I’ve been branching out lately and trying to learn Python. I got a Python book a while ago, and just didn’t take the time to work with it at all. However, the Fedora Project uses Python extensively for just about everything, from the build system, to web-hosting to smolt (A really cool HW tracker in Fedora), to Yum and the system-config-* tools, so I thought there must be something great in here somewhere and I just haven’t found it yet.

I normally write a lot of PHP and Perl. I use bash for the quick and dirty also. After working through a few hours of Python, I was amazed. The biggest plus, it’s readable. There’s nothing worse than having to maintain someone else’s Perl code. It’s just not nice, even if the person was a wonderful Perl coder, I normally find it takes me quite a while to figure out exactly how/where something is occurring. With Python, at least thus far, it’s readable, it requires indenting, and it’s easy. Now that I have worked through Beginning Python, and written a few administration scripts, I feel better. I can see a real future for my usage of Python as an administrative tool.

I’m also enjoying it having me think about problems from an OO approach. In my normal administrative world, problems are isolated and often times solved by a quick script and series of scripts that run procedurally and not in OO format. The OO format hopefully will offer a bit more code reuse, or at least help me recall some knowledge forgotten in CS classes of yesteryear.

It seems like only yesterday I attended the PyCon 2005 conference, which took place right around the same time that Ruby on Rails was really beginning to gather steam. Of all the talks I attended, one in particular stands out as attracting perhaps the largest crowd of the entire show. It was titled “PyWebOff: Mapping the python web application frameworks”, which was a rather ironic name (likely purposely so) in light of the complete lack of compelling Python-based framework solutions.

This was a fairly informal talk which generated a fair amount of discussion towards its conclusion regarding what could be done in order to satisfy the growing clamor. While it’s unclear whether any specific actions were undertaken as a direct result of that talk, what is clear is that soon thereafter several impressive projects were underway, most notably Django and Turbogears. Today I came across Pylons, which based on what I’ve read on the website, looks extremely promising.

To say the least, this is a pretty amazing turnaround, and indicative of how open source communities can quickly spring to action in order to adjust to market demands.

A few years back, Jython creator Jim Hugunin decided the best way to prove the theory that the CLI was not an ideal environment for scripting languages was by actually implementing one capable of running on it. He chose Python as the test language, and set about building a .NET compatible version. The result? His version, dubbed IronPython, ran far more efficiently than he’d ever imagined. In fact, in many cases IronPython now runs faster than the C-based version!

After several years of ongoing work, version 1.0 has been released.

Download version 1.0

The history of IronPython