class HacBotII(PokerBot, Htapi):
"""
Hac's policy-based bot.
This is a bot able to shoot moon. When I get a lot of big-rank cards. DO IT!
Don't want to avoid opponents shooting the moon, because I don't loss any money :-)
"""
SM_THOLD_PASS3 = 0.6
SM_THOLD_PICK = 0.1
def __init__(self, name, is_debug=False):
super(HacBotII, self).__init__(name)
self.htapi = Htapi(is_debug=is_debug)
self.players = {}
self.stat = {}
self.stat['roundCard'] = []
self.stat['usedCard'] = []
self.stat['nextPlayers'] = []
self.big_rank_cards = [
Card('AS'),
Card('KS'),
Card('JS'),
Card('TS'),
Card('AH'),
Card('KH'),
Card('QH'),
Card('JH'),
Card('TH'),
Card('AD'),
Card('KD'),
Card('QD'),
Card('JD'),
Card('TD'),
Card('AC'),
Card('KC'),
Card('QC'),
Card('JC'),
Card('TC'),
]
self.score_cards = [
Card("QS"),
Card("TC"),
Card("2H"),
Card("3H"),
Card("4H"),
Card("5H"),
Card("6H"),
Card("7H"),
Card("8H"),
Card("9H"),
Card("TH"),
Card("JH"),
Card("QH"),
Card("KH"),
Card("AH")
]
self.plz_rebuild_players = True
self.stat['expose_ah_mode'] = False
self.stat['shoot_moon_mode'] = True
def _rebuild_players(self, data):
"""
Configure player table by input data.
"""
self.htapi.dbg("Rebuild players: ", format(data))
data_players = data['players']
if len(data_players) != 4:
self.errmsg("Invalid player number " + str(len(data_players)) +
" in server msg")
self.players = {}
for dp in data_players:
self.players[dp['playerName']] = {
'playerName': dp['playerName'],
'score_accl': 0,
'score': 0,
'shoot_moon': 0,
'expose': False,
'shoot': [],
'suit_leak': [],
'pick': [],
}
def new_game(self, data):
"""
Event: The start of a new game.
"""
if self.plz_rebuild_players == True:
self._rebuild_players(data)
self.plz_rebuild_players = False
def receive_cards(self, data):
"""
Event: Receive my 13 cards.
"""
self.stat['hand'] = self.get_cards(data)
def _select_card2pass_shoot_moon_mode(self, my_hand_cards):
"""
Pop out 1 card to pass
The rule here is to pick the small cards in shortage
"""
#
# Find the non-empty suit but the card num is low.
#
# TODO: Try to remove non-heart suit can help shoot-moon ability.
#
card_num_stat = {'S': 0, 'H': 0, 'D': 0, 'C': 0}
for c in my_hand_cards:
card_num_stat[c.get_suit()] += 1
card_num_stat_sorted = sorted(card_num_stat.iteritems(),
key=lambda (k, v): (v, k))
#
# Try to remove a suit
#
for di in card_num_stat_sorted:
k, v = di
if v != 0:
pick_suit, v = di
tgt_cards = self.htapi.get_cards_by_suit(
my_hand_cards, pick_suit)
tgt_cards = self.htapi.arrange_cards(tgt_cards)
smaller_card = self.htapi.pick_smaller_card(
tgt_cards, [Card('9S')])
if smaller_card != None:
return self.htapi.remove_card(my_hand_cards, smaller_card)
#
# In case...
#
for di in card_num_stat_sorted:
k, v = di
if v != 0:
pick_suit, v = di
tgt_cards = self.htapi.get_cards_by_suit(
my_hand_cards, pick_suit)
tgt_cards = self.htapi.arrange_cards(tgt_cards)
card = tgt_cards[0]
return self.htapi.remove_card(my_hand_cards, card)
self.htapi.errmsg("BUG")
card = my_hand_cards[0]
return self.htapi.remove_card(my_hand_cards, card)
def _select_card2pass(self, my_hand_cards):
"""
Pop out 1 card to pass
"""
card = self.htapi.remove_card(my_hand_cards, Card('KS'))
if card != None:
return card
card = self.htapi.remove_card(my_hand_cards, Card('AS'))
if card != None:
return card
card = self.htapi.remove_card(my_hand_cards, Card('AC'))
if card != None:
return card
card = self.htapi.remove_card(my_hand_cards, Card('KC'))
if card != None:
return card
card = self.htapi.remove_card(my_hand_cards, Card('QC'))
if card != None:
return card
card = self.htapi.remove_card(my_hand_cards, Card('JC'))
if card != None:
return card
# Found the suit in shortage
suit_list = ['S', 'H', 'D', 'C']
pick_suit = None
pick_suit_num = 0
for suit in suit_list:
this_suit_num = self.htapi.calc_card_num_by_suit(
my_hand_cards, suit)
if this_suit_num == 0:
continue
if pick_suit_num == 0:
pick_suit_num = this_suit_num
pick_suit = suit
elif this_suit_num < pick_suit_num:
pick_suit_num = this_suit_num
pick_suit = suit
# Remove the most large card in the target suit
candidate_list = self.htapi.get_cards_by_suit(my_hand_cards, pick_suit)
candidate_list = self.htapi.arrange_cards(candidate_list)
card = candidate_list.pop()
return self.htapi.remove_card(my_hand_cards, card)
def _pass_cards_anti_score_mode(self, data):
"""
Pick 3 cards which can avoid taking score
"""
my_hand_cards = self.stat['hand']
output = []
for i in range(3):
card = self._select_card2pass(my_hand_cards)
if card == None:
self.htapi.errmsg("Cannot pick a card to pass")
output.append(card.toString())
self.htapi.dbg(self.get_name() + " pass 3 cards: " + format(output))
return output
def _pass_cards_shoot_moon_mode(self, data):
"""
Pick 3 cards which can help myself to shoot moon.
"""
my_hand_cards = self.stat['hand']
output = []
for i in range(3):
card = self._select_card2pass_shoot_moon_mode(my_hand_cards)
if card == None:
self.htapi.errmsg("Cannot pick a card to pass")
output.append(card.toString())
self.htapi.dbg(self.get_name() + " pass 3 cards: " + format(output))
return output
def _do_i_have_too_many_low_rank_cards(self):
my_hand_cards = self.stat['hand']
low_rank_card_num = 0
for c in my_hand_cards:
if c.get_rank_num() < Card("TS").get_rank_num:
low_rank_card_num += 1
# TBD
if low_rank_card_num >= 9:
return True
else:
return False
def pass_cards(self, data):
self.stat['hand'] = [Card(x) for x in data['self']['cards']]
self.stat['hand'] = self.htapi.arrange_cards(self.stat['hand'])
self.htapi.dbg("Select 3 cards from: " + format(self.stat['hand']))
sm_ability = self._calc_shoot_moon_ability(data)
if sm_ability > self.SM_THOLD_PASS3:
self.htapi.dbg("shoot moon mode pass3: " + str(sm_ability))
return self._pass_cards_shoot_moon_mode(data)
else:
self.htapi.dbg("anti score mode pass3")
return self._pass_cards_anti_score_mode(data)
def receive_opponent_cards(self, data):
"""
Event: Recv 3 cards from one opponent
"""
pass
def expose_my_cards(self, data):
"""
Event: The server asks me to expose AH...
"""
my_hand_cards = self.stat['hand']
output = []
candidates = data['cards']
if candidates == None or len(candidates) == 0:
return output
candidates = [Card(x) for x in candidates]
if self.htapi.find_card(candidates, Card('AH')) == None:
return output
if my_hand_cards == None or len(my_hand_cards) == 0:
return output
#
# TODO: Be smarter... calculate shoot-moon ability, or anti-score ability
# If I know I won't take any score, expose AH!
# If I want to shoot the moon, expose AH!
#
# Calculate the rate that I will get penalty...
# The simplest calculation is to check if I have too many big-rank cards
my_big_cards = self.htapi.find_cards(my_hand_cards,
self.big_rank_cards)
my_big_card_num = len(my_big_cards)
if my_big_card_num > 2 and my_big_card_num < 5:
"""
Have some but not many high-rank.
"""
return output
#
# Expose AH!
#
output = ['AH']
return output
def expose_cards_end(self, data):
"""
Event: Check if somebody expose AH. Damn.
