forked from caca2331/Autodori-BanGDream-Autoplay
-
Notifications
You must be signed in to change notification settings - Fork 0
/
SongInfo.py
362 lines (298 loc) · 14.3 KB
/
SongInfo.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
import re
import json
from sys import argv
from Util import *
#
# This file process the song data
#
#
#
class SongInfo:
# constants used for song play generation
F_DUR = 0.08 # expected duration of flick
STICK_DUR = 0.05 # expected duration of a touch on the the screen
SLIDE_STICK_DUR = 0.08 # min time for a finger to stay between two notes within a slide
IN_AIR_DUR = 0.08 # min time for a finger to stay in air between two touches
def __init__(self, song_name, screen_info=None, player_lv=None, sampling_period=None):
self.song_name = song_name
# by iOS default it should be 60Hz (except iPad pro of 120Hz)
self.sampling_period = float(sampling_period) if sampling_period else 1.0 / 60
self.screen_info = screen_info
# self.global_delay = global_delay
# normal good player(fc lv25 songs) will be 10,
# 20 will be very good player,
# 25 will be top 10 player,
# 30+ will be suspicious (risky!)
self.player_lv = int(player_lv) if player_lv else 10
# if actual_song_lv_in_the_game <= 20, difficulty = 0
# else difficulty = actual_song_lv_in_the_game - 20
self.difficulty = 10
# (40% 0-0.3 21% 0.3-0.5 19% 0.5-0.7 18% 0.72-0.95 2% 0.95-1) * p_perfect
# (50% 1-1.4 35% 1.4-1.7 10% 1.7-1.95 5% 1.95-2.0) * p_great
# (60% 2-3.5, miss otherwise) * p_miss
self.p_perfect = .85 + .15 * (1 - .8 ** self.player_lv * 1.2 ** self.difficulty)
self.great_miss_ratio = self.player_lv / 1.2 ** self.difficulty
self.score = []
self.music_info = {}
self.note_info = {}
self.timed_actions = []
''' constants '''
# note data (range in milliseconds for perfect)
self.note_time_range = {
('skill', 1): (-33, 50),
'bd': (-33, 50),
'flick': (-33, 50),
'fever_note': (),
'': ()
}
# delay between when song poster disappear and song begin to play
# self._post_poster_delay =
# The majority of this function is adapted from https://github.com/yp05327/BangScoreMaker/ by yp05327
# analysis song
def init_score(self):
music_info_key_word = {
'bpm': '#BPM (\\d*)',
'id': '#WAV01 bgm(\\d*).wav',
}
file_object = open('score/' + self.song_name + '.txt', 'r')
read_step = 0
try:
for line in file_object:
if read_step == 0:
# get music info
for key_word in music_info_key_word:
info = re.match(music_info_key_word[key_word], line)
if info:
self.music_info[key_word] = info.group(1)
if read_step < 2:
# get note info
info = re.match('#WAV(\\d\\w) (\\w*).wav', line)
if info:
if read_step == 0:
read_step = 1
self.note_info[info.group(1)] = info.group(2)
# get music score
if line == '*---------------------- MAIN DATA FIELD\n':
read_step = 2
continue
if read_step == 2:
info = re.match('#(\\d\\d\\d)(\\d)(\\d):(\\w*)', line)
if info:
self.score.append([
info.group(1), # unitId
info.group(2), # type
info.group(3), # row
info.group(4) # list
])
finally:
file_object.close()
# convert score into timestamped actions
def init_timed_actions(self):
"""
types of misses:
single touch:
totally misses a note
touched too early or late
didn't slide on flick
bar:
missed first one
didn't get in position on time in slide
release too early
didn't slide on flick
implementation v0:
single touch:
can only have all type
bar:
last one release too early
"""
# time offset from the beginning of the song to the time when teh action should be performed
def gen_time_offset():
if not randomTimeDelay:
local_offset = 0
else:
# for now, assume both bound are 40, and has same probably to go beyond each
l, h = -40, 40 # lower bound, upper(high) bound
if random.random() < self.p_perfect:
# (40% 0-0.3 21% 0.3-0.5 19% 0.5-0.72 18% 0.72-0.95 2% 0.95-1) * p_perfect
r = random.random() # range
if r < .4:
local_offset = random.random(0, .3) * h
elif r < .61:
local_offset = random.uniform(.3, .5) * h
elif r < .8:
