File size: 47,923 Bytes
36cbcc0 99bedf8 36cbcc0 c21f770 36cbcc0 c21f770 36cbcc0 efae661 99bedf8 efae661 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 efae661 36cbcc0 efae661 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 749a764 99bedf8 de31fe0 99bedf8 de31fe0 99bedf8 a727623 749a764 99bedf8 36cbcc0 99bedf8 de31fe0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 749a764 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 749a764 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 749a764 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 91dcd8f 36cbcc0 99bedf8 749a764 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 36cbcc0 99bedf8 56f75f1 99bedf8 749a764 99bedf8 36cbcc0 99bedf8 36cbcc0 de31fe0 99bedf8 a727623 99bedf8 de31fe0 8358864 7d10a66 3cb9e04 c21f770 99bedf8 8358864 3cb9e04 8358864 99bedf8 260483d 3cb9e04 de31fe0 c21f770 99bedf8 8358864 260483d 8358864 de31fe0 8358864 99bedf8 49d8e1d c42cf8e 99bedf8 3cb9e04 99bedf8 de31fe0 c42cf8e 49d8e1d 3cb9e04 49d8e1d 99bedf8 49d8e1d 749a764 49d8e1d de31fe0 49d8e1d c42cf8e 49d8e1d de31fe0 49d8e1d de31fe0 49d8e1d de31fe0 49d8e1d de31fe0 49d8e1d de31fe0 49d8e1d de31fe0 49d8e1d de31fe0 49d8e1d 3cb9e04 49d8e1d 3cb9e04 de31fe0 49d8e1d de31fe0 49d8e1d 3cb9e04 49d8e1d | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 | """Flow API Client for VideoFX (Veo)"""
import asyncio
import time
import uuid
import random
import base64
from typing import Dict, Any, Optional, List
from curl_cffi.requests import AsyncSession
from ..core.logger import debug_logger
from ..core.config import config
class FlowClient:
"""VideoFX API客户端"""
def __init__(self, proxy_manager, db=None):
self.proxy_manager = proxy_manager
self.db = db # Database instance for captcha config
self.labs_base_url = config.flow_labs_base_url # https://labs.google/fx/api
self.api_base_url = config.flow_api_base_url # https://aisandbox-pa.googleapis.com/v1
self.timeout = config.flow_timeout
# 缓存每个账号的 User-Agent
self._user_agent_cache = {}
# Default "real browser" headers (Android Chrome style) to reduce upstream 4xx/5xx instability.
# These will be applied as defaults (won't override caller-provided headers).
self._default_client_headers = {
"sec-ch-ua-mobile": "?1",
"sec-ch-ua-platform": "\"Android\"",
"sec-fetch-dest": "empty",
"sec-fetch-mode": "cors",
"sec-fetch-site": "cross-site",
"x-browser-channel": "stable",
"x-browser-copyright": "Copyright 2026 Google LLC. All Rights reserved.",
"x-browser-validation": "UujAs0GAwdnCJ9nvrswZ+O+oco0=",
"x-browser-year": "2026",
"x-client-data": "CJS2yQEIpLbJAQipncoBCNj9ygEIlKHLAQiFoM0BGP6lzwE="
}
def _generate_user_agent(self, account_id: str = None) -> str:
"""基于账号ID生成固定的 User-Agent
Args:
account_id: 账号标识(如 email 或 token_id),相同账号返回相同 UA
Returns:
User-Agent 字符串
"""
# 如果没有提供账号ID,生成随机UA
if not account_id:
account_id = f"random_{random.randint(1, 999999)}"
# 如果已缓存,直接返回
if account_id in self._user_agent_cache:
return self._user_agent_cache[account_id]
# 使用账号ID作为随机种子,确保同一账号生成相同的UA
import hashlib
seed = int(hashlib.md5(account_id.encode()).hexdigest()[:8], 16)
rng = random.Random(seed)
# Chrome 版本池
chrome_versions = ["130.0.0.0", "131.0.0.0", "132.0.0.0", "129.0.0.0"]
# Firefox 版本池
firefox_versions = ["133.0", "132.0", "131.0", "134.0"]
# Safari 版本池
safari_versions = ["18.2", "18.1", "18.0", "17.6"]
# Edge 版本池
edge_versions = ["130.0.0.0", "131.0.0.0", "132.0.0.0"]
# 操作系统配置
os_configs = [
# Windows
{
"platform": "Windows NT 10.0; Win64; x64",
"browsers": [
lambda r: f"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/{r.choice(chrome_versions)} Safari/537.36",
lambda r: f"Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:{r.choice(firefox_versions).split('.')[0]}.0) Gecko/20100101 Firefox/{r.choice(firefox_versions)}",
