change tetrislogic api
This commit is contained in:
parent
9a7aead918
commit
f025ad5fd8
@ -176,17 +176,6 @@ class MatrixSprites(MinoesSprites):
|
||||
|
||||
class TetrArcade(TetrisLogic, arcade.Window):
|
||||
|
||||
NB_LINES = NB_LINES
|
||||
NB_COLS = NB_COLS
|
||||
NB_NEXT = NB_NEXT
|
||||
LOCK_DELAY = LOCK_DELAY
|
||||
FALL_DELAY = FALL_DELAY
|
||||
AUTOREPEAT_DELAY = AUTOREPEAT_DELAY
|
||||
AUTOREPEAT_PERIOD = AUTOREPEAT_PERIOD
|
||||
MATRIX_PIECE_COORD = MATRIX_PIECE_COORD
|
||||
NEXT_PIECE_COORDS = NEXT_PIECE_COORDS
|
||||
HELD_PIECE_COORD = HELD_PIECE_COORD
|
||||
|
||||
def __init__(self):
|
||||
locale.setlocale(locale.LC_ALL, "")
|
||||
self.highlight_texts = []
|
||||
@ -203,7 +192,7 @@ class TetrArcade(TetrisLogic, arcade.Window):
|
||||
self.new_conf()
|
||||
self.load_conf()
|
||||
|
||||
super().__init__()
|
||||
super().__init__(NB_LINES, NB_COLS, NB_NEXT)
|
||||
arcade.Window.__init__(
|
||||
self,
|
||||
width=self.init_width,
|
||||
@ -234,8 +223,8 @@ class TetrArcade(TetrisLogic, arcade.Window):
|
||||
for path in MUSICS_PATHS
|
||||
)
|
||||
self.music.queue(playlist)
|
||||
except Exception as e:
|
||||
Warning("Can't play music :" + str(e))
|
||||
except:
|
||||
Warning("Can't play music.")
|
||||
self.play_music = False
|
||||
|
||||
def new_conf(self):
|
||||
@ -320,38 +309,37 @@ AGAIN""".format(
|
||||
|
||||
self.play_music = self.conf["MUSIC"].getboolean("play")
|
||||
|
||||
def new_game(self):
|
||||
def on_new_game(self):
|
||||
self.highlight_texts = []
|
||||
super().new_game()
|
||||
self.matrix.sprites = MatrixSprites(self.matrix)
|
||||
if self.play_music:
|
||||
self.music.seek(0)
|
||||
self.music.play()
|
||||
|
||||
def new_tetromino(self):
|
||||
tetromino = super().new_tetromino()
|
||||
tetromino.sprites = TetrominoSprites(tetromino, self)
|
||||
return tetromino
|
||||
def on_new_piece(self, piece):
|
||||
piece.sprites = TetrominoSprites(piece, self)
|
||||
|
||||
def new_matrix_piece(self):
|
||||
super().new_matrix_piece()
|
||||
def on_new_level(self, level):
|
||||
self.show_text("LEVEL\n{:n}".format(level))
|
||||
|
||||
def on_generation_phase(self, piece):
|
||||
self.matrix.ghost.sprites = TetrominoSprites(self.matrix.ghost, self, GHOST_ALPHA)
|
||||
for tetromino in [self.matrix.piece, self.matrix.ghost] + self.next.pieces:
|
||||
tetromino.sprites.refresh()
|
||||
|
||||
def move(self, movement, prelock=True):
|
||||
moved = super().move(movement, prelock)
|
||||
def on_falling_phase(self):
|
||||
self.matrix.piece.sprites.refresh()
|
||||
if moved:
|
||||
self.matrix.ghost.sprites.refresh()
|
||||
return moved
|
||||
|
||||
def rotate(self, rotation):
|
||||
rotated = super().rotate(rotation)
|
||||
if rotated:
|
||||
for tetromino in (self.matrix.piece, self.matrix.ghost):
|
||||
tetromino.sprites.refresh()
|
||||
return rotated
|
||||
def on_moved(self, moved):
|
||||
self.matrix.piece.sprites.refresh()
|
||||
self.matrix.ghost.sprites.refresh()
|
||||
|
||||
def on_rotated(self, direction):
|
||||
for tetromino in (self.matrix.piece, self.matrix.ghost):
|
||||
tetromino.sprites.refresh()
|
||||
|
||||
def on_lock_phase(self):
|
||||
