如何在python中使用sqlalchemy在查詢中創建sql server表變量

Question

我正在嘗試在SQL Server中創建一個表變量，查詢它，並將結果返回到pandas數據幀（參見示例）。 我想這樣做，以便我可以在將數據發送到pandas數據幀之前聚合數據庫中的數據。 我記得設置NOCOUNT ON將允許這個工作，因為它不會返回執行每個查詢的任何內容。 但這不起作用。 所以這顯然是一個示例代碼，但我已經能夠在這里重新創建錯誤。 按照建議的鏈接為您提供ProgrammingErrors的記錄 。 我覺得它沒有用。

import urllib
import sqlalchemy
import pandas as pd

quoted = urllib.parse.quote_plus('DRIVER={ODBC Driver 17 for SQL Server};Server=127.0.0.1;Database=mydb;UID=myuser;PWD=mypasswd;Port=1433;')
engine = sqlalchemy.create_engine('mssql+pyodbc:///?odbc_connect={}'.format(quoted))

query = """
SET NOCOUNT ON;

DECLARE @n_majors TABLE (id varchar(9), n_majors int)

INSERT INTO @n_majors
SELECT m.student_id_fk
, COUNT(DISTINCT dc.category) AS [N majors declared]
FROM msu_db.dbo.Majors AS m
JOIN department_categories AS dc
ON dc.dept_name = m.dept_name
WHERE m.Student_Level_Code = 'UN'
GROUP BY m.student_id_fk

DECLARE @grad_category TABLE (id varchar(9), category varchar(20))

INSERT INTO @grad_category
select m.student_id_fk
, MIN(dc.category)
from Majors AS m
join department_categories as dc
on dc.dept_name = m.dept_name
WHERE m.Student_Level_Code = 'UN'
and graduated = 'CONF'
GROUP BY m.student_id_fk

DECLARE @first_category TABLE (id varchar(9), category varchar(20))

INSERT INTO @first_category
select m.student_id_fk
, MIN(dc.category) as cat
from Majors AS m
join department_categories as dc
on dc.dept_name = m.dept_name
WHERE m.Student_Level_Code = 'UN'
and graduated IS NULL
GROUP BY m.student_id_fk

DECLARE @first_semester_grades TABLE (id varchar(9), avg_grade float, std_grade float, first_Semester_seq_id varchar(4))

INSERT INTO @first_semester_grades
SELECT c.student_id_fk
, AVG(c.Grade) AS [mean grade]
, STDEV(c.Grade) AS [stdev grade]
, MIN(c.Term_Seq_Id) AS Term_Seq_Id
FROM Courses AS c
WHERE c.Student_Level_Code = 'UN'
GROUP BY c.student_id_fk

SET NOCOUNT OFF;

SELECT  s.[student_id_fk]
      ,[gender]
      ,[ethnicity]
      ,[first_course_datetime]
      ,[hs_gpa]
      ,[math_placement_score]
      ,[math_act]
      ,[natsci_act]
      ,COUNT(c.[transfer institution name]) AS [N AP courses]
      , nm.n_majors AS [n-categories]
      , fc.category
      , gc.category AS [grad category]
      , fsg.avg_grade AS first_term_avg
      , fsg.std_grade AS first_term_std
      , fsg.first_Semester_seq_id
  FROM [msu_db].[dbo].[Students] AS s
  LEFT JOIN msu_db.dbo.Courses AS c
  ON s.student_id_fk = c.student_id_fk
  AND c.[transfer institution name] = 'Advanced Placement'

  LEFT JOIN @n_majors as nm
  ON s.student_id_fk = nm.id

  LEFT JOIN @grad_category as gc
  ON s.student_id_fk = gc.id

  LEFT JOIN @first_category AS fc
  ON s.student_id_fk = fc.id

  LEFT JOIN @first_semester_grades AS fsg
  ON s.student_id_fk = fsg.id

  WHERE s.first_course_datetime BETWEEN '1993' AND '2013'

  GROUP BY s.[student_id_fk]
      ,[gender]
      ,[ethnicity]
      ,[first_course_datetime]
      ,[hs_gpa]
      ,[math_placement_score]
      ,[math_act]
      ,[natsci_act]
      , nm.n_majors
      , fc.category
      , gc.category
      , fsg.avg_grade
      , fsg.std_grade
      , fsg.first_Semester_seq_id
    """
pd.read_sql_query(query, engine)

輸出的錯誤消息如下：

     --------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
~/anaconda3/envs/research/lib/python3.6/site-packages/sqlalchemy/engine/result.py in _fetchall_impl(self)
   1081         try:
-> 1082             return self.cursor.fetchall()
   1083         except AttributeError:

AttributeError: 'NoneType' object has no attribute 'fetchall'

