问题
I am trying to pass date1 and date 2 to a function in python that creates a dataframe from a sql server
import datetime as dt
import pandas as pd
import pyodbc
import sqlalchemy
from pandas import Series,DataFrame
cnx=sqlalchemy.create_engine("mssql+pyodbc://Omnius:MainBrain1@172.31.163.135:1433/Basis?driver=/opt/microsoft/sqlncli/lib64/libsqlncli-11.0.so.1790.0")
def func1(date1,date2):
sqlquery = "select top 10000 * from Pretty_Txns where Pay_Date between '"+date1+"' and '"+date2+"'"
df=pd.read_sql(sqlquery,cnx)
When I try and call this function from ipython
func1(dt.date(2015,05,01),dt.date(2015,05,02))
----> 8 sqlquery = "select top 10000 * from Pretty_Txns where Pay_Date between '"+date1+"' and '"+date2+"'"
TypeError: cannot concatenate 'str' and 'datetime.date' objects
How should I call the function so that I dont get the type error.
回答1:
This should get you started. Your SQL will likely not work unless the date formats match. Print out the sqlquery and test in CLI.
Example conversion:
t = datetime.datetime(2012, 2, 23, 0, 0)
t.strftime('%m/%d/%Y')
def func1(date1,date2):
strDate1 = date1.strftime('%m/%d/%Y')
strDate2 = date2.strftime('%m/%d/%Y')
sql = "select top 10000 * from Pretty_Txns where Pay_Date > '%s' and Pay_Date < '%s' " % (strDate1, strDate2)
return sql
sqlquery = func1( DateOne,DateTwo )
#print sqlquery.
df=pd.read_sql(sqlquery,cnx)
回答2:
sqlquery = "select top 10000 * from Pretty_Txns where Pay_Date between :date1 and :date2"
df=pd.read_sql(sqlquery,cnx, params=dict(date1=date1, date2=date2))
If your driver does not support named placeholders then use question marks as before and pass in a list of [date1, date2] as the value to params.
来源:https://stackoverflow.com/questions/37798229/passing-a-date-to-a-function-in-python-that-is-calling-sql-server