Configuration Files in Python

Do you even need config files?

If a developer has a module in their project that connects with an FTP server to download some files, then the developer would write a method to connect to FTP via FTP URL and use credentials like username and password for successful connection. If the developer uses these credentials in the code and hardcodes them in their code file and deploys the application, it could work fine, job done! Now, imagine that after two months the password for that FTP site is changed and the developer must again update that password in your application to make sure it does not break the existing functionality of the FTP connection. Now in this scenario, the developer would again change the code and redeploy it. Here, a developer must for a small configuration change, take the latest of the code, do the required change, make sure nothing else breaks, re-deploy the application and then test it. And this could be a repetitive task after 2~3 months when some configuration again changes. What if the application has a configuration file that has key-value pairs as “User”: “<username>”, “Password”: “password” and whenever some change is needed only that file is touched and configurations are updated rather than digging into the actual code.

Why config files?

Coming from a .Net background, I found it a bit challenging in figuring out how a developer can have a configuration file in the Python application which could be used to read setting values and one does not have to even touch the code to update or save settings. Config files are used to store key-value pairs or some configurable information that could be read or accessed in the code and at some point, of time. If that configuration changes, developers can just change the configuration in that config file and not worry about changing the code as it may require recompiling the code and deploying it. Not only this but using config files make your settings and code more reusable and keeps the settings information at a centralized location and segregated. Of course, sensitive information like passwords, secrets, and certificates should be kept more secure, maybe in cloud vaults. But basic settings used in the application could be part of a configuration file.

Microsoft.Net vs Python config files

Microsoft .Net applications sometimes by default provide files like appSettings, web.config, app.config when you start with a project template, based on what kind of application you are creating. These files serve the purpose of storing settings information.

Python applications do not by default provide a settings file, however, you can use an existing one and modify it. But the best way is to create your own from scratch and use it as needed. Read more...