forked from chrisbutner/ChessCoach
-
Notifications
You must be signed in to change notification settings - Fork 0
/
lichessbot.py
883 lines (741 loc) · 30 KB
/
lichessbot.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
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
# Based on https://github.com/ShailChoksi/lichess-bot
#
# AGPL3-or-later license (see lichess-bot/LICENSE and https://github.com/ShailChoksi/lichess-bot#license)
#
# Modifications by Chris Butner, 2021
import os
import json
import threading
import queue
import time
import random
import backoff
import requests
from requests.exceptions import ChunkedEncodingError, ConnectionError, HTTPError, ReadTimeout
from urllib3.exceptions import ProtocolError
from http.client import RemoteDisconnected
from urllib.parse import urljoin
import textwrap
from bs4 import BeautifulSoup
import numpy as np
try:
import chesscoach # See PythonModule.cpp
except:
pass
# --- Constants ---
WHITE = 0
BLACK = 1
STARTING_POSITION = "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"
# --- Data structures ---
class MissingPieces:
def __init__(self, below_rating, fens):
self.below_rating = below_rating
self.fens = fens
def is_supported(self, rating, fen):
return (fen in self.fens) and (rating < self.below_rating)
class Challenge:
def __init__(self, event):
self.id = event["id"]
self.rated = event["rated"]
self.variant = event["variant"]["key"]
self.perf_name = event["perf"]["name"]
self.speed = event["speed"]
self.increment = event.get("timeControl", {}).get("increment", -1)
self.base = event.get("timeControl", {}).get("limit", -1)
self.challenger = event.get("challenger")
self.challenger_title = self.challenger.get("title") if self.challenger else None
self.challenger_is_bot = self.challenger_title == "BOT"
self.challenger_master_title = self.challenger_title if not self.challenger_is_bot else None
self.challenger_name = self.challenger["name"] if self.challenger else "Anonymous"
self.challenger_rating_int = self.challenger["rating"] if self.challenger else 0
self.challenger_rating = self.challenger_rating_int or "?"
self.color = event["color"]
self.initial_fen = event.get("initialFen")
self.missing_pieces = [
# WHITE is missing pieces.
self.set_up_missing_pieces(lambda x: x.upper()),
# BLACK is missing pieces.
self.set_up_missing_pieces(lambda x: x),
]
# Opponent must have 2999 rating or below to allow pawn odds, 1499 rating or below to allow queen odds, etc.
# Go crazy and allow kingside pieces, any pawn combinations, first move, whatever.
def set_up_missing_pieces(self, filter):
return [
MissingPieces(3000, self.remove_piece(STARTING_POSITION, filter("p"))),
MissingPieces(2850, self.remove_two_pieces(STARTING_POSITION, filter("p"))),
MissingPieces(2500, self.remove_piece(STARTING_POSITION, filter("n"))),
MissingPieces(2500, self.remove_piece(STARTING_POSITION, filter("b"))),
MissingPieces(2000, self.remove_piece(STARTING_POSITION, filter("r"))),
MissingPieces(1500, self.remove_piece(STARTING_POSITION, filter("q"))),
]
def remove_piece(self, fen, piece):
return set((fen[:i] + "1" + fen[i + 1:]).replace("11", "2") for i, existing in enumerate(fen) if existing == piece)
def remove_two_pieces(self, fen, piece):
all = ([self.remove_piece(removed, piece) for removed in self.remove_piece(fen, piece)])
return set([item for sublist in all for item in sublist]) # The set de-duplicates.
