243 lines
8.4 KiB
Python
243 lines
8.4 KiB
Python
"""
|
|
腾讯天御点选验证码 (Tencent Turing Captcha) 样本纯协议采集器
|
|
通过模拟前端两步 HTTP 接口直接拉取高清原图：
|
|
1. GET /cap_union_prehandle 获取初始会话 sess
|
|
2. POST /cap_union_new_getsig 获取题目配置与背景/提示图 URL
|
|
3. 并行下载 672x480 背景大图与 170x50 提示图标图
|
|
"""
|
|
|
|
import argparse
|
|
import asyncio
|
|
import json
|
|
import logging
|
|
import os
|
|
import random
|
|
import re
|
|
import time
|
|
from pathlib import Path
|
|
from typing import Optional, Tuple
|
|
|
|
import httpx
|
|
|
|
logging.basicConfig(
|
|
level=logging.INFO,
|
|
format="%(asctime)s [%(levelname)s] %(message)s",
|
|
datefmt="%H:%M:%S",
|
|
)
|
|
logger = logging.getLogger("tcaptcha_collector")
|
|
|
|
|
|
class TencentCaptchaCollector:
|
|
BASE_URL = "https://turing.captcha.qcloud.com"
|
|
DEFAULT_AID = "2039519451"
|
|
|
|
def __init__(self, aid: str = DEFAULT_AID, output_dir: str = "tcaptcha/data/raw", timeout: float = 10.0):
|
|
self.aid = aid
|
|
self.output_dir = Path(output_dir)
|
|
self.output_dir.mkdir(parents=True, exist_ok=True)
|
|
self.index_file = self.output_dir / "index.jsonl"
|
|
self.timeout = timeout
|
|
|
|
self.client = httpx.Client(
|
|
base_url=self.BASE_URL,
|
|
trust_env=False,
|
|
headers={
|
|
"User-Agent": (
|
|
"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) "
|
|
"AppleWebKit/537.36 (KHTML, like Gecko) Chrome/152.0.0.0 Safari/537.36"
|
|
),
|
|
"Referer": "https://turing.captcha.qcloud.com/",
|
|
"Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8",
|
|
},
|
|
timeout=self.timeout,
|
|
follow_redirects=True,
|
|
)
|
|
|
|
def _fetch_session(self) -> Tuple[Optional[str], Optional[str]]:
|
|
"""第一步：预检请求获取 sess"""
|
|
cb_name = f"_aq_{random.randint(100000, 999999)}"
|
|
tkid = str(random.randint(100000000, 999999999))
|
|
ua_b64 = "TW96aWxsYS81LjAgKE1hY2ludG9zaDsgSW50ZWwgTWFjIE9TIFggMTBfMTVfNykgQXBwbGVXZWJLaXQvNTM3LjM2IChLSFRNTCwgbGlrZSBHZWNrbykgQ2hyb21lLzE1Mi4wLjAuMCBTYWZhcmkvNTM3LjM2"
|
|
|
|
params = {
|
|
"aid": self.aid,
|
|
"protocol": "https",
|
|
"accver": "1",
|
|
"showtype": "popup",
|
|
"ua": ua_b64,
|
|
"noheader": "1",
|
|
"fb": "1",
|
|
"grayscale": "1",
|
|
"clientype": "2",
|
|
"lang": "zh-cn",
|
|
"js": "/tcaptcha-frame.ba387dd1.js",
|
|
"wb": "2",
|
|
"tkid": tkid,
|
|
"subsid": "1",
|
|
"callback": cb_name,
|
|
}
|
|
|
|
resp = self.client.get("/cap_union_prehandle", params=params)
|
|
if resp.status_code != 200:
|
|
logger.error("prehandle failed with status %d", resp.status_code)
|
|
return None, None
|
|
|
|
match = re.search(rf"{cb_name}\((.*)\)", resp.text)
|
|
if not match:
|
|
logger.error("Failed to parse prehandle response: %s", resp.text[:100])
|
|
return None, None
|
|
|
|
try:
|
|
data = json.loads(match.group(1))
|
|
sess = data.get("sess")
|
|
return sess, tkid
|
|
except Exception as e:
|
|
logger.error("Failed to decode JSON from prehandle: %s", e)
|
|
return None, None
|
|
|
|
def _fetch_sig_and_images(self, sess: str, tkid: str) -> Optional[dict]:
|
|
"""第二步：请求题目信息并下载图片对"""
|
|
resp = self.client.post(
|
|
"/cap_union_new_getsig",
|
|
data={"sess": sess, "tkid": tkid},
|
|
headers={"Content-Type": "application/x-www-form-urlencoded"},
|
|
)
|
|
|
|
if resp.status_code != 200:
|
|
logger.error("getsig failed with status %d", resp.status_code)
|
|
return None
|
|
|
|
try:
|
|
res_json = resp.json()
|
|
except Exception as e:
|
|
logger.error("Failed to decode getsig JSON: %s", e)
|
|
return None
|
|
|
|
if str(res_json.get("ret")) != "0":
|
|
logger.warning("getsig returned non-zero ret: %s", res_json)
|
|
return None
|
|
|
|
cfg = res_json.get("data", {})
|
|
instruction = cfg.get("instruction", "请依次点击：")
|
|
bg_cfg = cfg.get("bg_elem_cfg", {})
|
|
