Tutorial of Python Azure features | by Kristjan Eljand | Aug, 2022

Publish your code as an API with out the normal server setup

Azure features is a “serverless resolution” which implies that you don’t have to fret in regards to the conventional server setup. It may be used for number of situations together with a) constructing an internet API, b) operating scheduled duties, c) reply to occasions like database modifications and so forth. On this tutorial, we’re going to use Azure features to publish an software as an internet API.

If you happen to haven’t used Azure features earlier than, then probably the most time-consuming step is the setting setup. I’ll recommend having the next setup:

1. An Azure account with an lively subscription. You possibly can create an account totally free.
2. The Azure Capabilities Core Instruments model 3.x.
3. Python model 3.9.
4. Visible Studio Code on one of many supported platforms (check out the supported platforms).
5. The Python extension for Visible Studio Code.
6. The Azure Capabilities extension for Visible Studio Code.

As you see, there are six predominant steps which could take as much as few hours of your time. Fortunately, you simply need to do them as soon as.

We begin this mission from the state the place we have already got the working app. Our software tells you, the place, when and what was the best electrical energy worth within the Baltic space over the last 7-days (examine the stay software).

The preliminary construction of the mission consists of the calculator.py file that features the operate for highest worth calculation and the necessities.txt file that defines two libraries: requests and azure-functions.

To begin wrapping our code into an Azure operate, we first must create the native Azure features mission. Comply with the steps under (or detailed directions by Microsoft):

1. Select the Azure icon within the Exercise bar. Then within the Workspace (native) space, choose the + button, select Create Perform within the dropdown. When prompted, select Create Perform.
2. Select the listing location on your mission workspace and select Choose.
3. Present the next data on the prompts:
   3.1 Choose a language: Python.
   3.2 Choose a Python interpreter to create a digital setting.
   3.3 choose a template on your mission’s operate: select HTTP set off.
   3.4 Present a operate identify: Enter my-calculator.
   3.5 Authorization stage: Select Nameless, which lets anybody name your operate endpoint.
   3.6 Choose the way you wish to open your mission: Select Add to workspace.

Visible Studio Code robotically creates the required information, and our mission listing will now have the construction as proven on picture 2 under:

For the time being, Azure operate spine is created however we haven’t linked our software to it.

Open __init__.py file contained in the my-calculator folder. This file contains the predominant operate that might be run as your Azure Perform. Presently, the file is stuffed with pattern code (picture 1 under). To run our calculator, we’re doing the next modifications to the file:

1. Import your operate (in our instance from calculator import highest_price_function,
2. Import json library to transform the dictionary into JSON format (a regular output format of REST APIs).
3. Exchange the code contained in the predominant operate with your individual logic:
   a. Run the operate and save the outcomes.
   b. Convert the end result (our result’s a dictionary) to JSON.
   c. Return the JSON end result.

To begin the operate regionally, press F5 or the play icon. If all the things goes nicely, you’ll see the URL to the endpoint in your terminal window:

Copy this URL and paste it into our browser or just click on on it proper inside VS Code:

We are actually able to deploy the operate to Azure. Comply with the steps under (or official directions by Microsoft):

1. Signal-in to Azure (if wanted),
2. Open Azure extension, choose Create useful resource and seek for “Create Perform App”.
3. Fill the prompts:
   3.1 Choose subscription — Select the subscription (you received’t see this if there is just one subscription).
   3.2 Enter globally distinctive identify — Enter the identify of your operate.
   3.3 Choose a runtime stack — Select the language model which you might be utilizing in your native machine.
   3.4 Choose a location for brand spanking new useful resource — Select some area close to you.
4. Within the WORKSPACE part of the Azure extension choose the mission folder and click on on Deploy.

The applying is now printed and you may share it with anyone who’s to make use of it (examine the stay software of this tutorial).

To summarize, Azure features is unquestionably a instrument to think about if you wish to share the prototype of your knowledge software to the world:

1. Documentation & tutorials — Superb
2. Preliminary setup complexity — Medium
3. Growth complexity — Low-to-medium
4. Deployment complexity — Low
5. Authorization prospects — Excessive
6. Authorization complexity — Low
7. Pace & scalability — Excessive stage
8. Monitoring — Extremely purposeful out-of-box
9. Pricing — Nearly free (for prototyping)

You possibly can learn my full analysis of Azure features from this publish. Have enjoyable prototyping!