枚举结果处理

2025-09-25 16:49:25 +08:00
parent 7132d07921
commit b9bf30a542
4 changed files with 65 additions and 8 deletions
--- a/main_service.py
+++ b/main_service.py
@@ -1,6 +1,7 @@
 from email.policy import default
 import logging

+import util.utils
 from logging_config import LOGGING_CONFIG
 from service.cus_vanna_srevice import CustomVanna, QdrantClient
 from decouple import config
@@ -97,18 +98,20 @@ def generate_sql_2():
        return jsonify({"type": "error", "error": "No question provided"})
    try:
        id = cache.generate_id(question=question)
+        logger.info(f"Generate sql for {question}")
        data = vn.generate_sql_2(question=question)
+        logger.info("Generate sql result is {0}".format(data))
        data['id'] = id
        sql = data["resp"]["sql"]
-        print("sql:", sql)
        cache.set(id=id, field="question", value=question)
        cache.set(id=id, field="sql", value=sql)
-        print("data---------------------------", data)
+        data["type"]="success"
        return jsonify(data)
    except Exception as e:
        return jsonify({"type": "error", "error": str(e)})


+
@app.flask_app.route("/api/v0/run_sql_2", methods=["GET"])
@app.requires_cache(["sql"])
 def run_sql_2(id: str, sql: str):
@@ -150,13 +153,15 @@ def run_sql_2(id: str, sql: str):
        df = vn.run_sql(sql=sql)
        logger.info("")
        app.cache.set(id=id, field="df", value=df)
-        x = df.head(10).to_dict(orient='records')
+        x = df.to_dict(orient='records')
        logger.info("df ---------------{0}   {1}".format(x,type(x)))
+        result = util.utils.deal_result(data=x)
+
        return jsonify(
            {
-                "type": "df",
+                "type": "success",
                "id": id,
-                "df": df.head(10).to_dict(orient='records'),
+                "df": result,
            }
        )

--- a/service/cus_vanna_srevice.py
+++ b/service/cus_vanna_srevice.py
@@ -184,6 +184,7 @@ class OpenAICompatibleLLM(VannaBase):

    def generate_sql_2(self, question: str, allow_llm_to_see_data=False, **kwargs) -> dict:
        try:
+            logger.info("Start to generate_sql_2 in cus_vanna_srevice")
            question_sql_list = self.get_similar_question_sql(question, **kwargs)
            ddl_list = self.get_related_ddl(question, **kwargs)
            doc_list = self.get_related_documentation(question, **kwargs)
@@ -202,9 +203,12 @@ class OpenAICompatibleLLM(VannaBase):
                [{'role': 'system', 'content': sys_temp}, {'role': 'user', 'content': user_temp}], **kwargs)
            logger.info(f"llm_response:{llm_response}")
            result = {"resp": orjson.loads(extract_nested_json(llm_response))}
+            logger.info(f"llm_response:{llm_response}")
            sql = check_and_get_sql(llm_response)
+            logger.info(f"sql:{sql}")
            # ---------------生成图表
            char_type = get_chart_type_from_sql_answer(llm_response)
+            logger.info(f"chart type:{char_type}")
            if char_type:
                sys_char_temp = char_temp['system'].format(engine=config("DB_ENGINE", default='mysql'),
                                                           lang='中文', sql=sql, chart_type=char_type)
@@ -215,8 +219,11 @@ class OpenAICompatibleLLM(VannaBase):
                print(llm_response2)
                result['chart'] = orjson.loads(extract_nested_json(llm_response2))
                logger.info(f"chart_response:{result}")
+
+            logger.info("Finish to generate_sql_2 in cus_vanna_srevice")
            return result
        except Exception as e:
+            logger.info("cus_vanna_srevice failed-------------------")
            raise e

