Most of the time I'm getting submission timeouts for the last couple hours although the site status indicates everything is fully operational.

I just added a major issue (as test) on http://www.codewars.com/kata/simple-maths-test, and I didn't even get prompted for a comment. However, even if I got prompted, I can usually simply close the window and therefore add a minor/major issue without comment, anonymously.

I propose that the three buttons (Ready, Major, Minor) don't do anything on their own (well, except for "Ready" -> upvote). Instead, add "major issue" as a new tag and have the major/minor issue buttons redirect the user to the comment box, where the appropriate tag is chosen.

(originally filed on CW by bkaes)

Error: Command failed: /tmp/csharp115224-22-lt0wpn/setup.cs(3,21): error CS0101: The namespace global::' already contains a definition forKata'
/tmp/csharp115224-22-lt0wpn/code.cs(3,21): (Location of the symbol related to previous error)

Trying to make a C# translation of an existing Kata in Kumite, continue to get this error. I can name my class anything in the world, will get the same error (definition for )

http://imgur.com/kUEU88e
Change class name:
http://imgur.com/ykcQ34m

(originally filed on CW by parabola949)

On my Profile page, my Honor graph (which is the circular thingy that shows the progress to the next rank) is missing. It does not show up in the latest versions of FF, Chrome and IE. Weirdly enough, it does appear on the mobile version of IE.

The kata is marked as solved, I can see everyone else's solution for a particular language, but if I choose My Solutions, I only get There are no solutions yet for this kata.
This happens for me at least for this one http://www.codewars.com/kata/i-liked-the-sql-better-dot-dot-dot/, others seem to be fine, but haven't checked all, obviously. As I remember, there were some problems on the server at the time I was trying to complete this kata, so that may be related, but in any case, check the integrity of the records and what not.

┆Issue is synchronized with this Clickup by Unito

At http://www.codewars.com/kata/the-enigma-machine-part-2-a-rotor/solutions I only see one (author's) solution and a number of obsolete invalid solutions. However, my new valid solution exists and is visible if I click "My solutions", but not visible again when I click "All solutions".

┆Issue is synchronized with this Clickup by Unito

Note: as a shorthand, all URLs in this post are relative to http://codewars.com/api/v1/users/ .

Today the bug that was displaying the :username/code-challenges/completed results in place of :username seemed to have been fixed. Then jhoffner told me how to find :username/code-challenges/completed and it all went to hell.

I went to MindWanderer/code-challenges/completed. Yay! That worked. Then I went back to MindWanderer. Huh, that now also shows my completed challenges as well.

I played around with different usernames, as well as :username/code-challenges/authored and it looks like some sort of caching issue. After an API call for a user, for several minutes, all calls for any data regarding that user return the same thing. It's server-side caching; if I make a call on one machine at one IP address, a different machine at a different IP address returns the same data. It seems to clear after several minutes of inactivity.

If a test method throws an exception with a null message, the error is ignored and the test passes. Many exceptions thrown by the VM, e.g. NullPointerException or StackOverflowError, have a null message.

I think the bug is in CwRunListener.java. CwRunListener.formatMessage() should probably return an empty string if the argument is null. Currently, it throws a NullPointerException.

Here's CwRunListener.testFailure() and CwRunListener.formatMessage():

public void testFailure(Failure failure)
{
  failed = true;
  System.out.println(String.format("<FAILED::>%s<:LF:>", formatMessage(failure.getMessage())));
}

private static String formatMessage(String s)
{
  return s.replaceAll("\n", "<:LF:>");
}

When the message is null, formatMessage() throws a NullPointerException, which flies right through testFailure(), so the <FAILED::> message doesn't get written. I guess the Codewars testing framework ignores the failed = true field, only checks what's written to System.out, doesn't see a <FAILED::> message, and marks the test as successful.

If a user A submits a comment, and user B tracks the topic (e.g. a kata where user B has commented already), user B will get a notification with an excerpt, even if user A marked the comment as spoiler.

The loading indicator in Firefox never stops, regardless of the used operating system (Windows 7, 8, 8.1, Ubuntu 12.04 LTS, 14.04 LTS), except for some occurences (no reloadable elements, e.g. no comments, no solutions, no kumites, …), see attachement.

EDIT: Also happens on kumites/kata editor.

https://www.codewars.com/api/v1/users/jhoffner is showing completed challenges list instead of profile information for the user.

