CR Weekend Challenge

Simon André Forsberg

12:43 AM

Current Weekend Challenge: Poker Hand Evaluation

retailcoder

Feel free to go overboard all you want!!

rolfl

2:25 AM

And ... I'm going to bed.... I'll stew on it.

retailcoder

6:29 AM

Ready to post :)

retailcoder

6:50 AM

Posted :D

Stack Exchange

0

Q: Poker Hand Evaluator Challenge

This week's review challenge is a poker hand evaluator. I started by enumerating the possible hands: public enum PokerHands { Pair, TwoPair, ThreeOfKind, Straight, Flush, FullHouse, FourOfKind, StraightFlush, RoyalFlush } Then I thought I was going to need c...

Simon André Forsberg

11:03 AM

@retailcoder You're done already? WTF?

retailcoder

1:32 PM

@sim went a little too fast :)

Simon André Forsberg

@retailcoder I'm still struggling with my code here. I have started it but I'm not sure if I am on a good track or not.

retailcoder

1:43 PM

My code sux, I might repost at one point..

Simon André Forsberg

I haven't watched your code yet, I try to avoid that until I post my own result

retailcoder

@sim I posted the minute all tests passed... which was dumb :/

Simon André Forsberg

So either it works fine but parts of it are not good code quality, or the tests passed but you had not tested everything?

I'm quite sure I am making this code more complicated than it needs to be, but I like flexibility so :)

retailcoder

@sim The bar is 6-7 notches below the quality of my RPLSL solution..

Simon André Forsberg

wow

Will be very interesting to read your solution later

retailcoder

1:54 PM

Lol yeah I coded crap

Simon André Forsberg

Well, at least it works

I don't even have a working solution yet

retailcoder

I just rushed delivery so I would post before @rolfl would post his :)

@SimonAndréForsberg oh sh!t, I only accounted for the [Ace] as a 14... my [Straight] fails with [A-2-3-4-5]...

Simon André Forsberg

So you missed a test after all

The Fastest Gun In The West is apparently not always The Best Gun In The West.

Well, the weekend challenges are about learning things, and I guess you will learn plenty of things from this week

retailcoder

Mostly about posting good CR questions - it should really be your best possible code, after you've tried everything you could to come up with good code - akin to SO's what have you tried? ...though we can't really enforce it.

Simon André Forsberg

Yeah, I've also thought about that one. I don't think we need a "What have you tried?" but I think we might need a "What does your code do?". Sometimes people just post their code without explaining it much, if they even explain something about it at all...

rolfl

2:11 PM

I have note yet started....

I have been burning my candles from both ends recently .... and today I am spending time 'napping' and with the kids.... mostly.

I may get to it this evening.

(so I am not looking at any solutions)

Simon André Forsberg

@rolfl Take the time you need (or they need) with your family :) The weekend challenge is not in a hurry :)

retailcoder

2:26 PM

Ugh. I shouldn't have rushed this. TBH I'm not very happy with this code, it's not my best possible code and I feel like saying "scratch that, I'll post again when it meets my quality bar" - I didn't even do a single pass of refactoring, it's really the 1st "draft"... mainly I wanted to post my solution before @rolfl posted his; and it's not a race. :/ — retailcoder 38 secs ago

I deserve a downvote or two.

Simon André Forsberg

@retailcoder Downvoted by request :)

rolfl

@retailcoder - best lessons are learned from other people's mistakes... lessons best learned are from your own mistakes:

2

in Code Review General Room, Nov 30 at 17:51, by Simon André Forsberg

OK, for next week's challenge, I'm definitely going to think more and not get stuck in the get-it-done-fast trap

You failed to learn frm 2 people's lessons .... huh... !!! ;-)

Now, don't do it again!

Simon André Forsberg

Haha, lol! I had forgot I made that mistake last week!

rolfl

hehe .... I down-voted it,..... but did not read it .... ;-)

Simon André Forsberg

Same here :)

rolfl

2:34 PM

bad, bad, bad CR etiquette

Simon André Forsberg

lol

Well, at least we did it on request. So I'd say it's good etiquette.

retailcoder

@rolfl lesson learned! And very well said / starred :)

Simon André Forsberg

2:58 PM

Yay, first tests passed!

