Flask Tutorial Series

This page is auto-generated from the markdown of our README.md files on Github. The full code is publicly available at:

Jerhub Flask Tutorial Series on Github

Jerhub Flask Tutorial Series

Author: Jeremy Ecker

Link: Jerhub Flask Tutorial Series

Part 3: Contact

In this section, we will add a functioning contact form which will send emails via AWS SES (Simple Email Service).

Step 1: Review AWS SES documentation and acquire a verified email

In order for this to work, you will require an AWS account, and a verified SES email. It is advised to review the AWS documentation on this subject prior to proceeding with this integration. The docs can be found here: AWS SES docs. Check out the tutorials there, which should provide a decent background for getting setup to send emails with SES.

IMPORTANT NOTE: You will need AWS access keys in order to interact programmatically with SES, and while the SES documentation does a great job of showing you how to setup SES, it doesn't go into much detail about the fact that there are different types of keys which are suited for different types of uses. It is highly recommended at this point to review the documentation: AWS Authentication and Access Docs, in order to select the best option for your case.

Once you have reviewed these, head over to sending emails programmatically and take a look at the Python example there - this is what we will use as the pattern for this tutorial.

The AWS documentation is fantastic, so rather than reinvent the wheel by repeating it here, go ahead and follow their instructions to get yourself setup, then come back here when you're done and continue with step 2.

Step 2: Add AWS keys as environment variables

In the AWS SES documentation on sending emails programmatically, you may have noticed that they suggest setting up a shared credentials file. While this approach is perfectly valid, we are going to instead add the keys as enviroment variables. For this tutorial, as we progress, we will add several other types of integrations which will require secret keys. Rather than having the various credentials stored all over the place in different files (which it might be easy to forget to exclude from your version control), we are going to keep them safe in one central place: the environment variables.

If you've been following along with the previous parts of this tutorial, you should be an old-hand at this by now. If you forgot how to do it, go back and review the readme in part 1 of our tutorial. For this example, we are naming the keys like this:


Step 3: Create the Python code to handle sending emails

In the example from the SES docs, they showed us everything we need to know in order to send an email using Python and Boto3. So all we really need to do is make it clean and useable from within our Flask application. The main things this class needs to do are:

We will create a new directory to house our SES code outside of core, let's call it utilities, and put our ses_config.yaml and ses.py files there. So our directory structure should resemble this now:

        (ses_config.yml and ses.py go here)

At this point, it is a good idea to add scaffold/utilities/ses_config.yml to your .gitignore file so that the email address is not in your source control.

Create an SES configuration file

Remember in step 2 when we said we were going to put all of our details in one place (environment variables)? Well, that is true as far as sensitive private credentials are concerned. In this case however, we are going to be declaring some variables which we might want to be able to easily change later on - specifically the email address and AWS region. So in this case, it makes sense to stash them in a yaml file which we will read from our app.

Since we are configuring SES, let's call this file ses_config.yml:

version: 1
email: your_email@your_domain.com
region: your-ses-region
charset: UTF-8

Create a class to manage SES

When you read through the SES docs about sending emails programmatically (you did read them, right?) you probably noticed the import boto3 and botocore lines. Since these are not part of the python standard library, you have to install them.

Go ahead and add a line to your requirements.txt file:


Botocore will be taken care of as a dependency.

Next, make a new file ses.py. Within this file, add these imports to the top:

import os
import logging

import yaml
import boto3
from botocore.exceptions import ClientError

Then, make a line to connect to your app's logger so that if there is a problem, we can log it:

logger = logging.getLogger('scaffold')

We'll use os to read the environment variables and open the config file, and yaml to parse the config data. Create a class called Ses, and an __init__ method that accomplishes this:

class Ses():
    def __init__(self):
        Reads the configuration details from the ses_config.yml file, and the
        credentials from env vars. Then instantiates a boto3 client for SES.
            with open(os.path.join(os.path.dirname(os.path.abspath(__file__)),
                      'ses_config.yml')) as infile:
                config = yaml.safe_load(infile)
                self.email = config['email']
                self.region = config['region']
                self.charset = config['charset']

                self.client = boto3.client(
        except Exception as e:
            logger.warning(f'SES failed due to {e}')

Lastly, let's implement a method to send the email. This is patterned after the AWS example, so it will be familiar by now:

def send_email(self, subject, body, body_html, client_address) -> bool:
        Sends an email using the boto3 client and the provided details.

        success = False

            response = self.client.send_email(
                    'ToAddresses': [client_address,],
                    'Body': {
                        'Html': {
                            'Charset': self.charset,
                            'Data': body_html,
                        'Text': {
                            'Charset': self.charset,
                            'Data': body,
                    'Subject': {
                        'Charset': self.charset,
                        'Data': subject,

        except ClientError as e:

            logger.warning(f'Email sent. Message ID: {response["MessageId"]}')
            success = True

        return success

Step 4: Create a Flask contact form

Compared to the last steps, this part is pretty easy. Open the core/forms.py file, and make a few adjustments. First, we are going to need a new import from wtforms:

from wtforms import TextAreaField

Next, we need a new class based on FlaskForm:

class ContactForm(FlaskForm):
    email = StringField('Email', validators=[DataRequired(), Email(), Length(min=6, max=64)])
    name = StringField('Name', validators=[DataRequired(), Length(min=1, max=128)])
    message = TextAreaField('Message', validators=[DataRequired(), Length(min=1, max=500)])
    submit = SubmitField('Submit')
    recaptcha = RecaptchaField()

And that's it; the form is ready to use.

Step 5: Add a view for displaying the contact form

Open up the scaffold/core/views.py file. We will be needing a couple of new imports:

from scaffold.core.forms import ContactForm
from scaffold.utilities.ses import Ses

Then we need to add the route:

@core.route('/contact', methods=['GET', 'POST'])
def contact():
    form = ContactForm()

    if form.validate_on_submit():
        email = form.email.data
        name = form.name.data + ' contact form submission'
        message = form.message.data

        # Instantiate the SES wrapper.
        ses = Ses()

        # Send an email to your verified SES email address.
        email_1 = ses.send_email(subject=name,
        if email_1:  # Only send user email if we got their message.
            # Send an email to the user acknowledging receipt of their message.
            subject = 'Thanks for contacting us.'
            body = f'''
                This is an automated response confirming our receipt of your
                contact form submission. Please do not reply to this message, as
                replies are not monitored for this address. Your message will be
                reviewed by a human and we'll get back to you soon!
            body_html = f'''
                <p>Best Regards,</p>
            email_2 = ses.send_email(subject=subject,

        if email_1 and email_2:
            return render_template('contact_thanks.html')
        else:  # If either email failed, user should know.
            return render_template('email_problem.html')
    return render_template('contact.html', form=form)

Step 6: Add new templates to support the contact view

In step 5 we added the view function which you probably noticed requires three new templates:

Since you already understand creating templates and forms from part 1 of the tutorial, these will be left as an excercise. If you get stuck, you can refer to the templates we have provided in this repo.

Step 7: Add link to contact page in the navigation bar

Whew! That may have felt like kind of a lot, and you aren't wrong. Luckily, the last step in this section is easy. All you need to do is locate the navigation bar in the base.html template, and add a new list item to the unordered list with a link to your shiny new contact page. All done!

Final thoughts

In this section, we learned to implement one of the most common and fundamental functionalities in web applications, the contact form. We used AWS SES to do the heavy lifting of the email sending, and we learned a lot in the process.

Stay tuned for Part 4 in this tutorial series, where we will learn to implement another common functionality of web apps: blog posts.

Next: Part 4

Copyright © Jerhub - All Rights Reserved