| 1 | from __future__ import annotations |
| 2 | |
| 3 | import asyncio |
| 4 | import random |
| 5 | from typing import Any, Dict, List, Optional, Tuple |
| 6 | from urllib.parse import urlencode |
| 7 | |
| 8 | import aiohttp |
| 9 | |
| 10 | from auth import MsTokenManager |
| 11 | from utils.cookie_utils import sanitize_cookies |
| 12 | from utils.logger import setup_logger |
| 13 | from utils.xbogus import XBogus |
| 14 | |
| 15 | try: |
| 16 | from utils.abogus import ABogus, BrowserFingerprintGenerator |
| 17 | except Exception: # pragma: no cover - optional dependency |
| 18 | ABogus = None |
| 19 | BrowserFingerprintGenerator = None |
| 20 | |
| 21 | logger = setup_logger("APIClient") |
| 22 | |
| 23 | _USER_AGENT_POOL = [ |
| 24 | ( |
| 25 | "Mozilla/5.0 (Windows NT 10.0; Win64; x64) " |
| 26 | "AppleWebKit/537.36 (KHTML, like Gecko) Chrome/139.0.0.0 Safari/537.36" |
| 27 | ), |
| 28 | ( |
| 29 | "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) " |
| 30 | "AppleWebKit/537.36 (KHTML, like Gecko) Chrome/139.0.0.0 Safari/537.36" |
| 31 | ), |
| 32 | ] |
| 33 | |
| 34 | |
| 35 | class DouyinAPIClient: |
| 36 | BASE_URL = "https://www.douyin.com" |
| 37 | _BROWSER_COOKIE_BLOCKLIST = { |
| 38 | "sessionid", |
| 39 | "sessionid_ss", |
| 40 | "sid_tt", |
| 41 | "sid_guard", |
| 42 | "uid_tt", |
| 43 | "uid_tt_ss", |
| 44 | "passport_auth_status", |
| 45 | "passport_auth_status_ss", |
| 46 | "passport_assist_user", |
| 47 | "passport_auth_mix_state", |
| 48 | "passport_mfa_token", |
| 49 | "login_time", |
| 50 | } |
| 51 | |
| 52 | def __init__(self, cookies: Dict[str, str], proxy: Optional[str] = None): |
| 53 | self.cookies = sanitize_cookies(cookies or {}) |
| 54 | self.proxy = str(proxy or "").strip() |
| 55 | self._session: Optional[aiohttp.ClientSession] = None |
| 56 | self._browser_post_aweme_items: Dict[str, Dict[str, Any]] = {} |
| 57 | self._browser_post_stats: Dict[str, int] = {} |
| 58 | selected_ua = random.choice(_USER_AGENT_POOL) |
| 59 | self.headers = { |
| 60 | "User-Agent": selected_ua, |
| 61 | "Referer": "https://www.douyin.com/?recommend=1", |
| 62 | "Accept": "*/*", |
| 63 | "Accept-Encoding": "gzip, deflate", |
| 64 | "Accept-Language": "zh-CN,zh;q=0.9,en-US;q=0.8,en;q=0.7", |
| 65 | } |
| 66 | self._signer = XBogus(self.headers["User-Agent"]) |
| 67 | self._ms_token_manager = MsTokenManager(user_agent=self.headers["User-Agent"]) |
| 68 | self._ms_token = (self.cookies.get("msToken") or "").strip() |
| 69 | self._abogus_enabled = ABogus is not None and BrowserFingerprintGenerator is not None |
| 70 | |
| 71 | async def __aenter__(self) -> "DouyinAPIClient": |
| 72 | await self._ensure_session() |
| 73 | return self |
| 74 | |
| 75 | async def __aexit__(self, exc_type, exc, tb): |
| 76 | await self.close() |
| 77 | |
| 78 | async def _ensure_session(self): |
| 79 | if self._session is None or self._session.closed: |
| 80 | self._session = aiohttp.ClientSession( |
| 81 | headers=self.headers, |
| 82 | cookies=self.cookies, |
| 83 | timeout=aiohttp.ClientTimeout(total=30), |
| 84 | raise_for_status=False, |
| 85 | ) |
| 86 | |
| 87 | async def close(self): |
| 88 | if self._session and not self._session.closed: |
| 89 | await self._session.close() |
| 90 | |
| 91 | async def get_session(self) -> aiohttp.ClientSession: |
| 92 | await self._ensure_session() |
| 93 | if self._session is None: |
| 94 | raise RuntimeError("Failed to create aiohttp session") |
| 95 | return self._session |
| 96 | |
| 97 | async def _ensure_ms_token(self) -> str: |
| 98 | if self._ms_token: |
| 99 | return self._ms_token |
| 100 | |
| 101 | token = await asyncio.to_thread( |
| 102 | self._ms_token_manager.ensure_ms_token, |
| 103 | self.cookies, |
| 104 | ) |
| 105 | self._ms_token = token.strip() |
| 106 | if self._ms_token: |
| 107 | self.cookies["msToken"] = self._ms_token |
| 108 | if self._session and not self._session.closed: |
| 109 | self._session.cookie_jar.update_cookies({"msToken": self._ms_token}) |
| 110 | return self._ms_token |
| 111 | |
| 112 | async def _default_query(self) -> Dict[str, Any]: |
| 113 | ms_token = await self._ensure_ms_token() |
| 114 | return { |
| 115 | "device_platform": "webapp", |
| 116 | "aid": "6383", |
| 117 | "channel": "channel_pc_web", |
| 118 | "update_version_code": "170400", |
| 119 | "pc_client_type": "1", |
| 120 | "pc_libra_divert": "Windows", |
| 121 | "version_code": "290100", |
| 122 | "version_name": "29.1.0", |
| 123 | "cookie_enabled": "true", |
| 124 | "screen_width": "1536", |
| 125 | "screen_height": "864", |
| 126 | "browser_language": "zh-CN", |
| 127 | "browser_platform": "Win32", |
| 128 | "browser_name": "Chrome", |
| 129 | "browser_version": "139.0.0.0", |
| 130 | "browser_online": "true", |
| 131 | "engine_name": "Blink", |
| 132 | "engine_version": "139.0.0.0", |
| 133 | "os_name": "Windows", |
| 134 | "os_version": "10", |
| 135 | "cpu_core_num": "16", |
| 136 | "device_memory": "8", |
| 137 | "platform": "PC", |
| 138 | "downlink": "10", |
| 139 | "effective_type": "4g", |
| 140 | "round_trip_time": "200", |
| 141 | "support_h265": "1", |
| 142 | "support_dash": "1", |
| 143 | "uifid": "", |