lambda r: f"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/{r.choice(chrome_versions)} Safari/537.36 Edg/{r.choice(edge_versions)}",
]
},
# macOS
{
"platform": "Macintosh; Intel Mac OS X 10_15_7",
"browsers": [
lambda r: f"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/{r.choice(chrome_versions)} Safari/537.36",
lambda r: f"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/{r.choice(safari_versions)} Safari/605.1.15",
lambda r: f"Mozilla/5.0 (Macintosh; Intel Mac OS X 14.{r.randint(0, 7)}; rv:{r.choice(firefox_versions).split('.')[0]}.0) Gecko/20100101 Firefox/{r.choice(firefox_versions)}",
]
},
# Linux
{
"platform": "X11; Linux x86_64",
"browsers": [
lambda r: f"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/{r.choice(chrome_versions)} Safari/537.36",
lambda r: f"Mozilla/5.0 (X11; Linux x86_64; rv:{r.choice(firefox_versions).split('.')[0]}.0) Gecko/20100101 Firefox/{r.choice(firefox_versions)}",
lambda r: f"Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:{r.choice(firefox_versions).split('.')[0]}.0) Gecko/20100101 Firefox/{r.choice(firefox_versions)}",
]
}
]
# 使用固定种子随机选择操作系统和浏览器
os_config = rng.choice(os_configs)
browser_generator = rng.choice(os_config["browsers"])
user_agent = browser_generator(rng)
# 缓存结果
self._user_agent_cache[account_id] = user_agent
return user_agent
async def _make_request(
self,
method: str,
url: str,
headers: Optional[Dict] = None,
json_data: Optional[Dict] = None,
use_st: bool = False,
st_token: Optional[str] = None,
use_at: bool = False,
at_token: Optional[str] = None,
timeout: Optional[int] = None
) -> Dict[str, Any]:
"""统一HTTP请求处理
Args:
method: HTTP方法 (GET/POST)
url: 完整URL
headers: 请求头
json_data: JSON请求体
use_st: 是否使用ST认证 (Cookie方式)
st_token: Session Token
use_at: 是否使用AT认证 (Bearer方式)
at_token: Access Token
timeout: 自定义超时时间(秒),不传则使用默认值
"""
proxy_url = await self.proxy_manager.get_proxy_url()
request_timeout = timeout or self.timeout
if headers is None:
headers = {}
# ST认证 - 使用Cookie
if use_st and st_token:
headers["Cookie"] = f"__Secure-next-auth.session-token={st_token}"
# AT认证 - 使用Bearer
if use_at and at_token:
headers["authorization"] = f"Bearer {at_token}"
# 确定账号标识(优先使用 token 的前16个字符作为标识)
account_id = None
if st_token:
account_id = st_token[:16] # 使用 ST 的前16个字符
elif at_token:
account_id = at_token[:16] # 使用 AT 的前16个字符
# 通用请求头 - 基于账号生成固定的 User-Agent
headers.update({
"Content-Type": "application/json",
"User-Agent": self._generate_user_agent(account_id)
})
# Add default Chromium/Android client headers (do not override explicitly provided values).
for key, value in self._default_client_headers.items():
headers.setdefault(key, value)
# Log request
if config.debug_enabled:
debug_logger.log_request(
method=method,
url=url,
headers=headers,
body=json_data,
proxy=proxy_url
)
start_time = time.time()
try:
async with AsyncSession() as session:
if method.upper() == "GET":
response = await session.get(
url,
headers=headers,
proxy=proxy_url,
timeout=request_timeout,
impersonate="chrome110"
)
else: # POST
response = await session.post(
url,
headers=headers,
json=json_data,
proxy=proxy_url,
timeout=request_timeout,
impersonate="chrome110"
)
duration_ms = (time.time() - start_time) * 1000
# Log response
if config.debug_enabled:
debug_logger.log_response(
status_code=response.status_code,
headers=dict(response.headers),
body=response.text,
duration_ms=duration_ms
)
# 检查HTTP错误
if response.status_code >= 400:
# 解析错误响应
error_reason = f"HTTP Error {response.status_code}"
try:
error_body = response.json()
# 提取 Google API 错误格式中的 reason
if "error" in error_body:
error_info = error_body["error"]
error_message = error_info.get("message", "")
# 从 details 中提取 reason
details = error_info.get("details", [])
for detail in details:
if detail.get("reason"):
error_reason = detail.get("reason")