local_offset = random.uniform(.5, .72) * h
elif r < .98:
local_offset = random.uniform(.72, .95) * h
else:
local_offset = random.uniform(.95, .1) * h
elif self.great_miss_ratio < random.uniform(0, 1 + self.great_miss_ratio):
pass
# (50% 1-1.4 35% 1.4-1.7 10% 1.7-1.95 5% 1.95-2.0) * p_great
# (60% 2-3.5, miss otherwise) * p_miss
else:
pass
if random.random() < .5:
local_offset = -local_offset
result = (four_beat * 4 + i // 2 * beat_per_division) * 60.0 / int(self.music_info["bpm"]) + local_offset
return result if result > 0 else 0
# generate a random location on a track
def gen_touch_loc(flick=False):
x1, x2, y1, y2 = self.screen_info.track_loc[track]
if flick:
return rand_loc(x1, x2, y1 + y1 + y1 - y2 - y2, y1 + y1 - y2)
return rand_loc(x1, x2, y1, y2)
# create continuous touch by multiple individual call with adb
def add_touch_cmd(action_list, finish_with_flick=False):
touch_command_list = []
time_offset, (x, y) = action_list[0]
# single touch
if len(action_list) == 1:
if finish_with_flick: # finger sticks on screen for 0.15s while flicking
x2, y2 = gen_touch_loc(True)
touch_command_list.append([x, y, x2, y2, rand_time_for_touch(SongInfo.F_DUR)])
else:
x2, y2 = rand_loc(x - 5, y - 5, x + 5, y + 5)
touch_command_list.append([x, y, x2, y2, rand_time_for_touch(SongInfo.STICK_DUR)])
# bar / slide
else:
for i in range(1, len(action_list)):
time_offset2, (x2, y2) = action_list[i]
if time_offset + SongInfo.SLIDE_STICK_DUR > time_offset2:
time_offset2 = time_offset + SongInfo.SLIDE_STICK_DUR
touch_command_list.append([x, y, x2, y2, time_offset2 - time_offset])
time_offset, x, y = time_offset2, x2, y2
if finish_with_flick: # finger sticks on screen for 0.15s while flicking
x2, y2 = gen_touch_loc(True)
touch_command_list.append([x, y, x2, y2, rand_time_for_touch(SongInfo.F_DUR)])
timed_actions_without_id.append((action_list[0][0], touch_command_list))
self.timed_actions = []
timed_actions_without_id = [] # reset
slide_a, slide_b = [], [] # slide a, slide b
v = [[] for _temp in range(9)] # vertical slide
for u in self.score:
four_beat = int(u[0])
# type of note.
# 0: command, ignore
# 1: normal note / non-vertical slide
# 5: vertical slide
note_type = int(u[1])
# which track?
# 6: left most track in the game
# 1-5: track 2-6 in the game (the center five)
# 8: right most track in the game
track = int(u[2])
# type of the note, and when the note falls
pattern = u[3]
if note_type == 0:
continue
beat_per_division = 4 / (len(pattern) / 2.0)
token = ""
for i in range(len(pattern)):
# extract two chars as a token
if i % 2 == 0:
token = pattern[i]
continue
else:
token += pattern[i]
# skip token if it is not recognized. otherwise re-assign its value (note_type string)
if token not in self.note_info:
continue
else:
token = self.note_info[token]
# generate time_offset and touch_loc based on current token
temp_cmd_of_token = gen_time_offset(), gen_touch_loc()
# vertical slide
if note_type == 5:
v[track].append(temp_cmd_of_token)
if len([track]) == 2: # end of vertical slide
add_touch_cmd(v[track], token in ["flick", "fever_note_flick"])
v[track] = []
# single note, or non-vertical slide
else:
if token in ["bd", "skill", "fever_note"]:
add_touch_cmd([temp_cmd_of_token])
elif token in ["flick", "fever_note_flick"]:
add_touch_cmd([temp_cmd_of_token], True)
elif token in ["slide_a"]:
slide_a.append(temp_cmd_of_token)
elif token in ["slide_b"]:
slide_b.append(temp_cmd_of_token)
elif token in ["slide_end_a", "slide_end_flick_a"]:
slide_a.append(temp_cmd_of_token)
add_touch_cmd(slide_a, token in ["slide_end_flick_a"])
slide_a = []
elif token in ["slide_end_b", "slide_end_flick_b"]:
slide_b.append(temp_cmd_of_token)
add_touch_cmd(slide_b, token in ["slide_end_flick_b"])
slide_b = []
''' *** *** '''
list.sort(timed_actions_without_id)
finger = [-1, -1] # the latest time that either finger is touched down (finger0 and finger1)
for start_time, actions in timed_actions_without_id:
