Problem
Solution
The structure is clear here: a
Deck contains n number of cards. So for Pocker, the Deck contains 52 cards.
A card is compose of suit and a rank. Suit can be like spade, hearts, diamond and club (in order) :
This can be treated as enum :
Each card has a numerical value from 1 to 13 in the order:
So, in the above example we have shown the deck containing the card from 2 to 14 for spades.
This should also be treated as enum :
If you think about a card game, different games differ from ways of dealing cards and putting cards back in. So we can have a set of abstract methods inside the class ‘Deck’ to allow sub-class implements its own way of dealing.
I think you can achieve what you want with a single deck class like below which is essentially a wrapper around Stack, I don't see why any particular deck/pile/hand would not want most if not all of the same methods.
The only real problem i see is with the
The class diagram I draw is here:
So, card will look like this:
and deck will contain cards:
You can find the detailed classes at my github (kinshuk4). Thanks.
Design the data structures for a generic deck of cards. Explain how you would sub-class it to implement particular card games.
Solution
The structure is clear here: a
Deck contains n number of cards. So for Pocker, the Deck contains 52 cards.
A card is compose of suit and a rank. Suit can be like spade, hearts, diamond and club (in order) :
This can be treated as enum :
public enum Suit { CLUBS(1), SPADES(2), HEARTS (3), DIAMONDS(4) ; private int suitValue; private Suit(int suitValue_){ suitValue = suitValue_; } }
Each card has a numerical value from 1 to 13 in the order:
So, in the above example we have shown the deck containing the card from 2 to 14 for spades.
This should also be treated as enum :
public enum Rank { TWO(2), THREE(3), ... KING(13), ACE(14) }
If you think about a card game, different games differ from ways of dealing cards and putting cards back in. So we can have a set of abstract methods inside the class ‘Deck’ to allow sub-class implements its own way of dealing.
I think you can achieve what you want with a single deck class like below which is essentially a wrapper around Stack, I don't see why any particular deck/pile/hand would not want most if not all of the same methods.
The only real problem i see is with the
deal()
method and for that we will a
Player class and a Dealer class that extends Player and implements the
logic of dealing a deckThe class diagram I draw is here:
So, card will look like this:
public class Card implements Comparable<Card>{ private Suit suit; private Rank rank; public Card(Suit suit, Rank rank) { this.suit = suit; this.rank = rank; } @Override public int compareTo(Card card2) { if(card2==null) return 1; int suit2 = card2.getSuit().getSuitValue(); int rank2 = card2.getRank().getCardValue(); if(suit.getSuitValue()!=suit2) return suit.getSuitValue() - suit2; if(rank.getCardValue() != rank2) return rank.getCardValue() - rank2; return 0; } }
and deck will contain cards:
class Deck { private Stack<Card> cards = new Stack<Card>(); public Deck() { } public Deck(int numberOfCards) { for (int i=0; i<numberOfCards; i++) { cards.push(CardFactory.createCard(i)); } } private void shuffle() { Collections.shuffle(this.cards); } public void sort() { Collections.sort(this.cards); } public void removeAllCards() { this.cards.removeAllElements(); } public void removeCard(Card c) { int i = this.cards.search(c); this.cards.remove(i); } public Card getCard(Card c) { int i = this.cards.search(c); return this.cards.get(i); } public Card getTopCard() { return this.cards.pop(); } public Card getNthCard(int i) { return this.cards.get(i); } public Card addCard(Card c) { this.cards.push(c); return c; } }
You can find the detailed classes at my github (kinshuk4). Thanks.
0 comments:
Post a Comment