break
if error_message:
error_reason = f"{error_reason}: {error_message}"
except:
error_reason = f"HTTP Error {response.status_code}: {response.text[:200]}"
# 失败时输出请求体和错误内容到控制台
debug_logger.log_error(f"[API FAILED] URL: {url}")
debug_logger.log_error(f"[API FAILED] Request Body: {json_data}")
debug_logger.log_error(f"[API FAILED] Response: {response.text}")
raise Exception(error_reason)
return response.json()
except Exception as e:
duration_ms = (time.time() - start_time) * 1000
error_msg = str(e)
# 如果不是我们自己抛出的异常,记录日志
if "HTTP Error" not in error_msg and not any(x in error_msg for x in ["PUBLIC_ERROR", "INVALID_ARGUMENT"]):
debug_logger.log_error(f"[API FAILED] URL: {url}")
debug_logger.log_error(f"[API FAILED] Request Body: {json_data}")
debug_logger.log_error(f"[API FAILED] Exception: {error_msg}")
raise Exception(f"Flow API request failed: {error_msg}")
# ========== 认证相关 (使用ST) ==========
async def st_to_at(self, st: str) -> dict:
"""ST转AT
Args:
st: Session Token
Returns:
{
"access_token": "AT",
"expires": "2025-11-15T04:46:04.000Z",
"user": {...}
}
"""
url = f"{self.labs_base_url}/auth/session"
result = await self._make_request(
method="GET",
url=url,
use_st=True,
st_token=st
)
return result
# ========== 项目管理 (使用ST) ==========
async def create_project(self, st: str, title: str) -> str:
"""创建项目,返回project_id
Args:
st: Session Token
title: 项目标题
Returns:
project_id (UUID)
"""
url = f"{self.labs_base_url}/trpc/project.createProject"
json_data = {
"json": {
"projectTitle": title,
"toolName": "PINHOLE"
}
}
result = await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_st=True,
st_token=st
)
# 解析返回的project_id
project_id = result["result"]["data"]["json"]["result"]["projectId"]
return project_id
async def delete_project(self, st: str, project_id: str):
"""删除项目
Args:
st: Session Token
project_id: 项目ID
"""
url = f"{self.labs_base_url}/trpc/project.deleteProject"
json_data = {
"json": {
"projectToDeleteId": project_id
}
}
await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_st=True,
st_token=st
)
# ========== 余额查询 (使用AT) ==========
async def get_credits(self, at: str) -> dict:
"""查询余额
Args:
at: Access Token
Returns:
{
"credits": 920,
"userPaygateTier": "PAYGATE_TIER_ONE"
}
"""
url = f"{self.api_base_url}/credits"
result = await self._make_request(
method="GET",
url=url,
use_at=True,
at_token=at
)
return result
# ========== 图片上传 (使用AT) ==========
def _detect_image_mime_type(self, image_bytes: bytes) -> str:
"""通过文件头 magic bytes 检测图片 MIME 类型
Args:
image_bytes: 图片字节数据
Returns:
MIME 类型字符串,默认 image/jpeg
"""
if len(image_bytes) < 12:
return "image/jpeg"
# WebP: RIFF....WEBP
if image_bytes[:4] == b'RIFF' and image_bytes[8:12] == b'WEBP':
return "image/webp"
# PNG: 89 50 4E 47
if image_bytes[:4] == b'\x89PNG':
return "image/png"
# JPEG: FF D8 FF
if image_bytes[:3] == b'\xff\xd8\xff':
return "image/jpeg"
# GIF: GIF87a 或 GIF89a
if image_bytes[:6] in (b'GIF87a', b'GIF89a'):
return "image/gif"
# BMP: BM
if image_bytes[:2] == b'BM':
return "image/bmp"
# JPEG 2000: 00 00 00 0C 6A 50
if image_bytes[:6] == b'\x00\x00\x00\x0cjP':
return "image/jp2"
return "image/jpeg"
def _convert_to_jpeg(self, image_bytes: bytes) -> bytes:
"""将图片转换为 JPEG 格式
Args:
image_bytes: 原始图片字节数据
Returns:
JPEG 格式的图片字节数据
"""
from io import BytesIO
from PIL import Image
img = Image.open(BytesIO(image_bytes))
# 如果有透明通道,转换为 RGB
if img.mode in ('RGBA', 'LA', 'P'):
img = img.convert('RGB')
output = BytesIO()
img.save(output, format='JPEG', quality=95)
return output.getvalue()
async def upload_image(
self,
at: str,
image_bytes: bytes,
aspect_ratio: str = "IMAGE_ASPECT_RATIO_LANDSCAPE"
) -> str:
"""上传图片,返回mediaGenerationId
Args:
at: Access Token
image_bytes: 图片字节数据
aspect_ratio: 图片或视频宽高比(会自动转换为图片格式)
Returns:
mediaGenerationId (CAM...)