self.matrix.piece.sprites.refresh()
|
||||
|
||||
def swap(self):
|
||||
super().swap()
|
||||
@ -360,20 +348,15 @@ AGAIN""".format(
|
||||
if tetromino:
|
||||
tetromino.sprites.refresh()
|
||||
|
||||
def lock(self):
|
||||
self.matrix.piece.prelocked = False
|
||||
self.matrix.piece.sprites.refresh()
|
||||
super().lock()
|
||||
self.matrix.sprites.refresh()
|
||||
|
||||
def enter_the_matrix(self):
|
||||
super().enter_the_matrix()
|
||||
for mino in self.matrix.piece:
|
||||
def on_locked(self, piece):
|
||||
piece.sprites.refresh()
|
||||
for mino in piece:
|
||||
self.matrix.sprites.append(mino.sprite)
|
||||
|
||||
def remove_line(self, y):
|
||||
self.matrix.sprites.remove_line(y)
|
||||
super().remove_line(y)
|
||||
self.matrix.sprites.refresh()
|
||||
|
||||
def pause(self):
|
||||
super().pause()
|
||||
@ -432,7 +415,7 @@ AGAIN""".format(
|
||||
if tetromino:
|
||||
tetromino.sprites.draw()
|
||||
|
||||
t = time.localtime(self.time)
|
||||
t = time.localtime(self.stats.time)
|
||||
font_size = STATS_TEXT_SIZE * self.scale
|
||||
for y, text in enumerate(("TIME", "LINES", "GOAL", "LEVEL", "HIGH SCORE", "SCORE")):
|
||||
arcade.draw_text(
|
||||
@ -448,11 +431,11 @@ AGAIN""".format(
|
||||
for y, text in enumerate(
|
||||
(
|
||||
"{:02d}:{:02d}:{:02d}".format(t.tm_hour - 1, t.tm_min, t.tm_sec),
|
||||
"{:n}".format(self.nb_lines_cleared),
|
||||
"{:n}".format(self.goal),
|
||||
"{:n}".format(self.level),
|
||||
"{:n}".format(self.high_score),
|
||||
"{:n}".format(self.score),
|
||||
"{:n}".format(self.stats.lines_cleared),
|
||||
"{:n}".format(self.stats.goal),
|
||||
"{:n}".format(self.stats.level),
|
||||
"{:n}".format(self.stats.high_score),
|
||||
"{:n}".format(self.stats.score),
|
||||
)
|
||||
):
|
||||
arcade.draw_text(
|
||||
|
10
test.py
10
test.py
@ -1,10 +1,18 @@
|
||||
# -*- coding: utf-8 -*-
|
||||
|
||||
from TetrArcade import TetrArcade, State
|
||||
from TetrArcade import TetrArcade, State, MinoSprite
|
||||
from tetrislogic import Mino, Color, Coord
|
||||
|
||||
game = TetrArcade()
|
||||
game.new_game()
|
||||
for x in range(game.matrix.collumns):
|
||||
mino = Mino(Color.ORANGE, Coord(x, 0))
|
||||
mino.sprite = MinoSprite(mino, game, 200)
|
||||
game.matrix[0][x] = mino
|
||||
game.matrix.sprites.append(mino.sprite)
|
||||
game.move_left()
|
||||
game.pause()
|
||||
game.resume()
|
||||
game.move_right()
|
||||
game.swap()
|
||||
game.rotate_clockwise()
|
||||
|
@ -1,5 +1,5 @@
|
||||
# -*- coding: utf-8 -*-
|
||||
from .consts import NB_LINES, NB_COLS, NB_NEXT
|
||||
from .consts import LINES, COLLUMNS, NEXT_PIECES
|
||||
from .utils import Movement, Rotation, Color, Coord
|
||||
from .tetromino import Mino, Tetromino
|
||||
from .tetrislogic import TetrisLogic, State, Matrix
|
||||
|
@ -3,9 +3,9 @@ from .utils import Coord
|
||||
|
||||
|
||||
# Matrix
|
||||
NB_LINES = 20
|
||||
NB_COLS = 10
|
||||
NB_NEXT = 5
|
||||
LINES = 20
|
||||
COLLUMNS = 10
|
||||
NEXT_PIECES = 5
|
||||
|
||||
# Delays (seconds)
|
||||
LOCK_DELAY = 0.5
|
||||
@ -14,6 +14,6 @@ AUTOREPEAT_DELAY = 0.300 # Official : 0.300 s
|
||||
AUTOREPEAT_PERIOD = 0.010 # Official : 0.010 s
|
||||
|
||||
# Piece init coord
|
