^ |

Lol same. I also thought I need to check the length of the only palindromic string I will add in the middle and changed my code in the last minute of the contest. And after 5 minutes comes the realization that all the string have the same length. Lost 500 points and went from +45 to -13.

→ Reply

KisekiPurin2019

5 years ago, # ^ |

I just found out that I also misread. Thank you.

→ Reply

Relief

5 years ago, # |

In problem D, you can solve for max LIS and re-use it to find min LIS by manipulating the string, ie. reversing and inversing — replacing $$$<$$$ and $$$>$$$ with one another. Here is my solution.

→ Reply

syh0313

5 years ago, # |

+15

Great contest！

→ Reply

SKhan

5 years ago, # |

My contest history for this contest is suddenly deleted.Why this happened?

→ Reply

KisekiPurin2019

5 years ago, # ^ |

Rating changes for the last round are temporarily rolled back. They will be returned soon.

→ Reply

Xellos

5 years ago, # |

apology for poor english

where were you when top rate dies

i was sat at work writin code when telegram ring

'top rated is die'

'no'

and you???????

→ Reply

someone8001

5 years ago, # |

hello in solution B try this 2 3 AAA AAA and it will answer you 3 AAA i think it should be 6 AAAAAA anyone can help me with this i can't see in description that say the same string doesn't count

→ Reply

5 years ago, # ^ |

It says "All strings are distinct."

→ Reply

towrist

4 years ago, # |

-10

great contest, but I would have liked if you could have removed problem F1. It is extremely trivial (I think div2C). F2 is the real deal and is a great problem. There was no reason to add F1. (and just because of the long legend and its place in the problemset, F1 is rated 2300, while it is more like 1500).

Again, great problems, especially E and F2, but don't add subtasks which add no value to the problemset.

→ Reply