I think I’ve figured it out.

# Create the user. Note that the (empty) `organizations` field has to be preloaded.
user_map = %{display_name: "User 1", login_name: "1@y.com", password: "password"}
changeset = User.password_setting_changeset(%User{}, user_map)
user = Repo.insert!(changeset) |> Repo.preload(:organizations)

# Do the same for the organization:
org = %Eecrit.Organization{short_name: "org1", full_name: "Organization 1"}
org = Repo.insert!(org) |> Repo.preload(:users)

# Update one of the two of them:
changeset = Ecto.Changeset.change(user) |> Ecto.Changeset.put_assoc(:organizations, [org])

# When you save this change to the user, the join table will have its foreign keys populated in both directions.
Repo.update!(changeset)

Pretty slick.

Thats a good start at documenting the many_to_many macro.
If anyone can point the way, I’m looking for some description of how to use many_to_many in CRUD operations with change_sets.

The example you give is the way I am using many_to_many also. But if we need to manage the join table manually like that, I’m not seeing how this is any different that defining has_many on both sides.

Is there a developers guide somewhere that shows usage for the many_to_many macro?

from: Ecto.Schema — Ecto v3.14.0
It gives examples on how to get data back out.

# Get all comments for a given post
post = Repo.get(Post, 42)
tags = Repo.all assoc(post, :tags)

# The comments can come preloaded on the post struct
[post] = Repo.all(from(p in Post, where: p.id == 42, preload: :tags)) 
post.tags #=> [%Tag{...}, ...]

But what I’m completely stumped about is how to add, update, and delete data from a many_to_many.

Hi everyone, sorry for being late to the discussion. If the documentation is incomplete, pull requests are always welcome! I have already improved those examples in the docs though.

I have also removed the restriction that associated entries need to be changesets. We will now convert them implicitly.