Heroku Connect : Sync Heroku app with Salesforce using Python Flask

  1. Introduction
  2. Prerequisites
  3. Clone the Source Code
  4. Heroku Login
  5. Requirements File
  6. Procfile
  7. DB Initialization
  8. Flask Controller
  9. Deploying to Heroku
  10. Add PostgreSQL Add-On
  11. Add Heroku Connect Add-On
  12. Configure Heroku Connect Add-On
  13. Home Page
  14. Contact List
  15. Create a New Contact
  16. Optional Step Show Contacts Locally
  17. Summary


This workshop shows how to Create and Run a Python app with psycopg2 which uses PostgreSQL based Heroku Connect

Figure 1 show how the HerokuConnect Add-On interacts with Heroku Postgres and force.com behind the scenes. Make sure you have Python installed. Also, install the Heroku Toolbelt

We will be using a basic Flask app available at flask_psycopg2_v2

This app has four rest endpoints

@app.route('/') which displays Hello world
@app.route('/contacts') which fetches list of contacts from Postgres table salesforce.contact
@app.route('/create_contact', methods=['POST','GET']) which creates a new contact
@app.route('/contactform') which serves an HTML Form for entering new contact details


This workshop assumes you have following setup

Clone the Source Code

$ git clone https://github.com/rajdeepd/flask-psycopg2-v2

Heroku Login

First download CLI, install it and run the following command.

$ heroku login

Requirements File

Note : This section is for information only

Notice that the Requirements file already exists, this will be used by Heroku to setup the dynos

    $ cat requirements.txt


Note : This section is for information only

There is already a Procfile which tells the Heroku what kind of Dyno is required and the source for the application.

    web: gunicorn app:app --log-file -

DB Initialization

Note : This section is for information only

We will parsing the DATABASE_URL environment variable to connect to PostgreSQL database. The PostgreSQL Python driver is already configured in the requirements file.

Steps are

  1. Parse DATABASE_URL into variable url
  2. Create a connection string db from the parsed url.
  3. Create a Database connection conn.
  4. Open a Database cursor cur.
url = urlparse.urlparse(os.environ.get('DATABASE_URL'))
db = "dbname=%s user=%s password=%s host=%s " % (url.path[1:], 
     url.username, url.password, url.hostname)
schema = "schema.sql"
conn = psycopg2.connect(db)
cur = conn.cursor()

Flask Controller

Note : This section is for information only

app.py is the main controller for our applications and code listing below shows the implementation of various use cases

List Contacts

def contacts():
        cur.execute("""SELECT name from salesforce.contact""")
        rows = cur.fetchall()
        response = ''
        my_list = []
        for row in rows:

        return render_template('template.html',  results=my_list)
    except Exception as e:
        return []

Create Contacts

Implementation of /createcontactform endpoint

def contactform():
   return render_template('contactform.html')

Implementation of /create_contact endpoint.

@app.route('/create_contact', methods=['POST','GET'])
def create_contact():

        if request.method == "POST":
            first_name = request.form["first-name"]
            last_name = request.form["last-name"]
            email = request.form["email"]

            statement = "insert into salesforce.contact(firstname, 
                        lastname, email) values ('" \
                + first_name + "','" + last_name + "','" + email + "');"
            errors = []
            return render_template('result.html', errors=errors, 
    except Exception as e:
        return []

Deploying to Heroku

Before moving on, create a Heroku account and run $ heroku login command to login to your created heroku account.

$ heroku create
$ git push heroku master
$ heroku open

Add PostgreSQL Add-On

Add Postgress Add-On as shown below.

  $ heroku addons:create heroku-postgresql:hobby-dev

Add Heroku Connect Add-On

Configure Heroku Connect Add-On. Command below configures Herok-Connect Add-On to the application.

  $ heroku addons:create herokuconnect

Configure Heroku Connect Add-On

  • Setup Connection

  • Enter Schema Name : This is the schema name underwhich database will be created.

  • Trigger OAuth

  • Enter Salesforce.com developer account credentials

  • Create Mappings

  • Create Mappings Contacts : Choose the fields in Salesforce Schema which need to be mapped to Postgres Database in the application.

  • Write Enable : Make sure you enable Write to Salesforce any updates to your database check box

  • Explore Contacts in the Dashboard

Home Page

Contact List

Browse to URL http://{your-app-name}.herokuapp.com/contacts to see the list of contact names.

Create a New Contact

Browse to URL http://{your-app-name}.herokuapp.com/createcontactform to see the list of contact names.

Optional Step Show Contacts Locally


  • Python 2.7
  • pip
  • virtualenv
  • PostgreSQL client (Optional if you want to run the application locally)
  1. Install Virtual Environment

    Go to the application Folder flask-psycopg2-sample and install a virtual environment in it.

    $ cd flask-psycopg2-v2
    $ virtualenv venv
    $ source venv/bin/activate
  2. Install Dependencies

     $ pip install flask gunicorn psycopg2
  3. Configure the DATABASE_URL in the local environment

      $ heroku config
      === fast-sands-40695 Config Vars
      DATABASE_URL:      postgres://<user_name>:<password>@<ipaddress>.compute-1.amazonaws.com:5432/<database_name>
  4. Export DATABASE_URL

     $ export DATABASE_URL=postgres://<user_name>:<password>@<ipaddress>.compute-1.amazonaws.com:5432/db

    Open the following URL http://localhost:5000/contacts you should be able see the contacts.

  5. Run the app using the following command

     $ python app.py

Your app should now be running on localhost:5000


In this workshop we learnt how to configure a Python Flask Application to work with Heroku Connect. We used Psycopg2 driver for talking to the PostgreSQL database deployed on Heroku.


Execute using PyCharm