How to connect to mainframe from Python?

Python for z/OS

As any programmer will tell you, not all computer languages are created equal. Ease of use—an important consideration for beginning and advanced programmers alike—can vary dramatically between languages. In the war of the computer languages, certain champions have emerged. Python, with its rigorous programming philosophy (which includes such maxims as “there should be one—and preferably only one—obvious way to do it”) and its emphasis on human readability, has become one of the most popular programming languages of the modern computing age. Python advocates wield many fine arguments to persuade users to join their ranks, but chief among these is that its intuitive take on the object-oriented approach makes it not only straightforward to decipher and easy to adapt, but also easy to reuse—a paramount concern of the computing world.

Python offers a unique value proposition for mainframe programming. The benefits of Python generally are well-documented, but any mainframe programmer struggling with mainframe programs, written in murky languages like assembly, will particularly appreciate the transparency of Rocket’s ported Python for IBM® z/OS®. The programmer working with mainframe environments will also appreciate Python’s user community. Its wide use in academia means that Python offers a vast talent pool. That talent pool enthusiastically supports Python and constitutes a thriving, helpful source.

Python for z/OS makes maintenance easier

One of the best arguments you’ll in hear favor of Python is the astounding productivity improvements it offers. Developers can be far more productive in Python than in languages like C, C++, or Java. Programs written in Python are often significantly shorter than programs written in other languages, usually one-fifth to one-third the size. 

This reduction in size translates to programs that are easier to read, debug, and maintain.

Python includes a vast array of support libraries, including its well-furnished standard library. Python’s standard library can facilitate such common application programming tasks as text pattern matching, or file and directory access. Python also works with a number of helpful third-party extensions that allow it to do just about anything. The NumPy library, for instance, is one of Python’s best known extensions. NumPy is a free and powerful numeric programming system used by professional programmers and scientists around the globe. The popularity of NumPy and its sister library, SciPy, demonstrates Python’s significance outside the classroom.

Python is also considered a great “glue language.” You can take old code, wrap it in Python, and have it up and running again—but this time with the enhanced functionality and readability of Python. Python works particularly well with code written in the C family: Python can invoke C and C++ libraries and can be called by C or C++ programs. Python also works well with many other popular programming tools and standards, including Java, .NET, COM, Silverlight, SOAP, and CORBA. 

Python has built-in interfaces to many operating system services, and it's ideal for writing code for system-administration tasks. It can handle databases, too: there are Python interfaces to all of the most popular relational databases. Python maintains a major presence in the online world, where web development framework packages like Django make it easy to construct websites. Instagram is written in Python, as is much of Google. Few programming languages have such faithful congregations as Python has, meaning that for any programming task you can think of, chances are someone has made Python an option.

Rocket Software brings even more functionality to z/OS

Our ported Python for z/OS is just one example of Rocket Software's wholehearted commitment to democratizing mainframe development. It's one of 30+ open source languages and tools Rocket has released for IBM z Systems® mainframes, each of which is absolutely free. Python for z/OS makes Python’s modern, object-oriented approach available to mainframe programmers, adding versatility and flexibility to the mainframe environment. 

Rocket has modified these languages and tools to operate on IBM z13® and z13s® systems in order to extend the main programming environment and make it friendly for programmers who lack conventional mainframe language skills. All 30+ tools are supported on z/OS 1.12 and up.

If you're going to be using ISPF then you will probably be screen scraping. This means you will be interacting with a 3270 emulator, possibly via HLLAPI.

An alternative, which may be available to you, is to use SSH. Many mainframe systems are now accessible via this protocol. If Python supports it, and your mainframe people allow it, SSH may be easier.