目的:
利用
read_sql_query()
的 to_sql()
和 pandas
方法,我的 python 3.7
脚本的目标是通过读取 .sql
文件来执行从一台服务器到另一台服务器的多个表的 ETL。两种方法中的连接参数都利用了 create_engine
的 sqlalchemy
模块。
出现错误:
成功提取并加载第一组表+事务后,出现第四个错误。
sqlalchemy.exc.DBAPIError: (pyodbc.Error) ('HY090', '[HY090] [Microsoft][ODBC Driver Manager] Invalid string or buffer length (0) (SQLExecDirectW)')
请参阅下文了解更多详情。
程序:
;
拆分。
'ExtractTables.sql'
SET NOCOUNT ON
SELECT
[ID1]
, [Name]
, [LastUpdated]
, [UpdatedBy]
INTO #table1
FROM DB1.dbo.table1
SELECT * FROM #table1
;
SET NOCOUNT ON
SELECT
[ID1]
, [ID2]
, [Descr]
INTO #table2 FROM DB1.dbo.table2
SELECT * FROM #table2
create_engine
模块为引擎设置的。两台服务器都是 MS-SQL 服务器。根据我的研究,我相信我的错误来自 fast_executemany
参数。 'connection.py'
import pyodbc
import urllib
from sqlalchemy import create_engine
#Use trusted connection to connect to server. fast_executemany is mssql specific. Allows for large data loads.
params_H = urllib.parse.quote_plus("DRIVER=ODBC Driver 17 for SQL Server;SERVER=SERVER1;DATABASE=DB1;Trusted_Connection=yes")
engine_H = create_engine(f'mssql+pyodbc:///?odbc_connect={params_H}', fast_executemany=True)
params_b = urllib.parse.quote_plus("DRIVER=ODBC Driver 17 for SQL Server;SERVER=SERVER2;DATABASE=DB2;Trusted_Connection=yes")
engine_b= create_engine(f'mssql+pyodbc:///?odbc_connect={params_b}', fast_executemany=True)
read_sql_query()
提取每个命令,并使用 to_sql()
加载每个表。
'LoadTables.py'
import pandas as pd
import conn
def readSQLFile_makeTables(filename):
# Open and read file
open_file = open(filename, 'r')
sql_file = open_file.read()
open_file.close()
#all SQL commands (split on ';')
sql_commands = sql_file.split(';')
# Execute every command from file
sql_tables = ['stg_table1', 'stg_table2', 'stg_table3']
i = 0
for command in sql_commands:
table = pd.read_sql_query(command, con=conn.engine_H)
print(table)
table.to_sql(sql_tables[i], con=conn.engine_b, chunksize=5000, index=False, if_exists='append')
i += 1
print('think this ran')
错误:
出于 StackOverflow 的目的,我将 sql 代码限制为 2 批,但实际上有 4 批。前 3 批成功通过了读取和写入。然而,第四个会引发写入错误。错误表批次与其他批次的主要区别在于其大小(700 万行 x 8 列),其次是大小(150 万行 x 6 列)。
故障排除:
我在该主题上研究的所有错误都表明这是 ODBC 连接的问题。两台服务器都是 64 位,我使用的是 pydobc 4.025,并且我只测试了提取整数值的字段。第一个能够成功加载的事务告诉我大多数事务正在工作,但最后一个事务存在阻止加载的问题。我假设大小,我相信这是由
chunksize=5000
和 fast_executemany = true
处理的,并且错误指向绑定参数。
https://github.com/mkleehammer/pyodbc/issues/548
追溯:
[6721864 rows x 8 columns]
Traceback (most recent call last):
File "C:\Users\x\AppData\Local\Continuum\anaconda3\envs\envname\lib\site-packages\sqlalchemy\engine\base.py", line 1244, in _execute_context
cursor, statement, parameters, context
File "C:\Users\x\AppData\Local\Continuum\anaconda3\envs\envname\lib\site-packages\sqlalchemy\engine\default.py", line 552, in do_execute
cursor.execute(statement, parameters)
pyodbc.Error: ('HY090', '[HY090] [Microsoft][ODBC Driver Manager] Invalid string or buffer length (0) (SQLExecDirectW)')
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "PYTHON\Testing\ETL\LoadTables.py", line 21, in readSQLFile_makeTables
table = pd.read_sql_query(command, con=conn.engine_H)
File "\lib\site-packages\pandas\io\sql.py", line 314, in read_sql_query
parse_dates=parse_dates, chunksize=chunksize)
File "\lib\site-packages\pandas\io\sql.py", line 1063, in read_query
result = self.execute(*args)
File "\lib\site-packages\pandas\io\sql.py", line 954, in execute
return self.connectable.execute(*args, **kwargs)
File "\lib\site-packages\sqlalchemy\engine\base.py", line 2166, in execute
return connection.execute(statement, *multiparams, **params)
File "\lib\site-packages\sqlalchemy\engine\base.py", line 982, in execute
return self._execute_text(object_, multiparams, params)
File "\lib\site-packages\sqlalchemy\engine\base.py", line 1155, in _execute_text
parameters,
File "\lib\site-packages\sqlalchemy\engine\base.py", line 1248, in _execute_context
e, statement, parameters, cursor, context
File "\lib\site-packages\sqlalchemy\engine\base.py", line 1466, in _handle_dbapi_exception
util.raise_from_cause(sqlalchemy_exception, exc_info)
File "\lib\site-packages\sqlalchemy\util\compat.py", line 383, in raise_from_cause
reraise(type(exception), exception, tb=exc_tb, cause=cause)
File "\lib\site-packages\sqlalchemy\util\compat.py", line 128, in reraise
raise value.with_traceback(tb)
File "\lib\site-packages\sqlalchemy\engine\base.py", line 1244, in _execute_context
cursor, statement, parameters, context
File "\lib\site-packages\sqlalchemy\engine\default.py", line 552, in do_execute
cursor.execute(statement, parameters)
sqlalchemy.exc.DBAPIError: (pyodbc.Error) ('HY090', '[HY090] [Microsoft][ODBC Driver Manager] Invalid string or buffer length (0) (SQLExecDirectW)')
(Background on this error at: http://sqlalche.me/e/dbapi)
我也遇到过类似的问题。发生的情况是,当将数据传输到 pyodbc 时,某些数据的信息格式不正确或 pyodb 不支持。解决方案是将数据传输到 CSV 并让 pyodbc 从 CSV 获取数据
尝试使用
cursor.setinputsizes
设置列的数据类型。并将 VARCHAR 类型设置为 INT、FLOAT 或 DateTime 列。
cursor.setinputsizes([(pyodbc.SQL_VARCHAR, 50, 0), ...])
这个错误是由
fast_executemany = True
造成的。这与记忆的一些魔法有关。这些文章将有所帮助 https://github.com/mkleehammer/pyodbc/issues/520,https://github.com/mkleehammer/pyodbc/pull/729