||||
MATRIX_PIECE_COORD = Coord(4, NB_LINES)
|
||||
NEXT_PIECE_COORDS = [Coord(NB_COLS + 4, NB_LINES - 4 * n - 3) for n in range(NB_NEXT)]
|
||||
HELD_PIECE_COORD = Coord(-5, NB_LINES - 3)
|
||||
MATRIX_PIECE_COORD = Coord(4, LINES)
|
||||
NEXT_PIECE_COORDS = [Coord(COLLUMNS + 4, LINES - 4 * n - 3) for n in range(NEXT_PIECES)]
|
||||
HELD_PIECE_COORD = Coord(-5, LINES - 3)
|
||||
|
@ -5,9 +5,9 @@ import pickle
|
||||
from .utils import Coord, Movement, Rotation, T_Spin
|
||||
from .tetromino import Tetromino, T_Tetrimino, I_Tetrimino
|
||||
from .consts import (
|
||||
NB_LINES,
|
||||
NB_COLS,
|
||||
NB_NEXT,
|
||||
LINES,
|
||||
COLLUMNS,
|
||||
NEXT_PIECES,
|
||||
LOCK_DELAY,
|
||||
FALL_DELAY,
|
||||
AUTOREPEAT_DELAY,
|
||||
@ -42,88 +42,58 @@ class HoldQueue(PieceContainer):
|
||||
|
||||
class Matrix(list, PieceContainer):
|
||||
|
||||
def __init__(self, *args, **kargs):
|
||||
list.__init__(self, *args, **kargs)
|
||||
def __init__(self, lines, collumns):
|
||||
list.__init__(self)
|
||||
PieceContainer.__init__(self)
|
||||
self.lines = lines
|
||||
self.collumns = collumns
|
||||
|
||||
def reset(self):
|
||||
self.clear()
|
||||
for y in range(self.lines + 3):
|
||||
self.append_new_line()
|
||||
|
||||
def append_new_line(self):
|
||||
self.append([None for x in range(self.collumns)])
|
||||
|
||||
def cell_is_free(self, coord):
|
||||
return 0 <= coord.x < NB_COLS and 0 <= coord.y and not self[coord.y][coord.x]
|
||||
return 0 <= coord.x < self.collumns and 0 <= coord.y and not self[coord.y][coord.x]
|
||||
|
||||
|
||||
class NextQueue(PieceContainer):
|
||||
|
||||
def __init__(self):
|
||||
def __init__(self, nb_pieces):
|
||||
super().__init__()
|
||||
self.nb_pieces = nb_pieces
|
||||
self.pieces = []
|
||||
|
||||
|
||||
class TetrisLogic:
|
||||
class Stats:
|
||||
|
||||
NB_LINES = NB_LINES
|
||||
NB_COLS = NB_COLS
|
||||
NB_NEXT = NB_NEXT
|
||||
LOCK_DELAY = LOCK_DELAY
|
||||
FALL_DELAY = FALL_DELAY
|
||||
AUTOREPEAT_DELAY = AUTOREPEAT_DELAY
|
||||
AUTOREPEAT_PERIOD = AUTOREPEAT_PERIOD
|
||||
MATRIX_PIECE_COORD = MATRIX_PIECE_COORD
|
||||
NEXT_PIECE_COORDS = NEXT_PIECE_COORDS
|
||||
HELD_PIECE_COORD = HELD_PIECE_COORD
|
||||
random_bag = []
|
||||
|
||||
def __init__(self, lines=NB_LINES, collumns=NB_COLS, next_pieces=NB_NEXT):
|
||||
self.load_high_score()
|
||||
self.state = State.STARTING
|
||||
self.held = HoldQueue()
|
||||
self.matrix = Matrix()
|
||||
self.matrix.ghost = None
|
||||
self.next = NextQueue()
|
||||
self.time = 0
|
||||
self.autorepeatable_actions = (self.move_left, self.move_right, self.soft_drop)
|
||||
self.pressed_actions = []
|
||||
self._score = 0
|
||||
|
||||
def get_score(self):
|
||||
def _get_score(self):
|
||||
return self._score
|
||||
|
||||
def set_score(self, new_score):
|
||||
def _set_score(self, new_score):
|
||||
self._score = new_score
|
||||
if self._score > self.high_score:
|
||||
self.high_score = self._score
|
||||
|
||||
score = property(get_score, set_score)
|
||||
score = property(_get_score, _set_score)
|
||||
|
||||
def new_game(self, level=1):
|
||||
self.level = level - 1
|
||||
self.score = 0
|
||||
self.nb_lines_cleared = 0