"""
data_players = data['players']
for dp in data_players:
local_player = self.players[dp['playerName']]
if len(dp['exposedCards']) > 0:
# This guy exposed AH.
local_player['expose'] = True
self.stat['expose_ah_mode'] = True
else:
local_player['expose'] = False
def _lead_play_monte_carlo_predict_penalty(self,
card2shoot,
opponent_cards,
round_cards=[]):
"""
Will I take the trick if I select the card2shoot!?
Output: The possibility to get score if I shoot the card: 0 ~ 100.0
"""
# TODO: Improve prediction by the players!
# next_players = self.stat['nextPlayers']
# next_player_num = len(next_players)
# Calculate the number of cards will... make me eat the trick.
opponent_same_suit_cards = self.htapi.get_cards_by_suit(
opponent_cards, card2shoot.get_suit())
if len(opponent_same_suit_cards) == 0:
opponent_score_cards = self.htapi.find_score_cards(opponent_cards)
if len(opponent_score_cards) > 0:
return 100.0
else:
# There's no penalty card now. Won't eat
return 0.0
# Danger level means the chance I will eat a trick.
# If the card2shoot is high-rank, the chance is higher, of course.
danger_level = 0
for ocard in opponent_same_suit_cards:
if ocard.get_rank_num() < card2shoot.get_rank_num():
# Oops...
danger_level += 1
percentage_max = float(len(opponent_same_suit_cards))
percentage = danger_level / percentage_max
return percentage
def _get_unused_cards_by_suits(self, my_cards, suits=[]):
"""
Get unused cards of other opponents so that I can know how many cards are left..
"""
all52cards = self.htapi.get52cards()
output = all52cards
self.htapi.remove_cards(output, self.stat['usedCard'])
self.htapi.remove_cards(output, my_cards)
if len(suits) == 0:
# Do not need to pick a specific suit
pass
else:
# Pick unused cards of target suits
output = self.htapi.get_cards_by_suits(output, suits)
return output
def __leadplay(self, data):
"""
I am the round leader. I want to avoid tacking the trick, so choose the best!
"""
my_hand_cards = self.stat['hand']
my_avail_cards = [Card(x) for x in data['self']['candidateCards']]
opponent_cards = self._get_unused_cards_by_suits(my_hand_cards)
selected = []
peak_percentage = 0
for c in my_avail_cards:
# If I pick this card, will I take the trick?
percentage = self._lead_play_monte_carlo_predict_penalty(
c, opponent_cards, round_cards=[])
if len(selected) == 0:
peak_percentage = percentage
selected = [c]
elif percentage < peak_percentage:
peak_percentage = percentage
selected = [c]
elif percentage == peak_percentage:
selected.append(c)
# Prefer a lower number suit
all_suit = []
for c in selected:
card_suit = c.get_suit()
if all_suit.count(card_suit) == 0:
all_suit.append(card_suit)
self.htapi.dbg("Selected candidates: " + format(selected) +
", the suits: " + format(all_suit) + ", all cards: " +
format(my_hand_cards))
prefer_suit = None
min_suit_num = 0
for suit in all_suit:
same_suit_num = self.htapi.get_cards_by_suit(my_hand_cards, suit)
if prefer_suit == None:
prefer_suit = suit
min_suit_num = same_suit_num
elif same_suit_num < min_suit_num:
prefer_suit = suit
min_suit_num = same_suit_num
prefer_cards = self.htapi.arrange_cards(
self.htapi.get_cards_by_suit(selected, prefer_suit))
self.htapi.dbg("Selected candidates: " + format(prefer_cards))
card2shoot = prefer_cards.pop(0)
self.htapi.dbg("Select card" + format(card2shoot))
return card2shoot
def __midplay(self, data):
round_cards = self.stat['roundCard']
my_hand_cards = self.stat['hand']
my_avail_cards = [Card(x) for x in data['self']['candidateCards']]
lead_card = round_cards[0]
my_lead_suit_num = self.htapi.calc_card_num_by_suit(
my_hand_cards, lead_card.get_suit())
if my_lead_suit_num > 0:
"""
Have the same suit. Choose a smaller card to avoid the trick.
"""
filtered_round_cards = self.htapi.get_cards_by_suit(
round_cards, lead_card.get_suit())
return self.htapi.pick_smaller_card(my_avail_cards,
filtered_round_cards,
auto_choose_big=True)
else:
"""
I don't have the same suit. Git rid of 'QS'.
"""
cardqs = self.htapi.find_card(my_avail_cards, Card('QS'))
if cardqs != None:
return cardqs
"""
I don't have the same suit. Throw dangerous high rank cards.
"""
high_card = self.htapi.pick_big_card(my_avail_cards)
if high_card.get_rank_num() >= Card('JS').get_rank_num():
return high_card
"""
I don't have the same suit. My chance to throw out point cards!
"""
candidate_list = []
candidate_list += self.htapi.find_cards(my_avail_cards,
[Card('TC')])
candidate_list += self.htapi.get_cards_by_suit(my_avail_cards, 'H')
if len(candidate_list) > 0:
return candidate_list.pop(0)
"""
Otherwise, pick the highest rank card.
"""
return high_card
def __lastplay(self, data):
round_cards = self.stat['roundCard']
my_hand_cards = self.stat['hand']
my_avail_cards = [Card(x) for x in data['self']['candidateCards']]
lead_card = round_cards[0]
my_lead_suit_num = self.htapi.calc_card_num_by_suit(
my_hand_cards, lead_card.get_suit())
score_card_num = self.htapi.find_cards(round_cards, self.score_cards)
if my_lead_suit_num > 0:
"""
Have the same suit. If there's no point on board, shoot the biggest card.
If there's a point, try to avoid.
"""
if score_card_num == 0:
return self.htapi.pick_big_card(my_avail_cards)
else:
return self.htapi.pick_smaller_card(my_avail_cards,
round_cards,
auto_choose_big=True)
else:
"""
Don't have the same suit. Play the trick.
"""
return self.__midplay(data)
def pick_card_anti_score_mode(self, data):
# roundPlayers is in the correct order of shooting cards.
round_players = data['roundPlayers']
# Identify my position in this round
my_pos = round_players.index(self.get_name())
# Get players in next turn.
self.stat['nextPlayers'] = data['roundPlayers'][(my_pos + 1):]
if my_pos == 0:
card = self.__leadplay(data)
elif my_pos == 3:
card = self.__lastplay(data)
else:
card = self.__midplay(data)
return card
def _calc_shoot_moon_ability(self, data):
"""
Detect if I can shoot the moon.
"""
if self.stat['shoot_moon_mode'] == False:
# I know it's unlikely to shoot the moon... Give up
return 0.0
# Detect factors: all played turn cards, hand_card
my_hand_cards = self.stat['hand']
all_used_cards = self.stat['usedCard']
opponent_unused_cards = self._get_unused_cards_by_suits(my_hand_cards)
opponent_unused_penalty_cards = self.htapi.find_penalty_cards(
opponent_unused_cards)
left_turn_num = len(my_hand_cards)
"""
The prediction is simple, calculate how many turns I can win.
"""
win_count = 0
for c in my_hand_cards:
unused_same_suit_cards = self.htapi.get_cards_by_suit(
opponent_unused_cards, c.get_suit())
if len(unused_same_suit_cards) > 0:
unused_same_suit_cards = self.htapi.arrange_cards(
unused_same_suit_cards)
if c.get_rank_num() > unused_same_suit_cards[-1].get_rank_num(
):
win_count += 1
self.htapi.dbg("shoot moon ability: " + str(win_count) + " / " +
str(left_turn_num))
return win_count / float(left_turn_num)
def __leadplay_shoot_moon_mode(self, data):
"""
I am the lead player. Try to eat heart and QS!
Make sure I have the a card that can win a score.
If I don't have, I will fail to shoot. Try Monte Carlo simulation.
