Bladeren bron

feat: add plugins for visa automation workflow

hujiarui 1 maand geleden
bovenliggende
commit
3a58bfe6f4
3 gewijzigde bestanden met toevoegingen van 103 en 8 verwijderingen
  1. 1 1
      configure.py
  2. 38 5
      plugins/tls_plugin.py
  3. 64 2
      tools/generate_clash_multi_port.py

+ 1 - 1
configure.py

@@ -3,7 +3,7 @@ CAPTCHA_LLM_CONFIG = None
 CAPTCHA_LLM_CONFIG = {
     "base_url": "https://dashscope.aliyuncs.com/compatible-mode/v1",
     "api_key": "sk-893e895724c6403d81374e515ffaf427",
-    "model": "qwen-vl-max",
+    "model": "qwen3-vl-plus",
     "timeout": 30.0,
 }
 

+ 38 - 5
plugins/tls_plugin.py

@@ -548,7 +548,9 @@ class TlsPlugin(IPlugin):
         res.success = False
         slots = []
         self._log(f"Executing silent JS fetch...") 
-        resp = self._perform_request("GET", self.page.url, retry_count=0)
+        resp = self._perform_request("GET", self.page.url, retry_count=0, allowed_status_codes=(200, 500))
+        if resp.status_code == 500:
+            self._log("⚠️ Query received HTTP 500 deception response, processing payload...")
         self._check_page_is_session_expired_or_invalid('Book your appointment', resp.text)
         slots = self._parse_appointment_slots(resp.text)
         
@@ -698,16 +700,38 @@ class TlsPlugin(IPlugin):
                 pass
         return res
 
-    def _perform_request(self, method, url, headers=None, data=None, json_data=None, params=None, retry_count=0):
+    def _perform_request(
+        self, 
+        method, 
+        url, 
+        headers=None, 
+        data=None, 
+        json_data=None, 
+        params=None, 
+        retry_count=0,
+        allowed_status_codes=None
+    ):
         """
         在浏览器上下文中注入 JS 执行 Fetch
         """
         resp = perform_browser_fetch(self.page, method, url, headers=headers, data=data, json_data=json_data, params=params)
         
-        if resp.status_code == 200:
+        valid_statuses = set(allowed_status_codes) if allowed_status_codes else {200}
+        
+        if resp.status_code in valid_statuses:
+            if resp.status_code != 200:
+                self._log(f"⚠️ Received HTTP {resp.status_code} (in allowed_status_codes), treating as success.")
             return resp
+
+        # 兼容 TLSContact 故意返回 HTTP 500 欺骗脚本的反爬机制:
+        # 如果返回 500,但响应体中实际包含正常的页面或预约槽位数据,则视为有效响应放行
+        if resp.status_code == 500 and resp.text:
+            text_lower = resp.text.lower()
+            if "availableappointments" in text_lower or "book your appointment" in text_lower:
+                self._log("⚠️ Server returned HTTP 500 (anti-scraper deception detected), but response contains valid appointment data. Proceeding.")
+                return resp
             
-        elif resp.status_code == 401:
+        if resp.status_code == 401:
             self.is_healthy = False
             raise SessionExpiredOrInvalidError()
             
@@ -718,7 +742,16 @@ class TlsPlugin(IPlugin):
                 self._log(f"HTTP {resp.status_code} Detected. Cloudflare block or rate limit? Attempting refresh (Try {retry_count+1}/2)...")
                 if self._refresh_firewall_session():
                     self._log("Firewall session refreshed. Retrying request...")
-                    return self._perform_request(method, url, headers, data, json_data, params, retry_count+1)
+                    return self._perform_request(
+                        method, 
+                        url, 
+                        headers=headers, 
+                        data=data, 
+                        json_data=json_data, 
+                        params=params, 
+                        retry_count=retry_count+1, 
+                        allowed_status_codes=allowed_status_codes
+                    )
                 else:
                     self._log("Failed to refresh firewall session.")    
             

+ 64 - 2
tools/generate_clash_multi_port.py

@@ -75,6 +75,25 @@ def detect_country(node_name: str) -> str:
     return "OTHER"
 