| 144 | "msToken": ms_token, |
| 145 | } |
| 146 | |
| 147 | def sign_url(self, url: str) -> Tuple[str, str]: |
| 148 | signed_url, _xbogus, ua = self._signer.build(url) |
| 149 | return signed_url, ua |
| 150 | |
| 151 | def build_signed_path(self, path: str, params: Dict[str, Any]) -> Tuple[str, str]: |
| 152 | query = urlencode(params) |
| 153 | base_url = f"{self.BASE_URL}{path}" |
| 154 | ab_signed = self._build_abogus_url(base_url, query) |
| 155 | if ab_signed: |
| 156 | return ab_signed |
| 157 | return self.sign_url(f"{base_url}?{query}") |
| 158 | |
| 159 | def _build_abogus_url(self, base_url: str, query: str) -> Optional[Tuple[str, str]]: |
| 160 | if not self._abogus_enabled: |
| 161 | return None |
| 162 | |
| 163 | try: |
| 164 | browser_fp = BrowserFingerprintGenerator.generate_fingerprint("Chrome") |
| 165 | signer = ABogus(fp=browser_fp, user_agent=self.headers["User-Agent"]) |
| 166 | params_with_ab, _ab, ua, _body = signer.generate_abogus(query, "") |
| 167 | return f"{base_url}?{params_with_ab}", ua |
| 168 | except Exception as exc: |
| 169 | logger.warning("Failed to generate a_bogus, fallback to X-Bogus: %s", exc) |
| 170 | return None |
| 171 | |
| 172 | async def _request_json( |
| 173 | self, |
| 174 | path: str, |
| 175 | params: Dict[str, Any], |
| 176 | *, |
| 177 | suppress_error: bool = False, |
| 178 | max_retries: int = 3, |
| 179 | ) -> Dict[str, Any]: |
| 180 | await self._ensure_session() |
| 181 | delays = [1, 2, 5] |
| 182 | last_exc: Optional[Exception] = None |
| 183 | |
| 184 | for attempt in range(max_retries): |
| 185 | signed_url, ua = self.build_signed_path(path, params) |
| 186 | try: |
| 187 | async with self._session.get( |
| 188 | signed_url, |
| 189 | headers={**self.headers, "User-Agent": ua}, |
| 190 | proxy=self.proxy or None, |
| 191 | ) as response: |
| 192 | if response.status == 200: |
| 193 | body = await response.read() |
| 194 | if not body: |
| 195 | # Empty 200 response is a common anti-bot signal |
| 196 | # from Douyin. Retry with a fresh signature. |
| 197 | logger.warning( |
| 198 | "Empty 200 response for %s (attempt %d/%d), " |
| 199 | "likely anti-bot; will retry", |
| 200 | path, |
| 201 | attempt + 1, |
| 202 | max_retries, |
| 203 | ) |
| 204 | last_exc = RuntimeError(f"Empty 200 response for {path} (anti-bot)") |
| 205 | if attempt < max_retries - 1: |
| 206 | delay = delays[min(attempt, len(delays) - 1)] |
| 207 | await asyncio.sleep(delay) |
| 208 | continue |
| 209 | try: |
| 210 | data = await response.json(content_type=None) |
| 211 | except Exception: |
| 212 | import json as _json |
| 213 | |
| 214 | try: |
| 215 | data = _json.loads(body) |
| 216 | except Exception: |
| 217 | logger.warning( |
| 218 | "Non-JSON 200 response for %s, length=%d", |
| 219 | path, |
| 220 | len(body), |
| 221 | ) |
| 222 | return {} |
| 223 | return data if isinstance(data, dict) else {} |
| 224 | if response.status < 500 and response.status != 429: |
| 225 | log_fn = logger.debug if suppress_error else logger.error |
| 226 | log_fn( |
| 227 | "Request failed: path=%s, status=%s", |
| 228 | path, |
| 229 | response.status, |
| 230 | ) |
| 231 | return {} |
| 232 | last_exc = RuntimeError(f"HTTP {response.status} for {path}") |
| 233 | except Exception as exc: |
| 234 | last_exc = exc |
| 235 | |
| 236 | if attempt < max_retries - 1: |
| 237 | delay = delays[min(attempt, len(delays) - 1)] |
| 238 | logger.debug( |
| 239 | "Request retry %d/%d for %s in %ds", |
| 240 | attempt + 1, |
| 241 | max_retries, |
| 242 | path, |
| 243 | delay, |
| 244 | ) |
| 245 | await asyncio.sleep(delay) |
| 246 | |
| 247 | log_fn = logger.debug if suppress_error else logger.error |
| 248 | log_fn("Request failed after %d attempts: path=%s, error=%s", max_retries, path, last_exc) |
| 249 | return {} |
| 250 | |
| 251 | @staticmethod |
| 252 | def _normalize_paged_response( |
| 253 | raw_data: Any, |
| 254 | *, |
| 255 | item_keys: Optional[List[str]] = None, |
| 256 | source: str = "api", |
| 257 | ) -> Dict[str, Any]: |
| 258 | raw = raw_data if isinstance(raw_data, dict) else {} |
| 259 | keys = item_keys or [] |
| 260 | keys = ["items", *keys, "aweme_list", "mix_list", "music_list"] |
| 261 | |
| 262 | items: List[Dict[str, Any]] = [] |
| 263 | for key in keys: |
| 264 | value = raw.get(key) |
| 265 | if isinstance(value, list): |
| 266 | items = value |
| 267 | break |
| 268 | |
| 269 | has_more_value = raw.get("has_more", False) |
| 270 | try: |
| 271 | has_more = bool(int(has_more_value)) |
| 272 | except (TypeError, ValueError): |
| 273 | has_more = bool(has_more_value) |
| 274 | |
| 275 | max_cursor_value = raw.get("max_cursor") |
| 276 | if max_cursor_value is None: |
| 277 | max_cursor_value = raw.get("cursor", 0) |
| 278 | try: |
| 279 | max_cursor = int(max_cursor_value or 0) |
| 280 | except (TypeError, ValueError): |
| 281 | max_cursor = 0 |
| 282 | |
| 283 | status_code_value = raw.get("status_code", 0) |
| 284 | try: |
| 285 | status_code = int(status_code_value or 0) |
| 286 | except (TypeError, ValueError): |
| 287 | status_code = 0 |
| 288 | |
| 289 | risk_flags = { |
| 290 | "login_tip": bool( |