"""
my_hand_cards = self.stat['hand']
my_avail_cards = [Card(x) for x in data['self']['candidateCards']]
all_used_cards = self.stat['usedCard']
opponent_cards = self._get_unused_cards_by_suits(my_hand_cards)
#
# Shoot the card that nobody has the same suit.
#
for c in my_avail_cards:
my_same_suit_card_num = self.htapi.get_cards_by_suit(
my_avail_cards, c.get_suit())
used_same_suit_card_num = self.htapi.get_cards_by_suit(
all_used_cards, c.get_suit())
if my_same_suit_card_num + used_same_suit_card_num == 13:
self.htapi.dbg("Select nobody have suit to play the trick.")
return c
#
# Heart eater.
#
my_heart_cards = self.htapi.get_cards_by_suit(my_avail_cards, 'H')
if len(my_heart_cards) > 0:
unused_heart_cards = self.htapi.get_cards_by_suit(
opponent_cards, 'H')
if len(unused_heart_cards) > 0:
unused_heart_cards = self.htapi.arrange_cards(
unused_heart_cards)
for c in my_heart_cards:
if c.get_rank_num() > unused_heart_cards[-1].get_rank_num(
):
return c
#
# Shoot the card that I will win the round.
#
for c in my_avail_cards:
unused_same_suit_cards = self.htapi.get_cards_by_suit(
opponent_cards, c.get_suit())
if len(unused_same_suit_cards) > 0:
unused_same_suit_cards = self.htapi.arrange_cards(
unused_same_suit_cards)
if c.get_rank_num() > unused_same_suit_cards[-1].get_rank_num(
):
return c
#
# If I don't have cards can win, ...shoot no score suit left a lot.
#
pick_suit = None
pick_suit_num = 0
for c in my_avail_cards:
unused_same_suit_cards = self.htapi.get_cards_by_suit(
opponent_cards, c.get_suit())
if pick_suit_num == 0:
pick_suit_num = len(unused_same_suit_cards)
pick_suit = c.get_suit()
elif len(unused_same_suit_cards) > pick_suit_num:
pick_suit_num = len(unused_same_suit_cards)
pick_suit = c.get_suit()
candidates = self.htapi.get_cards_by_suit(my_avail_cards, pick_suit)
candidates = self.htapi.arrange_cards(candidates)
return candidates[-1]
def __midplay_shoot_moon_mode(self, data):
"""
Check if the next players are in shortage of lead card suit.
If they are in shortage, shoot the biggest card I have to win the leader ship.
TBD: Use Monte Carlo simulation here?
"""
return self.__leadplay_shoot_moon_mode(data)
def __lastplay_shoot_moon_mode(self, data):
"""
I am the last player: See if there's a score card!
"""
my_avail_cards = [Card(x) for x in data['self']['candidateCards']]
round_cards = self.stat['roundCard']
lead_card = round_cards[0]
if self.htapi.calc_card_num_by_suit(my_avail_cards,
lead_card.get_suit()) == 0:
"""
I don't have the same suit. Pick a no-score small card :-)
"""
candidates = self.htapi.find_no_penalty_cards(my_avail_cards)
if len(candidates) == 0:
# Sorry, don't have no penalty card. The shoot-moon action will fail.
return self.pick_card_anti_score_mode(data)
#
# Detect the suit in shortage
#
card_num_stat = {'S': 0, 'H': 0, 'D': 0, 'C': 0}
for c in candidates:
card_num_stat[c.get_suit()] += 1
# Sort the card_num_stat
card_num_stat_sorted = sorted(card_num_stat.iteritems(),
key=lambda (k, v): (v, k))
# Fetch the suit in shortage, but ignore zero number suit.
for di in card_num_stat_sorted:
k, v = di
if v != 0:
pick_suit, v = di
break
candidates = self.htapi.get_cards_by_suit(candidates, pick_suit)
candidates = self.htapi.arrange_cards(candidates)
no_score_small_card = candidates.pop(0)
return no_score_small_card
elif self.htapi.calc_score(round_cards) > 0:
"""
I have the same suit. Try to win the score card.
"""
bigger_card = self.htapi.pick_bigger_card(my_avail_cards,
round_cards)
if bigger_card == None:
# Too bad, I can't win... switch to anti-score mode.
return self.pick_card_anti_score_mode(data)
else:
return bigger_card
else:
"""
I have the same suit. No score on board... should I take the leadership? Yes?! TBD?
"""
bigger_card = self.htapi.pick_bigger_card(my_avail_cards,
round_cards)
if bigger_card == None:
# Too bad, I can't win...
return self.htapi.pick_small_card(my_avail_cards)
else:
return bigger_card
def pick_card_shoot_moon_mode(self, data):
# roundPlayers is in the correct order of shooting cards.
round_players = data['roundPlayers']
# Identify my position in this round
my_pos = round_players.index(self.get_name())
# Get players in next turn.
self.stat['nextPlayers'] = data['roundPlayers'][(my_pos + 1):]
if my_pos == 0:
self.htapi.dbg("sm lead play")
card = self.__leadplay_shoot_moon_mode(data)
elif my_pos == 3:
self.htapi.dbg("sm last play")
card = self.__lastplay_shoot_moon_mode(data)
else:
self.htapi.dbg("sm mid play")
card = self.__midplay_shoot_moon_mode(data)
return card
def pick_card(self, data):
"""
Event: My turn to shoot a card.
"""
self.stat['hand'] = [Card(x) for x in data['self']['cards']]
if self._calc_shoot_moon_ability(data) >= self.SM_THOLD_PICK:
self.htapi.dbg("shoot moon mode")
card = self.pick_card_shoot_moon_mode(data)
else:
self.htapi.dbg("aniti-score mode")
card = self.pick_card_anti_score_mode(data)
# NOTE: Do not remove the card in hand card, do it ad turn end event.
self.htapi.dbg(self.get_name() + " shoot card: " + format(card) +
", from: " + format(data['self']['cards']) +
", next players: " + format(self.stat['nextPlayers']))
return card.toString()
def _detect_card_shortage(self, local_player, turn_card):
#
# If the player is not lead player and he shoots a card not the same as lead card,
# mark the player as card shortage.
#
# If I know the shortage status of players, can have advantage to predict.
#
round_cards = self.stat['roundCard']
if len(round_cards) <= 1:
# The turn player is leader. Ignore.
return False
lead_card = round_cards[0]
if lead_card.get_suit_num() != turn_card.get_suit_num():
if local_player['suit_leak'].count(lead_card.get_suit()):
local_player['suit_leak'].append(lead_card.get_suit())
self.htapi.dbg("Player: " + local_player['playerName'] +
" card leakage: " + lead_card.get_suit())
def turn_end(self, data):
"""
Event: turn end
"""
data_turn_player = data['turnPlayer']
data_turn_card = data['turnCard']
data_turn_card = Card(data_turn_card)
if data_turn_player != self.get_name():
self.htapi.dbg(data_turn_player + " shoot card: " +
format(data_turn_card))
local_player = self.players[data_turn_player]
local_player['shoot'].append(data_turn_card)
self.stat['roundCard'].append(data_turn_card)
self.stat['usedCard'].append(data_turn_card)
self._detect_card_shortage(local_player, data_turn_card)
def pick_history(self, data, is_timeout, pick_his):
"""
Event: turn end
"""
self.turn_end(data)
def _round_end_detect_shoot_moon_ability(self):
#
# Check if somebody get any score and give up shoot-moon mode.
#
if self.stat['shoot_moon_mode'] == False:
# Already disable shoot moon mode.
return
for key in self.players:
lp = self.players[key]
score = self.htapi.calc_score(
lp['pick'], is_expose_ah=self.stat['expose_ah_mode'])
if score != 0 and lp['playerName'] != self.get_name():
self.htapi.dbg("Give up shoot-moon mode...So sad.")