|
||||
self.goal = 0
|
||||
def __init__(self):
|
||||
self._score = 0
|
||||
self.time = 0
|
||||
|
||||
self.pressed_actions = []
|
||||
self.auto_repeat = False
|
||||
def new_game(self, level):
|
||||
self.level = level - 1
|
||||
self.score = 0
|
||||
self.lines_cleared = 0
|
||||
self.goal = 0
|
||||
self.time = 0
|
||||
self.combo = -1
|
||||
|
||||
self.lock_delay = self.LOCK_DELAY
|
||||
self.fall_delay = self.FALL_DELAY
|
||||
|
||||
self.matrix.clear()
|
||||
for y in range(self.NB_LINES + 3):
|
||||
self.append_new_line_to_matrix()
|
||||
self.next.pieces = [self.new_tetromino() for n in range(self.NB_NEXT)]
|
||||
self.held.piece = None
|
||||
self.state = State.PLAYING
|
||||
self.start(self.update_time, 1)
|
||||
|
||||
self.new_level()
|
||||
|
||||
def new_tetromino(self):
|
||||
if not self.random_bag:
|
||||
self.random_bag = list(Tetromino.shapes)
|
||||
random.shuffle(self.random_bag)
|
||||
return self.random_bag.pop()()
|
||||
|
||||
def append_new_line_to_matrix(self):
|
||||
self.matrix.append([None for x in range(self.NB_COLS)])
|
||||
self.lock_delay = LOCK_DELAY
|
||||
self.fall_delay = FALL_DELAY
|
||||
|
||||
def new_level(self):
|
||||
self.level += 1
|
||||
@ -132,23 +102,102 @@ class TetrisLogic:
|
||||
self.fall_delay = pow(0.8 - ((self.level - 1) * 0.007), self.level - 1)
|
||||
if self.level > 15:
|
||||
self.lock_delay = 0.5 * pow(0.9, self.level - 15)
|
||||
self.show_text("LEVEL\n{:n}".format(self.level))
|
||||
self.restart(self.fall, self.fall_delay)
|
||||
self.new_matrix_piece()
|
||||
|
||||
def new_matrix_piece(self):
|
||||
def update_time(self):
|
||||
self.time += 1
|
||||
|
||||
|
||||
class TetrisLogic:
|
||||
|
||||
LINES = LINES
|
||||
COLLUMNS = COLLUMNS
|
||||
NEXT_PIECES = NEXT_PIECES
|
||||
AUTOREPEAT_DELAY = AUTOREPEAT_DELAY
|
||||
AUTOREPEAT_PERIOD = AUTOREPEAT_PERIOD
|
||||
MATRIX_PIECE_COORD = MATRIX_PIECE_COORD
|
||||
NEXT_PIECE_COORDS = NEXT_PIECE_COORDS
|
||||
HELD_PIECE_COORD = HELD_PIECE_COORD
|
||||
random_bag = []
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
lines=LINES,
|
||||
collumns=COLLUMNS,
|
||||
next_pieces=NEXT_PIECES,
|
||||
):
|
||||
self.stats = Stats()
|
||||
self.load_high_score()
|
||||
self.state = State.STARTING
|
||||
self.held = HoldQueue()
|
||||
self.matrix = Matrix(lines, collumns)
|
||||
self.matrix.ghost = None
|
||||
self.next = NextQueue(next_pieces)
|
||||
self.autorepeatable_actions = (self.move_left, self.move_right, self.soft_drop)
|
||||
self.pressed_actions = []
|
||||
|
||||
def new_game(self, level=1):
|
||||
self.stats.new_game(level)
|
||||
|
||||
self.pressed_actions = []
|
||||
self.auto_repeat = False
|
||||
|
||||
self.matrix.reset()
|
||||
self.next.pieces = [self.new_piece() for n in range(self.next.nb_pieces)]
|
||||
self.held.piece = None
|
||||
self.state = State.PLAYING
|
||||
self.start(self.stats.update_time, 1)
|
||||
|
||||
self.on_new_game()
|
||||
|
||||
self.new_level()
|
||||
|
||||
def on_new_game(self):
|
||||
pass
|
||||
|
||||
def new_piece(self):
|
||||
if not self.random_bag:
|
||||
self.random_bag = list(Tetromino.shapes)
|
||||
random.shuffle(self.random_bag)
|