This is a suggestion that came from bkaes that I think is a pretty high priority item in terms of enhancements go. As the approval process is being pushed to more users, its important to make sure that all languages have been tested before a kata can be approved.

The idea here is that the author's solution does not count, and we should have at least 2 (maybe even more) non-author solutions before the language is considered verified.

┆Issue is synchronized with this Clickup by Unito

If one adds a issue (via comment), and then deletes the comment, the issue doesn't get resolved, see http://www.codewars.com/kata/last-digit-of-a-huge-number/discuss#55191fcacd82ff070f00078b.

Luckily, I can still click on the unhappy smiley to remove my issue, but I'll leave it for now.

I’m trying to pass the second lesson on the Javascript test to register. It’s the one where it uses the Person(name) construction.

1. There is nothing that says the expected names are Kate and Joe/Jane. That only becomes obvious with an error message.
2. If I set name = Joe, it errors and says it was expecting Jane. If I set name = Jane, it errors and says it was expecting Joe. This happens every single time in Chrome even after shutting down the browser and reopening it. And, it happens in Safari too.

Steps to reproduce:

• Go to status.codewars.com

Result:

• 406 Not Acceptable

See http://www.codewars.com/kata/reviews/552253df667a8dbf76000447/groups/55365b9986b7406e5e000071#553c8adc60ed0b9bbc000037. I've tried that "solution" on a random other kata, and it worked. That's quite frustrating.

Chrome 41 windows.
See attached screenshot.

I'm going to preface this by saying it may not be the browser, but my workplace's firewall, but I have no way of testing that. However, I don't know why the firewall would let me access the site and run tests, but not submit final code.

This issue appears when using IE 11. Before anyone shoots me, let me say I would certainly prefer to use Chrome or Firefox, but this is a Microsoft shop through and through, and black-suited men would come to cart me away if I attempted to install a more modern browser.

The issue: I click "Submit" and it fails immediately with a message flashing in red that it "failed to send to server." It flashes once, very quickly, then is gone. If this is a problem on my end, I apologize for submitting a bogus issue.

According to the statistics, the feature-request forum should have ~190 top-level comments. However, I can only see requests up to the following one:

A time estimate for how long it takes to solve each Kata would be useful for interview prep (coding under pressure). The estimate could either be submitted by the Kata author or based on historical averages. [by mxdubois, 3 months ago]

A quick check with $$('.comments-list > .comment').length ($$ = document.querySelectorAll) reveals that there are really just 94 top-level comments. So I can't view the additional ~100. They also don't load when I'm scrolling down.

When I receive an email about a new comment in a discourse I participate in, it looks like this:

SomeUser replied to your discourse on the "Some Kata" kata.
The comment reads:

Some text.

View in conversation

"View in conversation" is a link. In most cases this link contains an error - extra space somewhere in the middle of a link, that causes a 404 error when link is clicked. Here are couple of real examples, copied from real mails:

http://mandrillapp.com/track/click/12712279/www.codewars.com?p=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

get redirected to:

http://www.codewars.com/kata/the-e%20nigma-machine-part-4-a-rotor-part-ii/discuss?utm_campaign=code_challenge_comment_reply&utm_medium=email&utm_source=notifications#552e240d395cd0a351000034

Note the %20 after "http://www.codewars.com/kata/the-e"

Another example:

http://mandrillapp.com/track/click/12712279/www.codewars.com?p=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

gets redirected to:

http://www.codewars.com/kata/the-enigma-machi%20ne-part-3-the-reflector/discuss?utm_campaign=code_challenge_comment_reply&utm_medium=email&utm_source=notifications#552e245bab4a144c2500004b

Note the %20 after "http://www.codewars.com/kata/the-enigma-machi"

Some of the approved kata translations have not rewarded their honor to the author.

Sometimes when submitting Python katas, I receive an unknown error.
I have a feeling, that it happens when testing function that returns a list or None. Running the test for None manually reproduces the unknown error. eg. at this kata:
http://www.codewars.com/kata/54d81488b981293527000c8f/train
Submitting this solution: http://pastebin.com/1Pkn7nuZ
(i know, that it's not a good solution because of the timeout)
produces the error without going near the 6 sec timeout:
Unknown error
Process took 2909ms to complete

It seems now I should have the power.

I tried adding some sample test case on this kata, though, but to no avail: I keep getting a 500 Error.

