Kevin Cruijssen

@jdt Hacky or not, it works. Thanks again. :)

@jdt Thanks! I changed the | to || since otherwise inputting 0s beyond 4 characters was possible. Your use of e.key also fixes one of the two known issues as a bonus. :D

@jdt Thanks. I've been able to save 4 more bytes by removing the "", ;, and > as I had prior (and I've used !! + |/& instead of ||+&& due to personal preference). And now I also know why it didn't work earlier when l4m2 suggested returning false. Unlike Java, the programming challenge I work with daily, in JavaScript bitwise operations on booleans will convert them to bits - aka true&false will become false in Java, but 0 in JavaScript..

@l4m2 I've rolled back my answer and use event.preventDefault() again.

@l4m2 I did manage to get my answer working with a return instead of event.preventDefault(). I forgot to invert the checks.. Woops.

@l4m2 It does work regarding the return vs event.preventDefault(), but unfortunately it has a different issue (I can still type !@# and alike in your answer (aka Shift + 1,2,3). EDIT: Apparently after all the golfs my answer has the same issue rn.. -_-

@l4m2 Since I apparently can still continue typing even if it returns false. At least in the snippet on SE or a JSFiddle (I'm using Chrome if it's relevant)

@Shaggy That works as well, thanks. Btw, I just found another bug unfortunately: typing a bunch of ' (or other characters that are typed with a setup keypress of however you want to call it, like ", ^, ~, `````; etc.) still work. -_- Will see if I can fix it later on perhaps.

@Shaggy I've just added a !isNaN(...) around the value[3] for now. Still shower than maxlength=4.

@Shaggy For me the arrow key is indeed also blocked, but I can still click 'HOME' + 'DELETE' on my keyboard when it's -8.

@Shaggy Thanks. The -8 instead of !=8 still allows delete though. But thanks for all the other golfs. :)

@Shaggy I've been trying to find a browser that uses asterisk, but Chrome; Edge; FireFox; and Opera all use bullets unfortunately. As for your 69 byter: thanks for some golfs, but the onkeydown you've removed was to prevent 'delete' and other non-backspace key-presses.

@Simd I will fix the asterisks instead of circles later today.

@mousetail What does the input look like when you get that error? If I look at the 05AB1E source code, it seems to check for the trailing """ for multiline inputs. Do you also have a leading """? And which of the programs from the other answers was used as input?

@NobodyNeedsNames I barely program in Python, but your changes indeed make sense. Thanks! :)

@NobodyNeedsNames ^

It's undeleted again. So if you see any errors in it, lmk

That's why I deleted it, since I'm currently fixing it.. ;)

Could you upload and share the Controller somewhere on github? Since you mentioned fixing/modifying the existing controller on pypi.org.

For the rule "Cracking a submission consists of finding any programming language that prints the correct result, not just the intended one. If a submission is run in any language that was not declared or found to work, there are no requirements to do anything, and future answers must be valid in that language.": let's say I intended language X, but someone cracks it in language Y, should all the next answers run in language X or Y? The wording in your rule isn't exactly clear to me as is tbh.

I also noticed it depends on a lot on the position of the bots in the list, so I understand now why you had the shuffle() + sort() before.

By changing it's bonus stats a bit it became more competitive, so I've just posted that new bot. I initially had 6,1,1,2, but with the speed of 2 it was death a lot more often because of Destroyer.. But with 9,0,0,1 it's doing a lot better. :)

I did came up with a new more basic bot, but if I add it both itself and HpKing drop to 4th/5th place with Destroyed at the top. 😅 So not sure if I want to post it (just yet), haha.

Apparently the updated HpKing bot I had in mind doesn't make too much difference. In fact, the regular HpKing bot beats it by a few thousandth or so.

@Romanp ok, it should be fixed now. I've updated my answer. I now just use the class-Strings instead of Class directly, then the comparing should be fine. (And with a test run I seem to be in the lead, although I'll still continue working on my second version.)

