Some of the homework checkers haven't been updated in 3+ years and are extremely stringent. If you're a beginner at school or uni, many of the automated homework checkers only work up to a certain version (ex: Python 3.7, Python 3.8). Stay at least 1 sub-version behind (use 3.9 instead of 9.10). If this is just for personal fun projects and you don’t mind them breaking, then maybe leaning towards grasshopper is better. But, if this is part of what you want to do professionally, I’d lean towards the ant side. You’ll need to look at your situation and figure out which you think is right for you. Is it better to be constantly doing a little work to prepare for the future, or is it better to let the work build up in the hope that you’ll never have to do it? It’s probably not a binary choice. So, it ends up as the old problem of the grasshopper and the ant.
#Latest python download update
It’s not certain, but the likelihood that you’ll need to update the code for some reason increases over time. Maybe you need to move the code into an environment that only supports newer versions. Maybe there is a new library that substantially improves your code that only works on the new version of Python. Maybe there is a critical security issue that you have to fix. The longer you wait, the more these small changes build up into bigger changes that will take more time.īut, why not avoid the work by just never updating? Well, you can never be sure that something won’t happen at some point to force the update. Either way this equates into some amount of work being needed to get your code working on the new version of Python. If you use less well supported packages, the changes can be more substantial. If you only use the standard library and very popular and well supported third-party packages, the changes are going to be nothing or very minimal. At least, the code itself may not corrode, but it does build up something known as “technical debt.”Įach time Python updates, there is a chance that you need to change something about your code to make it work with the new version. There was a comment from someone saying that code doesn’t rust.
I’m probably not going to know the right answer for you, but here are some factors to think about when making the decision:Īre there new features that are useful to you?Īre you working with other people, and if so when are they planning on updating?Īre there fixes for bugs or security issues that are relevant to you?ĭo the packages you use support the new version yet? The answer depends a lot on what you are using Python for and how you are using it. In contrast to this, our CI compiles daily some bits of code not written by programmers from our company, many decades ago, in C89 (that's the name of a C language standard).
#Latest python download Patch
You'll have to patch and patch and patch it until you maybe will be able to get it to work.
10 or so years ago), it's pretty much broken beyond any conceivable measure. So, even if you are very happy to try to dig up something that's fairly recent (eg. For example, pip will not work without FFI. Some essential Python tools rely on this library and will not function without. FFI library will determine a lot about your communication with operating system. If you don't support the protocol that's used today, you cannot connect. OpenSSL will determine if you can connect to, basically, anywhere on the Internet.
Mostly, because Python source code is very sloppy, but there are other reasons too, especially, dependency on third-party libraries like OpenSSL or FFI library.īoth libraries mentioned above are detrimental to you being able to use your computer for almost any work.
3.6 will fail to compile with a moderately recent GCC or CLang. Here's the sad truth: on a moderately recent Linux, you cannot build Python older than 3.7. Python doesn't have a standard, and so, at this point, most of Python code ever written, most Python version ever released are beyond unusable / useless.