如何适配Spyke模型？如何扩展User模型支持user.company.name？

Hey there! Let's get your Spyke User model set up to handle nested company.name calls smoothly. There are two clean approaches you can take depending on your needs:

1. Use Spyke's has_one Association (Recommended) #

Since company is a nested resource tied to a user, the most idiomatic way in Spyke is to create a dedicated Company model and link it via a has_one association. This keeps your code organized and aligns with Spyke's resource-oriented design.

First, create a Company model:

class Company < Spyke::Base
  attributes :name
  # Add any other company attributes your API returns here
end

Then update your existing User model to include the association:

class User < Spyke::Base
  attributes :name, :email, :phone, :password, :errors
  has_one :company
end

That's it! If your API returns a JSON structure where the user object includes a nested company object (like below), Spyke will automatically instantiate a Company instance for you:

Example API response snippet:

{
  "user": {
    "name": "Jane Doe",
    "email": "jane@example.com",
    "company": {
      "name": "Tech Corp Inc."
    }
  }
}

You can now call user.company.name directly to access the company's name.

2. Define a Nested company Attribute (No Separate Model) #

If you don't want to create a separate Company model, you can simply add :company to your User's attributes and access the nested name directly. This works if you only need to read the company name and don't need additional company-related logic.

Update your User model like this:

class User < Spyke::Base
  attributes :name, :email, :phone, :password, :errors, :company

  # Optional: Add a helper method for safer access (handles nil company cases)
  def company_name
    company&.fetch(:name, nil)
  end
end

With this setup, you can either call user.company[:name] or use the helper method user.company_name to get the company name.

Both methods will let you access the nested company name—pick the one that fits how you plan to work with company data long-term!

内容的提问来源于stack exchange，提问作者Hatem Khattab