self.stat['shoot_moon_mode'] = False
def round_end(self, data):
"""
Event: round end
"""
round_player_name = data['roundPlayer']
round_cards = self.stat['roundCard']
local_player = self.players[round_player_name]
local_player['pick'] += round_cards
self.htapi.dbg(
"Player: " + round_player_name + " picked 4 cards: " +
format(round_cards),
" overall pick: " + format(local_player['pick']))
# Disable shoot moon mode if somebody rx a score.
self._round_end_detect_shoot_moon_ability()
#
# Reset round data
#
self.stat['roundCard'] = []
def deal_end(self, data):
"""
Event: deal end
"""
data_players = data['players']
self.htapi.dbg(format(data))
for player in data_players:
local_player = self.players[player['playerName']]
local_player['score_accl'] = player['gameScore']
if player['shootingTheMoon'] == True:
local_player['shoot_moon'] += 1
for key in self.players.keys():
p = self.players[key]
# Reset deal-specific data.
p['score'] = 0
p['shoot'] = []
p['expose'] = False
p['pick'] = []
self.stat['usedCard'] = []
self.stat['shoot_moon_mode'] = True
self.stat['expose_ah_mode'] = False
def game_over(self, data):
"""
Event: game end
"""
self.htapi.dbg(format(data))
class HacBotV(PokerBot, Htapi):
"""
Anti-Score Mode (AS)
Shoot-Moon Mode (SM)
...Player: HacBotV, score: 0, score_accl: 24516, score_negative_accl:-56136, shoot_moon_accl: 273, winner: [2234, 1790, 1280, 1096]
"""
SM_THOLD_PASS3 = 8.0
SM_THOLD_PICK = 9.5
AS_THOLD_PASS3 = 7.0
def __init__(self, name, is_debug=False):
super(HacBotV, self).__init__(name)
self.htapi = Htapi(is_debug=is_debug)
self.players = {}
self.stat = {}
self.stat['roundCard'] = []
self.stat['usedCard'] = []
self.stat['nextPlayers'] = []
self.stat['roundPlayers'] = []
self.big_rank_cards = [
Card('AS'),
Card('KS'),
Card('JS'),
Card('TS'),
Card('AH'),
Card('KH'),
Card('QH'),
Card('JH'),
Card('TH'),
Card('AD'),
Card('KD'),
Card('QD'),
Card('JD'),
Card('TD'),
Card('AC'),
Card('KC'),
Card('QC'),
Card('JC'),
Card('TC'),
]
self.score_cards = [
Card("QS"),
Card("TC"),
Card("2H"),
Card("3H"),
Card("4H"),
Card("5H"),
Card("6H"),
Card("7H"),
Card("8H"),
Card("9H"),
Card("TH"),
Card("JH"),
Card("QH"),
Card("KH"),
Card("AH")
]
self.plz_rebuild_players = True
self.stat['expose_ah_mode'] = False
self.stat['sm_mode'] = True
self.stat['sm_mode_started'] = False
self.stat['pass3_as_ability'] = 0
self.stat['pass3_as_ability_history'] = []
self.stat['pass3_sm_ability'] = 0
self.stat['pass3_sm_ability_history'] = []
def _rebuild_players(self, data):
"""
Configure player table by input data.
"""
self.htapi.dbg("Rebuild players: ", format(data))
data_players = data['players']
if len(data_players) != 4:
self.errmsg("Invalid player number " + str(len(data_players)) +
" in server msg")
self.players = {}
for dp in data_players:
self.players[dp['playerName']] = {
'playerName': dp['playerName'],
'score_accl': 0,
'score': 0,
'sm': 0,
'expose': False,
'shoot': [],
'suit_leak': [],
'pick': [],
}
def new_game(self, data):
"""
Event: The start of a new game.
"""
if self.plz_rebuild_players == True:
self._rebuild_players(data)
self.plz_rebuild_players = False
def receive_cards(self, data):
"""
Event: Receive my 13 cards.
"""
self.stat['hand'] = self.get_cards(data)
self.stat['hand'] = self.htapi.clone_cards(self.stat['hand'])
def _calc_hand_cards_num(self, my_hand_cards, reverse=False):
card_num_stat = {'S': 0, 'H': 0, 'D': 0, 'C': 0}
for c in my_hand_cards:
card_num_stat[c.get_suit()] += 1
card_num_stat_sorted = sorted(card_num_stat.iteritems(),
key=lambda (k, v): (v, k),
reverse=reverse)
return card_num_stat_sorted
def _select_card2pass_sm_mode(self, my_hand_cards):
"""
Pop out 1 card to pass
Reserve cards in long suit.
Remove small cards in shortage suit.
"""
card_num_stat_sorted = self._calc_hand_cards_num(my_hand_cards)
# Remove small cards in shortage suit.
for di in card_num_stat_sorted:
suit, num = di
if num == 0:
continue
if suit == 'H':
continue
same_suit_cards = self.htapi.get_cards_by_suit(my_hand_cards, suit)
smaller_card = self.htapi.pick_smaller_card(same_suit_cards,
[Card('9S')],
auto_choose_big=False)
if smaller_card != None:
return self.htapi.remove_card(my_hand_cards, smaller_card)
# Remove suit
for di in card_num_stat_sorted:
suit, num = di
if num == 0:
continue
if suit == 'H':
continue
same_suit_cards = self.htapi.get_cards_by_suit(my_hand_cards, suit)
small_card = self.htapi.pick_small_card(same_suit_cards)
return self.htapi.remove_card(my_hand_cards, small_card)
# Remove suit, in case I have 13 hearts... A.A
for di in card_num_stat_sorted:
suit, num = di
if num == 0:
continue
same_suit_cards = self.htapi.get_cards_by_suit(my_hand_cards, suit)
small_card = self.htapi.pick_small_card(same_suit_cards)
return self.htapi.remove_card(my_hand_cards, small_card)
self.htapi.errmsg("BUG")
def _select_card2pass(self, my_hand_cards):
"""
Pop out 1 card to pass. Anti-score mode.
- Remove big rank card from shortage suit
- Remove big rank card in long suit
"""
card_num_stat_sorted = self._calc_hand_cards_num(my_hand_cards)
# card_num_stat_sorted_dict = dict(card_num_stat_sorted)
if self.htapi.find_card(my_hand_cards, Card('QS')) == None:
card = self.htapi.remove_card(my_hand_cards, Card('AS'))
if card != None:
return card
# Spade in shortage. KS and AS will be dangerous.
card = self.htapi.remove_card(my_hand_cards, Card('KS'))
if card != None:
return card
# Remove big rank card from shortage suit
for di in card_num_stat_sorted:
suit, num = di
if num == 0:
continue
same_suit_cards = self.htapi.get_cards_by_suit(my_hand_cards, suit)
big_card = self.htapi.pick_big_card(same_suit_cards)
if big_card.get_rank_num() > Card('8S').get_rank_num():
return self.htapi.remove_card(my_hand_cards, big_card)
# Remove suit
for di in card_num_stat_sorted:
suit, num = di
if num == 0:
continue
same_suit_cards = self.htapi.get_cards_by_suit(my_hand_cards, suit)
big_card = self.htapi.pick_big_card(same_suit_cards)
return self.htapi.remove_card(my_hand_cards, big_card)
self.htapi.errmsg("BUG")
def _get_used_cards(self):
return self.stat['usedCard']
def _get_used_cards_by_suits(self, suits=[]):
used_cards = self.stat['usedCard']
output = self.htapi.get_cards_by_suits(used_cards, suits)
return output
def _get_unused_cards_by_suits(self, my_hand_cards, suits=[]):
"""
Get unused cards of other opponents so that I can know how many cards are left..
"""
all52cards = self.htapi.get52cards()
output = all52cards
self.htapi.remove_cards(output, self.stat['usedCard'])
self.htapi.remove_cards(output, my_hand_cards)
if len(suits) == 0:
# Do not need to pick a specific suit
pass
else:
# Pick unused cards of target suits
output = self.htapi.get_cards_by_suits(output, suits)
return output
def _get_unused_cards(self, my_hand_cards):
return self._get_unused_cards_by_suits(my_hand_cards, suits=[])
def _get_round_cards(self):
round_cards = self.stat['roundCard']
return round_cards
def _get_avail_cards(self):
my_avail_cards = self.stat['avail']
return my_avail_cards
def _get_hand_cards(self):
my_hand_cards = self.stat['hand']
return my_hand_cards
def _calc_as_point(self, card, oppo_unused_cards):
"""
Calculate the power of this card to avoid taking grades.
Output: 1.0 - This card won't take the trick. (Won't be next leader)
Output: 0.0 - This card will take the trick. Possibly a big card.