"""
# 转换视频aspect_ratio为图片aspect_ratio
# VIDEO_ASPECT_RATIO_LANDSCAPE -> IMAGE_ASPECT_RATIO_LANDSCAPE
# VIDEO_ASPECT_RATIO_PORTRAIT -> IMAGE_ASPECT_RATIO_PORTRAIT
if aspect_ratio.startswith("VIDEO_"):
aspect_ratio = aspect_ratio.replace("VIDEO_", "IMAGE_")
# 自动检测图片 MIME 类型
mime_type = self._detect_image_mime_type(image_bytes)
# 编码为base64 (去掉前缀)
image_base64 = base64.b64encode(image_bytes).decode('utf-8')
url = f"{self.api_base_url}:uploadUserImage"
json_data = {
"imageInput": {
"rawImageBytes": image_base64,
"mimeType": mime_type,
"isUserUploaded": True,
"aspectRatio": aspect_ratio
},
"clientContext": {
"sessionId": self._generate_session_id(),
"tool": "ASSET_MANAGER"
}
}
result = await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_at=True,
at_token=at
)
# 返回mediaGenerationId
media_id = result["mediaGenerationId"]["mediaGenerationId"]
return media_id
# ========== 图片生成 (使用AT) - 同步返回 ==========
async def generate_image(
self,
at: str,
project_id: str,
prompt: str,
model_name: str,
aspect_ratio: str,
image_inputs: Optional[List[Dict]] = None
) -> dict:
"""生成图片(同步返回)
Args:
at: Access Token
project_id: 项目ID
prompt: 提示词
model_name: GEM_PIX, GEM_PIX_2 或 IMAGEN_3_5
aspect_ratio: 图片宽高比
image_inputs: 参考图片列表(图生图时使用)
Returns:
{
"media": [{
"image": {
"generatedImage": {
"fifeUrl": "图片URL",
...
}
}
}]
}
"""
url = f"{self.api_base_url}/projects/{project_id}/flowMedia:batchGenerateImages"
# 403/reCAPTCHA 重试逻辑 - 最多重试3次
max_retries = 3
last_error = None
for retry_attempt in range(max_retries):
# 每次重试都重新获取 reCAPTCHA token
recaptcha_token, browser_id = await self._get_recaptcha_token(project_id, action="IMAGE_GENERATION")
if not recaptcha_token:
raise Exception("Failed to obtain reCAPTCHA token")
session_id = self._generate_session_id()
# 构建请求 - clientContext 只在外层,requests 内不重复
client_context = {
"recaptchaContext": {
"token": recaptcha_token,
"applicationType": "RECAPTCHA_APPLICATION_TYPE_WEB"
},
"sessionId": session_id,
"projectId": project_id,
"tool": "PINHOLE"
}
request_data = {
"seed": random.randint(1, 99999),
"imageModelName": model_name,
"imageAspectRatio": aspect_ratio,
"prompt": prompt,
"imageInputs": image_inputs or []
}
json_data = {
"clientContext": client_context,
"requests": [request_data]
}
try:
result = await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_at=True,
at_token=at
)
return result
except Exception as e:
error_str = str(e)
last_error = e
retry_reason = self._get_retry_reason(error_str)
if retry_reason and retry_attempt < max_retries - 1:
debug_logger.log_warning(f"[IMAGE] 生成遇到{retry_reason},正在重新获取验证码重试 ({retry_attempt + 2}/{max_retries})...")
await self._notify_browser_captcha_error(browser_id)
await asyncio.sleep(1)
continue
else:
raise e
# 所有重试都失败
raise last_error
async def upsample_image(
self,
at: str,
project_id: str,
media_id: str,
target_resolution: str = "UPSAMPLE_IMAGE_RESOLUTION_4K"
) -> str:
"""放大图片到 2K/4K
Args:
at: Access Token
project_id: 项目ID
media_id: 图片的 mediaId (从 batchGenerateImages 返回的 media[0]["name"])
target_resolution: UPSAMPLE_IMAGE_RESOLUTION_2K 或 UPSAMPLE_IMAGE_RESOLUTION_4K
Returns:
base64 编码的图片数据
"""
url = f"{self.api_base_url}/flow/upsampleImage"
# 获取 reCAPTCHA token - 使用 IMAGE_GENERATION action
recaptcha_token, _ = await self._get_recaptcha_token(project_id, action="IMAGE_GENERATION")
if not recaptcha_token:
raise Exception("Failed to obtain reCAPTCHA token")
session_id = self._generate_session_id()
json_data = {
"mediaId": media_id,
"targetResolution": target_resolution,
"clientContext": {
"recaptchaContext": {
"token": recaptcha_token,
"applicationType": "RECAPTCHA_APPLICATION_TYPE_WEB"
},
"sessionId": session_id,
"projectId": project_id,
"tool": "PINHOLE"
}
}
# 4K/2K 放大使用专用超时,因为返回的 base64 数据量很大
result = await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_at=True,
at_token=at,
