🌱Seeding

Seeding means populating the database with records. We do this in development to show what the web app looks like with data.

We do this in one script file: seeds.exs

seeds.exs
alias PetalPro.Accounts.User
alias PetalPro.Accounts.UserToken
alias PetalPro.Logs.Log
alias PetalPro.Accounts.UserSeeder
alias PetalPro.Orgs.OrgSeeder
alias PetalPro.Orgs.{Org, Membership, Invitation}

if Mix.env() == :dev do
  PetalPro.Repo.delete_all(Log)
  PetalPro.Repo.delete_all(Invitation)
  PetalPro.Repo.delete_all(Membership)
  PetalPro.Repo.delete_all(Org)
  PetalPro.Repo.delete_all(UserToken)
  PetalPro.Repo.delete_all(User)

  admin = UserSeeder.admin()

  normal_user =
    UserSeeder.normal_user(%{
      email: "user@example.com",
      name: "Sarah Cunningham",
      password: "password",
      confirmed_at: Timex.now() |> Timex.to_naive_datetime()
    })

  org = OrgSeeder.random_org(admin)
  PetalPro.Orgs.create_invitation(org, %{email: normal_user.email})

  UserSeeder.random_users(20)
end

We use the faker library for randomizing data:

def random_user_attributes(attrs \\ %{}) do
  Enum.into(attrs, %{
    name: Faker.Person.En.first_name() <> " " <> Faker.Person.En.last_name(),
    email: unique_user_email()
  })
end

Seeds should be regularly updated with any new tables you create. This allows you to better test and develop your app.

Last updated