"""
oppo_unused_same_suit_cards = self.htapi.get_cards_by_suit(
oppo_unused_cards, card.get_suit())
if len(oppo_unused_same_suit_cards) == 0:
# Opponent does not have same suit...
# oppo_no_score_cards = self.htapi.find_no_score_cards(oppo_unused_cards)
# return len(oppo_no_score_cards) / float(len(oppo_unused_cards))
return 0.0
lose_cnt = 0
for c in oppo_unused_same_suit_cards:
if card.get_rank_num() < c.get_rank_num():
lose_cnt += 1
point = lose_cnt / float(len(oppo_unused_same_suit_cards))
return point
def _calc_sm_point(self, card, oppo_unused_cards):
"""
Calculate the power of this card to win others.
"""
oppo_unused_same_suit_cards = self.htapi.get_cards_by_suit(
oppo_unused_cards, card.get_suit())
if len(oppo_unused_same_suit_cards) == 0:
# No same suit at opponent.
return 1.0
win_cnt = 0
for c in oppo_unused_same_suit_cards:
if card.get_rank_num() > c.get_rank_num():
win_cnt += 1
return win_cnt / float(len(oppo_unused_same_suit_cards))
def _pass_cards_calc_sm_ability(self, data=None):
"""
Detect if I can shoot the moon.
"""
my_hand_cards = self._get_hand_cards()
oppo_unused_cards = self._get_unused_cards(my_hand_cards)
if len(oppo_unused_cards) != (52 - 13) or len(my_hand_cards) != 13:
self.htapi.errmsg("BUG")
return self._pick_card_calc_sm_ability(data=None)
def _pass_cards_calc_as_ability(self, data=None):
"""
Detect if I can anti-score
"""
return self._pick_card_calc_as_ability(data=None)
def _pass_cards_sm_mode(self, data):
"""
Pick 3 cards which can help myself to shoot moon.
"""
my_hand_cards = self._get_hand_cards()
output = []
for i in range(3):
card = self._select_card2pass_sm_mode(my_hand_cards)
if card == None:
self.htapi.errmsg("Cannot pick a card to pass")
output.append(card.toString())
self.htapi.dbg(self.get_name() + " pass 3 cards: " + format(output))
return output
def _pass_cards_as_mode(self, data):
"""
Pick 3 cards which can avoid taking score
"""
my_hand_cards = self._get_hand_cards()
output = []
for i in range(3):
card = self._select_card2pass(my_hand_cards)
if card == None:
self.htapi.errmsg("Cannot pick a card to pass")
output.append(card.toString())
self.htapi.dbg(self.get_name() + " pass 3 cards: " + format(output))
return output
def pass_cards(self, data):
"""
Event: Pick 3 cards to pass to others
"""
self.stat['hand'] = self.htapi.clone_cards(
[Card(x) for x in data['self']['cards']])
self.stat['hand'] = self.htapi.arrange_cards(self.stat['hand'])
self.htapi.dbg("Select 3 cards from: " + format(self.stat['hand']))
sm_ability = self._pass_cards_calc_sm_ability()
as_ability = self._pass_cards_calc_as_ability()
self.stat['pass3_sm_ability'] = sm_ability
self.stat['pass3_as_ability'] = as_ability
self.htapi.dbg("Ability stat. sm: ", format(sm_ability), ", as: ",
format(self.stat['pass3_as_ability']))
if sm_ability > self.SM_THOLD_PASS3:
self.htapi.dbg("shoot moon mode pass3: " + str(sm_ability))
return self._pass_cards_sm_mode(data)
else:
self.htapi.dbg("anti score mode pass3")
return self._pass_cards_as_mode(data)
def receive_opponent_cards(self, data):
"""
Event: Recv 3 cards from one opponent
"""
pass
def ______pass3(self):
pass
def expose_my_cards(self, data):
"""
Event: The server asks me to expose AH...
"""
output = []
candidates = data['cards']
if candidates == None or len(candidates) == 0:
return output
candidates = [Card(x) for x in candidates]
if self.htapi.find_card(candidates, Card('AH')) == None:
return output
if self.stat['pass3_sm_ability'] > self.SM_THOLD_PASS3:
output = ['AH']
return output
if self.stat['pass3_as_ability'] > self.AS_THOLD_PASS3:
# I have confidence not taking score...
output = ['AH']
return output
return output
def expose_cards_end(self, data):
"""
Event: Check if somebody expose AH. Damn.
"""
data_players = data['players']
for dp in data_players:
local_player = self.players[dp['playerName']]
if len(dp['exposedCards']) > 0:
# This guy exposed AH.
local_player['expose'] = True
self.stat['expose_ah_mode'] = True
else:
local_player['expose'] = False
def ______expose_card(self):
pass
def _do_i_have_lead_suit(self):
"""
Cannot be called at lead play!!
"""
round_cards = self._get_round_cards()
if len(round_cards) == 0:
self.errmsg("BUG")
lead_card = round_cards[0]
my_avail_cards = self._get_avail_cards()
if len(
self.htapi.get_cards_by_suit(my_avail_cards,
lead_card.get_suit())) == 0:
return False
return True
def _pick_card_calc_sm_ability(self, data=None):
"""
Detect if I can shoot the moon.
"""
my_hand_cards = self._get_hand_cards()
oppo_unused_cards = self._get_unused_cards(my_hand_cards)
my_sm_point = 0.0
for c in my_hand_cards:
this_sm_point = self._calc_sm_point(c, oppo_unused_cards)
if this_sm_point >= 1.0:
# Power card
my_sm_point += this_sm_point
else:
my_sm_point += this_sm_point
my_sm_point *= (13 / float(len(my_hand_cards)))
my_sm_point = round(my_sm_point, 3)
print("my_hand_cards: ", my_hand_cards,
", sm ability -> " + format(my_sm_point))
return my_sm_point
def _pick_card_calc_as_ability(self, data=None):
"""
Detect if I can anti-score
"""
my_hand_cards = self._get_hand_cards()
oppo_unused_cards = self._get_unused_cards(my_hand_cards)
my_as_point = 0.0
for c in my_hand_cards:
this_as_point = self._calc_as_point(c, oppo_unused_cards)
if this_as_point >= 1.0:
my_as_point += this_as_point
else:
my_as_point += this_as_point
my_as_point *= (13 / float(len(my_hand_cards)))
my_as_point = round(my_as_point, 3)
print("my_hand_cards: ", my_hand_cards,
", as ability -> " + format(my_as_point))
return my_as_point
def _pick_card_should_i_sm(self, data):
"""
Predict my ability to shoot moon.
"""
sm_ability = self._pick_card_calc_sm_ability(data=None)
as_ability = self._pick_card_calc_as_ability(data=None)
if self.stat['sm_mode'] == False:
return False
if self.stat['sm_mode_started'] == True:
# Already starting to sm. Don't stop.
return True
my_hand_cards = self._get_hand_cards()
oppo_unused_cards = self._get_unused_cards(my_hand_cards)
my_heart_cards = self.htapi.get_cards_by_suit(my_hand_cards, 'H')
power_heart_num = 0
for c in my_hand_cards:
this_sm_point = self._calc_sm_point(c, oppo_unused_cards)
if this_sm_point >= 1.0 and c.get_suit() == 'H':
power_heart_num += 1
if power_heart_num == 0 and len(my_heart_cards) > 0:
sm_ability -= 1.0
if sm_ability > self.SM_THOLD_PICK:
self.stat['sm_mode_started'] = True
return True
return False
def _pick_card_sm_mode_leadplay(self, data):
"""
I am the leader. If I have sm ability. Eat! Eat! Eat you to the hole!
