添加最大数量限制

.env

@@ -8,10 +8,10 @@ PG_PASSWORD=feAeyR0u2fJGSS5ooFdHnSbyHQNY4WlV
 PG_DB=postgres
 
 # MySQL（业务库）
-DB_HOST=192.168.31.105
-DB_PORT=3306
+DB_HOST=8.163.14.142
+DB_PORT=30006
 DB_USER=root
-DB_PASSWORD=123456
+DB_PASSWORD=^CgDatabase2020
 DB_NAME=offerpie
 
 
@@ -19,10 +19,15 @@ DB_NAME=offerpie
 VOLCENGINE_API_KEY=fd065993-bee2-4f31-8bf2-56d5d3012c02
 VOLCENGINE_BASE_URL=https://ark.cn-beijing.volces.com/api/v3
 
+# Claude（Anthropic 风格）
+ANTHROPIC_API_KEY=sk-43ccdb29caa7e9ebe0db8ac0958c63f6d3a2d62e59064d3d26d94332055a9bc9
+ANTHROPIC_BASE_URL=https://code.warpdevloper.cloud
+
 # 岗位清洗参数
-CLEAN_BATCH_SIZE=100
-CLEAN_CONCURRENCY=50
-CLEAN_INTERVAL_SECONDS=180
+CLEAN_BATCH_SIZE=20
+CLEAN_CONCURRENCY=20
+CLEAN_INTERVAL_SECONDS=100
+CLEAN_TOTAL_LIMIT=0
 
 # 公司补充参数
 COMPANY_BATCH_SIZE=20

app/ai/model_config.py

@@ -9,14 +9,14 @@ from app.ai.models import LLM
 class JobCleanModel:
     """岗位清洗模块"""
     # 第一次AI：结构化提取岗位信息
-    STRUCTURE = LLM.DOUBAO_SEED_LITE.create(temperature=0)
+    STRUCTURE = LLM.CLAUDE_OPUS.create(temperature=0)
     # 第二次AI：专业匹配
-    MAJOR_MATCH = LLM.DOUBAO_SEED_LITE.create(temperature=0)
+    MAJOR_MATCH = LLM.CLAUDE_OPUS.create(temperature=0)
     # 第三次AI：技能提取
-    SKILL_EXTRACT = LLM.DOUBAO_SEED_LITE.create(temperature=0)
+    SKILL_EXTRACT = LLM.CLAUDE_OPUS.create(temperature=0)
 
 
 class CompanyCleanModel:
     """公司补充模块"""
     # 公司信息补充
-    ENRICH = LLM.DOUBAO_SEED_LITE.create(temperature=0)
+    ENRICH = LLM.CLAUDE_OPUS.create(temperature=0)

app/ai/models.py

@@ -4,36 +4,48 @@ Usage:
     from app.ai.models import LLM
 
     llm = LLM.DOUBAO_SEED_LITE.create(temperature=0)
+    llm = LLM.CLAUDE_OPUS.create(temperature=0)
 """
 
 from enum import Enum
 
+from langchain_anthropic import ChatAnthropic
+from langchain_core.language_models import BaseChatModel
 from langchain_openai import ChatOpenAI
 
 from app.config import settings
 
-# 供应商连接配置
+# 供应商连接配置 = (api_key函数, base_url函数)
 _VOLCENGINE = (lambda: settings.volcengine_api_key, lambda: settings.volcengine_base_url)
+_ANTHROPIC = (lambda: settings.anthropic_api_key, lambda: settings.anthropic_base_url)
 
 
 class LLM(Enum):
-    """所有可用模型，每个枚举值 = (模型名, api_key函数, base_url函数)"""
+    """所有可用模型，每个枚举值 = (模型名, 封装类, api_key函数, base_url函数)"""
 
-    # 火山引擎
-    DOUBAO_PRO_32K = ("doubao-1-5-pro-32k-250115", *_VOLCENGINE)
-    DOUBAO_LITE_32K = ("doubao-1-5-lite-32k-250115", *_VOLCENGINE)
-    DOUBAO_SEED_LITE = ("doubao-seed-2-0-lite-260215", *_VOLCENGINE)
-    DOUBAO_SEED_PRO = ("doubao-seed-2-0-pro-260215", *_VOLCENGINE)
-    DEEPSEEK_V4_FLASH = ("deepseek-v4-flash-260425", *_VOLCENGINE)
+    # 火山引擎（OpenAI 兼容）
+    DOUBAO_PRO_32K = ("doubao-1-5-pro-32k-250115", ChatOpenAI, *_VOLCENGINE)
+    DOUBAO_LITE_32K = ("doubao-1-5-lite-32k-250115", ChatOpenAI, *_VOLCENGINE)
+    DOUBAO_SEED_LITE = ("doubao-seed-2-0-lite-260215", ChatOpenAI, *_VOLCENGINE)
+    DOUBAO_SEED_PRO = ("doubao-seed-2-0-pro-260215", ChatOpenAI, *_VOLCENGINE)
+    DEEPSEEK_V4_FLASH = ("deepseek-v4-flash-260425", ChatOpenAI, *_VOLCENGINE)
 