def is_supported_variant(self):
return (
# Allow standard chess. Have to support arbitrary colors to allow for rematches.
(self.variant == "standard") or
# Allow piece odds, but only when the opponent has all pieces and we're missing some.
(self.variant == "fromPosition" and self.color == "white" and
any(m.is_supported(self.challenger_rating_int, self.initial_fen) for m in self.missing_pieces[BLACK])) or
(self.variant == "fromPosition" and self.color == "black" and
any(m.is_supported(self.challenger_rating_int, self.initial_fen) for m in self.missing_pieces[WHITE])))
def is_fast_enough(self):
return self.base <= 900 and self.increment <= 10
def is_slow_enough(self):
return self.base >= 60 or self.increment >= 1
# Returns (queue index, decline reason) tuple.
# Prefer rated games over casual games.
def is_supported(self):
if not self.is_supported_variant():
return None, "variant"
if not self.is_fast_enough():
return None, "tooSlow"
if not self.is_slow_enough():
return None, "tooFast"
return (0 if self.rated else 1), None
def mode(self):
return "rated" if self.rated else "casual"
def challenger_full_name(self):
return "{}{}".format(self.challenger_title + " " if self.challenger_title else "", self.challenger_name)
def __str__(self):
return "{} {} challenge from {}({})".format(self.perf_name, self.mode(), self.challenger_full_name(), self.challenger_rating)
def __repr__(self):
return self.__str__()
class Game:
comment_frequency_seconds = 30.0
abort_seconds = 30.0 # 10 for first white move, 10 for first black move, 10 buffer
def __init__(self, json, username, base_url):
self.username = username
self.id = json.get("id")
self.speed = json.get("speed")
clock = json.get("clock", {}) or {}
self.clock_initial = clock.get("initial", 1000 * 3600 * 24 * 365 * 10) # unlimited = 10 years
self.clock_increment = clock.get("increment", 0)
self.perf_name = json.get("perf").get("name") if json.get("perf") else "{perf?}"
self.variant_name = json.get("variant")["name"]
self.white = Player(json.get("white"))
self.black = Player(json.get("black"))
self.initial_fen = json.get("initialFen")
self.state = json.get("state")
self.is_white = bool(self.white.name and self.white.name.lower() == username.lower())
self.my_color = "white" if self.is_white else "black"
self.opponent_color = "black" if self.is_white else "white"
self.me = self.white if self.is_white else self.black
self.opponent = self.black if self.is_white else self.white
self.base_url = base_url
self.white_starts = self.initial_fen == "startpos" or self.initial_fen.split()[1] == "w"
self.abort_at = time.time() + self.abort_seconds
self.searching_moves = "(not searching yet)"
self.pending_comment = None
self.last_comment_time = time.time() - random.uniform(0.0, self.comment_frequency_seconds) - 0.5
def url(self):
return urljoin(self.base_url, "{}/{}".format(self.id, self.my_color))
def is_abortable(self):
return len(self.state["moves"]) < 6
def on_activity(self):
self.abort_at = time.time() + self.abort_seconds
def should_abort_now(self):
return self.is_abortable() and time.time() > self.abort_at
# Only allow commenting every 30 seconds, and delay after 429-errors.
def try_pop_comment(self):
comment = self.pending_comment
if not comment:
return None
now = time.time()
if (now - self.last_comment_time) < self.comment_frequency_seconds:
return None
if not throttle.ready():
return None
self.pending_comment = None
self.last_comment_time = now
return comment
# Additional optimization layer: only bother generating a comment when it can be immediately posted.
def can_comment(self):
now = time.time()
return (not self.pending_comment) and ((now - self.last_comment_time) >= self.comment_frequency_seconds)
def __str__(self):
return "{} {} vs {}".format(self.url(), self.perf_name, self.opponent.__str__())
def __repr__(self):
return self.__str__()
class Player:
def __init__(self, json):
self.id = json.get("id")
self.name = json.get("name")
self.title = json.get("title")
self.rating = json.get("rating")
self.provisional = json.get("provisional")
self.aiLevel = json.get("aiLevel")
def __str__(self):
if self.aiLevel:
return "AI level {}".format(self.aiLevel)
else:
rating = "{}{}".format(self.rating, "?" if self.provisional else "")
return "{}{}({})".format(self.title + " " if self.title else "", self.name, rating)
def __repr__(self):
return self.__str__()
class ChatLine:
def __init__(self, json):
self.room = json.get("room")
self.username = json.get("username")
self.text = json.get("text")
class Games:
def __init__(self):
self.queue = set()
self.in_progress = {}
self.last_finish_time = time.time()
def enqueue(self, id):
before = self.status()
self.queue.add(id)
if logging_enabled:
log(f"[Games][Enqueue][{id}] {before} -> {self.status()}")
def dequeue(self, id):
before = self.status()
try:
self.queue.remove(id)
except KeyError:
if logging_enabled:
log(f"[Games][Dequeue][{id}] Wasn't in queue (no action)")
return
if logging_enabled:
log(f"[Games][Dequeue][{id}] {before} -> {self.status()}")
def starting(self, id):
if id in self.in_progress:
if logging_enabled:
log(f"[Games][Starting][{id}] Already tracked (no action)")
return
before = self.status()
try:
self.queue.remove(id)
except KeyError:
if logging_enabled:
log(f"[Games][Starting][{id}] Wasn't in queue")
# Unfortunately, there's no direct correlation between a rematch challenge ID (which remains the same as the original game)
# and the resulting rematch game ID. So, rely on our only playing one game at a time and just dequeue a single game and log