| 291 | ((raw.get("not_login_module") or {}).get("guide_login_tip_exist")) |
| 292 | if isinstance(raw.get("not_login_module"), dict) |
| 293 | else False |
| 294 | ), |
| 295 | "verify_page": bool(raw.get("verify_ticket")), |
| 296 | } |
| 297 | |
| 298 | normalized = { |
| 299 | "items": items, |
| 300 | "aweme_list": items, # 兼容旧调用方 |
| 301 | "has_more": has_more, |
| 302 | "max_cursor": max_cursor, |
| 303 | "status_code": status_code, |
| 304 | "source": source, |
| 305 | "risk_flags": risk_flags, |
| 306 | "raw": raw, |
| 307 | } |
| 308 | for key, value in raw.items(): |
| 309 | if key not in normalized: |
| 310 | normalized[key] = value |
| 311 | return normalized |
| 312 | |
| 313 | async def _build_user_page_params( |
| 314 | self, sec_uid: str, max_cursor: int, count: int |
| 315 | ) -> Dict[str, Any]: |
| 316 | params = await self._default_query() |
| 317 | params.update( |
| 318 | { |
| 319 | "sec_user_id": sec_uid, |
| 320 | "max_cursor": max_cursor, |
| 321 | "count": count, |
| 322 | "locate_query": "false", |
| 323 | } |
| 324 | ) |
| 325 | return params |
| 326 | |
| 327 | # aid=1128 works for videos but filters out image/note content; |
| 328 | # aid=6383 works for notes/gallery but may miss some video content. |
| 329 | _DETAIL_AID_CANDIDATES = ("6383", "1128") |
| 330 | |
| 331 | async def get_video_detail( |
| 332 | self, aweme_id: str, *, suppress_error: bool = False |
| 333 | ) -> Optional[Dict[str, Any]]: |
| 334 | for aid in self._DETAIL_AID_CANDIDATES: |
| 335 | params = await self._default_query() |
| 336 | params.update( |
| 337 | { |
| 338 | "aweme_id": aweme_id, |
| 339 | "aid": aid, |
| 340 | } |
| 341 | ) |
| 342 | |
| 343 | data = await self._request_json( |
| 344 | "/aweme/v1/web/aweme/detail/", |
| 345 | params, |
| 346 | suppress_error=(suppress_error or aid != self._DETAIL_AID_CANDIDATES[-1]), |
| 347 | ) |
| 348 | if not data: |
| 349 | continue |
| 350 | |
| 351 | detail = data.get("aweme_detail") |
| 352 | if detail: |
| 353 | return detail |
| 354 | |
| 355 | # API returned data but aweme_detail is null — check if content was |
| 356 | # filtered (e.g. filter_reason="images_base" for note/gallery). |
| 357 | filter_info = data.get("filter_detail") |
| 358 | if isinstance(filter_info, dict) and filter_info.get("filter_reason"): |
| 359 | logger.info( |
| 360 | "Aweme %s filtered with aid=%s (reason=%s), retrying", |
| 361 | aweme_id, |
| 362 | aid, |
| 363 | filter_info["filter_reason"], |
| 364 | ) |
| 365 | continue |
| 366 | |
| 367 | # aweme_detail is null without a filter reason — no retry needed |
| 368 | break |
| 369 | |
| 370 | return None |
| 371 | |
| 372 | async def get_user_post( |
| 373 | self, sec_uid: str, max_cursor: int = 0, count: int = 18 |
| 374 | ) -> Dict[str, Any]: |
| 375 | params = await self._build_user_page_params(sec_uid, max_cursor, count) |
| 376 | params.update( |
| 377 | { |
| 378 | "show_live_replay_strategy": "1", |
| 379 | "need_time_list": "1", |
| 380 | "time_list_query": "0", |
| 381 | "whale_cut_token": "", |
| 382 | "cut_version": "1", |
| 383 | "publish_video_strategy_type": "2", |
| 384 | } |
| 385 | ) |
| 386 | raw = await self._request_json("/aweme/v1/web/aweme/post/", params) |
| 387 | return self._normalize_paged_response(raw, item_keys=["aweme_list"]) |
| 388 | |
| 389 | async def get_user_like( |
| 390 | self, sec_uid: str, max_cursor: int = 0, count: int = 20 |
| 391 | ) -> Dict[str, Any]: |
| 392 | params = await self._build_user_page_params(sec_uid, max_cursor, count) |
| 393 | raw = await self._request_json("/aweme/v1/web/aweme/favorite/", params) |
| 394 | return self._normalize_paged_response(raw, item_keys=["aweme_list"]) |
| 395 | |
| 396 | async def get_user_mix( |
| 397 | self, sec_uid: str, max_cursor: int = 0, count: int = 20 |
| 398 | ) -> Dict[str, Any]: |
| 399 | params = await self._build_user_page_params(sec_uid, max_cursor, count) |
| 400 | raw = await self._request_json("/aweme/v1/web/mix/list/", params) |
| 401 | return self._normalize_paged_response(raw, item_keys=["mix_list"]) |
| 402 | |
| 403 | async def get_user_music( |
| 404 | self, sec_uid: str, max_cursor: int = 0, count: int = 20 |
| 405 | ) -> Dict[str, Any]: |
| 406 | params = await self._build_user_page_params(sec_uid, max_cursor, count) |
| 407 | raw = await self._request_json("/aweme/v1/web/music/list/", params) |
| 408 | return self._normalize_paged_response(raw, item_keys=["music_list"]) |
| 409 | |
| 410 | async def get_following_page( |
| 411 | self, |
| 412 | sec_uid: str, |
| 413 | *, |
| 414 | max_time: int = 0, |
| 415 | count: int = 20, |
| 416 | ) -> Dict[str, Any]: |
| 417 | """Fetch a single page of the logged-in account's following list. |
| 418 | |
| 419 | Desktop-only: used by ``core/following.FollowingService`` to sync the |
| 420 | "My Following" tab. Douyin's web endpoint paginates via time-based |
| 421 | cursoring: the response contains ``min_time`` which must be passed as |
| 422 | ``max_time`` in the next request to get the next page. The ``count`` |
| 423 | parameter is capped at 20 by the server regardless of what we send. |
| 424 | |
| 425 | Returns a normalized dict with ``items``, ``has_more``, ``min_time``, |