Not checking everything yet though. Gotta add more tests. And more checks. Still not checking for straight for example.

retailcoder

3:14 PM

@sim hint: if you're checking all hands individually, some checks will be redundant :)

Simon André Forsberg

@retailcoder I know. I think I avoid that quite good.

I do, however, want my code to be quite flexible, which sometimes does not go hand in hand with avoiding redundancy..

Gotta eat now, will implement straight checks later.

Simon André Forsberg

4:00 PM

OK, time to check for straights!

retailcoder

4:40 PM

2

Q: Proposals for Weekly Challenge topics

To decide which subjects we should cover in the Code Review Weekend Challenges I am creating this meta-question. Propose one challenge per answer. The answer with the most up-votes each week gets chosen as the weekend challenge. (Answers containing already finished challenges is disregarded in c...

discussion featured fun

Added a note about the get-it-done-fast trap

retailcoder

5:51 PM

@SimonAndréForsberg Ugh getting that straight straight is fun eh?!

Simon André Forsberg

@retailcoder Your pun was funnier.

getting a straight straight....

retailcoder

:)

@SimonAndréForsberg I think I messed up with my enum-based CardNominalValue - works for aces high, but aces low can't be in the enum because it would break my CardDeck class (not published)... and it wouldn't make sense anyway. I think.

Simon André Forsberg

@retailcoder In that case it sounds like you're screwed! :P

I have the Ace High & Low thing working fine I believe

retailcoder

nanananah - if there's a will, there's a way.

Simon André Forsberg

I just gotta straighten out these straights...

retailcoder

6:11 PM

@SimonAndréForsberg I'm trying to make mine work with an unknown number of cards, i.e. IsStraight=*HasAnyFiveConsecutiveCards*. B!tch and a half.

Simon André Forsberg

@retailcoder I'm also trying to work with an unknown number of cards, and find the best PokerHand out of them

PokerHand = 5-card poker hand

to make it work with Texas Hold'em for example

retailcoder

Me too :)

Simon André Forsberg

Also, to make things even worse for me, I want to support wildcards as well. And that's where things are starting to go not so straight ahead anymore for the straights...

For example: 2, 3, WILDCARD, 5, 6 should be a straight... Right now, it's not :P

retailcoder

6:27 PM

@SimonAndréForsberg I thought of including jokers at first, but I changed my mind. Do casinos allow wildcards?

Simon André Forsberg

@retailcoder I have no idea, but I know some video-game poker programs that do. Although I have never seen it combined with more than five cards.

retailcoder

@SimonAndréForsberg then you might as well go all the way and have a random wildcard that changes with each deal! :D

Simon André Forsberg

@retailcoder Lol, well... Let's not overdo it, shall we? :P

For now, only jokers are wild.

I'm having enough problems straightening things out already...

retailcoder

I got the ace high / ace low straight now. Damn I'm struggling more than I thought I'd be on that one!!

Hi @Jamal!

Jamal

Hi! I've updated the tag wiki to refer to the new Meta post, which I mentioned this room second (as secondary). It seems better for the main location of problem submission to be Meta rather than chat (but the challenge can still be pinned here).

retailcoder

6:36 PM

Good!

@Jamal So you're not dating the tag wiki anymore? (nice typo :)

Jamal

Yes, I meant updated. :-P (I've edited it twice already, so I suppose you're not entirely wrong...)

Simon André Forsberg

Good work, Jamal

OK, I'm seriously considering changing approach to this method to a slower variant since I can't get this method to work correctly.

retailcoder

I somehow wish I'd still have that VB6 code from 10 years ago somewhere...

Simon André Forsberg

6:54 PM

@retailcoder Be careful what you wish for.

retailcoder

@SimonAndréForsberg yeah, well, that code is long lost. And wouldn't be applicable anyway - I remember it was a hard-coded 5-cards hand.

Simon André Forsberg

And it probably looked like codecrap as well?

Violà, changed approach and now that method is working.

@retailcoder Have you considered the case where you have 10 cards, and 5 of those give you a straight and the other five give you a flush? What would your code return then? (Not a straight flush I hope)

retailcoder

@SimonAndréForsberg I'll test that and return result; :)

@SimonAndréForsberg wow you just broke my IsFlush()... I definitely need more tests...