"""
my_hand_cards = self._get_hand_cards()
my_avail_cards = self._get_avail_cards()
oppo_unused_cards = self._get_unused_cards(my_hand_cards)
for c in my_avail_cards:
this_sm_point = self._calc_sm_point(c, oppo_unused_cards)
if this_sm_point >= 1.0 and c.get_suit() == 'S':
self.htapi.dbg("This card will win..." + format(c))
return c
for c in my_avail_cards:
this_sm_point = self._calc_sm_point(c, oppo_unused_cards)
if this_sm_point >= 1.0:
self.htapi.dbg("This card will win..." + format(c))
return c
for c in my_avail_cards:
this_sm_point = self._calc_sm_point(c, oppo_unused_cards)
if this_sm_point >= 1.0 and c.get_suit() == 'H':
self.htapi.dbg("This card will win..." + format(c))
return c
# Pick short suit candidates
self.htapi.dbg("Low strength stage... I am weak :-(.")
max_sm_point = 0
candidates = []
for c in my_avail_cards:
this_sm_point = self._calc_sm_point(c, oppo_unused_cards)
if len(candidates) == 0:
candidates = [c]
max_sm_point = this_sm_point
elif this_sm_point > max_sm_point:
candidates = [c]
max_sm_point = this_sm_point
elif this_sm_point < max_sm_point:
pass
else:
candidates.append(c)
card_num_stat_sorted = self._calc_hand_cards_num(my_hand_cards)
for di in card_num_stat_sorted:
suit, num = di
if num == 0:
continue
prefer_candidates = self.htapi.get_cards_by_suit(
my_avail_cards, suit)
if len(prefer_candidates) > 0:
prefer_candidates = self.htapi.arrange_cards(prefer_candidates)
return prefer_candidates[-1]
self.htapi.errmsg("BUG")
def _pick_card_sm_mode_freeplay(self, data):
"""
Do not have the same suit to follow. So I am free to shoot.
"""
my_hand_cards = self._get_hand_cards()
my_avail_cards = self._get_avail_cards()
oppo_unused_cards = self._get_unused_cards(my_hand_cards)
my_noscore_cards = self.htapi.find_no_score_cards(my_avail_cards)
if len(my_noscore_cards) == 0:
# Too bad... I have to give up sm. Turn to as mode.
return self._pick_card_as_mode(data)
min_sm_cards = []
min_sm_point = None
for c in my_noscore_cards:
this_sm_point = self._calc_sm_point(c, oppo_unused_cards)
if min_sm_point == None:
min_sm_point = this_sm_point
min_sm_cards = [c]
elif this_sm_point < min_sm_point:
min_sm_point = this_sm_point
min_sm_cards = [c]
elif this_sm_point > min_sm_point:
pass
else:
min_sm_cards.append(c)
card_num_stat_sorted = self._calc_hand_cards_num(my_hand_cards)
for di in card_num_stat_sorted:
suit, num = di
if num == 0:
continue
prefer_candidates = self.htapi.get_cards_by_suit(
min_sm_cards, suit)
if len(prefer_candidates) > 0:
prefer_candidates = self.htapi.arrange_cards(prefer_candidates)
return prefer_candidates[-1] # TBD: Use the big rank one!?
self.htapi.errmsg("BUG")
def _pick_card_sm_mode_midplay(self, data):
"""
I am the mid player.
Try to win...
"""
round_cards = self._get_round_cards()
lead_card = round_cards[0]
filtered_round_cards = self.htapi.get_cards_by_suit(
round_cards, lead_card.get_suit())
round_card_score = self.htapi.calc_score(round_cards)
my_hand_cards = self._get_hand_cards()
my_avail_cards = self._get_avail_cards()
oppo_unused_cards = self._get_unused_cards(my_hand_cards)
if self._do_i_have_lead_suit() == True:
# Try to win this round.
if self.htapi.pick_bigger_card(my_avail_cards,
filtered_round_cards) == None:
# ...I don't have bigger card to win.
self.htapi.dbg("Give up sm... Oops")
return self._pick_card_as_mode(data)
else:
# I have chance to win.
king_cards = []
max_sm_point = 0
my_avail_cards = self.htapi.arrange_cards(my_avail_cards)
for c in my_avail_cards:
this_sm_point = self._calc_sm_point(c, oppo_unused_cards)
if this_sm_point > max_sm_point:
max_sm_point = this_sm_point
king_cards = [c]
elif this_sm_point < max_sm_point:
pass
else:
king_cards.append(c)
print(" + king cards to win next players are: ",
format(king_cards))
if len(king_cards) == 0:
self.htapi.errmsg("BUG")
return king_cards[0]
else:
if round_card_score > 0:
# Cannot win this round. Have to give up shoot moon.
self.htapi.dbg("Give up sm... Oops")
return self._pick_card_as_mode(data)
else:
return self._pick_card_sm_mode_freeplay(data)
def _pick_card_sm_mode_lastplay(self, data):
"""
I am the last player. I can decide to win or not to win.
Try to win...maybe.
"""
round_cards = self._get_round_cards()
lead_card = round_cards[0]
filtered_round_cards = self.htapi.get_cards_by_suit(
round_cards, lead_card.get_suit())
round_card_score = self.htapi.calc_score(round_cards)
my_hand_cards = self._get_hand_cards()
my_avail_cards = self._get_avail_cards()
oppo_unused_cards = self._get_unused_cards(my_hand_cards)
if self._do_i_have_lead_suit() == True:
if round_card_score > 0:
#
# Have score on this round. I have to decide to take or not to take...
#
# Try to win...
bigger_card = self.htapi.pick_bigger_card(
my_avail_cards, filtered_round_cards)
if bigger_card != None:
return bigger_card
else:
# Cannot win this round. Have to give up shoot moon.
self.htapi.dbg("Give up sm... Oops")
return self._pick_card_as_mode(data)
else:
#
# Have no score, I can give up to win if I have a too weak point.
#
same_suit_card_num = self.htapi.calc_card_num_by_suit(
my_hand_cards + oppo_unused_cards, lead_card.get_suit())
min_sm_card = None
min_sm_point = None
for c in my_avail_cards:
this_sm_point = self._calc_sm_point(c, oppo_unused_cards)
if min_sm_point == None:
min_sm_point = this_sm_point
min_sm_card = c
elif this_sm_point < min_sm_point:
min_sm_point = this_sm_point
min_sm_card = c
if same_suit_card_num > 9:
return min_sm_card
else:
# Try to win this round...
bigger_card = self.htapi.pick_bigger_card(
my_avail_cards, filtered_round_cards)
if bigger_card != None:
return bigger_card
else:
# Cannot win this round.
return min_sm_card
else:
if round_card_score > 0:
# Cannot win this round. Have to give up shoot moon.
self.htapi.dbg("Give up sm... Oops")
return self._pick_card_as_mode(data)
else:
return self._pick_card_sm_mode_freeplay(data)
def _pick_card_sm_mode(self, data):
"""
Pick a card!
"""
# roundPlayers is in the correct order of shooting cards.
round_players = self.stat['roundPlayers']
# Identify my position in this round
my_pos = round_players.index(self.get_name())
if my_pos == 0:
card = self._pick_card_sm_mode_leadplay(data)
elif my_pos == 3:
card = self._pick_card_sm_mode_midplay(data)
else:
card = self._pick_card_sm_mode_lastplay(data)
return card
def _pick_card_as_mode_leadplay(self, data):
"""
I don't want to take a trick...
"""
my_avail_cards = self._get_avail_cards()
my_hand_cards = self._get_hand_cards()
oppo_unused_cards = self._get_unused_cards(my_hand_cards)
candidates = []
current_max_point = None
for c in my_avail_cards:
as_point = self._calc_as_point(c, oppo_unused_cards)
if current_max_point == None:
current_max_point = as_point
candidates = [c]
elif as_point > current_max_point:
candidates = [c]
current_max_point = as_point
elif as_point < current_max_point:
pass
else:
candidates.append(c)
if len(candidates) == 0:
self.htapi.errmsg("BUG")
#
# Try to make others eat 'QS'
#
spade_candidates = self.htapi.get_cards_by_suit(candidates, 'S')
if len(spade_candidates) > 0:
used_spade_cards = self._get_used_cards_by_suits(['S'])
if self.htapi.find_card(used_spade_cards, Card('QS')) != None:
for c in spade_candidates:
if self._calc_as_point(c, oppo_unused_cards) == 1.0:
return c
#
# Pick strong small heart cards first, so that I won't eat more later.
#
heart_candidates = self.htapi.get_cards_by_suit(candidates, 'H')
if len(heart_candidates) > 0:
for c in heart_candidates:
if self._calc_as_point(c, oppo_unused_cards) == 1.0:
return c
#
# Pick small rank cards from shortage suit first
#
card_num_stat_sorted = self._calc_hand_cards_num(my_hand_cards)
for di in card_num_stat_sorted:
suit, num = di
if num == 0:
continue
prefer_candidates = self.htapi.get_cards_by_suit(candidates, suit)
if len(prefer_candidates) > 0:
return prefer_candidates[0]
self.htapi.errmsg("BUG")
def _get_current_winner(self):
round_players = self.stat['roundPlayers']
round_cards = self._get_round_cards()
if len(round_cards) == 0:
self.errmsg("BUG")
lead_card = round_cards[0]
idx = 0
winner_idx = 0
for c in round_cards:
if c.get_suit() == lead_card.get_suit(
) and c.get_rank_num() > lead_card.get_rank_num():
winner_idx = idx
idx += 1
winner = round_players[winner_idx]
return winner
def _is_sm_possible(self):
#
# Check if there're 2 player having score. Then it's impossible to shoot moon.
#
have_score_player = 0
for key in self.players:
lp = self.players[key]
score = self.htapi.calc_score(
lp['pick'], is_expose_ah=self.stat['expose_ah_mode'])
if score != 0:
have_score_player += 1
if have_score_player >= 2:
return False # Not possible to shoot moon.
return True
def _check_current_winner_sm(self):
"""
Detect the round winner is now shooting moon.
