使用来自 Python (pyodbc) 的 SQL 服务器存储过程

Question

我有一个存储过程，代码：

DECLARE @RC int 
DECLARE @id varchar(13) 
DECLARE @pw varchar(13) 
DECLARE @depart varchar(32) 
DECLARE @class varchar(12) 
DECLARE @name varchar(12) 
DECLARE @birthday varchar(10) 
DECLARE @grade int 
DECLARE @subgrade int 
SELECT @id = 'test' 
SELECT @pw = '12345' 
SELECT @depart = 'none' 
SELECT @class = 'GM' 
SELECT @name = 'name' 
SELECT @birthday = 'None' 
SELECT @grade = 3 
SELECT @subgrade = 2 
EXEC @RC = [my_database].[dbo].[my_table] @id, @pw, @depart, @class, @name, @birthday, @grade, @subgrade 
DECLARE @PrnLine nvarchar(4000) 
PRINT 'Stored Procedure: my_database.dbo.my_table' 
SELECT @PrnLine = ' Return Code = ' + CONVERT(nvarchar, @RC)

我如何使用此过程进行原始 sql 查询以创建帐户？我正在使用烧瓶和 pyodbc。

Answer 1

来自pyodbc documentation

To call a stored procedure right now, pass the call to the execute method using either a format your database recognizes or using the ODBC call escape format. (The ODBC driver will then reformat the call for you to match the given database.)

For SQL Server you would use something like this:

# SQL Server format
cursor.execute("exec sp_dosomething(123, 'abc')")

# ODBC format
cursor.execute("{call sp_dosomething(123, 'abc')}")

所以调用你的程序

id_ = 'test' 
pw = '12345' 
depart = 'none' 
class_ = 'GM' 
name = 'name' 
birthday = 'None' 
grade = 3 
subgrade = 2 

sql = 'exec [my_database].[dbo].[my_table](?, ?, ?, ?, ?, ?, ?, ?)'
values = (id_, pw, depart, class_, name, birthday, grade, subgrade)

cursor.execute(sql, (values))

Answer 2

对于 MSSQL 正确的格式是这样的：

SQL = 'exec sp_UpdateUserGoogleAuthenticated ''?'', ''?'''

在 SQL 查询 window 中尝试运行 MSSQL 中的存储过程，每次都会失败， () 围绕着 ?分数。如果您转义单引号，它将允许其中包含空格的变量。

Answer 3

接受的答案没有解决从存储过程中捕获 return 值的问题，可以这样做：

id_ = 'test' 
pw = '12345' 
depart = 'none' 
class_ = 'GM' 
name = 'name' 
birthday = 'None' 
grade = 3 
subgrade = 2 

sql = """\
SET NOCOUNT ON;
DECLARE @RC int;
EXEC @RC = [my_database].[dbo].[my_sp] ?, ?, ?, ?, ?, ?, ?, ?;
SELECT @RC AS rc;
"""
values = (id_, pw, depart, class_, name, birthday, grade, subgrade)
cursor.execute(sql, values)
rc = cursor.fetchval()  # pyodbc convenience method similar to cursor.fetchone()[0]

Answer 4

不要忘记在存储过程中设置 NOCOUNT ON。

Answer 5

为了清楚起见，Gord 的另一种回答是使用 OUTPUT 和命名参数（将在存储过程中定义）。

id_ = 'test' 
pw = '12345' 
depart = 'none' 
class_ = 'GM' 
name = 'name' 
birthday = 'None' 
grade = 3 
subgrade = 2 

sql = """\
DECLARE @RC int;
EXEC [my_database].[dbo].[my_sp] @RC OUTPUT, @id_=?, @pw=?, @depart=?, @class_=?, @name=?, @birthday=?, @grade=?, @subgrade=?;
SELECT @RC AS rc;
"""
values = (id_, pw, depart, class_, name, birthday, grade, subgrade)
cursor.execute(sql, values)
rc = cursor.fetchval()

Answer 6

使用连接初始化的游标，sp可以直接调用如下

sql = " exec your_SP @codemp = ?, @fecha = ? "
prm = (dict['param1'], dict['param2'])
cursor.execute(qry, params)

Answer 7

到处搜索这个解决方案后，我找不到简化版本。所有的结果似乎都使这本应该很容易做到的事情变得过于复杂。这是我的解决方案。

 import pyodbc
 import pandas as pd
 import datetime as d


  conn = pyodbc.connect('Driver=;'
                  'Server=;'
                  'Database=;'
                  'UID=;'
                  'PWD=;')


     # define parameters to be passed in and out

     quarter_date = d.date(year=2020, month=10, day=1)

     SQL = r'exec TERRITORIES_SP @quarterStart = ' + "'" + str(quarter_date) + "'"

     print(SQL)

     try:
         cursor = conn.cursor()
          cursor.execute(SQL)
          cursor.close()
          conn.commit()
    finally:
          conn.close()

使用来自 Python (pyodbc) 的 SQL 服务器存储过程

Using SQL Server stored procedures from Python (pyodbc)

python

sql-server

pyodbc