Let me know if you require screenshots, additional info or the like. Thanks for your kind attention :)

The authorization on github's side worked, but then coming back only yields "Failed to connect account" if I click on "Link Your GitHub Account" again.

Could it be the length of my account? Maybe codewars has a minimum >= 3?

I'm not sure whether this is a problem with CamingoCode, gecko's font rendering (FF 33 and higher on Windows 7, 8, 8.1, Ubuntu 12.04, 14.04) or the syntax highlighter + custom stylesheets, but the code isn't completely monospaced. You'll notice this in Haskell often, since people try to align the code manually, like:

-- http://www.codewars.com/kata/reviews/55169d3a891547bedd000681/groups/551856df048fd53df1000004
module Codewars.Kata.Spinning where

spinWords :: String -> String
spinWords = unwords . map rev . words
  where
    rev w | 4 < length w = reverse w
          | otherwise    = w

However, the results aren't that nice:

Example 1

Example 2

Kumite/Translations are not represented within the honor breakdown chart.

Didn't realize how much I really loved this site until I couldn't submit a solution for going on two weeks now.

Steps to reproduce:

• Run tests for any kata
• Submit solution for any kata

Outcome:

• "Submission Timed Out"

Hopefully killing off the issue forum gives the team time to get to the bottom of it.

The kata author for this one and me had some issue while trying to approve the translation (you can read it in more detail in the relevant discussion).

I have tried to unpublish it, but the button doesn't seem to do a thing.

I also tried creating a new translation, but it refuses to make me create another Python version: that is surely wanted, but isn't it a bit dangerous? Suppose user X goes around publishing poor and/or not working of everything with language Y, de facto preventing other users to translate a kata in Y again, and you get a nasty picture.

Wouldn't it be better to have a warning if you try to translate a given kata in a language that has already a pending (of course, not an approved) translation?

Finally, I would like a way to permanently delete my translation drafts (I created one of the same kata in Coffeescript while trying stuff around this issue and it has no meaning to me, of course).

Thanks again for your kind attention :)

My code shows it is correct when i hit run tests and when i try to submit it it shows the "Process was terminated. It took longer than 10000ms to complete".

Please help me fix it I am pretty new here.

Thanks

If I submit a working solution, I get honor and the kata is marked as completed, but my solution doesn't get shown in the list.

Background: I wanted to check whether #48 really works, and tried it on Wintermute's http://www.codewars.com/kata/simple-interactive-interpreter/solutions/javascript/me/best_practice. It worked, and I wanted to cancel the final submit, so that I won't get honor and/or completion. However, both honor and completion got granted, but the cheating solution doesn't show up.

┆Issue is synchronized with this Clickup by Unito

Currently, code is saved only when you press Run tests.
When you write a lot of a code and then your browser got a crash, goes backward in the history etc
Then your user has a such feelings:

A ARGARGARHH!!!

Oh no! What a FUCK? I hate you

Best regards.

┆Issue is synchronized with this Clickup by Unito

I don't know the Ruby equivalent of htmlspecialchars, but it gets applied to often, since ampersands (&) are encoded in the kata excerpt on the dash board, see screenshot.

The source code shows that > has been encoded as > which indicates that it was already encoded as >, but the ampersand got encoded again:

<div class="markdown"><p>Say we have a function of this type:</p>
<pre><code class="lang-haskell"><span class="hljs-title">f</span> :: <span class="hljs-type">Ord</span> a =&amp;gt; a -&amp;gt; <span class="hljs-type">Either</span> b ([a], [b] -&amp;gt; b)</code></pre>

When I publish a kata, my solution needs to pass both the hidden and the example tests (if they've been provided). However, if I publish a translation, only the hidden test cases get checked, not the example ones, so it's possible to publish a translation with completely wrong example test cases.

┆Issue is synchronized with this Clickup by Unito

Déjà vu Duplicates has no author. Well, it has one (lkspencer), but his name isn't shown in the status bar, see below.

(Sorry for the marked text, I forgot to deselect my selection before I took the screenshot)

┆Issue is synchronized with this Clickup by Unito

I recently completed the "text align justify" Kata (4 kyu), but noticed that my honor only went up by 1. This has happened many times in the past, but recently all my points were adjusted. Is anyone else seeing this? Or am I missing something?

http://www.codewars.com/kata/paginationhelper/edit/ruby

Once a translation is approved, it is no longer available in the public areas (only in the authors translations list). Older translations should still be visible to help reason about things.

