Python dataframe value insertion to database table column

Question

Is it possible to insert python dataframe values to database table column?

I am using snowflake as my database.

CommuteTime is the table which contains the StudentID column. "add_col" is the python dataframe. I need to insert the df values to StudentID column.

Below is my code which i have tried to insert df values to table column.

c_col = pd.read_sql_query('insert into "SIS_WIDE"."PUBLIC"."CommuteTime" ("StudentID") VALUES ("add_col")', engine)

When I execute the above its not accepting the dataframe. Its throwing the below error.

ProgrammingError: (snowflake.connector.errors.ProgrammingError) 000904 (42000): SQL compilation error: error line 1 at position 68
invalid identifier '"add_col"' [SQL: 'insert into "SIS_WIDE"."PUBLIC"."CommuteTime" ("StudentID") VALUES ("add_col")'] 
(Background on this error at: http://sqlalche.me/e/f405)

Please provide suggestions to fix this..

Answer 1

You cannot make it with pd.read_sql_query .

First, you need to create Snowflake cursor .

eg

import snowflake.connector

cursor = snowflake.connector.connect(
user='username',
password='password',
database='database_name',
schema='PUBLIC',
warehouse='warehouse_name'
).cursor()

Once you have a cursor, you can query like this: cursor.execute("SELECT * from "CommuteTime")

To insert data into tables, you need to use INSERT INTO from Snowflake

Please provide more info about your dataframe, to help you further.

Answer 2

I was only able to do that using SQL Alchemy , not Snowflake Python Connector.

from sqlalchemy import create_engine

# Establish the connection to the Snowflake database
sf = 'snowflake://{}:{}@{}{}'.format(user, password, account, table_location)
engine = create_engine(sf)
# Write your data frame to a table in database 
add_col.to_sql(table_name, con=engine, if_exists='replace', index=False)

See here to learn how to establish a connection to Snowflake by passing username , password , account , and table location .

Explore here to learn about the arguments you can pass as if_exists to the function to_sql() .