init: RAG 知识库服务初始提交

- 后端 API（Flask + Gunicorn） - RAG 引擎（混合检索 + 云端 Reranker + 引用溯源） - 文档解析（MinerU + 多格式支持） - Docker 生产部署配置 - 排除前端项目、敏感配置、模型文件
2026-06-04 17:35:27 +08:00
commit 100d1a06eb
158 changed files with 64534 additions and 0 deletions
--- a/api/audit_routes.py
+++ b/api/audit_routes.py
@@ -0,0 +1,133 @@
+"""
+审计日志 API
+
+路由:
+- GET /audit/logs - 查询审计日志（管理员）
+"""
+
+from flask import Blueprint, request, jsonify
+import logging
+
+logger = logging.getLogger(__name__)
+from auth.gateway import require_gateway_auth
+from data.db import get_connection
+
+audit_bp = Blueprint('audit', __name__)
+
+
+@audit_bp.route('/audit/logs', methods=['GET'])
+@require_gateway_auth
+def get_audit_logs():
+    """
+    查询审计日志
+
+    参数:
+    - limit: 返回条数（默认50）
+    - days: 查询天数（默认7）
+    - action: 操作类型过滤（可选）
+
+    返回:
+    {
+        "logs": [
+            {
+                "id": 1,
+                "user_id": "admin001",
+                "username": "admin",
+                "action": "rag_query",
+                "query": "xxx",
+                "result_summary": "...",
+                "role": "admin",
+                "department": "管理部",
+                "ip_address": "127.0.0.1",
+                "duration_ms": 1234,
+                "timestamp": "2025-01-01 12:00:00"
+            }
+        ],
+        "total": 100
+    }
+    """
+    limit = request.args.get('limit', 50, type=int)
+    days = request.args.get('days', 7, type=int)
+    action_filter = request.args.get('action', '')
+
+    try:
+        with get_connection("session") as conn:
+            # 构建查询
+            where_clauses = ["created_at >= datetime('now', ?)"]
+            params = [f'-{days} days']
+
+            if action_filter:
+                where_clauses.append("action = ?")
+                params.append(action_filter)
+
+            where_sql = " AND ".join(where_clauses)
+
+            # 查询总数
+            count_sql = f"SELECT COUNT(*) FROM audit_logs WHERE {where_sql}"
+            total = conn.execute(count_sql, params).fetchone()[0]
+
+            # 查询日志
+            query_sql = f"""
+                SELECT id, user_id, username, action, query, result_summary,
+                       role, department, ip_address, duration_ms, created_at
+                FROM audit_logs
+                WHERE {where_sql}
+                ORDER BY created_at DESC
+                LIMIT ?
+            """
+            params.append(limit)
+            rows = conn.execute(query_sql, params).fetchall()
+
+        logs = []
+        for row in rows:
+            logs.append({
+                "id": row[0],
+                "user_id": row[1],
+                "username": row[2],
+                "action": row[3],
+                "query": row[4],
+                "result_summary": row[5],
+                "role": row[6],
+                "department": row[7],
+                "ip_address": row[8],
+                "duration_ms": row[9],
+                "timestamp": row[10]
+            })
+
+        return jsonify({"logs": logs, "total": total})
+
+    except Exception as e:
+        return jsonify({"error": str(e), "logs": [], "total": 0}), 500
+
+
+def log_audit_event(user_id: str, username: str, action: str,
+                    query: str = None, result_summary: str = None,
+                    role: str = None, department: str = None,
+                    ip_address: str = None, duration_ms: int = None):
+    """
+    记录审计日志（供其他模块调用）
+
+    Args:
+        user_id: 用户ID
+        username: 用户名
+        action: 操作类型（rag_query, chat, feedback, sync 等）
+        query: 查询内容
+        result_summary: 结果摘要
+        role: 用户角色
+        department: 部门
+        ip_address: IP地址
+        duration_ms: 耗时（毫秒）
+    """
+    try:
+        with get_connection("session") as conn:
+            conn.execute("""
+                INSERT INTO audit_logs
+                (user_id, username, action, query, result_summary,
+                 role, department, ip_address, duration_ms)
+                VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?)
+            """, (user_id, username, action, query, result_summary,
+                  role, department, ip_address, duration_ms))
+    except Exception as e:
+        # 审计日志写入失败不应影响主流程
+        logger.debug(f"审计日志写入失败: {e}")
+        pass