| 426 | ``max_time``, ``status_code``, and ``raw`` (the full response). |
| 427 | """ |
| 428 | params = await self._default_query() |
| 429 | params.update( |
| 430 | { |
| 431 | "user_id": sec_uid, |
| 432 | "sec_user_id": sec_uid, |
| 433 | "offset": 0, |
| 434 | "count": count, |
| 435 | "source_type": "1", |
| 436 | "gps_access": "0", |
| 437 | "address_book_access": "0", |
| 438 | "min_change": "0", |
| 439 | } |
| 440 | ) |
| 441 | if max_time > 0: |
| 442 | params["max_time"] = max_time |
| 443 | raw = await self._request_json("/aweme/v1/web/user/following/list/", params) |
| 444 | normalized = self._normalize_paged_response( |
| 445 | raw, |
| 446 | item_keys=["followings", "follow_list", "user_list"], |
| 447 | ) |
| 448 | # Expose the time-based pagination fields for the sync loop. |
| 449 | normalized["min_time"] = int(raw.get("min_time") or 0) if isinstance(raw, dict) else 0 |
| 450 | normalized["max_time_resp"] = int(raw.get("max_time") or 0) if isinstance(raw, dict) else 0 |
| 451 | return normalized |
| 452 | |
| 453 | async def _build_collect_page_params(self, max_cursor: int, count: int) -> Dict[str, Any]: |
| 454 | params = await self._default_query() |
| 455 | params.update( |
| 456 | { |
| 457 | "cursor": max_cursor, |
| 458 | "count": count, |
| 459 | "version_code": "170400", |
| 460 | "version_name": "17.4.0", |
| 461 | } |
| 462 | ) |
| 463 | return params |
| 464 | |
| 465 | async def get_user_collects( |
| 466 | self, sec_uid: str, max_cursor: int = 0, count: int = 10 |
| 467 | ) -> Dict[str, Any]: |
| 468 | if sec_uid and sec_uid != "self": |
| 469 | logger.warning("Collect folders currently require self sec_uid, got=%s", sec_uid) |
| 470 | return self._normalize_paged_response({}, item_keys=["collects_list"], source="api") |
| 471 | |
| 472 | params = await self._build_collect_page_params(max_cursor, count) |
| 473 | raw = await self._request_json("/aweme/v1/web/collects/list/", params) |
| 474 | return self._normalize_paged_response(raw, item_keys=["collects_list"]) |
| 475 | |
| 476 | async def get_collect_aweme( |
| 477 | self, collects_id: str, max_cursor: int = 0, count: int = 10 |
| 478 | ) -> Dict[str, Any]: |
| 479 | params = await self._build_collect_page_params(max_cursor, count) |
| 480 | params.update({"collects_id": collects_id}) |
| 481 | raw = await self._request_json("/aweme/v1/web/collects/video/list/", params) |
| 482 | return self._normalize_paged_response(raw, item_keys=["aweme_list"]) |
| 483 | |
| 484 | async def get_user_collect_mix( |
| 485 | self, sec_uid: str, max_cursor: int = 0, count: int = 12 |
| 486 | ) -> Dict[str, Any]: |
| 487 | if sec_uid and sec_uid != "self": |
| 488 | logger.warning("Collect mix currently require self sec_uid, got=%s", sec_uid) |
| 489 | return self._normalize_paged_response({}, item_keys=["mix_infos"], source="api") |
| 490 | |
| 491 | params = await self._build_collect_page_params(max_cursor, count) |
| 492 | raw = await self._request_json("/aweme/v1/web/mix/listcollection/", params) |
| 493 | return self._normalize_paged_response(raw, item_keys=["mix_infos"]) |
| 494 | |
| 495 | async def get_user_info(self, sec_uid: str) -> Optional[Dict[str, Any]]: |
| 496 | params = await self._default_query() |
| 497 | params.update({"sec_user_id": sec_uid}) |
| 498 | |
| 499 | data = await self._request_json("/aweme/v1/web/user/profile/other/", params) |
| 500 | if data: |
| 501 | return data.get("user") |
| 502 | return None |
| 503 | |
| 504 | async def get_self_info(self) -> Optional[Dict[str, Any]]: |
| 505 | """Fetch the logged-in user's own profile. |
| 506 | |
| 507 | Uses the ``/aweme/v1/web/user/profile/self/`` endpoint which |
| 508 | identifies the user from the session cookies — no ``sec_uid`` |
| 509 | parameter needed. Returns the ``user`` dict (containing |
| 510 | ``sec_uid``, ``uid``, ``nickname``, etc.) or ``None`` on failure. |
| 511 | |
| 512 | Desktop-only: used by the Following sync to resolve the |
| 513 | logged-in user's ``sec_uid`` before calling |
| 514 | ``get_following_page``. |
| 515 | """ |
| 516 | params = await self._default_query() |
| 517 | data = await self._request_json( |
| 518 | "/aweme/v1/web/user/profile/self/", params |
| 519 | ) |
| 520 | if data: |
| 521 | return data.get("user") |
| 522 | return None |
| 523 | |
| 524 | async def get_mix_detail(self, mix_id: str) -> Optional[Dict[str, Any]]: |
| 525 | params = await self._default_query() |
| 526 | params.update({"mix_id": mix_id}) |
| 527 | data = await self._request_json("/aweme/v1/web/mix/detail/", params) |
| 528 | if not data: |
| 529 | return None |
| 530 | return data.get("mix_info") or data.get("mix_detail") or data |
| 531 | |
| 532 | async def get_mix_aweme(self, mix_id: str, cursor: int = 0, count: int = 20) -> Dict[str, Any]: |
| 533 | params = await self._default_query() |
| 534 | params.update({"mix_id": mix_id, "cursor": cursor, "count": count}) |
| 535 | raw = await self._request_json("/aweme/v1/web/mix/aweme/", params) |
| 536 | return self._normalize_paged_response(raw, item_keys=["aweme_list"]) |
| 537 | |
| 538 | async def get_music_detail(self, music_id: str) -> Optional[Dict[str, Any]]: |