||||
piece = self.random_bag.pop()()
|
||||
self.on_new_piece(piece)
|
||||
return piece
|
||||
|
||||
def on_new_piece(self, piece):
|
||||
pass
|
||||
|
||||
def new_level(self):
|
||||
self.stats.new_level()
|
||||
self.restart(self.fall, self.stats.fall_delay)
|
||||
|
||||
self.on_new_level(self.stats.level)
|
||||
|
||||
self.generation_phase()
|
||||
|
||||
def on_new_level(self, level):
|
||||
pass
|
||||
|
||||
def generation_phase(self):
|
||||
self.matrix.piece = self.next.pieces.pop(0)
|
||||
self.matrix.piece.coord = self.MATRIX_PIECE_COORD
|
||||
self.matrix.ghost = self.matrix.piece.ghost()
|
||||
self.move_ghost()
|
||||
self.next.pieces.append(self.new_tetromino())
|
||||
self.next.pieces.append(self.new_piece())
|
||||
self.next.pieces[-1].coord = self.NEXT_PIECE_COORDS[-1]
|
||||
for tetromino, coord in zip(self.next.pieces, self.NEXT_PIECE_COORDS):
|
||||
tetromino.coord = coord
|
||||
|
||||
self.on_generation_phase(self.matrix.piece)
|
||||
self.on_falling_phase()
|
||||
|
||||
if not self.can_move(self.matrix.piece.coord, (mino.coord for mino in self.matrix.piece)):
|
||||
self.game_over()
|
||||
|
||||
def on_generation_phase(self, piece):
|
||||
pass
|
||||
|
||||
def on_falling_phase(self):
|
||||
pass
|
||||
|
||||
def move_left(self):
|
||||
self.move(Movement.LEFT)
|
||||
|
||||
@ -171,12 +220,12 @@ class TetrisLogic:
|
||||
def soft_drop(self):
|
||||
moved = self.move(Movement.DOWN)
|
||||
if moved:
|
||||
self.score += 1
|
||||
self.stats.score += 1
|
||||
return moved
|
||||
|
||||
def hard_drop(self):
|
||||
while self.move(Movement.DOWN, prelock=False):
|
||||
self.score += 2
|
||||
self.stats.score += 2
|
||||
self.lock()
|
||||
|
||||
def fall(self):
|
||||
@ -186,35 +235,47 @@ class TetrisLogic:
|
||||
potential_coord = self.matrix.piece.coord + movement
|
||||
if self.can_move(potential_coord, (mino.coord for mino in self.matrix.piece)):
|
||||
if self.matrix.piece.prelocked:
|
||||
self.restart(self.lock, self.lock_delay)
|
||||
self.restart(self.lock, self.stats.lock_delay)
|
||||
self.matrix.piece.coord = potential_coord
|
||||
if not movement == Movement.DOWN:
|
||||
self.matrix.piece.last_rotation_point = None
|
||||
self.move_ghost()
|
||||
self.on_moved(movement)
|
||||
return True
|
||||
else:
|
||||
if prelock and not self.matrix.piece.prelocked and movement == Movement.DOWN:
|
||||
self.matrix.piece.prelocked = True
|
||||
self.start(self.lock, self.lock_delay)
|
||||
self.start(self.lock, self.stats.lock_delay)
|
||||
self.on_lock_phase()
|
||||
return False
|
||||
|
||||
def on_moved(self, movement):
|
||||
pass
|
||||
|
||||
def rotate(self, rotation):
|
||||
rotated_coords = tuple(Coord(rotation * mino.coord.y, -rotation * mino.coord.x) for mino in self.matrix.piece)
|
||||
for rotation_point, liberty_degree in enumerate(self.matrix.piece.SRS[rotation][self.matrix.piece.orientation], start=1):
|
||||
potential_coord = self.matrix.piece.coord + liberty_degree
|
||||
if self.can_move(potential_coord, rotated_coords):
|
||||
if self.matrix.piece.prelocked:
|
||||
self.restart(self.lock, self.lock_delay)
|
||||
self.restart(self.lock, self.stats.lock_delay)
