Add yaml to the standard library

[invisibleroads (Roy Hyunjin Han)]

Should we add yaml to the standard library? Then it can be used in environments that do not permit third-party packages.

Personally, I prefer the ruaml.yaml implementation by Anthon van der Neut because it is more feature complete, though the PyYAML implementation is more popular.

Please see here for an earlier discussion.

Does anyone want to talk about this at PyCon 2024?

[elis.byberi]

Hi @invisibleroads ,

Welcome to the Python forum! I’ve moved this thread to Help because, in my opinion, it doesn’t fit the Idea category. It’s not clear why the library is being proposed to be added. What are the other candidate libraries? Are there similar libraries for other formats that have been added to the standard library, etc.?

@invisibleroads

Does anyone want to talk about this at PyCon 2024?

There is another category for this type of question, perhaps Events.

[sinoroc]

@invisibleroads I guess it might make sense for you to read up on the whole process that lead to getting toml in the standard library. A lot of the questions and steps will be the same for yaml.

PEP 680: "tomllib" Support for parsing TOML in the Standard Library

Some previous discussion at: Adopting/recommending a toml parser? - #84 by brettcannon Issue 40059: Provide a toml module in the standard library - Python tracker [Python-Dev] Adding a toml module to the standard lib? Mailman 3 Python standard library TOML module - Python-ideas - python.org Please consider pushing tomli into stdlib · Issue #141 · hukkin/tomli · GitHub

[fungi]

Should we add yaml to the standard library? Then it can be used
in environments that do not permit third-party packages.

I suggest finding and revisiting all the discussion that transpired
in the lead-up to PEP 518, which got summarized at
PEP 518 – Specifying Minimum Build System Requirements for Python Projects | peps.python.org eventually (but the
original discussion was long and heated). YAML was initially under
consideration, and a big reason we eventually ended up with TOML
instead is that none of the existing YAML libraries were suitable
implementations for inclusion in the CPython standard library. The
reasons for that are almost certainly still the same today.

[tjreedy (Terry Jan Reedy)]

The core devs can barely and in some cased inadequately maintain what we already have. We don’t want much more.

@invisibleroads

Then it can be used in environments that do not permit third-party packages.

This is not our fault. Talk to your environment manager. There are sumo distributions maintained by others that include many things besides the stdlib vetted and selected by the distribution maintainers.