| 539 | params = await self._default_query() |
| 540 | params.update({"music_id": music_id}) |
| 541 | data = await self._request_json("/aweme/v1/web/music/detail/", params) |
| 542 | if not data: |
| 543 | return None |
| 544 | return data.get("music_info") or data.get("music_detail") or data |
| 545 | |
| 546 | async def get_music_aweme( |
| 547 | self, music_id: str, cursor: int = 0, count: int = 20 |
| 548 | ) -> Dict[str, Any]: |
| 549 | params = await self._default_query() |
| 550 | params.update({"music_id": music_id, "cursor": cursor, "count": count}) |
| 551 | raw = await self._request_json("/aweme/v1/web/music/aweme/", params) |
| 552 | return self._normalize_paged_response(raw, item_keys=["aweme_list"]) |
| 553 | |
| 554 | async def get_live_room_info( |
| 555 | self, room_id: str, *, sec_user_id: str = "" |
| 556 | ) -> Optional[Dict[str, Any]]: |
| 557 | """通过房间号（web_rid）拉取直播间信息。 |
| 558 | |
| 559 | 返回包含 room_info + stream_url 的 dict；若房间不在直播中或接口失败返回 None。 |
| 560 | """ |
| 561 | params = await self._default_query() |
| 562 | params.update( |
| 563 | { |
| 564 | "web_rid": room_id, |
| 565 | "room_id_str": room_id, |
| 566 | "enter_source": "", |
| 567 | "is_need_double_stream": "false", |
| 568 | "cookie_enabled": "true", |
| 569 | } |
| 570 | ) |
| 571 | if sec_user_id: |
| 572 | params["sec_user_id"] = sec_user_id |
| 573 | |
| 574 | raw = await self._request_json( |
| 575 | "/webcast/room/web/enter/", |
| 576 | params, |
| 577 | suppress_error=True, |
| 578 | ) |
| 579 | if not raw: |
| 580 | return None |
| 581 | |
| 582 | data_section = raw.get("data") if isinstance(raw.get("data"), dict) else raw |
| 583 | if not isinstance(data_section, dict): |
| 584 | return None |
| 585 | |
| 586 | room_list = data_section.get("data") |
| 587 | room = None |
| 588 | if isinstance(room_list, list) and room_list: |
| 589 | first = room_list[0] |
| 590 | if isinstance(first, dict): |
| 591 | room = first |
| 592 | elif isinstance(data_section.get("room"), dict): |
| 593 | room = data_section.get("room") |
| 594 | elif isinstance(raw.get("room"), dict): |
| 595 | room = raw.get("room") |
| 596 | |
| 597 | if not isinstance(room, dict): |
| 598 | return None |
| 599 | |
| 600 | user = data_section.get("user") if isinstance(data_section, dict) else None |
| 601 | return { |
| 602 | "room": room, |
| 603 | "user": user if isinstance(user, dict) else {}, |
| 604 | "raw": raw, |
| 605 | } |
| 606 | |
| 607 | async def get_hot_search_board(self) -> Dict[str, Any]: |
| 608 | """获取抖音热搜榜。返回归一化 dict，items 为热搜词条列表。""" |
| 609 | params = await self._default_query() |
| 610 | params.update({"detail_list": "1", "source": "6"}) |
| 611 | raw = await self._request_json( |
| 612 | "/aweme/v1/web/hot/search/list/", params, suppress_error=True |
| 613 | ) |
| 614 | # 热榜返回结构中数据在 data.word_list 或 word_list |
| 615 | data_root = raw.get("data") if isinstance(raw.get("data"), dict) else raw |
| 616 | word_list = data_root.get("word_list") if isinstance(data_root, dict) else None |
| 617 | status_code = int(raw.get("status_code") or 0) |
| 618 | items = word_list if isinstance(word_list, list) else [] |
| 619 | # 响应为空 + 非正常状态码时显式告警，方便排查 cookie 失效/签名失败 |
| 620 | if not items and (status_code or not raw): |
| 621 | logger.warning( |
| 622 | "Hot search board returned no items (status_code=%s). " |
| 623 | "Check cookies / signature; Douyin may be rejecting the request.", |
| 624 | status_code, |
| 625 | ) |
| 626 | return { |
| 627 | "items": items, |
| 628 | "has_more": False, |
| 629 | "max_cursor": 0, |
| 630 | "status_code": status_code, |
| 631 | "raw": raw, |
| 632 | } |
| 633 | |
| 634 | async def search_aweme( |
| 635 | self, |
| 636 | keyword: str, |
| 637 | *, |
| 638 | offset: int = 0, |
| 639 | count: int = 10, |
| 640 | sort_type: int = 0, |
| 641 | publish_time: int = 0, |
| 642 | ) -> Dict[str, Any]: |
| 643 | """搜索作品。 |
| 644 | |
| 645 | Args: |
| 646 | sort_type: 0 综合 / 1 最多点赞 / 2 最新发布 |
| 647 | publish_time: 0 不限 / 1 一天内 / 7 一周内 / 182 半年内 |
| 648 | """ |
| 649 | params = await self._default_query() |
| 650 | params.update( |
| 651 | { |
| 652 | "keyword": keyword, |
| 653 | "search_channel": "aweme_video_web", |
| 654 | "sort_type": sort_type, |
| 655 | "publish_time": publish_time, |
| 656 | "search_source": "normal_search", |
| 657 | "query_correct_type": "1", |
| 658 | "is_filter_search": 1 if (sort_type or publish_time) else 0, |
| 659 | "offset": offset, |
| 660 | "count": count, |
| 661 | } |
| 662 | ) |
| 663 | raw = await self._request_json( |
| 664 | "/aweme/v1/web/general/search/single/", params, suppress_error=True |
| 665 | ) |
| 666 | # 搜索结果每条在 data[].aweme_info；需要拍平 |
| 667 | data_list = raw.get("data") if isinstance(raw.get("data"), list) else [] |
| 668 | items: List[Dict[str, Any]] = [] |
| 669 | for entry in data_list: |
| 670 | if not isinstance(entry, dict): |
| 671 | continue |
| 672 | aweme_info = entry.get("aweme_info") |
| 673 | if isinstance(aweme_info, dict): |
| 674 | items.append(aweme_info) |
| 675 | |
| 676 | has_more_value = raw.get("has_more", 0) |
| 677 | try: |
| 678 | has_more = bool(int(has_more_value)) |
| 679 | except (TypeError, ValueError): |