|
||||
self.matrix.piece.coord = potential_coord
|
||||
for mino, coord in zip(self.matrix.piece, rotated_coords):
|
||||
mino.coord = coord
|
||||
self.matrix.piece.orientation = (self.matrix.piece.orientation + rotation) % 4
|
||||
self.matrix.piece.last_rotation_point = rotation_point
|
||||
self.move_ghost()
|
||||
self.on_rotated(rotation)
|
||||
return True
|
||||
else:
|
||||
return False
|
||||
|
||||
def on_rotated(self, direction):
|
||||
pass
|
||||
|
||||
def on_lock_phase(self):
|
||||
pass
|
||||
|
||||
SCORES = (
|
||||
{LINES_CLEAR_NAME: "", T_Spin.NONE: 0, T_Spin.MINI: 1, T_Spin.T_SPIN: 4},
|
||||
{LINES_CLEAR_NAME: "SINGLE", T_Spin.NONE: 1, T_Spin.MINI: 2, T_Spin.T_SPIN: 8},
|
||||
@ -232,13 +293,12 @@ class TetrisLogic:
|
||||
return
|
||||
|
||||
# Game over
|
||||
if all((mino.coord + self.matrix.piece.coord).y >= self.NB_LINES for mino in self.matrix.piece):
|
||||
if all((mino.coord + self.matrix.piece.coord).y >= self.matrix.lines for mino in self.matrix.piece):
|
||||
self.game_over()
|
||||
return
|
||||
|
||||
if self.pressed_actions:
|
||||
self.auto_repeat = False
|
||||
self.restart(self.repeat_action, self.AUTOREPEAT_DELAY)
|
||||
|
||||
# T-Spin
|
||||
if type(self.matrix.piece) == T_Tetrimino and self.matrix.piece.last_rotation_point is not None:
|
||||
@ -255,17 +315,21 @@ class TetrisLogic:
|
||||
else:
|
||||
t_spin = T_Spin.NONE
|
||||
|
||||
self.enter_the_matrix()
|
||||
for mino in self.matrix.piece:
|
||||
coord = mino.coord + self.matrix.piece.coord
|
||||
if coord.y <= self.matrix.lines + 3:
|
||||
self.matrix[coord.y][coord.x] = mino
|
||||
self.on_locked(self.matrix.piece)
|
||||
|
||||
# Clear complete lines
|
||||
nb_lines_cleared = 0
|
||||
lines_cleared = 0
|
||||
for y, line in reversed(list(enumerate(self.matrix))):
|
||||
if all(mino for mino in line):
|
||||
nb_lines_cleared += 1
|
||||
lines_cleared += 1
|
||||
self.remove_line(y)
|
||||
self.append_new_line_to_matrix()
|
||||
if nb_lines_cleared:
|
||||
self.nb_lines_cleared += nb_lines_cleared
|
||||
self.matrix.append_new_line()
|
||||
if lines_cleared:
|
||||
self.stats.lines_cleared += lines_cleared
|
||||
|
||||
# Scoring
|
||||
lock_strings = []
|
||||
@ -273,37 +337,37 @@ class TetrisLogic:
|
||||
|
||||
if t_spin:
|
||||
lock_strings.append(t_spin)
|
||||
if nb_lines_cleared:
|
||||
lock_strings.append(self.SCORES[nb_lines_cleared][LINES_CLEAR_NAME])
|
||||
self.combo += 1
|
||||
if lines_cleared:
|
||||
lock_strings.append(self.SCORES[lines_cleared][LINES_CLEAR_NAME])
|
||||
self.stats.combo += 1
|
||||
else:
|
||||
self.combo = -1
|
||||
self.stats.combo = -1
|
||||
|
||||
if nb_lines_cleared or t_spin:
|
||||
ds = self.SCORES[nb_lines_cleared][t_spin]
|
||||
self.goal -= ds
|
||||
ds *= 100 * self.level
|
||||
if lines_cleared or t_spin:
|
||||
ds = self.SCORES[lines_cleared][t_spin]
|
||||
self.stats.goal -= ds
|
||||
ds *= 100 * self.stats.level
|
||||
lock_score += ds
|
||||
lock_strings.append(str(ds))
|
||||
self.show_text("\n".join(lock_strings))
|
||||
|
||||
if self.combo >= 1:
|
||||
ds = (20 if nb_lines_cleared == 1 else 50) * self.combo * self.level
|
||||
if self.stats.combo >= 1:
