优化Dify工具逻辑，调整知识提取和重排序流程，增强API调用的重试机制，更新意图识别API以支持更好的错误处理和日志记录，改进多线程检索功能

2025-07-01 18:56:10 +08:00
parent 603c8122d4
commit f76f44640a
6 changed files with 149 additions and 34 deletions
@@ -0,0 +1,54 @@
+# from gevent import monkey
+# monkey.patch_all()
+
+from flask import Flask, request, Response
+import os
+from dotenv import load_dotenv
+import json
+import time
+from gevent.lock import RLock
+
+
+import datetime
+import logging
+# 加载环境变量
+load_dotenv()
+from DifyQueryRetrieval import DifyQueryRetrieval
+
+
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
+    handlers=[
+        logging.StreamHandler()
+    ]
+)
+logging.getLogger('httpx').setLevel(logging.WARNING)
+logging.getLogger('openai').setLevel(logging.WARNING)
+
+logger = logging.getLogger(__name__)
+
+app = Flask(__name__)
+
+dify_query_retrieval = DifyQueryRetrieval(api_key="dataset-skLjmPVonjHo119OWNf3kAmY", base_url="http://10.1.16.39/v1")
+@app.route('/retrieve', methods=['POST'])
+def retrieve():
+    data = request.get_json(force=True)
+    original_query_str = data.get('original_query')
+    query_list_str = data.get('query_list')
+    data_set_list_str = data.get('data_set_list')
+    try:
+        query_list = query_list_str.split("<sub_query>")
+        data_set_list = data_set_list_str.split("<dataset>")
+        results = dify_query_retrieval.retrieve_api(original_query_str, query_list, data_set_list)
+        return Response(json.dumps(results, ensure_ascii=False), content_type='application/json; charset=utf-8')
+    except Exception as e:
+        logger.error(f"检索出错: {str(e)}", exc_info=True)
+        return Response(json.dumps({"error": str(e)}, ensure_ascii=False), content_type='application/json; charset=utf-8', status=500)
+
+
+if __name__ == "__main__":
+    # 开发环境使用Flask内置服务器
+    # 生产环境使用gunicorn支持高并发 uv run gunicorn -w 10 -k gevent -b 0.0.0.0:8001 rag2_0.dify.DifyQueryRetrieval_api:app
+    # uv run gunicorn -w 10 -k gevent --preload -b 0.0.0.0:8001 rag2_0.dify.DifyQueryRetrieval_api:app
+    app.run(host="0.0.0.0", port=8002, threaded=True)