| 680 | has_more = bool(has_more_value) |
| 681 | |
| 682 | cursor_value = raw.get("cursor") or raw.get("offset") or 0 |
| 683 | try: |
| 684 | next_offset = int(cursor_value) |
| 685 | except (TypeError, ValueError): |
| 686 | next_offset = 0 |
| 687 | |
| 688 | status_code = int(raw.get("status_code") or 0) |
| 689 | if not items and (status_code or not raw): |
| 690 | logger.warning( |
| 691 | "Search returned no items for keyword=%r (status_code=%s, offset=%s). " |
| 692 | "Possible causes: cookies expired, signature rejected, or query blocked.", |
| 693 | keyword, |
| 694 | status_code, |
| 695 | offset, |
| 696 | ) |
| 697 | |
| 698 | return { |
| 699 | "items": items, |
| 700 | "has_more": has_more, |
| 701 | "max_cursor": next_offset, |
| 702 | "status_code": status_code, |
| 703 | "raw": raw, |
| 704 | } |
| 705 | |
| 706 | async def get_aweme_comments( |
| 707 | self, |
| 708 | aweme_id: str, |
| 709 | *, |
| 710 | cursor: int = 0, |
| 711 | count: int = 20, |
| 712 | include_replies: bool = False, |
| 713 | ) -> Dict[str, Any]: |
| 714 | """获取作品评论列表（一页）。 |
| 715 | |
| 716 | Args: |
| 717 | aweme_id: 作品 ID |
| 718 | cursor: 分页游标（首次传 0） |
| 719 | count: 每页数量（抖音上限一般为 20） |
| 720 | include_replies: 是否拉取每条评论的二级回复（额外请求） |
| 721 | Returns: |
| 722 | 归一化后的分页响应 dict，items 为评论列表。 |
| 723 | """ |
| 724 | params = await self._default_query() |
| 725 | params.update( |
| 726 | { |
| 727 | "aweme_id": aweme_id, |
| 728 | "cursor": cursor, |
| 729 | "count": count, |
| 730 | "item_type": "0", |
| 731 | "insert_ids": "", |
| 732 | "whale_cut_token": "", |
| 733 | "cut_version": "1", |
| 734 | "rcFT": "", |
| 735 | } |
| 736 | ) |
| 737 | raw = await self._request_json("/aweme/v1/web/comment/list/", params) |
| 738 | normalized = self._normalize_paged_response(raw, item_keys=["comments"]) |
| 739 | |
| 740 | if include_replies: |
| 741 | comments = normalized.get("items") or [] |
| 742 | for comment in comments: |
| 743 | if not isinstance(comment, dict): |
| 744 | continue |
| 745 | comment_id = comment.get("cid") or comment.get("comment_id") |
| 746 | if not comment_id or int(comment.get("reply_comment_total") or 0) <= 0: |
| 747 | continue |
| 748 | try: |
| 749 | reply_page = await self.get_aweme_comment_replies( |
| 750 | aweme_id=aweme_id, comment_id=str(comment_id), count=count |
| 751 | ) |
| 752 | comment["_replies"] = reply_page.get("items") or [] |
| 753 | except Exception as exc: # noqa: BLE001 |
| 754 | logger.debug("Fetch reply for comment %s failed: %s", comment_id, exc) |
| 755 | return normalized |
| 756 | |
| 757 | async def get_aweme_comment_replies( |
| 758 | self, |
| 759 | *, |
| 760 | aweme_id: str, |
| 761 | comment_id: str, |
| 762 | cursor: int = 0, |
| 763 | count: int = 20, |
| 764 | ) -> Dict[str, Any]: |
| 765 | """获取某条评论的二级回复列表。""" |
| 766 | params = await self._default_query() |
| 767 | params.update( |
| 768 | { |
| 769 | "item_id": aweme_id, |
| 770 | "comment_id": comment_id, |
| 771 | "cursor": cursor, |
| 772 | "count": count, |
| 773 | } |
| 774 | ) |
| 775 | raw = await self._request_json("/aweme/v1/web/comment/list/reply/", params) |
| 776 | return self._normalize_paged_response(raw, item_keys=["comments"]) |
| 777 | |
| 778 | async def resolve_short_url( |
| 779 | self, short_url: str, *, timeout_seconds: float = 10.0 |
| 780 | ) -> Optional[str]: |
| 781 | """跟随短链 302，返回最终 URL。失败时返回 None。 |
| 782 | |
| 783 | 单独设置较短超时（默认 10s），避免被目标站挂死后拖慢整轮下载。 |
| 784 | HTTP 状态码 ≥ 400 时视为解析失败，返回 None 以避免把错误页 URL |
| 785 | 继续喂给下游 parser，从而在下游触发更隐晦的 "Unsupported URL" 噪声。 |
| 786 | """ |
| 787 | try: |
| 788 | await self._ensure_session() |
| 789 | async with self._session.get( |
| 790 | short_url, |
| 791 | allow_redirects=True, |
| 792 | timeout=aiohttp.ClientTimeout(total=timeout_seconds), |
| 793 | proxy=self.proxy or None, |
| 794 | ) as response: |
| 795 | final_url = str(response.url) |
| 796 | if response.status >= 400: |
| 797 | logger.warning( |
| 798 | "Short URL resolved with HTTP %s (treated as failure): %s -> %s", |
| 799 | response.status, |
| 800 | short_url, |
| 801 | final_url, |
| 802 | ) |
| 803 | return None |
| 804 | return final_url |
| 805 | except asyncio.TimeoutError: |
| 806 | logger.error( |
| 807 | "Timeout resolving short URL after %.1fs: %s", |
| 808 | timeout_seconds, |
| 809 | short_url, |
| 810 | ) |
| 811 | return None |
| 812 | except Exception as e: |
| 813 | logger.error("Failed to resolve short URL: %s, error: %s", short_url, e) |
| 814 | return None |
| 815 | |
| 816 | async def collect_user_post_ids_via_browser( |
| 817 | self, |
| 818 | sec_uid: str, |
| 819 | *, |
| 820 | expected_count: int = 0, |
| 821 | headless: bool = False, |
| 822 | max_scrolls: int = 240, |
| 823 | idle_rounds: int = 8, |
| 824 | wait_timeout_seconds: int = 600, |
| 825 | ) -> List[str]: |
| 826 | try: |
| 827 | from playwright.async_api import async_playwright |
| 828 | except Exception as exc: |
| 829 | logger.warning("Playwright not available, browser fallback disabled: %s", exc) |
| 830 | return [] |
| 831 | |
| 832 | target_url = f"{self.BASE_URL}/user/{sec_uid}" |
| 833 | timeout_ms = max(30, int(wait_timeout_seconds)) * 1000 |