Simon André Forsberg

@retailcoder I had a feeling about that :) I have thought about the problem but don't have a solution for it yet

retailcoder

Fixed IsFlush(): return Suits(cards).Any(suit => suit.Count() == 5);

I think the result should be the highest winning hand then

Simon André Forsberg

7:04 PM

Yes

And Flush is better than Straight, so it should be a flush.

retailcoder

Ahem. Fixed IsFlush(): return Suits(cards).Any(suit => suit.Count() >= 5);

Simon André Forsberg

haha

return Suits(cards).Any(suit => suit.Count() >= 5); seems like a lovely statement, but such beauty does not exists in Java yet. suit => suit.Count()... is some kind of Lambda expression, right? Or what is it called?

retailcoder

Should IsFourOfKind() assume there's only 1 deck of cards in the game?

lambda indeed

Simon André Forsberg

@retailcoder Why prevent it?

Java 8 will get Lambdas I have heard (looking forward to it)

retailcoder

I've never written in .net 2.0 ever since LINQ appeared in C# :)

Simon André Forsberg

7:10 PM

Oh, it's LINQ that adds that? You make me want to code C# :P

Simon reminds himself about Visual Studio - Desire to code C# disappeared

retailcoder

@SimonAndréForsberg no, LINQ is essentially a set of extension methods on IEnumerable<T> - that's Any(), GroupBy(), Where() and many, many others.

And they mostly work with lambdas

Doesn't Java have delegates?

Simon André Forsberg

@retailcoder Not in the same way. Java have interfaces. So even though you could have the same functionality, you'd need a whole lot of code to do it.

But, Java 8 will be an improvement there.

retailcoder

Hmm so Java is more akin to COM-style event handling right?

Simon André Forsberg

COM?

retailcoder

The windows COMponent model. See com :)

Simon André Forsberg

7:16 PM

Yeah, I guess it is. (@rolfl probably knows more about this than I do)

retailcoder

@SimonAndréForsberg back to Poker, if your game has two decks with jokers (i.e. 108 cards), and your hand is A-A-A-A-A, that should be a FourOfKind + HiCard(Ace).

Simon André Forsberg

@retailcoder My code is not dependent on how many decks there are, or what they contain.

It would say that it has Four of a kind + Ace "kicker"

retailcoder

But if your hand has 10 cards and you get A-A-A-A-A-2-2-Q-Q-Q, it should output "CHEATER!" right?

Simon André Forsberg

@retailcoder I don't have any cheater condition check.

retailcoder

RollPlayerInTar(); RollPlayerInFeathers(); NowDance();

Simon André Forsberg

7:20 PM

Don't see any reason to include cheater condition check here either. That is up to the calling methods to determine if the hand should even be possible or not.

at least that's my approach

retailcoder

So it would be FourOfKind(Ace)+FullHouse+Ace.

Simon André Forsberg

Well, a poker hand can only be 5 cards (in my code at least). So I would only return FourOfKind(Ace) + Ace. But yes, IsFullHouse (or however your code looks) would return true as well.

retailcoder

Makes sense

Simon André Forsberg

Now I gotta add more tests. I think my code works fine now. But I don't want to do the same mistakes as others have done ;)

rolfl

7:52 PM

Saw my name pop up here.... this is why I am not writing a game at the moment:

0

A: Find elements occurring even number of times in an integer array

Wiki Answer - discussion of scalability. After some investigation I thought it would be interesting to publish some of the performance results I have seen.... there are some interesting observations. I have taken two algorithms, both on this page, the 200_Success one, and the rolfl one. I have ...

And I should be ** working** !!!!

Simon André Forsberg

8:14 PM

@rolfl Impressive

Simon André Forsberg

8:29 PM

Feels like my test code is longer than my main code... :P

retailcoder

@SimonAndréForsberg that means your code is well covered :)

I have like 15 tests.... doesn't feel right, but I'm somewhat sure my code works

Simon André Forsberg

@retailcoder Now you just said something! I gotta make an actual covering analysis as well!

rolfl

cobertura?

retailcoder

If you want to...

