i18n in 2024

[petterroea]

I am reading through Pyramid docs trying to determine how to deal with internationalization. What I find is inconsistencies and dead ends:

• pyramid-jinja2 suggests a babel-based workflow: https://docs.pylonsproject.org/projects/pyramid_jinja2/en/latest/#pcreate-template-i18n
• Pyramid suggests a getttext/lingua-based workflow: https://docs.pylonsproject.org/projects/pyramid/en/latest/narr/i18n.html
  • But lingua last had a commit 3 years ago: https://github.com/wichert/lingua
  • Lingua isn't compatible with Python 3.12: Add Python 3.12 related updates wichert/lingua#110
  • And lingva seems to be the community-approved fork: https://github.com/vacanza/lingva

A lot can change in a few years 😅

So which one is it? The pyramid_jinja2 docs refer to pcreate, which seems to be a Pyramid 1.x version of cookiecutter. The sqlalchemy cookiecutter template doesn't seem to set up anything language related at least. What is the recommended workflow for i18n in 2024?

I'm currently trying to find "correct" babel config that i can use to generate .pot files

[russellballestrini]

pyramid is a choose your own adventure. Some people don't use jinja2 at all.

Maybe put a message on the other repo too because it's often different maintainers.

I've never done this work myself so I'm keen to learn what strategy you use.