| 834 | ids: List[str] = [] |
| 835 | seen: set[str] = set() |
| 836 | post_api_ids: List[str] = [] |
| 837 | post_api_seen: set[str] = set() |
| 838 | post_api_aweme_items: Dict[str, Dict[str, Any]] = {} |
| 839 | post_api_page_hits = 0 |
| 840 | self._browser_post_aweme_items = {} |
| 841 | self._browser_post_stats = {} |
| 842 | |
| 843 | def _merge(new_ids: List[str]): |
| 844 | for aweme_id in new_ids: |
| 845 | if aweme_id and aweme_id not in seen: |
| 846 | seen.add(aweme_id) |
| 847 | ids.append(aweme_id) |
| 848 | |
| 849 | logger.warning( |
| 850 | "API翻页受限，启动浏览器兜底采集（可在弹出页面手动通过验证码/登录）：%s", |
| 851 | target_url, |
| 852 | ) |
| 853 | |
| 854 | async with async_playwright() as playwright: |
| 855 | browser = await playwright.chromium.launch( |
| 856 | headless=headless, |
| 857 | args=[ |
| 858 | "--disable-blink-features=AutomationControlled", |
| 859 | "--disable-dev-shm-usage", |
| 860 | "--no-sandbox", |
| 861 | ], |
| 862 | ) |
| 863 | context = await browser.new_context( |
| 864 | user_agent=self.headers.get("User-Agent", ""), |
| 865 | locale="zh-CN", |
| 866 | viewport={"width": 1600, "height": 900}, |
| 867 | ) |
| 868 | cookies = self._browser_cookie_payload() |
| 869 | if cookies: |
| 870 | await context.add_cookies(cookies) |
| 871 | |
| 872 | page = await context.new_page() |
| 873 | pending_response_tasks: List[asyncio.Task] = [] |
| 874 | |
| 875 | async def _handle_response(response): |
| 876 | nonlocal post_api_page_hits |
| 877 | url = response.url or "" |
| 878 | if "/aweme/v1/web/aweme/post/" not in url: |
| 879 | return |
| 880 | try: |
| 881 | data = await response.json() |
| 882 | except Exception: |
| 883 | return |
| 884 | aweme_items = data.get("aweme_list") if isinstance(data, dict) else None |
| 885 | if isinstance(aweme_items, list): |
| 886 | post_api_page_hits += 1 |
| 887 | extracted: List[str] = [] |
| 888 | for item in aweme_items: |
| 889 | if not isinstance(item, dict): |
| 890 | continue |
| 891 | aweme_id = item.get("aweme_id") |
| 892 | if not aweme_id: |
| 893 | continue |
| 894 | aweme_id_str = str(aweme_id) |
| 895 | extracted.append(aweme_id_str) |
| 896 | if aweme_id_str not in post_api_aweme_items: |
| 897 | post_api_aweme_items[aweme_id_str] = item |
| 898 | _merge(extracted) |
| 899 | for aweme_id in extracted: |
| 900 | if aweme_id not in post_api_seen: |
| 901 | post_api_seen.add(aweme_id) |
| 902 | post_api_ids.append(aweme_id) |
| 903 | |
| 904 | def _on_response(response): |
| 905 | pending_response_tasks.append(asyncio.create_task(_handle_response(response))) |
| 906 | |
| 907 | page.on("response", _on_response) |
| 908 | |
| 909 | try: |
| 910 | try: |
| 911 | await page.goto(target_url, wait_until="domcontentloaded", timeout=timeout_ms) |
| 912 | except Exception as exc: |
| 913 | logger.warning( |
| 914 | "Browser goto timeout or error, continue with current page state: %s", |
| 915 | exc, |
| 916 | ) |
| 917 | |
| 918 | title = "" |
| 919 | try: |
| 920 | title = await page.title() |
| 921 | except Exception: |
| 922 | pass |
| 923 | if "验证码" in title: |
| 924 | if headless: |
| 925 | logger.warning( |
| 926 | "检测到验证码页面且当前为 headless 模式，无法人工验证。" |
| 927 | "请将 browser_fallback.headless 设为 false。" |
| 928 | ) |
| 929 | return [] |
| 930 | logger.warning("检测到验证码页面，请在浏览器中完成验证，程序会自动继续采集。") |
| 931 | await self._wait_for_manual_verification( |
| 932 | page, wait_timeout_seconds=wait_timeout_seconds |
| 933 | ) |
| 934 | if not page.is_closed(): |
| 935 | try: |
| 936 | await page.goto( |
| 937 | target_url, |
| 938 | wait_until="domcontentloaded", |
| 939 | timeout=timeout_ms, |
| 940 | ) |
| 941 | except Exception as exc: |
| 942 | logger.warning("Reload user page after verification failed: %s", exc) |
| 943 | |
| 944 | try: |
| 945 | warmup_seconds = min(20, max(3, int(wait_timeout_seconds))) |
| 946 | for _ in range(warmup_seconds): |
| 947 | if page.is_closed(): |
| 948 | logger.warning("Browser page closed during warmup") |
| 949 | break |
| 950 | _merge(await self._extract_aweme_ids_from_page(page)) |
| 951 | if ids: |
| 952 | break |
| 953 | await page.wait_for_timeout(1000) |
| 954 | |
| 955 | stable_rounds = 0 |
| 956 | max_scroll_rounds = max(1, int(max_scrolls)) |
| 957 | idle_stop_rounds = max(1, int(idle_rounds)) |
| 958 | |
| 959 | for _ in range(max_scroll_rounds): |
| 960 | if page.is_closed(): |
| 961 | logger.warning("Browser page closed during scrolling") |
| 962 | break |
| 963 | await page.mouse.wheel(0, 3800) |
| 964 | await page.wait_for_timeout(1200) |
| 965 | |
| 966 | before = len(ids) |
| 967 | _merge(await self._extract_aweme_ids_from_page(page)) |
| 968 | if len(ids) == before: |
| 969 | stable_rounds += 1 |
| 970 | else: |
| 971 | stable_rounds = 0 |
| 972 | |
| 973 | if expected_count > 0 and len(ids) >= expected_count: |
| 974 | break |
| 975 | if expected_count <= 0 and stable_rounds >= idle_stop_rounds: |
| 976 | break |
| 977 | except Exception as exc: |
| 978 | logger.warning( |
| 979 | "Browser collection interrupted, use collected ids so far: %s", |
| 980 | exc, |
| 981 | ) |
| 982 | finally: |
| 983 | if pending_response_tasks: |
| 984 | await asyncio.gather(*pending_response_tasks, return_exceptions=True) |