"""
if self._is_sm_possible() == False:
return False
winner = self._get_current_winner()
pl = self.players[winner]
winner_picked_cards = pl['pick']
winner_picked_score_cards = self.htapi.find_score_cards(
winner_picked_cards)
if len(winner_picked_score_cards) > 7:
return True # Possibly want to shoot moon.
return False
def _pick_card_as_mode_freeplay(self, data):
"""
I am in midplay or last play. I don't have lead suit, so I can shoot anything.
"""
my_avail_cards = self.htapi.arrange_cards(self._get_avail_cards())
my_hand_cards = self._get_hand_cards()
round_cards = self._get_round_cards()
if len(round_cards) == 0:
self.errmsg("BUG")
oppo_unused_cards = self._get_unused_cards(my_hand_cards)
if self._check_current_winner_sm() == True:
# Current winner is a sm player. Prefer no score card here.
self.htapi.dbg("Current winner player is a suspicous pig.")
if self.htapi.find_card(oppo_unused_cards, Card('QS')) != None:
# I don't want to eat QS
card = self.htapi.find_card(my_avail_cards, Card('AS'))
if card != None:
return card
card = self.htapi.find_card(my_avail_cards, Card('KS'))
if card != None:
return card
# # Shoot heart card but reserve the biggest rank
# my_heart_cards = self.htapi.get_cards_by_suit(my_avail_cards, 'H')
# if len(my_heart_cards) >= 2:
# my_heart_cards = self.htapi.arrange_cards(my_heart_cards)
#
# my_2nd_heart_card = my_heart_cards[-2]
# score = self._calc_as_point(my_2nd_heart_card, oppo_unused_cards)
#
# if score < 0.5:
# # Remove the 2nd dangerous heart.
# return my_2nd_heart_card
# Shoot no-score cards
candidates = []
as_point_min = None
for c in my_avail_cards:
this_as_point = self._calc_as_point(c, oppo_unused_cards)
if self.htapi.calc_card_num_by_suit(oppo_unused_cards,
c.get_suit()) == 0:
# Don't worry. The oppo won't send the suit again... so I won't eat the trick.
# But if I am the lead... I will eat 100%.
continue
if len(self.htapi.find_score_cards([c])) > 0:
# Avoid giving out score cards for sm player
continue
if as_point_min == None:
as_point_min = this_as_point
candidates = [c]
elif this_as_point < as_point_min:
as_point_min = this_as_point
candidates = [c]
elif this_as_point > as_point_min:
pass
else:
candidates.append(c)
if len(candidates) > 0:
card_num_stat_sorted = self._calc_hand_cards_num(my_hand_cards)
# Remove small cards in shortage suit.
for di in card_num_stat_sorted:
suit, num = di
if num == 0:
continue
prefer_candidates = self.htapi.get_cards_by_suit(
candidates, suit)
if len(prefer_candidates) > 0:
prefer_candidates = self.htapi.arrange_cards(
prefer_candidates)
return prefer_candidates[-1]
#
# Shoot QS out if I have chance.
#
card = self.htapi.find_card(my_avail_cards, Card('QS'))
if card != None:
return card
#
# Prefer AS, KS if the oppo has QS. Otherwise, KS, AS are not so dangerous...
#
if self.htapi.find_card(oppo_unused_cards, Card('QS')) != None:
card = self.htapi.find_card(my_avail_cards, Card('AS'))
if card != None:
return card
card = self.htapi.find_card(my_avail_cards, Card('KS'))
if card != None:
return card
#
# Shoot dangerous heart while opponents still have hearts.
#
oppo_heart_cards = self.htapi.get_cards_by_suit(oppo_unused_cards, 'H')
if len(oppo_heart_cards) > 0:
my_heart_cards = self.htapi.get_cards_by_suit(my_avail_cards, 'H')
my_heart_cards = self.htapi.arrange_cards(my_heart_cards)
for c in reversed(my_heart_cards):
point = self._calc_as_point(c, oppo_unused_cards)
if point < 0.5:
return c
#
# Shoot 'TC'
#
card = self.htapi.find_card(my_avail_cards, Card('TC'))
if card != None:
return card
#
# Avoid eating 'TC'
#
if self.htapi.find_card(oppo_unused_cards, Card('TC')) != None:
my_club_cards = self.htapi.get_cards_by_suit(my_hand_cards, 'C')
if len(my_club_cards) > 0:
my_small_club_card = self.htapi.pick_small_card(my_club_cards)
if my_small_club_card.get_rank_num() < Card('TC'):
# I still have small club. Won't eat 'TC'.
pass
else:
card = self.htapi.find_card(my_avail_cards, Card('AC'))
if card != None:
return card
card = self.htapi.find_card(my_avail_cards, Card('KC'))
if card != None:
return card
card = self.htapi.find_card(my_avail_cards, Card('QC'))
if card != None:
return card
card = self.htapi.find_card(my_avail_cards, Card('JC'))
if card != None:
return card
#
# Choose the most dangerous cards
#
candidates = []
as_point_min = None
for c in my_avail_cards:
this_as_point = self._calc_as_point(c, oppo_unused_cards)
if self.htapi.calc_card_num_by_suit(oppo_unused_cards,
c.get_suit()) == 0:
# Don't worry. The oppo won't send the suit again... so I won't eat the trick.
# But if I am the lead... I will eat 100%.
continue
if as_point_min == None:
as_point_min = this_as_point
candidates = [c]
elif this_as_point < as_point_min:
as_point_min = this_as_point
candidates = [c]
elif this_as_point > as_point_min:
pass
else:
candidates.append(c)
if len(candidates) > 0:
card_num_stat_sorted = self._calc_hand_cards_num(my_hand_cards)
# Remove small cards in shortage suit.
for di in card_num_stat_sorted:
suit, num = di
if num == 0:
continue
prefer_candidates = self.htapi.get_cards_by_suit(
candidates, suit)
if len(prefer_candidates) > 0:
prefer_candidates = self.htapi.arrange_cards(
prefer_candidates)
return prefer_candidates[-1]
#
# Shoot the min as point card
#
candidates = []
as_point_min = None
for c in my_avail_cards:
this_as_point = self._calc_as_point(c, oppo_unused_cards)
if as_point_min == None:
as_point_min = this_as_point
candidates = [c]
elif this_as_point < as_point_min:
as_point_min = this_as_point
candidates = [c]
elif this_as_point > as_point_min:
pass
else:
candidates.append(c)
if len(candidates) == 0:
self.htapi.errmsg("BUG")
candidates = self.htapi.arrange_cards(candidates)
return candidates[-1]
def _pick_card_as_mode_midplay(self, data):
my_hand_cards = self._get_hand_cards()
my_avail_cards = self._get_avail_cards()
round_cards = self._get_round_cards()
lead_card = round_cards[0]
filtered_round_cards = self.htapi.get_cards_by_suit(
round_cards, lead_card.get_suit())
filtered_round_cards_sorted = self.htapi.arrange_cards(
filtered_round_cards)
round_score_cards = self.htapi.find_score_cards(round_cards)
oppo_same_suit_cards = self._get_unused_cards_by_suits(
my_hand_cards, [lead_card.get_suit()])
if self._do_i_have_lead_suit() == True:
