Python Web Development: Cost-Effective for Business, Convenient for Developers. KT.Team Experience

In this article, you will learn more about our experience with Python: what products we build in this programming language, why we choose it for these tasks, and what benefits our development clients get from it. We will be helped by: - Egor, Senior Software Engineer; - Anton, Team Lead, Partner; - Anton, Project Manager, Executive Director, Partner.

Accelerated MVP development. With the growth of agile approaches and the philosophy Agile development cycles are getting shorter and shorter. Software engineers do not need to spend years building projects in isolation while referring to a thousand-page specification. Development begins with creating an MVP.

The priority is to launch a minimum viable product within one or two months and get feedback from real users (you can read more about MVP development in our blog article: "MVP, or how to avoid endless developmentFor fast MVP development, Python is an excellent choice. Its speed comes from a syntax that is as close as possible to natural English.

Speed is influenced by the large number of open-source libraries and frameworks, as well as minimalist microframeworks such as Flask. They are designed for different needs, have a large and friendly community, and attract developers with the simplicity and flexibility of their syntax. One caveat: this high speed is maintained only when the project is relatively simple (small and medium e-commerce, CRM, and straightforward integrations).

If development is for enterprise projects, then all work takes roughly the same amount of time regardless of the technology stack. It does not matter whether you write in Java, PHP, or Python, because the total number of hours will not differ significantly. At the same time, the time can be distributed differently across types of work. One stack may require more time for refactoring, another for the MVP, and a third for testing, and so on.

For example, to work quickly in Python, you need to think through the architecture in advance and prepare the environment properly. Also, you cannot judge the complexity of the work by the number of lines of code or how fast it is written. A minimal amount of code means high quality, and for a developer to fit a task into three lines, they must carefully think through how to do it, where in the program to place the code, and so on.

Easy integration with other technologies and software. Today, any software product requires many integrations: with various information systems used by the company and its partners, with banking and marketing systems, and many others. Python is well suited for integrations and can control different software components, including those written in another language. It is versatile across all platforms.

This was one of the reasons why we chose Python specifically to build the print service API in one of our current projects. It works very well with the Linux CUPS print server, and from Python it is easy to access the OS kernel to perform the required operations. Scalability and the ability to add new features. This advantage does not happen on its own; it requires a well-thought-out project architecture and clean code.

It can be said that with Python and its simple, clear syntax constructs, it is easier to write code that scales well than in most other languages. This was important to us when we chose a stack for managing print services. With Python, we did not have to change the main print service module when its parts changed.

Most Python-specific features are visible only to developers and create difficulties for them, not for clients or product users. Code execution speed is probably the most painful topic for Python developers. Python is considered a fairly slow programming language, which is due to the structure of its interpreter and its dynamic nature.

To improve performance, developers either rewrite the slow part of the code in C or Cython, a Python-to-C translator, or use PyPy, an alternative Python interpreter whose name stands for Python written in Python. Developers often say that you can write anything in Python, even Python itself. In principle, this is normal in today's practice: the Rust compiler is also written in Rust, and the Golang compiler in Golang.

PyPy updates lag slightly behind updates to the main Python interpreter. For example, if the current Python version is 3.8, PyPy is closer to version 3.5 with some features from 3.6.

The following principles can be used to speed up Python development: - Python has many libraries, so you need to choose one that is suitable and proven for solving similar tasks, since some of them are already written in C and perform very well; - to speed up a small problematic part of the code, you can rewrite it in Cython, a translator and intermediary layer for Python in

C/C++); - to speed up large parts of code written in Python, PyPy can help if the code is not sensitive to the language version (enterprise-level Django projects often start on PyPy).

PyPy is powerful because of JIT compilation (just-in-time compilation). Functions in the program are compiled into machine code on the fly, speeding up the program for future use. You do not need to rewrite the code, you just need to keep the version downgrade in mind. One-line anonymous functions and pipelines. In Python, anonymous functions, or lambda functions, are single-line.

Developers sometimes miss multiline lambdas, but they will never appear because that would go against Python's philosophy. Pipelines are the sequential execution of functions with subsequent transfer of data from the previous step. In Python, the sequence has to be written roughly like this. _a = func1() a = func2(a) a = func3(a) result = func4(a)___ In some other programming language, say Elixir, this would look simpler.

func4()___ This does not affect code performance, but it causes some inconvenience for developers by making it harder to follow functional programming methodology.

Python is ideal for projects in CV & ML (computer vision and machine learning), analytics, corporate websites, and corporate training systems. It is well suited for building small microservices and short-lived projects when a client needs to launch an MVP quickly and test hypotheses. Yes, this versatile language has many use cases, but there are also cases where another stack is a better choice.

For example, Magento (PHP) may be preferable for developing high-load online stores. Whether Python should be used in your project should be decided after a joint discussion between developers and clients, based on an analysis of the product characteristics and business goals.