# when the IDs don't match.
if len(self.queue) == 1:
existing_id = next(iter(self.queue))
self.queue.clear()
if logging_enabled:
log(f"[Games][Starting][{id}] Assuming a rematch, dequeuing {existing_id}")
self.in_progress[id] = None
if logging_enabled:
log(f"[Games][Starting][{id}] {before} -> {self.status()}")
def started(self, game):
id = game.id
assert not self.in_progress.get(id, None) # A None in the dictionary is okay (starting) but not a real Game.
before = self.status()
self.in_progress[id] = game
if logging_enabled:
log(f"[Games][Started][{id}] {before} -> {self.status()}")
def finished(self, id):
if id not in self.in_progress:
if logging_enabled:
log(f"[Games][Finished][{id}] Wasn't tracked (no action)")
return
before = self.status()
self.in_progress.pop(id)
self.last_finish_time = time.time()
if logging_enabled:
log(f"[Games][Finished][{id}] {before} -> {self.status()}")
def status(self):
return f"(queue: {self.queue}, in_progress: {self.in_progress})"
# --- Protocol ---
ENDPOINTS = {
"profile": "/api/account",
"playing": "/api/account/playing",
"stream": "/api/bot/game/stream/{}",
"stream_event": "/api/stream/event",
"bots": "/player/bots",
"status": "/api/users/status",
"game": "/api/bot/game/{}",
"move": "/api/bot/game/{}/move/{}",
"chat": "/api/bot/game/{}/chat",
"abort": "/api/bot/game/{}/abort",
"accept": "/api/challenge/{}/accept",
"decline": "/api/challenge/{}/decline",
"challenge": "/api/challenge/{}",
"cancel": "/api/challenge/{}/cancel",
"upgrade": "/api/bot/account/upgrade",
"resign": "/api/bot/game/{}/resign",
}
# docs: https://lichess.org/api
class Lichess:
def __init__(self, token, url):
self.header = {
"Authorization": "Bearer {}".format(token),
"User-Agent": "ChessCoach",
}
self.baseUrl = url
self.session = requests.Session()
self.session.headers.update(self.header)
def is_final(exception):
return isinstance(exception, HTTPError) and exception.response.status_code < 500
@backoff.on_exception(backoff.constant,
(RemoteDisconnected, ConnectionError, ProtocolError, HTTPError, ReadTimeout),
max_time=60,
interval=1.0,
giveup=is_final)
def api_get(self, path, params=None):
url = urljoin(self.baseUrl, path)
if logging_enabled:
log(url)
response = self.session.get(url, params=params, timeout=2)
response.raise_for_status()
return response.json()
@backoff.on_exception(backoff.constant,
(RemoteDisconnected, ConnectionError, ProtocolError, HTTPError, ReadTimeout),
max_time=60,
interval=1.0,
giveup=is_final)
def api_post(self, path, data=None, headers=None):
url = urljoin(self.baseUrl, path)
if logging_enabled:
log(url)
response = self.session.post(url, data=data, headers=headers, timeout=2)
response.raise_for_status()
return response.json()
def get_game(self, game_id):
return self.api_get(ENDPOINTS["game"].format(game_id))
def upgrade_to_bot_account(self):
return self.api_post(ENDPOINTS["upgrade"])
def make_move(self, game_id, move):
return self.api_post(ENDPOINTS["move"].format(game_id, move))
def chat(self, game_id, room, text):
payload = {'room': room, 'text': text}
return self.api_post(ENDPOINTS["chat"].format(game_id), data=payload)
def abort(self, game_id):
return self.api_post(ENDPOINTS["abort"].format(game_id))