| 985 | try: |
| 986 | browser_cookies = await context.cookies(self.BASE_URL) |
| 987 | self._sync_browser_cookies(browser_cookies) |
| 988 | except Exception as exc: |
| 989 | logger.debug("Sync browser cookies skipped: %s", exc) |
| 990 | await context.close() |
| 991 | await browser.close() |
| 992 | |
| 993 | selected_ids: List[str] = [] |
| 994 | selected_seen: set[str] = set() |
| 995 | for aweme_id in post_api_ids + ids: |
| 996 | if aweme_id and aweme_id not in selected_seen: |
| 997 | selected_seen.add(aweme_id) |
| 998 | selected_ids.append(aweme_id) |
| 999 | self._browser_post_aweme_items = post_api_aweme_items |
| 1000 | self._browser_post_stats = { |
| 1001 | "merged_ids": len(ids), |
| 1002 | "post_api_ids": len(post_api_ids), |
| 1003 | "selected_ids": len(selected_ids), |
| 1004 | "post_items": len(post_api_aweme_items), |
| 1005 | "post_pages": post_api_page_hits, |
| 1006 | } |
| 1007 | logger.warning( |
| 1008 | "浏览器兜底采集 aweme_id: merged=%s, from_post_api=%s, selected=%s, post_items=%s", |
| 1009 | len(ids), |
| 1010 | len(post_api_ids), |
| 1011 | len(selected_ids), |
| 1012 | len(post_api_aweme_items), |
| 1013 | ) |
| 1014 | return selected_ids |
| 1015 | |
| 1016 | def pop_browser_post_aweme_items(self) -> Dict[str, Dict[str, Any]]: |
| 1017 | items = self._browser_post_aweme_items |
| 1018 | self._browser_post_aweme_items = {} |
| 1019 | return items |
| 1020 | |
| 1021 | def pop_browser_post_stats(self) -> Dict[str, int]: |
| 1022 | stats = self._browser_post_stats |
| 1023 | self._browser_post_stats = {} |
| 1024 | return stats |
| 1025 | |
| 1026 | def _browser_cookie_payload(self) -> List[Dict[str, str]]: |
| 1027 | payload: List[Dict[str, str]] = [] |
| 1028 | for name, value in self.cookies.items(): |
| 1029 | if not name: |
| 1030 | continue |
| 1031 | if name in self._BROWSER_COOKIE_BLOCKLIST: |
| 1032 | continue |
| 1033 | payload.append( |
| 1034 | { |
| 1035 | "name": str(name), |
| 1036 | "value": str(value or ""), |
| 1037 | "url": f"{self.BASE_URL}/", |
| 1038 | } |
| 1039 | ) |
| 1040 | return payload |
| 1041 | |
| 1042 | async def _extract_aweme_ids_from_page(self, page) -> List[str]: |
| 1043 | script = """ |
| 1044 | () => { |
| 1045 | const result = []; |
| 1046 | const seen = new Set(); |
| 1047 | const push = (id) => { |
| 1048 | if (!id || seen.has(id)) return; |
| 1049 | seen.add(id); |
| 1050 | result.push(id); |
| 1051 | }; |
| 1052 | |
| 1053 | const collectFrom = (text, pattern) => { |
| 1054 | if (!text) return; |
| 1055 | let match; |
| 1056 | while ((match = pattern.exec(text)) !== null) { |
| 1057 | push(match[1]); |
| 1058 | } |
| 1059 | }; |
| 1060 | |
| 1061 | const links = document.querySelectorAll("a[href]"); |
| 1062 | for (const node of links) { |
| 1063 | const href = node.getAttribute("href") || ""; |
| 1064 | collectFrom(href, /\\/video\\/(\\d{15,20})/g); |
| 1065 | collectFrom(href, /\\/note\\/(\\d{15,20})/g); |
| 1066 | } |
| 1067 | |
| 1068 | const html = document.documentElement ? document.documentElement.innerHTML : ""; |
| 1069 | collectFrom(html, /"aweme_id":"(\\d{15,20})"/g); |
| 1070 | collectFrom(html, /"group_id":"(\\d{15,20})"/g); |
| 1071 | |
| 1072 | return result; |
| 1073 | } |
| 1074 | """ |
| 1075 | try: |
| 1076 | data = await page.evaluate(script) |
| 1077 | if isinstance(data, list): |
| 1078 | return [str(x) for x in data if x] |
| 1079 | except Exception as exc: |
| 1080 | logger.debug("Extract aweme_id from page failed: %s", exc) |
| 1081 | return [] |
| 1082 | |
| 1083 | async def _wait_for_manual_verification(self, page, *, wait_timeout_seconds: int) -> None: |
| 1084 | deadline = asyncio.get_running_loop().time() + max(30, int(wait_timeout_seconds)) |
| 1085 | while asyncio.get_running_loop().time() < deadline: |
| 1086 | if page.is_closed(): |
| 1087 | logger.warning("Browser page closed while waiting manual verification") |
| 1088 | return |
| 1089 | title = "" |
| 1090 | try: |
| 1091 | title = await page.title() |
| 1092 | except Exception: |
| 1093 | pass |
| 1094 | if "验证码" not in title: |
| 1095 | logger.warning("验证码页面已退出，继续采集。") |
| 1096 | return |
| 1097 | await page.wait_for_timeout(1000) |
| 1098 | |
| 1099 | logger.warning("等待手动验证超时（%ss），继续按当前页面状态采集。", wait_timeout_seconds) |
| 1100 | |
| 1101 | def _sync_browser_cookies(self, browser_cookies: List[Dict[str, Any]]) -> None: |
| 1102 | merged: Dict[str, str] = {} |
| 1103 | for cookie in browser_cookies or []: |
| 1104 | if not isinstance(cookie, dict): |
| 1105 | continue |
| 1106 | name = str(cookie.get("name") or "").strip() |
| 1107 | value = str(cookie.get("value") or "").strip() |
| 1108 | domain = str(cookie.get("domain") or "") |
| 1109 | if not name or not value: |
| 1110 | continue |
| 1111 | if "douyin.com" not in domain: |
| 1112 | continue |
| 1113 | merged[name] = value |
| 1114 | |
| 1115 | if not merged: |
| 1116 | return |
| 1117 | |
| 1118 | self.cookies.update(merged) |
| 1119 | if self._session and not self._session.closed: |
| 1120 | self._session.cookie_jar.update_cookies(merged) |
| 1121 | logger.warning("Synced %s browser cookie(s) back to API client", len(merged)) |
| 1122 |