 
+def extract_rate(node_name: str) -> Optional[float]:
+    """
+    从节点名称中提取倍率数字 (例如 '丨10x HK' -> 10.0, '5x' -> 5.0)
+    """
+    m = re.search(r'(?:丨|\||\s|\[|\()(\d+(?:\.\d+)?)\s*[xX]', node_name)
+    if m:
+        try:
+            return float(m.group(1))
+        except ValueError:
+            pass
+    m2 = re.search(r'(\d+(?:\.\d+)?)\s*[xX]', node_name)
+    if m2:
+        try:
+            return float(m2.group(1))
+        except ValueError:
+            pass
+    return None
+
+
 def fetch_config(source: str) -> dict:
     """从网络链接或本地文件读取 yaml 配置"""
     if source.startswith("http://") or source.startswith("https://"):
@@ -99,6 +118,8 @@ def main():
     parser.add_argument("--sub", type=str, required=True, help="服务商提供的订阅链接 (URL) 或本地 yaml 文件路径")
     parser.add_argument("--start-port", type=int, default=10001, help="起始监听端口 (默认: 10001)")
     parser.add_argument("--count", type=int, default=0, help="最多导出的节点总数量 (默认: 0 表示全部节点)")
+    parser.add_argument("--exclude-rates", type=str, default="5,10", help="排除指定倍率的节点,逗号隔开 (默认: '5,10',支持 '5,10' 或 '5x,10x',留空或设为 'none' 则不过滤)")
+    parser.add_argument("--max-rate", type=float, default=None, help="最大允许倍率,超过该倍率的节点将被过滤 (例如: 3.0 则自动排除 5x, 10x 等)")
     parser.add_argument("--filter-kw", type=str, default="", help="节点关键词过滤 (如: '专线', '家宽',留空表示不限制)")
     parser.add_argument("--countries", type=str, default="", help="仅导出的国家代码列表,逗号分隔 (例如: 'DE,IE,GB,US',留空表示全量导出)")
     parser.add_argument("--pool-prefix", type=str, default="local", help="国家池子的前缀 (默认: 'local' -> 生成 'local_de', 'local_ie' 等)")
@@ -125,8 +146,20 @@ def main():
             if c:
                 target_countries.add(c)
 
+    # 倍率过滤规则解析
+    exclude_rates_set = set()
+    if args.exclude_rates and args.exclude_rates.lower() != "none":
+        for r_str in args.exclude_rates.split(","):
+            r_str = r_str.strip().lower().rstrip("x")
+            if r_str:
+                try:
+                    exclude_rates_set.add(float(r_str))
+                except ValueError:
+                    pass
+
     # 2. 节点过滤与国家识别
     selected_proxies = []
+    filtered_by_rate_count = 0
     for p in proxies:
         name = p.get("name", "")
         # 排除不可用的系统提示节点
@@ -134,6 +167,13 @@ def main():
             continue
         if args.filter_kw and args.filter_kw.lower() not in name.lower():
             continue
+
+        # 倍率过滤
+        node_rate = extract_rate(name)
+        if node_rate is not None:
+            if node_rate in exclude_rates_set or (args.max_rate is not None and node_rate > args.max_rate):
+                filtered_by_rate_count += 1
+                continue
             
         # 国家过滤
         c_code = detect_country(name)
@@ -143,6 +183,14 @@ def main():
         p["_country_code"] = c_code
         selected_proxies.append(p)
 
+    if filtered_by_rate_count > 0:
+        rules_desc = []
+        if exclude_rates_set:
+            rules_desc.append(f"排除倍率: {[f'{r:g}x' for r in sorted(exclude_rates_set)]}")
+        if args.max_rate is not None:
+            rules_desc.append(f"倍率上限 <= {args.max_rate:g}x")
+        print(f"[*] 已根据倍率规则自动过滤掉 {filtered_by_rate_count} 个高倍率节点 ({', '.join(rules_desc)})。")
+
     if args.count > 0:
         selected_proxies = selected_proxies[:args.count]
 
@@ -219,11 +267,16 @@ def main():
         except Exception:
             pass
 
-    # 汇总所有现有节点和新节点到 nodes 列表 (按节点名称去重)
+    # 汇总所有现有节点和新节点到 nodes 列表 (按节点名称去重,并同步过滤掉需要排除的高倍率节点)
     existing_nodes_dict = {}
     for n in all_proxies_data["nodes"]:
         if isinstance(n, dict) and "name" in n:
-            existing_nodes_dict[n["name"]] = n
+            n_name = n["name"]
+            n_rate = extract_rate(n_name)
+            if n_rate is not None:
+                if n_rate in exclude_rates_set or (args.max_rate is not None and n_rate > args.max_rate):
+                    continue
+            existing_nodes_dict[n_name] = n
 
     for item in all_proxies_list:
         existing_nodes_dict[item["name"]] = item
@@ -243,6 +296,15 @@ def main():
         port_range_str = f"{min(ports)}" if len(ports) == 1 else f"{min(ports)}~{max(ports)}"
         generated_country_pools.append((pool_key, c_code.upper(), len(items), port_range_str))
 
+    # (3) 清理所有历史池子中的高倍率节点名称,并移除变空的废弃池子
+    valid_node_names = set(existing_nodes_dict.keys())
+    for pool_key in list(all_proxies_data["pools"].keys()):
+        cleaned_pool = [n for n in all_proxies_data["pools"][pool_key] if n in valid_node_names]
+        if cleaned_pool:
+            all_proxies_data["pools"][pool_key] = cleaned_pool
+        else:
+            del all_proxies_data["pools"][pool_key]
+
     with open(proxies_yaml_path, "w", encoding="utf-8") as f:
         yaml.safe_dump(all_proxies_data, f, allow_unicode=True, sort_keys=False)