timeout=config.upsample_timeout
)
# 返回 base64 编码的图片
return result.get("encodedImage", "")
# ========== 视频生成 (使用AT) - 异步返回 ==========
async def generate_video_text(
self,
at: str,
project_id: str,
prompt: str,
model_key: str,
aspect_ratio: str,
user_paygate_tier: str = "PAYGATE_TIER_ONE"
) -> dict:
"""文生视频,返回task_id
Args:
at: Access Token
project_id: 项目ID
prompt: 提示词
model_key: veo_3_1_t2v_fast 等
aspect_ratio: 视频宽高比
user_paygate_tier: 用户等级
Returns:
{
"operations": [{
"operation": {"name": "task_id"},
"sceneId": "uuid",
"status": "MEDIA_GENERATION_STATUS_PENDING"
}],
"remainingCredits": 900
}
"""
url = f"{self.api_base_url}/video:batchAsyncGenerateVideoText"
# 403/reCAPTCHA 重试逻辑 - 最多重试3次
max_retries = 3
last_error = None
for retry_attempt in range(max_retries):
# 每次重试都重新获取 reCAPTCHA token - 视频使用 VIDEO_GENERATION action
recaptcha_token, browser_id = await self._get_recaptcha_token(project_id, action="VIDEO_GENERATION")
if not recaptcha_token:
raise Exception("Failed to obtain reCAPTCHA token")
session_id = self._generate_session_id()
scene_id = str(uuid.uuid4())
json_data = {
"clientContext": {
"recaptchaContext": {
"token": recaptcha_token,
"applicationType": "RECAPTCHA_APPLICATION_TYPE_WEB"
},
"sessionId": session_id,
"projectId": project_id,
"tool": "PINHOLE",
"userPaygateTier": user_paygate_tier
},
"requests": [{
"aspectRatio": aspect_ratio,
"seed": random.randint(1, 99999),
"textInput": {
"prompt": prompt
},
"videoModelKey": model_key,
"metadata": {
"sceneId": scene_id
}
}]
}
try:
result = await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_at=True,
at_token=at
)
return result
except Exception as e:
error_str = str(e)
last_error = e
retry_reason = self._get_retry_reason(error_str)
if retry_reason and retry_attempt < max_retries - 1:
debug_logger.log_warning(f"[VIDEO T2V] 生成遇到{retry_reason},正在重新获取验证码重试 ({retry_attempt + 2}/{max_retries})...")
await self._notify_browser_captcha_error(browser_id)
await asyncio.sleep(1)
continue
else:
raise e
# 所有重试都失败
raise last_error
async def generate_video_reference_images(
self,
at: str,
project_id: str,
prompt: str,
model_key: str,
aspect_ratio: str,
reference_images: List[Dict],
user_paygate_tier: str = "PAYGATE_TIER_ONE"
) -> dict:
"""图生视频,返回task_id
Args:
at: Access Token
project_id: 项目ID
prompt: 提示词
model_key: veo_3_0_r2v_fast
aspect_ratio: 视频宽高比
reference_images: 参考图片列表 [{"imageUsageType": "IMAGE_USAGE_TYPE_ASSET", "mediaId": "..."}]
user_paygate_tier: 用户等级
Returns:
同 generate_video_text
"""
url = f"{self.api_base_url}/video:batchAsyncGenerateVideoReferenceImages"
# 403/reCAPTCHA 重试逻辑 - 最多重试3次
max_retries = 3
last_error = None
for retry_attempt in range(max_retries):
# 每次重试都重新获取 reCAPTCHA token - 视频使用 VIDEO_GENERATION action
recaptcha_token, browser_id = await self._get_recaptcha_token(project_id, action="VIDEO_GENERATION")
if not recaptcha_token:
raise Exception("Failed to obtain reCAPTCHA token")
session_id = self._generate_session_id()
scene_id = str(uuid.uuid4())
json_data = {
"clientContext": {
"recaptchaContext": {
"token": recaptcha_token,
"applicationType": "RECAPTCHA_APPLICATION_TYPE_WEB"
},
"sessionId": session_id,
"projectId": project_id,
"tool": "PINHOLE",
"userPaygateTier": user_paygate_tier
},
"requests": [{
"aspectRatio": aspect_ratio,
"seed": random.randint(1, 99999),
"textInput": {
"prompt": prompt
},
"videoModelKey": model_key,
"referenceImages": reference_images,
"metadata": {
"sceneId": scene_id
}
}]
}
try:
result = await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_at=True,
at_token=at
)
return result
except Exception as e:
error_str = str(e)
last_error = e
retry_reason = self._get_retry_reason(error_str)
if retry_reason and retry_attempt < max_retries - 1:
debug_logger.log_warning(f"[VIDEO R2V] 生成遇到{retry_reason},正在重新获取验证码重试 ({retry_attempt + 2}/{max_retries})...")