# determine which finger to use
# if finger0 not in use, use it.
if finger[0] + SongInfo.IN_AIR_DUR <= start_time:
finger_to_use = 0
# else if finger1 not in use, use it.
elif finger[1] + SongInfo.IN_AIR_DUR <= start_time:
finger_to_use = 1
# else use whichever finger that is available sooner.
elif finger[1] + SongInfo.IN_AIR_DUR <= finger[0]:
finger_to_use = 1
start_time = min(finger) + SongInfo.IN_AIR_DUR
else:
finger_to_use = 0
start_time = min(finger) + SongInfo.IN_AIR_DUR
ttl_dur = 0
for i in range(len(actions)):
# format: [ [time_in_sec, finger_id, x1, y1, x2, y2, duration_in_sec,
# is_start_with_finger_attached(1 or 0), is_end_with_finger_attached(1 or 0)] ... ]
self.timed_actions.append([start_time + ttl_dur, finger_to_use,
round(actions[i][0]), round(actions[i][1]),
round(actions[i][2]), round(actions[i][3]), actions[i][4],
0 if i == 0 else 1, 0 if i == len(actions) - 1 else 1])
ttl_dur += actions[i][4]
finger[finger_to_use] = start_time + ttl_dur
# write timestamped commands to a json file.
# format 1: [ [time_in_sec, finger_id, x1, y1, x2, y2, duration_in_sec,
# is_start_with_finger_attached(1 or 0), is_end_with_finger_attached(1 or 0)] ... ]
# format 2: [ [time_in_sec, finger_id, x, y, finger_attaching/moving/leaving(0/1/2)], ... ]
def write_to_file(self, name=None):
if name is None:
name = self.song_name + '-timed_actions'
with open('build/' + name + '.json', 'w') as f:
json.dump(self.timed_actions, f)
def transform_timed_actions_type(self, timed_actions_type=None):
if timed_actions_type == 1:
transformed_timed_actions = []
for s_time, f_id, x1, y1, x2, y2, dur, is_s_attached, is_e_attached in self.timed_actions:
print(s_time, f_id, x1, y1, x2, y2, dur, is_s_attached, is_e_attached)
if not is_s_attached:
transformed_timed_actions.append([s_time, f_id, x1, y1, 0])
time_sep = dur / self.sampling_period
dx, dy = (x2 - x1) / time_sep, (y2 - y1) / time_sep
while time_sep > 1:
s_time += self.sampling_period
x1 += dx
y1 += dy
time_sep -= 1
transformed_timed_actions.append([s_time, f_id, x1, y1, 1])
s_time += time_sep * self.sampling_period
transformed_timed_actions.append([s_time, f_id, x2, y2, 1])
if not is_e_attached:
transformed_timed_actions.append([s_time, f_id, x2, y2, 2])
list.sort(transformed_timed_actions)
self.timed_actions = transformed_timed_actions
@staticmethod
def gen_timed_actions(song_name, w, h, player_lv=None, filename=None, timed_actions_type=0,
sampling_period=None):
song = SongInfo(song_name, ScreenInfo(w, h), player_lv, sampling_period)
song.init_score()
song.init_timed_actions()
song.transform_timed_actions_type(int(timed_actions_type))
song.write_to_file(filename)
if __name__ == "__main__":
print("Usage: song_name, [screen width] [screen height] [player_lv]"
"song_name is the name of the score file in score directory, without the suffix of '.txt'"
"screen width")
SongInfo.gen_timed_actions(*argv[1:])