Ok, I'm back. I've just downloaded your github project and will see what the issue is and fix it. I guess it still gives the same error because of the containsKey now that I think about it.. :/

I'm going to eat dinner now, so if there are any other issues, I'l try to fix them when I get back. :)

You can change the TreeMap<Class<?>> back. And use var shotByClazz = getShotBy().getClass(); \n shotBy.put(shotByClazz, 1 + (shotBy.containsKey(shotByClazz) ? shotBy.get(shotByClazz) : 0)); instead of that entire merge-line.

@Romanp I've updated my answer with a fix. Didn't knew Class wasn't Comprable and can't be used with Map's merge. I now do it manually with a regular map.put.

@R

I'm also working on an updated version of HpKing, since I found a flaw in my general approach. :) Will post it as a separated entry when it's done, though.

@Romanp Ah. I think you can fix it by changing TreeMap<Class<?> to TreeMap<Class<PlayerClass> and shotBy.merge(getShotBy().getClass() to shotBy.merge((Class<PlayerClass>) getShotBy().getClass() (not 100% sure, though)

@Romanp Ah ok, that makes sense. One more thing: you may want to remove the reload(); in the default makeMove() method of the PlayerClass. Right now I could do super.makeMove(); to have infinite ammo, and then do another action like healing/shooting in that same move.

Why is there a Collections.sort(players); after the shuffle in MainClass? Or was that added just for testing purposes?

@ovs JDF also works (in theory given enough time and resources), but simply takes waaayyy to long in the first few iterations. It will basically loop the 100101010100101001... amount of times, timing out on TIO. With the ¤ it will only loop 0 or 1 times.

This time I'll wait with changing my answer until you've confirmed it. But F (and the other ranged loops) error on empty strings, so using ¤F at the end can be used in this case. :)

@ovs Ah, stopping with an error is pretty smart. I don't know if we can do it with ¿, but this might work by the looks of it

@ovs Ok, I think you're indeed correct that the Δ won't work. With the wrap it's still the same issue as before "000000" being equal to "000". I've tried to fix it like this, but unfortunately it will always loop one additional time before the Δ terminates, so it'll output four As instead of 3.

Ah, but the issue still remains. It outputs two AA, but you said it should output three. Let me check what's going on

PS: I usually add comments by wrapping them in "..."\ :)

@ovs The q in your footer is part of the loop now, so it terminates the entire program right after the first iteration. If you change the q to }q it will work as intended.

@ovs Yeah, ik. I tried that initially before I posted my answer. Unfortunately something like 0001 is equal to 1 in 05AB1E's eyes.. Most of the times useful, but in this case the opposite of useful unfortunately.

@ovs Actually, your approach of putting the J at the front might make Δ`JD...¸ possible for -1 by the looks of it. :)

@ovs Thanks again! That's again a lot better. ;) I also have the feeling the infinite loop could be golfed somehow with a fixed number of iterations, but I'm unable to save bytes thus far. I either also output the length at the cost of 2 bytes in the compressor, and replace [ÐõQ# with IFD in that case. Or I change [ÐõQ# to something like DvÐdi. If I just change [ÐõQ# to DvD, it will output trailing characters equal to the first character of the input-string unfortunately.

@ovs Oh, that's indeed a lot better. Thanks for the -18!

@DominicvanEssen I think that would indeed make the challenge more interesting. As is I have the feeling this challenge will be over rather quickly otherwise... :/ After turn 5, the "Running programs must be irreducible." is gone, which seems to be the most annoying one to deal with tbh. The other four rules thus far (must contain C"pr; may not contains brackets of any kind; must be at least 19 bytes; and may not contain whitespaces) aren't too hard to apply in most golfing languages for example.

Then again, it might now give an unfair advantage to the existing answers over new ones..

Just to make sure I understand it correctly. Let's say someone posts the turn-5 answer in a bit. After that the starting rules are dropped, and if turn-6 would output Codington Crescent this challenge would be over already?

Someone should try to finish the game in the language Mornington Crescent. ;)