Simon André Forsberg

@rolfl EclEmma

I need tests! Can someone give me a bunch of cards to test? @retailcoder?

retailcoder

8:44 PM

@SimonAndréForsberg DAMN! the (ding!) made me click ths [Post your Question] button!!

0

Q: Poker Hand Evaluator, take 2

This is following up on my previous attempt, which was admittedly done fast and not-so-well. This one is more tested, and abstracts the dirt away into an interface: public interface IPokerHandEvaluator { bool IsPair(IEnumerable<PlayingCard> cards); bool IsTwoPair(IEnumerable<PlayingCard>...

c# linq weekend-challenge

I think it's fine..

Simon André Forsberg

@retailcoder Then click "Edit", then "Delete", then continue editing? :)

retailcoder

lol

nah, I was just re-reading the last bits

@SimonAndréForsberg Don't you have a CardDeck class that you can Shuffle() and Draw() poker hands out of?

Simon André Forsberg

@retailcoder Good thinking!! (I do have such a class, didn't think about integrating it into this project :P)

a is incorrect though, you'd be surprised at how many Card-related classes I have...

Stack Exchange

0

Q: Poker Hand Evaluator, take 2

This is following up on my previous attempt, which was admittedly done fast and not-so-well. This one is more tested, and abstracts the dirt away into an interface: public interface IPokerHandEvaluator { bool IsPair(IEnumerable<PlayingCard> cards); bool IsTwoPair(IEnumerable<PlayingCard>...

c# linq weekend-challenge

Simon André Forsberg

Today's song should really be: youtube.com/watch?v=bESGLojNYSo

retailcoder

8:55 PM

lol I knew it was that song before I even clicked the link!

retailcoder

9:25 PM

Should I vote to close my first post?

Simon André Forsberg

for what reason?

(Still haven't read it btw)

retailcoder

not-working code...

it doesn't account for aces low

Simon André Forsberg

Ah

I don't know... I guess it works partially

you weren't aware of the problem at the time of posting

retailcoder

..not at the time of posting, indeed. (ignorance is bliss)

Simon André Forsberg

Maybe I will change my mind later when actually reading your question

Found a problem here. [JACK-DIAMONDS, ACE-HEARTS, 7-SPADES, 2-DIAMONDS, 2-HEARTS, 4-DIAMONDS, WILDCARD] should be three of a kind (2s), but it becomes two pair (ace + 2s)

Gotta fix this

Next challenge, let's do something a bit easier, shall we? Something that even I can't over-complicate...

retailcoder

9:46 PM

@SimonAndréForsberg I guess @rolfl's (ding! still here?) Sudoku Solver doesn't really fall in that category does it?

Looking at @radarbob's answer on the first post, I suspect upcoming weekend-challenge posts will push it to determining the winning hand out of two... — retailcoder 9 mins ago

Jamal

Can't we just do a super-sophisticated and over-engineered Hello, World!?

2

retailcoder

lol

Simon André Forsberg

@retailcoder Since I already have a flexible tree-solving algorithm around here, I actually don't think it will be that complicated :)

Jamal

Classes for "hello" and "world", at least.

retailcoder

@Jamal OMG.

Simon André Forsberg

9:48 PM

@Jamal Sure! I've already done a super-sophisticated Fizz-Buzz-Banana in Java a while ago, so Hello World should be the next step! (This is why I wanted to make Fizz-Buzz-Banana in another language)

Jamal

...or would that just delve into bad code, therefore being off-topic?

Simon André Forsberg

@Jamal Naah, one class for each letter. And at least two interfaces.

The worst part is, it would actually work...

retailcoder

@Jamal Not necessarily. Of course things like what @SimonAndréForsberg has just mentioned would be, but a super-overkill implementation of Dependency Injection could eventually turn into a nice CW.

Jamal

I have a feeling making a class for each letter would just be downvote-bait.

Simon André Forsberg

And one class for each word wouldn't?

rolfl

9:51 PM

@retailcoder I am working I am on my work computer, I have ding! turned off.

Jamal

It could, but two classes are better than... however many characters are in that string.

Simon André Forsberg

HELOWRD, 7. Or 8 if including space.

And don't forget the interfaces.

rolfl

Hey Jamal, did you ever follow that 99bottlesofbeer link I put in ideone ... ?

