Python 3.12 Now Under Feature Freeze With Beta 1 Released

[ssokolow]

In what world are rust and python competing languages

In the world of people spending what time they can spare rewriting their Python projects into Rust to gain stronger compile-time guarantees than strict-mode MyPy can provide and an ecosystem that cares more about "fearless upgrades".

[schmidtbag]

Bit of a subdued way to say "Who knows how many packages will be incompatible with this because they ripped a bunch of deprecated modules out of stdlib and the maintainers of dependent packages may not have time to do what might be a major rework when you include QA testing."

Ever since they surreptitiously announced this "Python 4.x, but try to avoid getting people mad by not bumping thee major version number" strategy, I've been glad my strategy had already shifted to "code in Rust where possible for the fearless upgrades, minimize the Python side of a Rust-Python hybrid codebase where not possible".

(Seriously. Would you still have called it GTK+ 2.x if the GTK people had decided to do their migration of various functions from GDK to Cairo without bumping the version number? As an example, Python 3.12 deletes the asyncore framework module.)

Comments like this are great examples of getting angry over your own willful ignorance.
Python and Rust target such very different markets. It's like saying "why would I buy a sedan when I could have a pickup with more room for hauling?" as if your way of doing things is the only way.
As for whether Python should have bumped up a major version, yeah, probably. Is it that big of a difference from 2.x to 3.x? No, not really.

[ssokolow]

Comments like this are great examples of getting angry over your own willful ignorance.
Python and Rust target such very different markets. It's like saying "why would I buy a sedan when I could have a pickup with more room for hauling?" as if your way of doing things is the only way.

I'm quite literally a person who used Python as primary development language since 2003 in the process of switching to Rust for stronger compile-time guarantees than MyPy can provide, easier deployment than Poetry can provide, and faster startup times than CPython can provide.

"Why would I [continue to use Python] when I could have a [language that's better in every way except availability of certain third-party libraries in the ecosystem]?"

The main things I still use Python for are PyQt GUIs (slapped onto a Rust backend via PyO3), Django (for its ecosystem), and the maturity of either Django ORM or SQLAlchemy+Alembic when I need to do SQL stuff. ...and the former two aren't a case of "Python is a good language"; They're a case of "Python and Java are the only languages with memory-safe QWidget bindings and I hate Java more" and "Django's ecosystem is more or less unique."

[schmidtbag]

"Why would I [continue to use Python] when I could have a [language that's better in every way except availability of certain third-party libraries in the ecosystem]?"

Rust is slower to get from point A to point B, and it also requires you to compile code. Python is popular because it is interpreted and because it is quick to code with. You ought to pick the right tool for the job, so if you prioritize performance, security, and compatibility, use something else. While I have nothing against Rust, people like you act like it's the only language that ever makes sense for any purpose.

[Vistaus]

(Seriously. Would you still have called it GTK+ 2.x if the GTK people had decided to do their migration of various functions from GDK to Cairo without bumping the version number? As an example, Python 3.12 deletes the asyncore framework module.)

LOL. Python already discouraged the use of asyncore since 2016. By the time 3.12 is released, that's about 8 years ago since the inital announcement. What's next: complaining that Microsoft announces 8 years in advance that the next Windows version is going to be obsolete eventually? Come on.

[Weasel]

Documentation says asyncore was deprecated in 3.6.0 released in December 2016. If someone didn't took action yet, they are never going to do it.

Point?

I'm fucking sick of trying to compile something and it fails because it can't run the dumbass python script used for building some bullshit. And why it fails? Oh, because i updated my Python, so I have to keep 3.4, 3.9 and 3.11 versions around, because of retarded monkeys who drop backwards compatibility.

[cynic]

Rust is slower to get from point A to point B

I disagree with that.

sometimes in Python you THINK you got from A to B.
...then you find yourself debugging your code in production for the next 6 months because obscure bugs continue to pop out.

in Rust when you think you got from A to B, more or less you got it.

[fluke]

I disagree with that.

sometimes in Python you THINK you got from A to B.
...then you find yourself debugging your code in production for the next 6 months because obscure bugs continue to pop out.

in Rust when you think you got from A to B, more or less you got it.

A thousand times this!

I'm in the process of rewriting a major embedded QT app at work and I can tell you that not only is the same true for that, but it is overall *easier* because we have *confidence* that the code works as it should not only because of the nature of rust, but because it's a damned sight easier to write both doc-comments, doc examples that are tested, unit tests, and CI tests (and even run those tests on the embedded hardware).

I am having a harder and harder time justifying using any other language for anything at all if rust can be used as first choice.

[oleid]

I am having a harder and harder time justifying using any other language for anything at all if rust can be used as first choice.

To be fair: rust is probably not the only language with a similar properties, however probably the most widespread one.

[ssokolow]

I disagree with that.

sometimes in Python you THINK you got from A to B.
...then you find yourself debugging your code in production for the next 6 months because obscure bugs continue to pop out.

in Rust when you think you got from A to B, more or less you got it.

Exactly my rationale. I'm tired of burning out trying to replicate the kinds of guarantees I get from Rust in Python unit tests.