🃏 FreeCell 🃏

Rate this contribution

+45

Make the contribution public (lvl 20+)

Comments

37 Comments

about 3 years ago

I would like to ask ffgth, Konstant and VivienBillot - have you solved this puzzle? Did you spot any issues with it, or inconsistencies compared to any other puzzle on the platform? I genuinely love this contribution, but there were still few small things that needed tweaking and now this is approved with useless comments like "good", "well done"? Are we just hunting contribution quest, really?

JBM Unnamed CodinGamer

about 3 years ago

I'd guess not, seeing as *nobody* has solved this puzzle yet.

VivienBillot Unnamed CodinGamer

over 3 years ago

✅ Approved the contribution. Comment: good !

Konstant Unnamed CodinGamer

over 3 years ago

✅ Approved the contribution. Comment: Well done

bradix Unnamed CodinGamer

over 3 years ago

Freecell & minesweeper... my personal tools for procrastination! LOL

ffgth Unnamed CodinGamer

over 3 years ago

✅ Approved the contribution. Comment: fun

theAlphaGuardian Unnamed CodinGamer

over 3 years ago

Nice work. Good Luck <3

heldwin Unnamed CodinGamer

over 3 years ago

Thanks for this. I am still looking at this project, but it made me discover the coding game SDK also :)

Gabbek Unnamed CodinGamer

over 3 years ago

One more question as I've fixed my bugs with home moves. The testcase called "Impossible Game" can't be solved, is this correct? I've tried using fc-solve and it runs out of iterations, same for my algorithm. Wondering how to have it "green", or we're not supposed to solve it? I've never encountered a testcase which is supposed to fail (unless I've missed it in description and we somehow can output "unsolvable" or something like that).

JBM Unnamed CodinGamer

over 3 years ago

It's MS game #11982. No one has ever found a solution for it using the MS ruleset. But to the best of my knowledge it hasn't ever been proven unsolvable, so you can still become famous by being the first. You don't need to solve it to pass the validators.

Gabbek Unnamed CodinGamer

over 3 years ago

Let me start off by saying that this is incredibly well done. I had some issues with the input initially - but it has been resolved. It's not the most friendly puzzle to debug, but there's lots of information, so that's to be expected. I fought much more with autoplay than anything else to be honest. I finally have solution which is able to solve some games, but it seems I still have something wrong with home moves removal so I had to remove few home moves manually to see if my solution works - luckily it does. Feedback for the puzzle: 1) In my opinion the most important one - ability to turn off autoplay, as currently there's much more issue than gain with it. 2) Detailed description about how the autoplay works - in order to solve the puzzle you have to remove those moves which will be auto-played, unless someone is trying to find a new solution each turn and providing only one turn per move (guess that's possible, but I've used python and haven't tried that approach) 3) Information about turn time limit and maximum allowed moves in total in description - it's really nice to have as it's a great and quick way to judge how advanced solution you probably have to write (for example with easy puzzles you often know that you can use bruteforce if the puzzle spec has small inputs) 4) Very minor - maybe add information in the output which move is being executed on a certain turn (and when a turn is auto-play)? Mostly an issue with readability while trying to watch the game slowly and outputing multiple moves in a single turn, easily fixed by player when he outputs a single move per turn. More than happy to approve in few days, once I'll have a bit more time to remove my home moves issue in my code and maybe some optimizations are still required to solve harder games. Great job!

JBM Unnamed CodinGamer

over 3 years ago

Thanks a lot for trying it out and taking the time to provide feedback! To address your points: (1) has been in the plan since the start (along with netcell-style). It'll get there. Eventually. (2) ditto. (3) they're the platform defaults. I haven't actually thought about them, what settings do you think would make sense? IIRC the platform constraints are: <=30s aggregated per game, >=50ms per turn. (4) great idea. That might actually come up first as it's easiest to implement.

ChampionCoder Unnamed CodinGamer

over 3 years ago

Works well. Nice contribution.

Nicolas_BREANT Unnamed CodinGamer

over 3 years ago

Nice !

SamyMessouak Unnamed CodinGamer

over 3 years ago

nice work

CrazyCoder12 Unnamed CodinGamer

over 3 years ago

Very Good

Starfighter Unnamed CodinGamer

over 3 years ago

J'adore ce jeu . C'est bien de le mettre en exercise :)

Potato_God2 Unnamed CodinGamer

over 3 years ago

POG

JBM Unnamed CodinGamer

over 3 years ago

I've updated the identifiers to ridiculously descriptive "foundationCountPlusOne". So now it compiles in all languages but Go (but who cares, it's the stub generator's choice), and runs in all remaining but Clojure (Timeout, I don't know why, please tell me), F# (IndexOutOfRangeException, I don't know why, please tell me) and Ocaml (Timeout, I don't know why, please tell me).

Djoums Unnamed CodinGamer

over 3 years ago

For F# the problem is that you reuse an array (words) that is initially sized at 1 and is then supposed to contain 8 elements. Strangely enough F# won't throw an error here, but it won't do anything either. Hence the index out of range exception in the cascade loop.

RABII_BEL_HADJ Unnamed CodinGamer

over 3 years ago

java code does not compile

JBM Unnamed CodinGamer

over 3 years ago

We know. See other comments. Replace "_1" with "+1" in the meantime.

eulerscheZahl Unnamed CodinGamer

over 3 years ago

Me, looking at the "Hard Game" testcase: how do these moves work, I don't understand them. Let's print the 5th one and watch the replay. string line = ""; for (int i = 0; i < 14; i++) { line = Console.ReadLine(); Console.Error.WriteLine(line); } Console.Error.WriteLine("action: " + line.Split()[4]); Console.WriteLine(line.Split()[4]); Game: The game has crashed. Please contact the author and enclose the following error: java.lang.IllegalArgumentException: Trying to remove more than can be at once at re.desast.freecell.Cascade.removeCards(Cascade.java:32) at re.desast.freecell.Move.performRetrieve(Move.java:34) at com.codingame.game.Referee.performAndViewMove(Referee.java:126) at com.codingame.game.Referee.humanPlay(Referee.java:98) at com.codingame.game.Referee.gameTurn(Referee.java:38) at com.codingame.gameengine.core.GameManager.start(GameManager.java:122) at com.codingame.gameengine.core.RefereeMain.start(RefereeMain.java:67) at com.codingame.gameengine.core.RefereeMain.main(RefereeMain.java:50)

JBM Unnamed CodinGamer

over 3 years ago

Full code http://chat.codingame.com/... On it, thanks!

Illedan Unnamed CodinGamer

over 3 years ago

Stub crashes in C#

Unnamed CodinGamer

over 3 years ago

In some languages "fc+1" became "fc_1"... Stub magic. ^^

JBM Unnamed CodinGamer

over 3 years ago

The gameplay is complete. The graphics are complete. A few backward-compatible cosmetic and playing options are in the pipe, but as far as solving goes, it's ready and good to try out, and actually publishable as far as my standards go. So I'm taking it out of WIP. Now maybe I could start working on a solver of my own. My todo-lists: ☑ cleanup ☐ more cleanup ☑ graphics ☑ game number ☑ a tolerable icon ☑ end screen ☑ animation ☑ card counter ☐ supermove split ☐ bonus end screen ☑ game protocol ☐ autoplay selector ☑ move queueing

ABOUT US CAREERS PRIVACY FAQ BLOG

The #1 tech hiring platform

SCREENING INTERVIEWING