img_url_bg = bg_cfg.get("img_url", "")
|
|
|
|
if not img_url_bg:
|
|
logger.warning("No img_url found in getsig response")
|
|
return None
|
|
|
|
img_url_prompt = img_url_bg.replace("img_index=1", "img_index=0")
|
|
|
|
# 下载背景大图 (672x480) 与 提示图 (170x50)
|
|
bg_resp = self.client.get(img_url_bg)
|
|
prompt_resp = self.client.get(img_url_prompt)
|
|
|
|
if bg_resp.status_code != 200 or prompt_resp.status_code != 200:
|
|
logger.error("Failed to download image pairs")
|
|
return None
|
|
|
|
return {
|
|
"instruction": instruction,
|
|
"bg_bytes": bg_resp.content,
|
|
"prompt_bytes": prompt_resp.content,
|
|
"img_url_bg": img_url_bg,
|
|
"img_url_prompt": img_url_prompt,
|
|
"sess": sess,
|
|
"tkid": tkid,
|
|
}
|
|
|
|
def fetch_one(self) -> Optional[str]:
|
|
"""采集单组样本并保存到本地"""
|
|
sess, tkid = self._fetch_session()
|
|
if not sess or not tkid:
|
|
return None
|
|
|
|
data = self._fetch_sig_and_images(sess, tkid)
|
|
if not data:
|
|
return None
|
|
|
|
sample_id = f"{int(time.time() * 1000)}_{random.randint(1000, 9999)}"
|
|
bg_filename = f"{sample_id}_bg.png"
|
|
prompt_filename = f"{sample_id}_prompt.png"
|
|
meta_filename = f"{sample_id}_meta.json"
|
|
|
|
bg_path = self.output_dir / bg_filename
|
|
prompt_path = self.output_dir / prompt_filename
|
|
meta_path = self.output_dir / meta_filename
|
|
|
|
bg_path.write_bytes(data["bg_bytes"])
|
|
prompt_path.write_bytes(data["prompt_bytes"])
|
|
|
|
meta_info = {
|
|
"sample_id": sample_id,
|
|
"timestamp": time.time(),
|
|
"aid": self.aid,
|
|
"instruction": data["instruction"],
|
|
"bg_file": bg_filename,
|
|
"prompt_file": prompt_filename,
|
|
"bg_size": len(data["bg_bytes"]),
|
|
"prompt_size": len(data["prompt_bytes"]),
|
|
"sess": data["sess"],
|
|
"tkid": data["tkid"],
|
|
}
|
|
|
|
meta_path.write_text(json.dumps(meta_info, ensure_ascii=False, indent=2), encoding="utf-8")
|
|
|
|
# 写入 index.jsonl
|
|
with open(self.index_file, "a", encoding="utf-8") as f:
|
|
f.write(json.dumps(meta_info, ensure_ascii=False) + "\n")
|
|
|
|
logger.info(
|
|
"✓ 采集成功 [%s]: 大图 %d KB, 提示图 %d KB | 提示: %s",
|
|
sample_id,
|
|
len(data["bg_bytes"]) // 1024,
|
|
len(data["prompt_bytes"]) // 1024,
|
|
data["instruction"],
|
|
)
|
|
return sample_id
|
|
|
|
def collect_batch(self, count: int = 10, delay: float = 0.8) -> int:
|
|
"""批量采集样本"""
|
|
logger.info("开始批量采集点选验证码样本，目标数量: %d, 间隔: %.2fs", count, delay)
|
|
success_count = 0
|
|
|
|
for i in range(count):
|
|
try:
|
|
sid = self.fetch_one()
|
|
if sid:
|
|
success_count += 1
|
|
else:
|
|
logger.warning("第 %d/%d 个样本采集失败，准备重试", i + 1, count)
|
|
except Exception as e:
|
|
logger.error("采集异常: %s", e)
|
|
|
|
if i < count - 1 and delay > 0:
|
|
# 随机微弱浮动，模拟正常采集
|
|
actual_delay = delay + random.uniform(-0.2, 0.3)
|
|
time.sleep(max(0.2, actual_delay))
|
|
|
|
logger.info("批量采集完成: 成功 %d/%d (存储目录: %s)", success_count, count, self.output_dir)
|
|
return success_count
|
|
|
|
def close(self):
|
|
self.client.close()
|
|
|
|
|
|
def main():
|
|
parser = argparse.ArgumentParser(description="腾讯天御点选验证码纯协议批量采集器")
|
|
parser.add_argument("--count", type=int, default=10, help="采集样本数量 (默认: 10)")
|
|
parser.add_argument("--delay", type=float, default=0.8, help="每次请求间隔秒数 (默认: 0.8)")
|
|
parser.add_argument("--aid", type=str, default=TencentCaptchaCollector.DEFAULT_AID, help="业务 AppID")
|
|
parser.add_argument("--output-dir", type=str, default="tcaptcha/data/raw", help="保存路径")
|
|
|
|
args = parser.parse_args()
|
|
collector = TencentCaptchaCollector(aid=args.aid, output_dir=args.output_dir)
|
|
try:
|
|
collector.collect_batch(count=args.count, delay=args.delay)
|
|
finally:
|
|
collector.close()
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|