In this article, we will be looking at some methods to write Pandas dataframes to PostgreSQL tables in the Python.
Method 1: Using to_sql() function
to_sql function is used to write the given dataframe to a SQL database.
Syntax
df.to_sql(‘data’, con=conn, if_exists=’replace’, index=False)
Parameters :
- data: name of the table.
- con: connection to the database.
- if_exists: if table exists or not. “replace” or “append”.
- index: True or False.
Example:
In the example demonstrated below, we import the required packages and modules, establish a connection to the PostgreSQL database and convert the dataframe to PostgreSQL table by using the to_sql() method. Finally, all rows are fetched using fetchall() method.
To access the CSV file used click here.
Python3
# import packages import psycopg2 import pandas as pd from sqlalchemy import create_engine # establish connections db = create_engine(conn_string) conn = db.connect() conn1 = psycopg2.connect( database = "Airlines_Database" , user = 'postgres' , password = 'pass' , host = '127.0.0.1' , port = '5432' ) conn1.autocommit = True cursor = conn1.cursor() # drop table if it already exists cursor.execute( 'drop table if exists airlines_final' ) sql = '''CREATE TABLE airlines_final(id int , day int ,airline char(20),destination char(20));''' cursor.execute(sql) # import the csv file to create a dataframe data = pd.read_csv( "airlines_final.csv" ) data = data[[ "id" , "day" , "airline" , "destination" ]] # Create DataFrame print (data) # converting data to sql data.to_sql( 'airlines_final' , conn, if_exists = 'replace' ) # fetching all rows sql1 = '''select * from airlines_final;''' cursor.execute(sql1) for i in cursor.fetchall(): print (i) conn1.commit() conn1.close() |
Output:
Method 2: Using execute_values() function
The execute_values() function from the psycopg2 library is used to get the postgres table of the given data frame.
Syntax:
psycopg2.extras.execute_values(cur, sql, argslist, template=None, page_size=100, fetch=False)
Parameters:
- cur – the cursor that will be used to run the query.
- sql – the query that will be run. A single percent s placeholder must be present, which will be substituted by a VALUES list.
- argslist — a list of series or dictionaries containing the query’s arguments. The template must be followed in terms of font and content.
- template – the snippet that will be merged into each item in the argslist to form the query
- page size – the maximum amount of argslist items that each statement can have.
- fetch – it’s similar to fetchall. the values can be “True” or “False”
Example:
In this example, the same CSV file is used in this method. code begins with importing packages, then we form a custom function execute_values, where the given dataframe, connection, and table name are given as arguments. The dataframe rows and values are updated into the PostgreSQL table using the execute_values() method. The defined method contains an exception handling block, if there’s no exception “execute_values() done” is printed.
Python3
# import packages import psycopg2 import psycopg2.extras as extras import pandas as pd def execute_values(conn, df, table): tuples = [ tuple (x) for x in df.to_numpy()] cols = ',' .join( list (df.columns)) # SQL query to execute query = "INSERT INTO %s(%s) VALUES %%s" % (table, cols) cursor = conn.cursor() try : extras.execute_values(cursor, query, tuples) conn.commit() except (Exception, psycopg2.DatabaseError) as error: print ( "Error: %s" % error) conn.rollback() cursor.close() return 1 print ( "execute_values() done" ) cursor.close() # establishing connection conn = psycopg2.connect( database = "Airlines_Database" , user = 'postgres' , password = 'sherlockedisi' , host = '127.0.0.1' , port = '5432' ) sql = '''CREATE TABLE airlines_final1(id int ,day char(20) ,airline char(20),destination char(20));''' # creating a cursor cursor = conn.cursor() cursor.execute(sql) data = pd.read_csv( "airlines_final.csv" ) data = data[[ "id" , "day" , "airline" , "destination" ]] # using the function defined execute_values(conn, data, 'airlines_final1' ) |
Output:
execute_values() done