|
||||
ds = (20 if lines_cleared == 1 else 50) * self.stats.combo * self.stats.level
|
||||
lock_score += ds
|
||||
self.show_text("COMBO x{:n}\n{:n}".format(self.combo, ds))
|
||||
self.show_text("COMBO x{:n}\n{:n}".format(self.stats.combo, ds))
|
||||
|
||||
self.score += lock_score
|
||||
self.stats.score += lock_score
|
||||
|
||||
if self.goal <= 0:
|
||||
if self.stats.goal <= 0:
|
||||
self.new_level()
|
||||
else:
|
||||
self.new_matrix_piece()
|
||||
self.generation_phase()
|
||||
|
||||
def enter_the_matrix(self):
|
||||
for mino in self.matrix.piece:
|
||||
coord = mino.coord + self.matrix.piece.coord
|
||||
if coord.y <= self.NB_LINES + 3:
|
||||
self.matrix[coord.y][coord.x] = mino
|
||||
if self.auto_repeat:
|
||||
self.restart(self.repeat_action, self.AUTOREPEAT_DELAY)
|
||||
|
||||
def on_locked(piece):
|
||||
pass
|
||||
|
||||
def remove_line(self, y):
|
||||
self.matrix.pop(y)
|
||||
@ -334,7 +398,7 @@ class TetrisLogic:
|
||||
self.matrix.ghost = self.matrix.piece.ghost()
|
||||
self.move_ghost()
|
||||
else:
|
||||
self.new_matrix_piece()
|
||||
self.generation_phase()
|
||||
|
||||
def pause(self):
|
||||
self.state = State.PAUSED
|
||||
@ -345,10 +409,10 @@ class TetrisLogic:
|
||||
|
||||
def resume(self):
|
||||
self.state = State.PLAYING
|
||||
self.start(self.fall, self.fall_delay)
|
||||
self.start(self.fall, self.stats.fall_delay)
|
||||
if self.matrix.piece.prelocked:
|
||||
self.start(self.lock, self.lock_delay)
|
||||
self.start(self.update_time, 1)
|
||||
self.start(self.lock, self.stats.lock_delay)
|
||||
self.start(self.stats.update_time, 1)
|
||||
|
||||
def game_over(self):
|
||||
self.state = State.OVER
|
||||
@ -358,10 +422,7 @@ class TetrisLogic:
|
||||
def stop_all(self):
|
||||
self.stop(self.fall)
|
||||
self.stop(self.lock)
|
||||
self.stop(self.update_time)
|
||||
|
||||
def update_time(self):
|
||||
self.time += 1
|
||||
self.stop(self.stats.update_time)
|
||||
|
||||
def do_action(self, action):
|
||||
action()
|
||||
@ -369,7 +430,7 @@ class TetrisLogic:
|
||||
self.auto_repeat = False
|
||||
self.pressed_actions.append(action)
|
||||
if action == self.soft_drop:
|
||||
delay = self.fall_delay / 20
|
||||
delay = self.stats.fall_delay / 20
|
||||
else:
|
||||
delay = self.AUTOREPEAT_DELAY
|
||||
self.restart(self.repeat_action, delay)
|
||||
@ -398,16 +459,16 @@ class TetrisLogic:
|
||||
def load_high_score(self, crypted_high_score=None):
|
||||
if crypted_high_score:
|
||||
crypted_high_score = int(pickle.loads(crypted_high_score))
|
||||
self.high_score = crypted_high_score ^ CRYPT_KEY
|
||||
self.stats.high_score = crypted_high_score ^ CRYPT_KEY
|
||||
else:
|
||||
raise Warning(
|
||||
"""TetrisLogic.load_high_score not implemented.
|
||||
High score is set to 0"""
|
||||
)
|
||||
self.high_score = 0
|
||||
self.stats.high_score = 0
|
||||
|
||||
def save_high_score(self):
|
||||
crypted_high_score = self.high_score ^ CRYPT_KEY
|
||||
crypted_high_score = self.stats.high_score ^ CRYPT_KEY
|
||||
crypted_high_score = pickle.dumps(crypted_high_score)
|
||||
return crypted_high_score
|
||||
|
||||
|
Loading…
x
Reference in New Issue
Block a user