await self._notify_browser_captcha_error(browser_id)
await asyncio.sleep(1)
continue
else:
raise e
# 所有重试都失败
raise last_error
async def generate_video_start_end(
self,
at: str,
project_id: str,
prompt: str,
model_key: str,
aspect_ratio: str,
start_media_id: str,
end_media_id: str,
user_paygate_tier: str = "PAYGATE_TIER_ONE"
) -> dict:
"""收尾帧生成视频,返回task_id
Args:
at: Access Token
project_id: 项目ID
prompt: 提示词
model_key: veo_3_1_i2v_s_fast_fl
aspect_ratio: 视频宽高比
start_media_id: 起始帧mediaId
end_media_id: 结束帧mediaId
user_paygate_tier: 用户等级
Returns:
同 generate_video_text
"""
url = f"{self.api_base_url}/video:batchAsyncGenerateVideoStartAndEndImage"
# 403/reCAPTCHA 重试逻辑 - 最多重试3次
max_retries = 3
last_error = None
for retry_attempt in range(max_retries):
# 每次重试都重新获取 reCAPTCHA token - 视频使用 VIDEO_GENERATION action
recaptcha_token, browser_id = await self._get_recaptcha_token(project_id, action="VIDEO_GENERATION")
if not recaptcha_token:
raise Exception("Failed to obtain reCAPTCHA token")
session_id = self._generate_session_id()
scene_id = str(uuid.uuid4())
json_data = {
"clientContext": {
"recaptchaContext": {
"token": recaptcha_token,
"applicationType": "RECAPTCHA_APPLICATION_TYPE_WEB"
},
"sessionId": session_id,
"projectId": project_id,
"tool": "PINHOLE",
"userPaygateTier": user_paygate_tier
},
"requests": [{
"aspectRatio": aspect_ratio,
"seed": random.randint(1, 99999),
"textInput": {
"prompt": prompt
},
"videoModelKey": model_key,
"startImage": {
"mediaId": start_media_id
},
"endImage": {
"mediaId": end_media_id
},
"metadata": {
"sceneId": scene_id
}
}]
}
try:
result = await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_at=True,
at_token=at
)
return result
except Exception as e:
error_str = str(e)
last_error = e
retry_reason = self._get_retry_reason(error_str)
if retry_reason and retry_attempt < max_retries - 1:
debug_logger.log_warning(f"[VIDEO I2V] 首尾帧生成遇到{retry_reason},正在重新获取验证码重试 ({retry_attempt + 2}/{max_retries})...")
await self._notify_browser_captcha_error(browser_id)
await asyncio.sleep(1)
continue
else:
raise e
# 所有重试都失败
raise last_error
async def generate_video_start_image(
self,
at: str,
project_id: str,
prompt: str,
model_key: str,
aspect_ratio: str,
start_media_id: str,
user_paygate_tier: str = "PAYGATE_TIER_ONE"
) -> dict:
"""仅首帧生成视频,返回task_id
Args:
at: Access Token
project_id: 项目ID
prompt: 提示词
model_key: veo_3_1_i2v_s_fast_fl等
aspect_ratio: 视频宽高比
start_media_id: 起始帧mediaId
user_paygate_tier: 用户等级
Returns:
同 generate_video_text
"""
url = f"{self.api_base_url}/video:batchAsyncGenerateVideoStartImage"
# 403/reCAPTCHA 重试逻辑 - 最多重试3次
max_retries = 3
last_error = None
for retry_attempt in range(max_retries):
# 每次重试都重新获取 reCAPTCHA token - 视频使用 VIDEO_GENERATION action
recaptcha_token, browser_id = await self._get_recaptcha_token(project_id, action="VIDEO_GENERATION")
if not recaptcha_token:
raise Exception("Failed to obtain reCAPTCHA token")
session_id = self._generate_session_id()
scene_id = str(uuid.uuid4())
json_data = {
"clientContext": {
"recaptchaContext": {
"token": recaptcha_token,
"applicationType": "RECAPTCHA_APPLICATION_TYPE_WEB"
},
"sessionId": session_id,
"projectId": project_id,
"tool": "PINHOLE",
"userPaygateTier": user_paygate_tier
},
"requests": [{
"aspectRatio": aspect_ratio,
"seed": random.randint(1, 99999),
"textInput": {
"prompt": prompt
},
"videoModelKey": model_key,
"startImage": {
"mediaId": start_media_id
},
# 注意: 没有endImage字段,只用首帧
"metadata": {
"sceneId": scene_id
}
}]
}
try:
result = await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_at=True,
at_token=at
)
return result
except Exception as e:
error_str = str(e)
last_error = e
retry_reason = self._get_retry_reason(error_str)
if retry_reason and retry_attempt < max_retries - 1:
debug_logger.log_warning(f"[VIDEO I2V] 首帧生成遇到{retry_reason},正在重新获取验证码重试 ({retry_attempt + 2}/{max_retries})...")