def get_event_stream(self):
url = urljoin(self.baseUrl, ENDPOINTS["stream_event"])
if logging_enabled:
log(url)
return requests.get(url, headers=self.header, stream=True)
def get_game_stream(self, game_id):
url = urljoin(self.baseUrl, ENDPOINTS["stream"].format(game_id))
if logging_enabled:
log(url)
return requests.get(url, headers=self.header, stream=True)
def get_bots(self):
url = urljoin(self.baseUrl, ENDPOINTS["bots"])
if logging_enabled:
log(url)
return requests.get(url)
def get_status(self, ids):
return self.api_get(ENDPOINTS["status"], params={ "ids": ids })
def accept_challenge(self, challenge_id):
return self.api_post(ENDPOINTS["accept"].format(challenge_id))
def decline_challenge(self, challenge_id, reason="generic"):
payload = { "reason": reason }
return self.api_post(ENDPOINTS["decline"].format(challenge_id), data=payload)
def send_challenge(self, username, clock_limit, clock_increment):
payload = { "rated": "true", "clock.limit": clock_limit, "clock.increment": clock_increment }
return self.api_post(ENDPOINTS["challenge"].format(username), data=payload)
def cancel_challenge(self, challenge_id):
return self.api_post(ENDPOINTS["cancel"].format(challenge_id))
def get_profile(self):
profile = self.api_get(ENDPOINTS["profile"])
return profile
def get_ongoing_games(self):
ongoing_games = self.api_get(ENDPOINTS["playing"])["nowPlaying"]
return ongoing_games
def resign(self, game_id):
self.api_post(ENDPOINTS["resign"].format(game_id))
# --- Chat ---
class Conversation:
command_prefix = "!"
def __init__(self, game, lichess, challenge_queues):
self.game = game
self.lichess = lichess
self.challenge_queues = challenge_queues
def react(self, line, game):
if (line.text[0] == self.command_prefix):
self.command(line, game, line.text[1:].lower())
def command(self, line, game, cmd):
pass
def send_reply(self, line, reply):
self.lichess.chat(self.game.id, line.room, reply)
class Throttle:
delay_seconds = 65.0
def __init__(self):
self.lock = threading.Lock()
self.ready_time = time.time() - 1000.0
def on_exception(self, exception):
with self.lock:
if isinstance(exception, HTTPError) and (exception.response.status_code == 429):
if logging_enabled:
log(f"*** 429 *** for {exception.request.url}: delaying optional requests by {self.delay_seconds} seconds")
self.ready_time = (time.time() + self.delay_seconds)
def ready(self):
with self.lock:
return (time.time() >= self.ready_time)
# --- Sending challenges when idle ---
class OutgoingChallenges:
idle_seconds_before_send = 65.0
unanswered_seconds_before_cancel = 15.0
minimum_best_rating = 1800
rating_sample_temperature = 0.5
time_controls = [
(30, 3),
(60, 0),
(120, 1),
(180, 0),
(180, 2),
(300, 0),
(300, 3),
(600, 0),
(600, 5),
(900, 0),
(900, 10),
]
def __init__(self, lichess, username, games):
self.challenges_active = {}
self.challenges_declined = set()
self.challenges_forbidden = set()
self.lichess = lichess
self.username = username
self.games = games
self.rng = np.random.default_rng()
self.last_send_attempt = time.time()
def upkeep(self):
# Send a new challenge? (Wait 60 seconds in case players want to challenge us, and delay between sends, and after 429-errors.)
