Clash of Code : Bugs & Improvement

Give a score bonus if a codinGamer can solve the clash without Paste-ing .
(but the cheaters will use a macro)
then give another bonus if the gamer never loses focus on the editor.

or just print a Donkey image at the side of the clash winner if he copy-pasted everything!

1 Like

How about those using their own editor?

Iā€™ve just got an empty code window in CoC reverse.
Wasted precious time trying to switch to another lang and back and to write all that code that was always available before.
stuff like
import java.util.*;
like public class Solution {
public static void main(String[] args) {
Scanner s = new Scanner
etc (yes. i like java)

Is that another new feature?

I actually like this feature :slight_smile: Good for shortest mode.
P.S. Looks like stubs are empty everywhere, not just CoC. Thatā€™s sad.

Code generator are broken, you may not play CoC till someone fix it.

1 Like

Problem solved, feel free to clash.

1 Like

Hello there,

Seems the topic is dead. But i will try my luck anyway :slight_smile:

Itā€™s only my opinion and feel free to criticize me.
I think is not encouraging at all to use java language( C# or other similar languages) in clash of code shortest mode. Because it has a lot of key words and the result code length is large a.f.

Just look at the example of adding 2 numbers:
Java

import java.util.*;
class Solution {

public static void main(String args[]) {
    Scanner i = new Scanner(System.in);
    int A = i.nextInt();
    int B = i.nextInt();
    System.out.println(A+B);
}

}

Python

a,b=[int(i)for i in input().split()]
print(str(a+b))

Same logic steps but 226 chars vs 61 chars.

I suggest some improvements for that kind of competition. Now it calculates every char in the submitted code.

  • Maybe it will be better to count only the effective code lines?
  • Do not count spaces and tabs.

Just make it more fair for every language. So, opinions?

Best regards
Mefodii

1 Like

Impossible.

Can be abused easily. So impossible.

In all fairness, competing in short mode with different languages will never be fair. This has been discussed a lot of times. Unfortunately, we donā€™t have a perfect solution.

You can see it as an opportunity to:

  • learn a less verbose language
  • learn tricks of your language to reduce the number of characters
1 Like

Please investigate scripting languages for Clash of Code, especially JavaScript. At least once (and possibly more) have I gotten 0% on test cases with JavaScript, but after I ported it to Java with identical logic and submitted, did I get 100%. And JavaScript is actually my expertise. I met others with similar issues.

Maximum value for a number in Javascript is 2^53 - 1. The maximum value for a long in Java is 2^63 - 1.
If you have to manipulate an integer above 2^53 - 1, you canā€™t do this in Javascript.

As i always said, this kind of puzzle should not exists in CoC. You canā€™t copy/paste a BigInteger library in your javascript code just for that ā€¦

Nope, it was just something basic without need of libraries, etc.
IIRC, one was this clash with just string concatenation where I got 100% Java after porting from 0% JavaScript:

EIDT: My final Java:
public static void main(String args[]) {
Scanner in = new Scanner(System.in);
int s = in.nextInt();

String output = ā€œā€;
for (int i = 0; i < s; i++) {
for (int j = 0; j < s; j++) {
output += ā€œ+ā€;
}
output += ā€œ\nā€;
}
System.out.print(output);
}

JavaScript was very similar where I used var instead of int.

print() in js adds a newline in the end doesnā€™t it?

Hmm, thatā€™s probably why. Then that should be on Clash of Code to fix. However, on any other problems requiring output of a single line, itā€™s been fine most of the time too, so I donā€™t think thatā€™s the case.

Also, since my original post, I havenā€™t encountered these issues again with JavaScript. If thatā€™s because an actual fix has been made, then thanks a lot. However, I do still see some weird results from others.

For example, recently thereā€™s this Reverse Mode averaging problem that I think is very trivial to deduce and solve in a short time. However, other players seem to struggle and even the 6th player that used to beat me quickly a lot on both harder and simpler problems before without issues:

JavaScript

var N = parseInt(readline());
var inputs = readline().split(ā€™ ');
sum=0;
for (var i = 0; i < N; i++) {
sum+= parseInt(inputs[i]);
}
print(sum/N);

I think thereā€™s a leaderboard bug.

Leaderboard shows me that martin is first and unihedron is 3rd, but after going to their profiles i see something completly different :open_mouth:

Martin is 17th

and Unihedron is first

Didnā€™t have time to check all forum posts but after quick look didnā€™t find any corresponding to this.

Using Safari Browser :stuck_out_tongue:

Cheers :slight_smile:

not sure, but Iā€™d say that the profile panel is updated once a day (like the achievements) whereas the leaderboard is updated continuously.

I donā€™t know if it is said before but here is my suggestion:

Now there are 3 modes: fastest, shortest and reverse

I think reverse mode is not a mode because the winner is the fastest like in fastest mode.

What about letting the winner be the player who needed less compilations in order to pass all the test and in the case of a tie, take the fastest?

1 Like

I second scaulerā€™s suggestion.

another minor nitpick:
the text still reads ā€œJoin a 5min coding battle!ā€

anyone ever had a 5min one? all i get are 15minsā€¦ (which is perfect for me!)

4 Likes

I consider some clashes to be very annoying.

Many of them were up-voted, approved and went through contribution acceptance process, but it seems that they are not liked.

How about removing some of the clashes in post acceptance process based on ā€œFunā€ and feedback ratings?

Simple solution: after 6 months ā€œFunā€ rating is check and if mean in below ā€˜**ā€™ (two stars) it is removed.

there is already a bot which removes unliked contributions:
https://www.codingame.com/playgrounds/40701/contribute---help/moderation-of-contributions

1 Like