Python 服务器无法恢复事务。描述：980000002。（3971）（SQLEndTran）“）

我试图做的是使用pd.to_SQL方法在SQL server中推送数据帧。 langyuage-python 3.6，pandas版本0.25.1

发动机变量

engine=create\U enginemssql+pyodbc://？odbc\U connect=%s%PARAM，池大小=20，最大溢出=10，池回收=60，池超时=30

这是调用staging_标准化方法的主脚本

标准化方法

def staging_standardisation(engine, table_name, column_names, column_obj, **kwargs):
    try:
        query = "SELECT * FROM  dbo.License_Search"
        sql_df = pd.read_sql(query, engine)
        df = sql_df[column_names]
        df = df.rename(index=str, columns=column_obj)
        try:
            if 'authority_id' in column_obj.values():
                pass
            else:

                df['authority_id'] =53

        except KeyError:
            df['authority_id'] = None
    try:
        df.to_sql(name=staging_table, con=engine, index=False, if_exists='append') # this line causing error.

    except Exception as exe:
        print(table_name, exe)

投掷误差

(pyodbc.ProgrammingError) ('42000', '[42000] [Microsoft][ODBC Driver 17 for SQL Server][SQL Server]
The server failed to resume the transaction. Desc:8c00000007. 
(3971) (SQLEndTran); [42000] [Microsoft][ODBC Driver 17 for SQL Server][SQL Server]The 
transaction active in this session has been committed or aborted by another session. (3926)')
(Background on this error at: http://sqlalche.me/e/f405)

---

问题在于铸造。在dataframe中，datecolumn类型是字符串。然后我将列类型转换为DateTime

仅供参考：vs代码显示的错误有限。我们在jupyter笔记本中运行了一个单独的测试，其中出现了全部错误，并发现是日期列

def staging_standardisation(engine, table_name, column_names, column_obj, **kwargs):
    try:
        query = "SELECT * FROM  dbo.License_Search"
        sql_df = pd.read_sql(query, engine)
        df = sql_df[column_names]
        df = df.rename(index=str, columns=column_obj)
        df['Date'] = pd.to_datetime(df['Date'])
        try:
            if 'authority_id' in column_obj.values():
                pass
            else:

                df['authority_id'] =53

        except KeyError:
            df['authority_id'] = None
    try:
        df.to_sql(name=staging_table, con=engine, index=False, if_exists='append') # this line causing error.

    except Exception as exe:
        print(table_name, exe)

---

query=从表_name+中选择id+，其中name=+'+{0}；真是一团糟。它还暴露于SQL注入。但是从表_name中选择id+让我感到困惑；为什么你要把这些字符串合并成一个字符串呢？query=SELECT id FROM table\u name，其中name=%s，authority\u id。但是我完全不能理解它。@roganjosh我修改了set df['authority\u id']=53，query=SELECT*FROM dbo.License\u搜索代码，然后面对同样的错误再次运行它

def staging_standardisation(engine, table_name, column_names, column_obj, **kwargs):
    try:
        query = "SELECT * FROM  dbo.License_Search"
        sql_df = pd.read_sql(query, engine)
        df = sql_df[column_names]
        df = df.rename(index=str, columns=column_obj)
        df['Date'] = pd.to_datetime(df['Date'])
        try:
            if 'authority_id' in column_obj.values():
                pass
            else:

                df['authority_id'] =53

        except KeyError:
            df['authority_id'] = None
    try:
        df.to_sql(name=staging_table, con=engine, index=False, if_exists='append') # this line causing error.

    except Exception as exe:
        print(table_name, exe)