if (not self.games.queue
and not self.games.in_progress
and not self.challenges_active
and (time.time() >= self.games.last_finish_time + self.idle_seconds_before_send)
and (time.time() >= self.last_send_attempt + self.idle_seconds_before_send)
and throttle.ready()):
key = self.choose()
if key:
username, (limit, increment) = key
try:
self.last_send_attempt = time.time()
response = self.lichess.send_challenge(username, limit, increment)
id = response["challenge"]["id"]
time_sent = time.time()
self.challenges_active[id] = key, time_sent
self.games.enqueue(id)
if logging_enabled:
log(f"*** OUTGOING CHALLENGE SENT *** with ID {id} to {username} with time control {limit}+{increment}")
except HTTPError as e:
if e.response.status_code == 400:
if logging_enabled:
log(f"Forbidding challenges to {username}: {e.response.content} ({e.response.status_code})")
self.challenges_forbidden.add(username)
if e.response.status_code == 429:
# Assume that all 429s sending challenges are per-day limits and wait an hour.
self.last_send_attempt = time.time() + 3600.0
throttle.on_exception(e)
except Exception:
pass
# Cancel an unanswered challenge? (Wait until next upkeep to send another.)
now = time.time()
remove = []
for id, (key, time_sent) in self.challenges_active.items():
if (now - time_sent) >= self.unanswered_seconds_before_cancel:
username, (limit, increment) = key
try:
self.lichess.cancel_challenge(id)
if logging_enabled:
log(f"*** CANCELED OUTGOING CHALLENGE *** with ID {id} to {username} with time control {limit}+{increment}")
except Exception as e:
throttle.on_exception(e)
if logging_enabled:
log(f"*** ABANDONING OUTGOING CHALLENGE *** with ID {id} to {username} with time control {limit}+{increment} (cancel failed)")
remove.append(id)
for id in remove:
self.games.dequeue(id)
self.challenges_active.pop(id, None)
def game_start(self, id):
# Most details are handled elsewhere: we just need to untrack active outgoing challenges.
self.challenges_active.pop(id, None)
def declined(self, event):
id = event["challenge"]["id"]
reason = event["challenge"]["declineReason"]
self.games.dequeue(id)
challenge = self.challenges_active.pop(id, None)
if challenge:
key, _ = challenge
self.challenges_declined.add(key)
username, (limit, increment) = key
if logging_enabled:
log(f"*** OUTGOING CHALLENGE DECLINED *** with ID {id} to {username} with time control {limit}+{increment}")
if (("I'm not accepting challenges from bots." in reason) or
("I'm not willing to play this variant right now." in reason) or
("Please send me a casual challenge instead." in reason) or
("only accepts challenges from friends." in reason) or
("does not accept challenges." in reason) or
("Please register to send challenges." in reason)):
if logging_enabled:
log(f"Forbidding challenges to {username}: {reason}")
self.challenges_forbidden.add(username)
def list_users(self):
response = self.lichess.get_bots()
soup = BeautifulSoup(response.content, "html.parser")
users = {}
for link, ratings in zip(soup.select(".user-link"), soup.select(".rating")):
username = link["href"][3:]
if username.lower() == self.username.lower() or username in self.challenges_forbidden:
continue
best = 0
for rating in ratings.select("span"):
text = rating.get_text().replace("-", "").replace("?", "").strip()
if text:
best = max(best, int(text))
if best >= self.minimum_best_rating:
users[username] = best
return users
def filter_out_playing(self, users):
status_limit = 50
usernames = list(users.keys())
for i in range(0, len(usernames), status_limit):
ids = ",".join(usernames[i:i + status_limit])
time.sleep(1.0)
status = self.lichess.get_status(ids)
for user_status in status:
if user_status.get("playing", False):
users.pop(user_status["name"])
def sample_users(self):
if not throttle.ready():
return None
users = None
try:
users = self.list_users()
self.filter_out_playing(users)
if not users:
return None
except Exception as e:
throttle.on_exception(e)
return None
usernames = list(users.keys())
bests = list(users.values())
bests = (bests / np.max(bests)) ** (1.0 / self.rating_sample_temperature)
bests /= np.sum(bests)
return self.rng.choice(usernames, p=bests).item()
def sample_time_controls(self):
return random.choice(self.time_controls)
def choose(self):