During handling of the above exception, another exception occurred:

ResourceClosedError                       Traceback (most recent call last)
<ipython-input-3-2a0ea765a8e2> in <module>()
----> 1 df = pd.read_sql_query(query, engine)

~/anaconda3/envs/research/lib/python3.6/site-packages/pandas/io/sql.py in read_sql_query(sql, con, index_col, coerce_float, params, parse_dates, chunksize)
    312     return pandas_sql.read_query(
    313         sql, index_col=index_col, params=params, coerce_float=coerce_float,
--> 314         parse_dates=parse_dates, chunksize=chunksize)
    315 
    316 

~/anaconda3/envs/research/lib/python3.6/site-packages/pandas/io/sql.py in read_query(self, sql, index_col, coerce_float, parse_dates, params, chunksize)
   1070                                         parse_dates=parse_dates)
   1071         else:
-> 1072             data = result.fetchall()
   1073             frame = _wrap_result(data, columns, index_col=index_col,
   1074                                  coerce_float=coerce_float,

~/anaconda3/envs/research/lib/python3.6/site-packages/sqlalchemy/engine/result.py in fetchall(self)
   1135             self.connection._handle_dbapi_exception(
   1136                 e, None, None,
-> 1137                 self.cursor, self.context)
   1138 
   1139     def fetchmany(self, size=None):

~/anaconda3/envs/research/lib/python3.6/site-packages/sqlalchemy/engine/base.py in _handle_dbapi_exception(self, e, statement, parameters, cursor, context)
   1414                 )
   1415             else:
-> 1416                 util.reraise(*exc_info)
   1417 
   1418         finally:

~/anaconda3/envs/research/lib/python3.6/site-packages/sqlalchemy/util/compat.py in reraise(tp, value, tb, cause)
    185         if value.__traceback__ is not tb:
    186             raise value.with_traceback(tb)
--> 187         raise value
    188 
    189 else:

~/anaconda3/envs/research/lib/python3.6/site-packages/sqlalchemy/engine/result.py in fetchall(self)
   1129 
   1130         try:
-> 1131             l = self.process_rows(self._fetchall_impl())
   1132             self._soft_close()
   1133             return l

~/anaconda3/envs/research/lib/python3.6/site-packages/sqlalchemy/engine/result.py in _fetchall_impl(self)
   1082             return self.cursor.fetchall()
   1083         except AttributeError:
-> 1084             return self._non_result([])
   1085 
   1086     def _non_result(self, default):

~/anaconda3/envs/research/lib/python3.6/site-packages/sqlalchemy/engine/result.py in _non_result(self, default)
   1087         if self._metadata is None:
   1088             raise exc.ResourceClosedError(
-> 1089                 "This result object does not return rows. "
   1090                 "It has been closed automatically.",
   1091             )

ResourceClosedError: This result object does not return rows. It has been closed automatically.

似乎只要NoneType對象被傳遞，它就會失敗。 我不明白的原因是為什么首先傳遞一個NoneType對象。 不應該傳遞查詢結果？

Answer 1

你在變量聲明中拼錯了表 - 它有1而不是l。 如果您認為某些工作不起作用，請先檢查您的假設。

更新：

import urllib
import sqlalchemy
import pandas as pd

quoted = urllib.parse.quote_plus('DRIVER={ODBC Driver 17 for SQL Server};Server=127.0.0.1;Database=mydb;UID=myuser;PWD=mypasswd;Port=1433;')
engine = sqlalchemy.create_engine('mssql+pyodbc:///?odbc_connect={}'.format(quoted))

query = """
SET NOCOUNT ON
DECLARE @table TABLE (id int, value float)
INSERT INTO @table VALUES (1, 2.7)
INSERT INTO @table VALUES (2, 4.5)
INSERT INTO @table VALUES (3, 1.2)

SELECT * FROM @table
"""
pd.read_sql_query(query, engine)

Answer 2

在返回查詢結果之前，必須先關閉NOCOUNT，以便從SQL Server返回正確的行受影響的消息：

import urllib
import sqlalchemy
import pandas as pd

quoted = urllib.parse.quote_plus('DRIVER={ODBC Driver 17 for SQL Server};Server=127.0.0.1;Database=mydb;UID=myuser;PWD=mypasswd;Port=1433;')
engine = sqlalchemy.create_engine('mssql+pyodbc:///?odbc_connect={}'.format(quoted))

query = """
SET NOCOUNT ON
DECLARE @table TABLE (id int, value float)
INSERT INTO @table VALUES (1, 2.7)
INSERT INTO @table VALUES (2, 4.5)
INSERT INTO @table VALUES (3, 1.2)

SET NOCOUNT OFF
SELECT * FROM @table
"""
pd.read_sql_query(query, engine)