While trying Don't rely on luck kata in Python (first by trying to bind the variable to a function and then by overwriting the function) I discovered a leak that lets you pass every level. Plugging it in appears to trick into into a successful completion and terminate the program:

Test.assert_equals(1, 1, "Hello", True)
exit()

Was not my intention, but then again the level was sort-of hackish anyway so it was encouraging thought in that direction. Just figured you perhaps wanted to patch this hole up :)

Fortunately, I've got a notification about Jake's comment and could find the translation: http://www.codewars.com/kumite/553a556b54f6034493000022?sel=553a556b54f6034493000022. Still, Jake's comment gets shown on the dashboard, but the anchor doesn't have a href attribute:

<ul class="bulleted-text">
 <li><a href="/users/jhoffner">jhoffner</a></li>
 <li><span class="bullet"></span><span>commented on</span>
       <a class="js-commentable"> ""Make them bark!" CoffeeScript Translation"</a>
 <!-- ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ -->
 </li>
 <li><span class="bullet"></span><time data-text-time="1429898932">5 minutes ago</time></li>
</ul>

EDIT: This also holds for links in the "authored" list on the user's page, but not notifications.
EDIT2: See http://www.codewars.com/users/%20imagostorm/authored_comments for an example for an approved translation that doesn't have the correct href attribute.

┆Issue is synchronized with this Clickup by Unito

Firefox (Version 36.0.4 | Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:36.0) Gecko/20100101 Firefox/36.0) almost freezes on Ubuntu 14.04 if I visit the front page (not the dashboard, but www.codewars.com without being logged in).

(Originally filed on CW by bkaes)

I was wondering if more advanced levels are available, since the documentation says there are dan levels.

Hi,
I don't like using mouses a lot, almost anything I do is done using keyboard. I appreciate your hotkeys, but it's annoying that I have to focus the code in order to submit/test my code, so I'd like to be able to use hotkeys without focusing.

Thanks.

┆Issue is synchronized with this Clickup by Unito

Navigate to user profile page and view followers. Actual followers will be shown underneith repeated followers loading text.

I'm writing a test case that is very random and so I want to get the result from the complete solution.

Something like this:

Test.expect( myFunction(val), userFunction(val), "Wrong value for " + val);

Type: Enhancement
TL;DR: Fix or drop the "you can't rank" message for the kata author.

Apparently, there's something wrong with my current honor, since I'm greeted with the following message on my own kata:

You must have at least 100 honor before you can give your rank assessment.

See the following screenshot for a proof:

I guess something along the following should be a little bit more appropriate ;)

Message for normal user with less than 100 honor

You must have at least 100 honor before you can give your rank assessment.

Message for normal author (or just drop the message)

Kata authors cannot vote on their own katas. Instead, change the field "estimated rank" in the kata editor.

Some of the katas never gets marked as completed for me.

The funny thing is that the kata "Metaprogramming: Lisp-style Generic Functions" [http://www.codewars.com/kata/metaprogramming-lisp-style-generic-functions] works as some kind of divider. I still haven't solved this one.

When the kata list is sorted by "Latest", katas that appear before the divider kata are correctly marked as completed when solved, while katas that appear after it never gets marked as completed.

I've tried to solve the katas multiple times and in different languages, but it doesn't make any difference to this odd behaviour.

It's no hurry, just a bit inconvenient when I'm looking for new katas to solve. Let me know if I can do anything to help you figure it out.

Thanks for a wonderful site!!!

I've been getting "Process was terminated. It took longer than 10000ms to complete" errors on any java code I try to run. How can I fix it?

Not sure if Jake read this already, in case sorry for the repetition, but of all the translation I have done, two apparently seem gone MIA in this kata.

Or, rather: they can't be found by either me or the kata creator, but I can read them in my profile page (browsing A LOT of other stuff, but ok) and the kata creator somehow manages to find them among forks or something like that, but can't approve them.

The above link should provide you the full discussion (expand my message). Now, I have no big problem copy&pasting said translations into two new ones (provided this work), but I was wondering if that could hinder your search for the problem, or, worse, do some more damage.

Let us (me and suuuzi) about how to proceed and thanks in advance for your attention :)

While using the site, I will sometimes get logged out. Any attempt to do something fails and I have to reload the page, which then asks me to log in again.