for _ in range(0, 10):
username = self.sample_users()
time_control = self.sample_time_controls()
if not username or not time_control:
return None
key = (username, time_control)
if key not in self.challenges_declined:
return key
return None
# --- Bot ---
def is_final(exception):
return isinstance(exception, HTTPError) and exception.response.status_code < 500
def decode_json(line):
return json.loads(line.decode("utf-8")) if line else { "type": "ping" }
def watch_control_stream(event_queue, li):
while True:
try:
response = li.get_event_stream()
lines = response.iter_lines()
for line in lines:
event_queue.put(decode_json(line))
except Exception as e:
throttle.on_exception(e)
pass
time.sleep(1.0)
@backoff.on_exception(backoff.constant, BaseException, max_time=600, interval=1.0, giveup=is_final)
def watch_game_stream(li, game_id, event_queue, user_profile):
watch_game = None
response = li.get_game_stream(game_id)
lines = response.iter_lines()
for line in lines:
try:
event = decode_json(line)
if event["type"] == "gameFull":
watch_game = Game(event, user_profile["username"], li.baseUrl)
elif event["type"] == "gameState":
watch_game.state = event
event_queue.put(event)
if watch_game and is_game_over(watch_game):
break
except (HTTPError, ReadTimeout, RemoteDisconnected, ChunkedEncodingError, ConnectionError, ProtocolError) as e:
throttle.on_exception(e)
if game_id not in (ongoing_game["gameId"] for ongoing_game in li.get_ongoing_games()):
break
raise
event_queue.put({ "type": "local_game_done", "id": game_id })
def loop(li, user_profile):
challenge_queues = [queue.Queue(), queue.Queue()]
control_stream = threading.Thread(target=watch_control_stream, args=[event_queue, li])
control_stream.start()
# We only play one game at a time.
games = Games()
game = None
outgoing_challenges = OutgoingChallenges(li, user_profile["username"], games)
while True:
event = event_queue.get()
if event["type"] != "ping":
if logging_enabled:
log(event)
if event["type"] == "local_game_done":
# We only play one game at a time, so make sure that no games are in progress now.
games.finished(event["id"])
game = None
assert not games.in_progress
# Stop any search/ponder in progress.
chesscoach.bot_search(b"", b"", b"", 0, False, 0, 0, 0, 0, 0)
elif event["type"] == "local_play_move":
if game and game.id == event["id"]:
try:
li.make_move(game.id, event["move"])
except Exception as e:
throttle.on_exception(e)
elif event["type"] == "gameFull":
if game:
# We reconnected to the game stream.
# We only play one game at a time, so make sure that it's the same game.
assert game.id == event["id"]
game.state = event["state"] # Only update state: maintain "searching_moves".
else:
game = Game(event, user_profile["username"], li.baseUrl)
games.started(game)
conversation = Conversation(game, li, challenge_queues)
on_game_state(li, game)
elif event["type"] == "chatLine":
if game:
conversation.react(ChatLine(event), game)
elif event["type"] == "gameState":
if game:
game.state = event
on_game_state(li, game)
elif event["type"] == "challenge":
challenge = Challenge(event["challenge"])
# We may have sent the challenge.
if challenge.challenger_name.lower() != user_profile["username"].lower():
queue_index, decline_reason = challenge.is_supported()
if not decline_reason:
challenge_queues[queue_index].put(challenge)
else:
try:
if logging_enabled:
log("Decline {} for reason '{}'".format(challenge, decline_reason))
li.decline_challenge(challenge.id, reason=decline_reason)
except Exception as e:
throttle.on_exception(e)
elif event["type"] == "challengeDeclined":
challenge = Challenge(event["challenge"])
# We may *not* have sent the challenge.
if challenge.challenger_name.lower() == user_profile["username"].lower():
outgoing_challenges.declined(event)
elif event["type"] == "gameStart":
game_id = event["game"]["id"]
outgoing_challenges.game_start(game_id)
if games.in_progress and game_id not in games.in_progress:
# We accidentally sent or accepted too many challenges and can't handle this game.
try:
li.abort(game_id)
except Exception as e:
throttle.on_exception(e)
else:
games.starting(game_id)
game_stream = threading.Thread(target=watch_game_stream, args=[li, game_id, event_queue, user_profile])
game_stream.start()