await self._notify_browser_captcha_error(browser_id)
await asyncio.sleep(1)
continue
else:
raise e
# 所有重试都失败
raise last_error
# ========== 视频放大 (Video Upsampler) ==========
async def upsample_video(
self,
at: str,
project_id: str,
video_media_id: str,
aspect_ratio: str,
resolution: str,
model_key: str
) -> dict:
"""视频放大到 4K/1080P,返回 task_id
Args:
at: Access Token
project_id: 项目ID
video_media_id: 视频的 mediaId
aspect_ratio: 视频宽高比 VIDEO_ASPECT_RATIO_PORTRAIT/LANDSCAPE
resolution: VIDEO_RESOLUTION_4K 或 VIDEO_RESOLUTION_1080P
model_key: veo_3_1_upsampler_4k 或 veo_3_1_upsampler_1080p
Returns:
同 generate_video_text
"""
url = f"{self.api_base_url}/video:batchAsyncGenerateVideoUpsampleVideo"
# 403/reCAPTCHA 重试逻辑 - 最多重试3次
max_retries = 3
last_error = None
for retry_attempt in range(max_retries):
recaptcha_token, browser_id = await self._get_recaptcha_token(project_id, action="VIDEO_GENERATION")
if not recaptcha_token:
raise Exception("Failed to obtain reCAPTCHA token")
session_id = self._generate_session_id()
scene_id = str(uuid.uuid4())
json_data = {
"requests": [{
"aspectRatio": aspect_ratio,
"resolution": resolution,
"seed": random.randint(1, 99999),
"videoInput": {
"mediaId": video_media_id
},
"videoModelKey": model_key,
"metadata": {
"sceneId": scene_id
}
}],
"clientContext": {
"recaptchaContext": {
"token": recaptcha_token,
"applicationType": "RECAPTCHA_APPLICATION_TYPE_WEB"
},
"sessionId": session_id
}
}
try:
result = await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_at=True,
at_token=at
)
return result
except Exception as e:
error_str = str(e)
last_error = e
retry_reason = self._get_retry_reason(error_str)
if retry_reason and retry_attempt < max_retries - 1:
debug_logger.log_warning(f"[VIDEO UPSAMPLE] 放大遇到{retry_reason},正在重新获取验证码重试 ({retry_attempt + 2}/{max_retries})...")
await self._notify_browser_captcha_error(browser_id)
await asyncio.sleep(1)
continue
else:
raise e
raise last_error
# ========== 任务轮询 (使用AT) ==========
async def check_video_status(self, at: str, operations: List[Dict]) -> dict:
"""查询视频生成状态
Args:
at: Access Token
operations: 操作列表 [{"operation": {"name": "task_id"}, "sceneId": "...", "status": "..."}]
Returns:
{
"operations": [{
"operation": {
"name": "task_id",
"metadata": {...} # 完成时包含视频信息
},
"status": "MEDIA_GENERATION_STATUS_SUCCESSFUL"
}]
}
"""
url = f"{self.api_base_url}/video:batchCheckAsyncVideoGenerationStatus"
json_data = {
"operations": operations
}
result = await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_at=True,
at_token=at
)
return result
# ========== 媒体删除 (使用ST) ==========
async def delete_media(self, st: str, media_names: List[str]):
"""删除媒体
Args:
st: Session Token
media_names: 媒体ID列表
"""
url = f"{self.labs_base_url}/trpc/media.deleteMedia"
json_data = {
"json": {
"names": media_names
}
}
await self._make_request(
method="POST",
url=url,
json_data=json_data,
use_st=True,
st_token=st
)
# ========== 辅助方法 ==========
def _get_retry_reason(self, error_str: str) -> Optional[str]:
"""判断是否需要重试,返回日志提示内容"""
error_lower = error_str.lower()
if "403" in error_lower:
return "403错误"
if "recaptcha evaluation failed" in error_lower:
return "reCAPTCHA 验证失败"
if "recaptcha" in error_lower:
return "reCAPTCHA 错误"
return None
async def _notify_browser_captcha_error(self, browser_id: int = None):
"""通知有头浏览器打码切换指纹(仅当使用 browser 打码方式时)
Args:
browser_id: 要标记为 bad 的浏览器 ID
"""
if config.captcha_method == "browser":
try:
from .browser_captcha import BrowserCaptchaService
service = await BrowserCaptchaService.get_instance(self.db)
await service.report_error(browser_id)
except Exception:
pass
def _generate_session_id(self) -> str:
"""生成sessionId: ;timestamp"""
return f";{int(time.time() * 1000)}"
def _generate_scene_id(self) -> str:
"""生成sceneId: UUID"""
return str(uuid.uuid4())
async def _get_recaptcha_token(self, project_id: str, action: str = "IMAGE_GENERATION") -> tuple[Optional[str], Optional[int]]:
"""获取reCAPTCHA token - 支持多种打码方式
Args:
project_id: 项目ID
action: reCAPTCHA action类型
- IMAGE_GENERATION: 图片生成和2K/4K图片放大 (默认)
- VIDEO_GENERATION: 视频生成和视频放大
Returns:
(token, browser_id) 元组,browser_id 用于失败时调用 report_error
对于非 browser 打码方式,browser_id 为 None
"""
captcha_method = config.captcha_method
# 内置浏览器打码 (nodriver)
if captcha_method == "personal":