#
# I have the lead suit... Don't have many choice...
#
filtered_round_cards = self.htapi.get_cards_by_suit(
round_cards, lead_card.get_suit())
card2shoot = self.htapi.pick_smaller_card(my_avail_cards,
filtered_round_cards,
auto_choose_big=False)
if card2shoot != None:
return card2shoot
else:
# I don't have smaller cards, so I am possible to eat the trick. Avoid picking QS here, or I will eat it myself.
my_no_score_cards = self.htapi.find_no_score_cards(
my_avail_cards)
if len(my_no_score_cards) > 0:
# Prefer no score card to decrease the score I will eat.
if len(oppo_same_suit_cards) > 9 and len(
round_score_cards) == 0:
# Gamble! Pick a big card and guess the next player have the same suit.
# TODO: Improve...
return self.htapi.pick_big_card(my_no_score_cards)
else:
return self.htapi.pick_small_card(my_no_score_cards)
else:
# All cards left are score cards... what can I do...
return self.htapi.pick_small_card(my_avail_cards)
else:
#
# I don't have the same suit. Can do anything.
#
return self._pick_card_as_mode_freeplay(data)
self.htapi.errmsg("BUG")
def _pick_card_as_mode_lastplay(self, data):
my_hand_cards = self._get_hand_cards()
my_avail_cards = self._get_avail_cards()
round_cards = self._get_round_cards()
lead_card = round_cards[0]
filtered_round_cards = self.htapi.get_cards_by_suit(
round_cards, lead_card.get_suit())
filtered_round_cards_sorted = self.htapi.arrange_cards(
filtered_round_cards)
oppo_same_suit_cards = self._get_unused_cards_by_suits(
my_hand_cards, [lead_card.get_suit()])
round_score_cards = self.htapi.find_score_cards(round_cards)
if self._do_i_have_lead_suit() == True:
#
# I have the lead suit... Don't have many choice...
#
if len(round_score_cards) > 0:
card2shoot = self.htapi.pick_smaller_card(
my_avail_cards,
filtered_round_cards,
auto_choose_big=False)
if card2shoot != None:
return card2shoot
else:
if self.stat['sm_mode'] == False:
return self.htapi.pick_big_card(my_avail_cards)
else:
return self.htapi.pick_bigger_card(
my_avail_cards, filtered_round_cards)
else:
# No score card on table. Choose any no score card first.
# Pick the score card if I have no choice.
current_winner_card = filtered_round_cards_sorted[-1]
candidates = []
for c in my_avail_cards:
if len(self.htapi.find_score_cards(
[c])) > 0 and c.get_rank_num(
) > current_winner_card.get_rank_num():
# This is a score card and will eat the trick. Don't shoot it.
pass
else:
candidates.append(c)
if len(candidates) > 0:
candidates = self.htapi.arrange_cards(candidates)
score_candidates = self.htapi.find_score_cards(candidates)
if len(score_candidates) > 0:
return self.htapi.pick_big_card(score_candidates)
else:
return self.htapi.pick_big_card(candidates)
else:
# I don't have good candidates. Usually means I will eat the trick myself.
return self.htapi.pick_big_card(my_avail_cards)
else:
return self._pick_card_as_mode_freeplay(data)
def _pick_card_as_mode(self, data):
"""
Pick a card!
"""
# roundPlayers is in the correct order of shooting cards.
round_players = self.stat['roundPlayers']
# Identify my position in this round
my_pos = round_players.index(self.get_name())
if my_pos == 0:
card = self._pick_card_as_mode_leadplay(data)
elif my_pos == 3:
card = self._pick_card_as_mode_lastplay(data)
else:
card = self._pick_card_as_mode_midplay(data)
return card
def pick_card(self, data):
"""
Event: My turn to shoot a card.
"""
self.stat['hand'] = [Card(x) for x in data['self']['cards']]
self.stat['hand'] = self.htapi.arrange_cards(self.stat['hand'])
self.stat['avail'] = self.htapi.arrange_cards(
[Card(x) for x in data['self']['candidateCards']])
self.stat['roundPlayers'] = data['roundPlayers'][:]
# Get players in next turn.
round_players = self.stat['roundPlayers']
my_pos = round_players.index(self.get_name())
self.stat['nextPlayers'] = data['roundPlayers'][(my_pos + 1):]
if self._pick_card_should_i_sm(data) == True:
self.htapi.dbg("sm mode")
card2shoot = self._pick_card_sm_mode(data)
else:
self.htapi.dbg("as mode")
card2shoot = self._pick_card_as_mode(data)
self.htapi.dbg(self.get_name() + " shoot card: " + format(card2shoot) +
", from: " + format(data['self']['cards']) +
", next players: " + format(self.stat['nextPlayers']))
return card2shoot.toString()
def _detect_card_shortage(self, local_player, turn_card):
#
# If the player is not lead player and he shoots a card not the same as lead card,
# mark the player as card shortage.
#
# If I know the shortage status of players, can have advantage to predict.
#
round_cards = self._get_round_cards()
if len(round_cards) <= 1:
# The turn player is leader. Ignore.
return False
lead_card = round_cards[0]
if lead_card.get_suit_num() != turn_card.get_suit_num():
if local_player['suit_leak'].count(lead_card.get_suit()):
local_player['suit_leak'].append(lead_card.get_suit())
self.htapi.dbg("Player: " + local_player['playerName'] +
" card leakage: " + lead_card.get_suit())
def turn_end(self, data):
"""
Event: turn end
"""
data_turn_player = data['turnPlayer']
data_turn_card = data['turnCard']
data_turn_card = Card(data_turn_card)
if data_turn_player != self.get_name():
self.htapi.dbg(data_turn_player + " shoot card: " +
format(data_turn_card))
local_player = self.players[data_turn_player]
local_player['shoot'].append(data_turn_card)
self.stat['roundCard'].append(data_turn_card)
self.stat['usedCard'].append(data_turn_card)
self._detect_card_shortage(local_player, data_turn_card)
def pick_history(self, data, is_timeout, pick_his):
"""
Event: turn end
"""
self.turn_end(data)
def ______pick(self):
pass
def _round_end_detect_sm_ability(self):
#
# Check if somebody get any score and give up shoot-moon mode.
#
if self.stat['sm_mode'] == False:
# Already disable shoot moon mode.
return
for key in self.players:
lp = self.players[key]
score = self.htapi.calc_score(
lp['pick'], is_expose_ah=self.stat['expose_ah_mode'])
if score != 0 and lp['playerName'] != self.get_name():
self.htapi.dbg("Give up shoot-moon mode...So sad.")
self.stat['sm_mode'] = False
def round_end(self, data):
"""
Event: round end
"""
round_player_name = data['roundPlayer']
round_cards = self._get_round_cards()
local_player = self.players[round_player_name]
local_player['pick'] += round_cards
self.htapi.dbg(
"Player: " + round_player_name + " picked 4 cards: " +
format(round_cards),
" overall pick: " + format(local_player['pick']))
# Disable shoot moon mode if somebody rx a score.
self._round_end_detect_sm_ability()
#
# Reset round data
#
self.stat['roundCard'] = []
self.stat['nextPlayers'] = []
self.stat['roundPlayers'] = []
def deal_end(self, data):
"""
Event: deal end
"""
data_players = data['players']
self.htapi.dbg(format(data))
for player in data_players:
local_player = self.players[player['playerName']]
local_player['score_accl'] = player['gameScore']
if player['shootingTheMoon'] == True:
local_player['sm'] += 1
if player['playerName'] == self.get_name():
self.stat['pass3_sm_ability_history'].append(
self.stat['pass3_sm_ability'])
if player['gameScore'] == 0 and self.get_name(
) == player['playerName']:
self.stat['pass3_as_ability_history'].append(
self.stat['pass3_as_ability'])
for key in self.players.keys():
p = self.players[key]
# Reset deal-specific data.
p['score'] = 0
p['shoot'] = []
p['expose'] = False
p['pick'] = []
self.stat['usedCard'] = []
self.stat['sm_mode'] = True
self.stat['sm_mode_started'] = False
self.stat['expose_ah_mode'] = False
self.stat['pass3_sm_ability'] = 0.0
self.stat['pass3_as_ability'] = 0.0
def game_over(self, data):
"""
Event: game end
"""
self.htapi.dbg(format(data))
print(self.stat)
print(self.players[self.get_name()])