    def generate_rewritten_question(self, last_question: str, new_question: str, **kwargs) -> str:
--- a/template.yaml
+++ b/template.yaml
@@ -28,11 +28,17 @@ template:
          你只能生成查询用的SQL语句，不得生成增删改相关或操作数据库以及操作数据库数据的SQL
        </rule>
        <rule>
-          如果只涉及查询人员信息，但没说具体哪些信息，可以不用查询所有信息，主要查询相关性较强的五个字段即可
+          如果因为客观原因无法生成sql，请合理分析无法生成的原因并反馈给用户
+        </rule>
+        <rule>
+          涉及查询人员信息时，如果用户没明确指出要查询哪些字段，主要查询相关性较强的10个字段即可，如果指定要查询所有信息，请返回所有字段信息
        </rule>
        <rule>
          不要编造<m-schema>内没有提供给你的表结构
        </rule>
+        <rule>
+          当需要计算的字段类型为varchar或者text时，请根据需求转换为合理的类型格式进行计算
+        </rule>
        <rule>
          生成的SQL必须符合<db-engine>内提供数据库引擎的规范
        </rule>
@@ -43,7 +49,7 @@ template:
          请注意区分'哪些'和'多少'的区别，哪些是指具体信息，多少是指数量，请注意甄别
        </rule>
        <rule>
-          如遇字符串类型的日期要计算时，务必转化为合理的格式进行计算
+          如遇字符串类型的日期要计算时，务必转化为合理的日期格式进行计算
        </rule>
        <rule>
          请使用JSON格式返回你的回答:
@@ -75,6 +81,9 @@ template:
        <rule>
          SQL查询的字段若是函数字段，如 COUNT(),CAST() 等，必须加上别名
        </rule>
+        <rule>
+          SQL查询的如果用了聚合函数，如SUM(),COUNT()等，必须配合GROUP BY使用
+        </rule>
        <rule>
          计算占比，百分比类型字段，保留两位小数，以%结尾
        </rule>
@@ -167,7 +176,15 @@ template:
              <user-question>今天天气如何？</user-question>
            </input>
            <output>
-              {{"success":false,"message":"我是智能问数小助手，我无法回答您的问题。"}}
+              {{"success":false,"message":"我是智能问数小助手，我无法回答您的问题，该问题与当前数据库问数不相关，数据库中无天气等信息。","status":0}}
+            </output>
+          </example>
+          <example>
+            <input>
+              <user-question>张三的年龄是多大</user-question>
+            </input>
+            <output>
+              {{"success":true,"sql":"SELECT name,  FLOOR(MONTHS_BETWEEN(SYSDATE, birthday) / 12) AS age FROM YJOA_APPSERVICE_DB.t_pr3rl2oj_yj_person_database","tables":["t_pr3rl2oj_yj_person_database"],"chart-type":"columns"}}
            </output>
          </example>
          <example>
--- a/util/utils.py
+++ b/util/utils.py
@@ -2,6 +2,15 @@ from typing import Optional

 from orjson import orjson

+keywords = {
+    "gender":{"1":"男","2":"女"},
+    "person_status":{"1":"草稿","2":"审批中","3":"制卡中","4":"已入库","5":"停用"},
+    "pass_type":{"1":"集团公司员工","2":"借调人员","3":"借用人员","4":"外部监管人员","5":"外协服务人员","6":"工勤人员","7":"来访人员"},
+    "person_type": {"YG":"正式员工","PQ":"劳务派遣人员","QT":"其他柔性引进人员","WHZ":"合作单位","WLS":"临时访客","WQT":"其他外部人员"},
+    "id_card_type":{"1":"身份证","2":"护照","3":"港澳通行证"},
+    "highest_education": {"1":"初中","2":"高中","3":"中专","4":"技校","5":"职高","6":"大专","7":"本科","8":"硕士","9":"博士"},
+    "highest_degree":{"1":"学士学位","2":"硕士学位","3":"博士学位","4":"无"},
+}

 def check_and_get_sql(res: str) -> str:
    json_str = extract_nested_json(res)
@@ -11,17 +20,21 @@ def check_and_get_sql(res: str) -> str:
    sql: str
    data: dict
    try:
+        print("check_and_get_sql1----------------------------")
        data = orjson.loads(json_str)

        if data['success']:
            sql = data['sql']
            return sql
        else:
+            print("check_and_get_sql2----------------------------")
            message = data['message']
            raise Exception(message)
    except Exception as e:
+        print("check_and_get_sql3----------------------------")
        raise e
    except Exception:
+        print("check_and_get_sql4----------------------------")
        raise Exception(orjson.dumps({'message': 'Cannot parse sql from answer',
                                      'traceback': "Cannot parse sql from answer:\n" + res}).decode())

@@ -69,3 +82,18 @@ def get_chart_type_from_sql_answer(res: str) -> Optional[str]:
    except Exception:
        return None
    return chart_type
+
+
+def deal_result(data: list) -> list:
+    try:
+        for item in data:
+            for key, map_value in keywords.items():
+                if key in item:
+                    new_key = item.get(key)
+                    item[key] = map_value[new_key]
+        print("data----------{0}".format(data))
+        return data
+    except Exception as e:
+        raise Exception(f"sql执行结果处理失败：{str(e)}")
+
+