# We arrive here after regular, "local" and "ping" events.
# Handle game upkeep.
if game and game.should_abort_now():
if logging_enabled:
log("Aborting {} by lack of activity".format(game.url()))
try:
li.abort(game.id)
except Exception as e:
throttle.on_exception(e)
# Handle control upkeep
while not games.queue and not games.in_progress:
try:
challenge = pop_challenge(challenge_queues)
if not challenge:
break
if logging_enabled:
log("Accept {}".format(challenge))
li.accept_challenge(challenge.id)
games.enqueue(challenge.id)
except (HTTPError, ReadTimeout) as exception:
if isinstance(exception, HTTPError) and exception.response.status_code == 404: # ignore missing challenge
if logging_enabled:
log("Skip missing {}".format(challenge))
throttle.on_exception(exception)
except Exception:
break
# Send commentary to player and spectator chat (use a separate thread).
comment = game and game.try_pop_comment()
if comment:
max_line = 140
lines = textwrap.wrap(comment, width=max_line) if (len(comment) > max_line) else [comment]
def comment_sync():
try:
for line in lines:
li.chat(game.id, "spectator", line)
except Exception as e:
throttle.on_exception(e)
comment_async = threading.Thread(target=comment_sync)
comment_async.start()
# Send and keep track of outgoing challenges when idle.
outgoing_challenges.upkeep()
def pop_challenge(challenge_queues):
for challenge_queue in challenge_queues:
try:
return challenge_queue.get_nowait()
except queue.Empty:
continue
return None
def on_game_state(li, game):
# Update the abort-game timeout upon activity.
game.on_activity()
# No need to search when the game's over.
if is_game_over(game):
return
# We may already be searching this position (draw offers also trigger game state events).
if game.searching_moves == game.state["moves"]:
return
game.searching_moves = game.state["moves"]
# C++ expects bytes
game_id = game.id.encode("utf-8")
fen = game.initial_fen.encode("utf-8")
moves = game.state["moves"].encode("utf-8")
bot_side = (WHITE if game.is_white else BLACK)
can_comment = game.can_comment()
limit_seconds = (10 if game.is_abortable() else 0)
# Start the search/ponder.
log("Preparing to search/ponder")
status, ply, san, comment = chesscoach.bot_search(game_id, fen, moves, bot_side, can_comment, limit_seconds,
game.state["wtime"], game.state["btime"], game.state["winc"], game.state["binc"])
if logging_enabled:
log(f"*** {status.upper()} *** ply: {ply}, SAN: {san}, comment: {comment}")
# Store the comment and process it in the main loop (we only comment every 30 seconds).
if comment:
assert (ply > 0)
ply_before_move = (ply - 1)
move_number = ((ply_before_move // 2) + 1)
suffix = "." if (ply_before_move % 2 == 0) else "..."
game.pending_comment = f'{move_number}{suffix} {san}: "{comment}"'
else:
game.pending_comment = None
def is_game_over(game):
return game.state["status"] != "started"
# --- Logging ---
logging_enabled = True
def log(message):
print(round(time.time(), 2), message)
# --- Global state ---
event_queue = queue.Queue()
throttle = Throttle()
# --- Public API ---
def run():
li = Lichess(os.environ["LICHESS_API_KEY"], "https://lichess.org/")
user_profile = li.get_profile()
if logging_enabled:
log(user_profile)
loop(li, user_profile)
def play_move(game_id, move):
if logging_enabled:
log(f"Posting move, id: {game_id}, move: {move}")
event_queue.put({ "type": "local_play_move", "id": game_id, "move": move })