try:
from .browser_captcha_personal import BrowserCaptchaService
service = await BrowserCaptchaService.get_instance(self.db)
return await service.get_token(project_id, action), None
except RuntimeError as e:
# 捕获 Docker 环境或依赖缺失的明确错误
error_msg = str(e)
debug_logger.log_error(f"[reCAPTCHA Personal] {error_msg}")
print(f"[reCAPTCHA] ❌ 内置浏览器打码失败: {error_msg}")
return None, None
except ImportError as e:
debug_logger.log_error(f"[reCAPTCHA Personal] 导入失败: {str(e)}")
print(f"[reCAPTCHA] ❌ nodriver 未安装,请运行: pip install nodriver")
return None, None
except Exception as e:
debug_logger.log_error(f"[reCAPTCHA Personal] 错误: {str(e)}")
return None, None
# 有头浏览器打码 (playwright)
elif captcha_method == "browser":
try:
from .browser_captcha import BrowserCaptchaService
service = await BrowserCaptchaService.get_instance(self.db)
return await service.get_token(project_id, action)
except RuntimeError as e:
# 捕获 Docker 环境或依赖缺失的明确错误
error_msg = str(e)
debug_logger.log_error(f"[reCAPTCHA Browser] {error_msg}")
print(f"[reCAPTCHA] ❌ 有头浏览器打码失败: {error_msg}")
return None, None
except ImportError as e:
debug_logger.log_error(f"[reCAPTCHA Browser] 导入失败: {str(e)}")
print(f"[reCAPTCHA] ❌ playwright 未安装,请运行: pip install playwright && python -m playwright install chromium")
return None, None
except Exception as e:
debug_logger.log_error(f"[reCAPTCHA Browser] 错误: {str(e)}")
return None, None
# API打码服务
elif captcha_method in ["yescaptcha", "capmonster", "ezcaptcha", "capsolver"]:
token = await self._get_api_captcha_token(captcha_method, project_id, action)
return token, None
else:
debug_logger.log_info(f"[reCAPTCHA] 未知的打码方式: {captcha_method}")
return None, None
async def _get_api_captcha_token(self, method: str, project_id: str, action: str = "IMAGE_GENERATION") -> Optional[str]:
"""通用API打码服务
Args:
method: 打码服务类型
project_id: 项目ID
action: reCAPTCHA action类型 (IMAGE_GENERATION 或 VIDEO_GENERATION)
"""
# 获取配置
if method == "yescaptcha":
client_key = config.yescaptcha_api_key
base_url = config.yescaptcha_base_url
task_type = "RecaptchaV3TaskProxylessM1"
elif method == "capmonster":
client_key = config.capmonster_api_key
base_url = config.capmonster_base_url
task_type = "RecaptchaV3TaskProxyless"
elif method == "ezcaptcha":
client_key = config.ezcaptcha_api_key
base_url = config.ezcaptcha_base_url
task_type = "ReCaptchaV3TaskProxylessS9"
elif method == "capsolver":
client_key = config.capsolver_api_key
base_url = config.capsolver_base_url
task_type = "ReCaptchaV3EnterpriseTaskProxyLess"
else:
debug_logger.log_error(f"[reCAPTCHA] Unknown API method: {method}")
return None
if not client_key:
debug_logger.log_info(f"[reCAPTCHA] {method} API key not configured, skipping")
return None
website_key = "6LdsFiUsAAAAAIjVDZcuLhaHiDn5nnHVXVRQGeMV"
website_url = f"https://labs.google/fx/tools/flow/project/{project_id}"
page_action = action
try:
async with AsyncSession() as session:
create_url = f"{base_url}/createTask"
create_data = {
"clientKey": client_key,
"task": {
"websiteURL": website_url,
"websiteKey": website_key,
"type": task_type,
"pageAction": page_action
}
}
result = await session.post(create_url, json=create_data, impersonate="chrome110")
result_json = result.json()
task_id = result_json.get('taskId')
debug_logger.log_info(f"[reCAPTCHA {method}] created task_id: {task_id}")
if not task_id:
error_desc = result_json.get('errorDescription', 'Unknown error')
debug_logger.log_error(f"[reCAPTCHA {method}] Failed to create task: {error_desc}")
return None
get_url = f"{base_url}/getTaskResult"
for i in range(40):
get_data = {
"clientKey": client_key,
"taskId": task_id
}
result = await session.post(get_url, json=get_data, impersonate="chrome110")
result_json = result.json()
debug_logger.log_info(f"[reCAPTCHA {method}] polling #{i+1}: {result_json}")
status = result_json.get('status')
if status == 'ready':
solution = result_json.get('solution', {})
response = solution.get('gRecaptchaResponse')
if response:
debug_logger.log_info(f"[reCAPTCHA {method}] Token获取成功")
return response
time.sleep(3)
debug_logger.log_error(f"[reCAPTCHA {method}] Timeout waiting for token")
return None
except Exception as e:
debug_logger.log_error(f"[reCAPTCHA {method}] error: {str(e)}")
return None
|