-    def __init__(self, model_name: str, api_key_fn, base_url_fn):
+    # Claude（Anthropic 风格）
+    CLAUDE_OPUS = ("claude-opus-4-6", ChatAnthropic, *_ANTHROPIC)
+
+    def __init__(self, model_name: str, cls, api_key_fn, base_url_fn):
         self.model_name = model_name
+        self._cls = cls
         self._api_key_fn = api_key_fn
         self._base_url_fn = base_url_fn
 
-    def create(self, **kwargs) -> ChatOpenAI:
-        """创建 LLM 实例，kwargs 透传给 ChatOpenAI（temperature, max_tokens 等）"""
-        return ChatOpenAI(
+    def create(self, **kwargs) -> BaseChatModel:
+        """创建 LLM 实例，kwargs 透传给底层封装（temperature, max_tokens 等）
+
+        封装类（ChatOpenAI / ChatAnthropic）均实现 langchain BaseChatModel 接口，
+        对上层调用方完全透明。
+        """
+        return self._cls(
             model=self.model_name,
             api_key=self._api_key_fn(),
             base_url=self._base_url_fn(),

app/config/settings.py

@@ -26,13 +26,19 @@ class Settings(BaseSettings):
     mysql_max_overflow: int = 20
 
     # ──────────── AI 供应商 ────────────
+    # 火山引擎（OpenAI 兼容风格）
     volcengine_api_key: str =  "fd065993-bee2-4f31-8bf2-56d5d3012c02"
     volcengine_base_url: str = "https://ark.cn-beijing.volces.com/api/v3"
 
+    # Claude（Anthropic 风格）
+    anthropic_api_key: str = "sk-43ccdb29caa7e9ebe0db8ac0958c63f6d3a2d62e59064d3d26d94332055a9bc9"
+    anthropic_base_url: str = "https://code.warpdevloper.cloud"
+
     # ──────────── 岗位清洗参数 ────────────
     clean_batch_size: int = 100
     clean_concurrency: int = 80
     clean_interval_seconds: int = 200
+    clean_total_limit: int = 0  # 累计清洗总数上限，达到后停止清洗任务；0 = 不限制
 
     # ──────────── 公司补充参数 ────────────
     company_batch_size: int = 20

app/services/ai_tool.py

@@ -4,7 +4,7 @@ import re
 from typing import Any
 
 from json_repair import repair_json
-from langchain_openai import ChatOpenAI
+from langchain_core.language_models import BaseChatModel
 from langchain_core.messages import SystemMessage, HumanMessage
 
 from app.core.logger import log
@@ -28,7 +28,7 @@ def parse_llm_json(text: str) -> Any:
     return repair_json(cleaned, return_objects=True)
 
 
-async def ai_chat(llm: ChatOpenAI, system_prompt: str, user_message: str) -> str:
+async def ai_chat(llm: BaseChatModel, system_prompt: str, user_message: str) -> str:
     """异步调用 LLM，返回原始文本"""
     messages = [
         SystemMessage(content=system_prompt),
@@ -38,7 +38,7 @@ async def ai_chat(llm: ChatOpenAI, system_prompt: str, user_message: str) -> str
     return response.content
 
 
-async def ai_chat_json(llm: ChatOpenAI, system_prompt: str, user_message: str) -> Any:
+async def ai_chat_json(llm: BaseChatModel, system_prompt: str, user_message: str) -> Any:
     """异步调用 LLM，返回解析后的 JSON 对象"""
     raw = await ai_chat(llm, system_prompt, user_message)
     if not raw or not raw.strip():

app/services/job_clean_service.py

@@ -24,9 +24,23 @@ _id_gen = SnowflakeGenerator(instance=1)
 # 公司创建锁（防止并发重复插入同一公司）
 _company_lock = asyncio.Lock()
 
+# 累计已清洗总数（用于总数上限控制，进程内统计）
+_cleaned_total = 0
+
+
+def is_clean_limit_reached() -> bool:
+    """是否已达到累计清洗总数上限（clean_total_limit=0 表示不限制）"""
+    return 0 < settings.clean_total_limit <= _cleaned_total
+
 
 async def run_job_clean() -> None:
     """一次批量清洗任务"""
+    global _cleaned_total
+
+    # 总数上限：已达上限直接跳过
+    if is_clean_limit_reached():
+        return
+
     # 1. 从 PG 锁定一批待清洗数据
     async with PgSession() as pg:
         result = await pg.execute(
@@ -62,7 +76,11 @@ async def run_job_clean() -> None:
 
     # 汇总
     errors = sum(1 for r in results if isinstance(r, Exception))
-    log.info("岗位清洗：本批完成，共{}条，异常{}条", len(rows), errors)
+    _cleaned_total += len(rows)
+    log.info("岗位清洗：本批完成，共{}条，异常{}条，累计{}条", len(rows), errors, _cleaned_total)
+
+    if is_clean_limit_reached():
+        log.info("岗位清洗：已达累计上限 {} 条，停止清洗任务", settings.clean_total_limit)
 
 
 async def _clean_one(sem: asyncio.Semaphore, data: dict) -> None:

requirements.txt

@@ -7,6 +7,7 @@ apscheduler>=3.10
 
 # AI
 langchain-openai>=0.3
+langchain-anthropic>=0.3
 langchain-core>=0.3
 
 # 工具