why in hexpm's migrations files it use `execute` for create and drop table?-CodePudding

In etco doc, it use its independent create table statement by mix phx.gen.schema.

The command is as follows:

mix phx.gen.schema User users name:string email:string

The migration file is as follows:

defmodule Pen.Repo.Migrations.CreateUsers do
  use Ecto.Migration

  def change do
    create table(:users) do
      add :name, :string
      add :email, :string
      add :password, :string

      timestamps()
    end
  end
end

But in hexpm project, the migrations file is as follows: My question is what's the difference between them? Is etco hasn't drop table method?

defmodule Hexpm.Repo.Migrations.AddUsersTable do
  use Ecto.Migration

  def up() do
    execute("""
      CREATE TABLE users (
        id serial PRIMARY KEY,
        username text,
        email text UNIQUE,
        password text,
        created_at timestamp,
        updated_at timestamp)
    """)

    execute("CREATE UNIQUE INDEX ON users (lower(username))")
  end

  def down() do
    execute("DROP TABLE IF EXISTS users")
  end
end

CodePudding user response：

Ecto.Migration has extensive documentation, explicitly stating:

Change

Having to write both up/0 and down/0 functions for every migration is tedious and error-prone. For this reason, Ecto allows you to define a change/0 callback with all of the code you want to execute when migrating and Ecto will automatically figure out the down/0 for you.

— https://hexdocs.pm/ecto_sql/Ecto.Migration.html#module-change

For this to work, the up migration should be clean enough for ecto to figure out down counterpart.

Why is it done that way in hexpm? Well, legacy (the project was started before ecto reached any mature state,) or because authors liked explicitness, or whatever else.