Set a date variable to SQL query in Python [duplic

2019-08-16 16:19发布

问题:

This question already has an answer here:

  • How to pass variable values dynamically in pandas sql query 1 answer

I want to do sql query in python. I could use cx_oracle to connection database in python:

# Build connection
conn_str = u'username/password@host:1521/sid'
conn = cx_Oracle.connect(conn_str)

Now I'm trying to retrieve data from the database by using SQL query in Python:

sql_select_statement = """SELECT * FROM TABLE 
WHERE DATE BETWEEN '20-oct-2017' AND '30-oct-2017'"""

Assume we don't know the starting date, we only have a date variable called starting_time, and its value is a datetime %m/%d/%Y. Also, ending_time is yesterday, I would like to modify my SQL query as:

sql_select_statement = """SELECT * FROM TABLE 
WHERE DATE BETWEEN '20-oct-2017' AND sysdate-1"""

df = pd.read_sql(sql_select_statement, conn)

It works and generate a new df, but how to replace '20-oct-2017' with the variable starting_time? It's inside the sql query, and it's datetime format, so general python method like 'd%' % variable doesn't work. How to solve this problem? Thanks!

回答1:

Consider SQLAlchemy to connect pandas and use the params argument of pandas.read_sql to bind variable to SQL statement:

from sqlalchemy import create_engine

engine = create_engine("username/password@host:1521/sid")

sql_select_statement = "SELECT * FROM TABLE WHERE DATE BETWEEN :my_date AND sysdate-1"

my_var = '20-oct-2017'
df = pd.read_sql(sql_select_statement, engine, params={'my_date':my_var})

Alternatively, continue to use the raw connection with parameterization:

sql_select_statement = "SELECT * FROM TABLE WHERE DATE BETWEEN :my_date AND sysdate-1"

my_var = '20-oct-2017'
df = pd.read_sql(sql_select_statement, conn, params={'my_date':my_var})