09:21目的：将Secretflow通过读取CSV文件获取数据，修改为 传入的CSV文件中包含了数据库的连接串信息，底层通过连接串信息去查询MYSQL，然后返回DF

[secretflowcode]

思路：修改secretflow/data/core/io.py，通过修改_read_csv方法，分析CSV文件得到连接串，然后请求MYSQL获取数据
实现代码：
secretflow/data/core/io.py修改如下
def _read_csv(_filepath, _backend, *_args, **_kwargs):
import select_sql
df = pd.read_csv(_filepath)
url = df.iloc[0, 0]
port = df.iloc[0, 1]
username = df.iloc[0, 2]
password = df.iloc[0, 3]
database = df.iloc[0, 4]
table = df.iloc[0, 5]
columns = list(df.columns)
return select_sql.fetch_data_from_mysql(url, port, username, password, database, table, columns)

	# 以下模块不执行
    if _backend == "pandas":
        from secretflow.data.core.pandas.util import read_pandas_csv

        return read_pandas_csv(_filepath, *_args, **_kwargs)

    elif _backend == "polars":
        from secretflow.data.core.polars.util import read_polars_csv

        return read_polars_csv(_filepath, *_args, **_kwargs)
    else:
        raise RuntimeError(f"Unknown data backend {_backend}")

select_sql.py实现如下
import pymysql
import pandas as pd
def fetch_data_from_mysql(url, port, username, password, database, table, columns):
connection = pymysql.connect(
host=url,
port=int(port),
user=username,
password=password,
database=database,
charset='utf8mb4'
)

[zimu-yuxi]

此issue中回复解决#1436