From now on my #neovim null-ls #pylint configuration uses a dynamic command for pylint. This trick allowed me to remove all the nasty stuff I was doing to work with per-project #PythonPoetry #Python #virtualenv
No more 𝘜𝘯𝘢𝘣𝘭𝘦 𝘵𝘰 𝘪𝘮𝘱𝘰𝘳𝘵... messages 🎇🥳 #programming

@timdnewman
Yea, Ive made the same mistake myself many times... When I added this one line to my .bashrc it prevents me from borking my system and makes it easier to recover when I bork my nonbase default environment. I just delete it and my #bashrc or .profile recreates it on next login. TBH #anaconda (and #virtualenv and virtualenvwrapper) should do this automatically.
@hl

Oh, great: #spaCy requires #click <8.1.0, but #flask requires #click 8.1.3. Hopefully, I won't need spaCy and flask in the same project / #virtualenv.

@b_draad @kkarhan @nixCraft

The bummer is there's not enough direct messaging to entry-level users that the first rule of #python is NEVER USE SYSTEM python.

It takes getting burned several times and then getting screamed at from a friend to start going down the virtualenv, pyenv, etc rabbit holes.

Bloggers aren't doing people any favors when they introduce users with a simple "pip install" and not a massive #virtualenv disclaimer up front either.

#PyPA #virtualenv is yet another #Python project that "optimizes" its test suite by using time-machine. Since time-machine works on CPython only, it also keeps supporting #freezegun but you are allowed to use it on PyPy only now.

I really do wonder if this "optimization" actually makes anything faster, and if it outweighs the added complexity. The complexity also falls on downstreams (#Gentoo).

But that's modern development for you! "Shiny! Must have it!"

https://github.com/pypa/virtualenv/commit/fd93dd79be89b21e6e9d43ca2dd1b02b811f6d6f

I think that's a good think, disallowing messing native packages with pip packages. Anyway, I always use virtualenv (viritualenvwrapper namely) for every separated development, or pip --user #python #pip #virtualenv

https://www.omgubuntu.co.uk/2023/04/pip-install-error-externally-managed-environment-fix

In case you still use Python 2.7, 3.5 or 3.6, please note that #virtualenv 20.22.0 dropped support for these EOL #Python versions.

As #tox uses #virtualenv under its hood, it is also affected, at least when you want to continue using the above mentioned EOL #Python versions.

If so, please have a look at our FAQ

https://tox.wiki/en/latest/faq.html#testing-end-of-life-python-versions

Well, that was fast…

So if you are concerned about being unable to upgrade #virtualenv reliably in #Gentoo without installing the old version first, please voice your concerns upstream:

https://github.com/pypa/virtualenv/issues/2554#issuecomment-1518023016

Not that they care.

Ok, so it turns out that #Python #virtualenv uses old importlib.metadata API on Python < 3.10 that merges entry points from multiple installations of the same package across sys.path.

In practical terms: when we're testing virtualenv 2.22.0 in a --system-site-packages venv on #Gentoo and 2.21.0 is already installed on the system, 2.22.0 doesn't work. It tries to load plugins defined by the old version and since they don't exist anymore, it spews AttributeErrors.

https://github.com/pypa/virtualenv/issues/2554

I've been fighting with #conda and #mamba all day long, and let's just say I'm this close to never using either ever again.

I'm usually an apologist when it comes to #Python packaging, but really that's just because #pip and #virtualenv are serviceable enough for 90% of use cases. mamba does a pretty good job of papering over the unusable mess that is conda, but it's still lipstick on a pig.

In terms of the brokenness of python packaging tools, I think its

conda > virtualenv > mamba = pip

you know, when you've been working in a #python 3.8.7 #virtualenv and it starts running 3.10 causing your application to die a horrific and error-ridden death? Oh no, only me?

I'm back up and running but there were a few maddening minutes there...especially when I reinstalled the 3.8 venv and it decided to still run 3.10. 😳

Some time ago I was wondering...

...how many #artists, either amateurs or professionals, have to run #Python #virtualenv to publish their works in the internet :blobcatjoy:​

#selfhosted

#debian #python #pip #virtualenv
Debian releases may not allow pip outside virtual environments.

Works out fine for me but.....

https://www.linuxuprising.com/2023/03/next-debianubuntu-releases-will-likely.html

No sé si te has preguntado qué son esos ficheros activate.nu y activate.fish que aparecen en los entornos virtuales #python. Sirven para configurar el entorno para nushell y fish shell, respectivamente.

Por ejemplo, en el caso de #nushell 0.74.0 y #virtualenv 20.21.0, para activar el entorno sería algo así:

> overlay use .venv/bin/activate.nu

También se podría hacer con poetry, que depende de virtualenv para la gestión de los entornos virtuales. Lamentablente, aún usa una versión vieja de virtualenv que no gestiona correctamente los entornos de nushell. Habrá que esperar a que actualice la versión que emplea.

Using /usr/bin/env python3 to point to virtual environment in 22.04 #python #2204 #cron #python3 #virtualenv

https://askubuntu.com/q/1459024/612

Tried out #nix to manage some packages on my system today that I wanted to keep from polluting my main operating system and boy.... That was an awful experience alright.

I don't care what people say about it, the system was not made for #developers who want to sit down and just get shit done.

The space is wide open for anyone wanting to develop a tool that is not a #containerization solution but is instead the equivalent of #Python's #virtualenv but at an OS level.

A few thoughts on this

Does someone here use `poetry` for python dependency management ? I am trying to replicate the environment for a module. I left it running overnight, and 10 hours later, dependencies are still unresolved ...

`pyproject.toml` is very friendly. But ease of use without actually fulling said use is, well, useless.

Go on, share your python dependency-management strategies! Anything more efficient than `poetry` will be awesome.

#python #poetry #dev #virtualenv

What do you think of #Docker as the foundation for your #ML dev environment as opposed to #conda or other #virtualenv? Pros outweigh cons?

#PythonForSciComp day 3 has finished, and it went great!

Tomorrow we switch and focus on the big picture of #ReproducibleResearch, using and reusing #python code.
- Dependency management with #pip #conda #mamba #virtualenv - keep your work organized!
- #MyBinder / #BinderHub for making code easy-to-run by others in the cloud
- Packaging and sharing Python libs
- A #PanelDiscussion where you can ask us anything
- Wrap-up and after-party

Tomorrow 9:50 EET / 8:50 CET, https://scicomp.aalto.fi/training/scip/python-for-scicomp-2022/

#python #venv #virtualenv

How to find a virtual environment without installing any packages:

`sudo find / -name "*activate" -type f`

This will look in the entire system structure and will look for the `activate` file.