Let's write a program that makes that 'readable'.

Jamal

I didn't understand a bit of it. :-P

rolfl

@retailcoder I answered a C# question ... can you please check it and curse me out if I am wrong?

0

A: Running a block of code on every interval except the first iteration

Initialize the first byte, and then start the loop from 1..... You may also want to do a bit-check instead of modulo, but they may be equally fast... if (dataSize > 0) { data[0] ^= key; } for (int i = 1; i < dataSize; i++) { // Increase the key as per the algorithm if (i & 3 == 0) { ...

Or anyone else for that matter.

Jamal

9:55 PM

Anyone brave enough to write something in Brainfuck? :P

2

rolfl

I am not brave enough to put those four letters together online.... ;-)

Although, I think I can use that language as a euphamism..... like all you brains here should get the bugs out of your code.

Jamal

It's not like I made up the language. :-P Okay, I created CR's tag for it, but still.

retailcoder

@Jamal >+++++++4+[>++++++<-]>++++.----.++++.*/

rolfl

Well, that was a brained up thing to do.

Jamal

Does that generate all the prime numbers from 2 to 1000?

rolfl

9:59 PM

1000 is prime ?

retailcoder

lol. prints "404"

Jamal

Less than 1000. Whatever...

rolfl

Did you guys know that the largest signed 32-bit value is prime?

Simon André Forsberg

Welcome, @syb0rg

2147483647 ? Really?

(Memorized that number 10 years ago)

rolfl

It's quite an easy large prime number to remember....

syb0rg

10:01 PM

@SimonAndréForsberg Hi, that response almost seemed automated.

Haha

Simon André Forsberg

@syb0rg Trust me, it wasn't :)

Wow, it really is... wolframalpha.com/input/?i=2147483647&t=mfftb01

syb0rg

I believe you, just coming over for a visit.

rolfl

math.com/students/calculators/source/prime-number.htm

type in 2147483647

syb0rg

I was checking to see if there was a weekend challenge in the making really.

rolfl

Wait a fe seconds

Simon André Forsberg

10:02 PM

@syb0rg Yup, there is. @retailcoder has posted his code already, and I am working on mine.

The current challenge is: Poker Hand Evaluation

rolfl

@retailcoder first posted his broken code, then is partially broken code.....

retailcoder

@syb0rg Hello! codereview.stackexchange.com/questions/36873/…

syb0rg

The Poker Hand thing? That's what I thought it was, but I was just checking the chat logs to make sure it was it.

retailcoder

@rolfl lol.

syb0rg

@retailcoder Hi! Unfortunately, I don't program in C#. Java and C are my main languages.

retailcoder

10:04 PM

@syb0rg the current challenge will always be pinned in the starred posts list, and you can implement it in the language you want!

Simon André Forsberg

@syb0rg You can choose any language you want for the challenge :) I'm doing it in Java myself.

syb0rg

@retailcoder Thanks, I didn't notice that at first. Good tip!

Simon André Forsberg

@rolfl Thanks for instructing me to hang my browser for a couple of seconds. (I think I'll use wolframalpha to check/calculate primes for me instead)

Jamal

11:15 PM

5 messages deleted

Simon André Forsberg

@retailcoder My code seems to be working fine now, but perhaps I haven't added enough "equals" checks. For example, [2, 3, King, 10, 10, 12, 12] vs. [4, 5, King, 10, 10, 12, 12] should be equal in rank, right?

retailcoder

11:32 PM

@SimonAndréForsberg depends the suits of the pairs would be a tiebreaker

Simon André Forsberg

@retailcoder Why would the suits matter?

retailcoder

@SimonAndréForsberg "Just one more thing" here:

1

A: Poker Hand Evaluator Challenge

The Design may be Incomplete The readability concern in the PokerGame class can be addressed by encapsulation instead of syntactical slight of hand. Push details down is the mantra. Addresses the concern about a fungible Dictionary by enveloping (encapsulating) it. Would a HandEvaluator class...

Simon André Forsberg

No no no, I don't want to read your code yet!

I don't want to be corrupted

retailcoder

corrupted

Simon André Forsberg

that's what I said :P

Transcript for

CR Weekend Challenge