Skip to Content

A Flask application on Azure App Service that queries HANA, express edition.


A Flask application on Azure App Service that queries HANA, express edition.

By csoane

A How-To that demonstrates how to create a Flask/Python application for Azure's App Service that queries a HANA, express edition database.

How-To Details

This How-To shows how to create a web application with Flask for Azure’s App Service that queries a HANA, express edition database. It requires that you have a HANA, express edition database that is accessible from Azure’s App Service.

Step 1: Setup a HANA Database Accessible from Azure's App Service

Make sure that you have the HANA, express edition database running, and that it is accessible from Azure’s App Service. For instructions on how to perform a binary installation of HANA, express edition on an Azure virtual machine see the Installing SAP HANA, express edition, on Microsoft Azure and openSUSE.

Please log in to access this content.
Step 2: Update Application to Access HXE

You will now modify the Flask application to enable it to query the HANA database and displays an html page containing the contents of the HANA M_DATABASE view. Your application will replace the original and templates/hello.html source files:

`` - the python source file. The file contains the application code that queries the HANA Express database, places the results in a data structure then merges that data with `template.html`.

`templates/hello.html` - An HTML formatted template that uses the flask rendering functionality to merge data generated results from `` to create the html page. `hello.html` must be in the `templates` directory.

Follow procedure defined by Azure to create an application using the Flask web framework. Do not configure Git publishing for your newly created web app as instructed in the Azure Doc Tutorial step #6. You will clone the generated Git repository later to have access to the application code.

Azure Python Web App - follow steps 1-5

Confirm the application framework is working by clicking the Browse link in the Application Services Overview window or follow the link provided in the URL field.

Results in Browser


Clone the generated application git repository using the provided Git clone url in the Application Services Overview window:

git clone <Git clone url from Application Services Overview>

Navigate to the application directory to find the files cloned from the Git repository, if you don’t find the applications including the requirements.txt your clone command didn’t work, verify the Deployment Credentials, or Deployment options, you may need to Disconnect and setup a new Git Local repository:

Results from <code>ls</code> in application folder


Edit the requirements.txt file to have the following packages:


Delete the file .skipPythonDeployment in the application directory, this is required to assure Azure does a full deployment of the application.

Navigate to the folder named FlaskWebProject1 and modify the file to include the following code. Update the HXE host IP Address and HXE system DB password in the below code:

    # [START application]
    import logging
    from flask import Flask
    from flask import render_template
    import pyhdb

    app = Flask(__name__)

    def hello_world():
      # define connection to the HXE database
      user_provided_host='<HXE host IP Address>'
      user_provided_password='<HXE system DB password>'
      connection = pyhdb.connect(
              host = user_provided_host,
              port = user_provided_port,
              user = user_provided_user,
              password = user_provided_password
      if not connection.isconnected():
          return 'HXE Server is not accessible'
      cursor.execute ("SELECT * from M_DATABASE")
      allrows = ''
      row = cursor.fetchone()
      while row is not None:
          print (row)
          allrows = allrows + str(row)
          row = cursor.fetchone()
      if row is None:
          return render_template(
                  host = user_provided_host,
                  port = user_provided_port,
                  user = user_provided_user,
                  name = allrows)
          return 'Select failed'

    def server_error(e):
        #log the error and stacktrace
        logging.exception ('an error occurred during a request.')
        return 'An internal error occured.', 500

    if __name__ == '__main__':'', port=8080, debug=True)
    #[END application]

Delete the file, this file is not needed and will conflict with the new code in

Navigate to the templates folder and create a new file hello.html with the below content:

<!doctype html>
<title>HXE Select from HXE</title>
{% if host %}
<h1>IP Address: {{ host }}</h1>
{% else %}
<h1>IP Address missing</h1>
{% endif %}
{% if port %}
<h1>HXE Port Number: {{ port }}</h1>
{% else %}
<h1>HXE Port Number missing</h1>
{% endif %}
{% if user %}
<h1>HXE User Name: {{ user }}</h1>
{% else %}
<h1>HXE User Name missing</h1>
{% endif %}
{% if name %}
<h1>M_DATABASE contents:</h1>
  {% for n in name %}</h1>
  {% endfor %}
{% else %}
<h1>M_DATABASE contents missing</h1>
{% endif %}

Add the hello.html file to the git repository:

git add hello.html

Commit all of the changes to the repository:

$ git commit -a -m "updates for HXE application"

Push the updated files to Azure to the Git repository:

git push
Please log in to access this content.
Step 3: Test Flask Application Test Your Application to Azure's App Service
  1. Click the Browse link on the Application Services Overview Window to view the web application. Example URL will be something like this:
    http://<your application name>

  2. After you push your code successfully to Azure using Git, go to the Azure dashboard and select the URL for your application. When you do so, you should see results from the select of the M_DATABASE table on your HANA, express edition like the following:

Results in Browser


Please log in to access this content.

Next Steps

Updated 08/04/2017

Time to Complete

30 Min.



  • A HANA, express edition database server must be running and accessible from Azure App Service.
  • Familiarity with the Python language.
  • Familiarity with Azure App Service(AAS).
  • An account on Azure. If you do not already have an Azure account, you can go to the Azure Portal Page and set one up.
  • Git installed on your client system. You can follow these instructions to download git.
  • Follow the Azure procedure to create a simple Python web service application, Azure Python Web App

Next Steps

  • Complete other tutorials.
Back to top