Agile Web Development Using Rails 6/

...

/

Managing Tables

Managing Tables

Get hands-on experience in managing SQLite tables in different ways.

We'll cover the following...

So far we’ve been using migrations to manipulate the columns in existing tables. Now let’s look at creating and dropping tables:

Press + to interact
class CreateOrderHistories < ActiveRecord::Migration 
  def change
    create_table :order_histories do |t| 
      t.integer :order_id, null: false 
      t.text :notes
      
      t.timestamps 
    end
  end 
end

The create_table() method in line 3 takes the name of a table (remember, table names are plural) and a block. It also takes some optional parameters that we’ll look at in a minute. The block is passed a table definition object which we use to define the columns in the table.

Generally, the call to drop_table() is not needed, as create_table() is reversible. The drop_table() method accepts a single parameter, which is the name of the table to drop.

The calls to the various table definition methods should look familiar. They’re similar to the add_column method we used previously except these methods don’t take the name of the table as the first parameter, and the name of the method itself is the data type desired. This reduces repetition.

Note that we don’t define the id column for our new table. Unless we say otherwise, Rails migrations automatically add a primary key called id to all tables they create.

The timestamps method creates both the created_at and updated_at columns, with the correct timestamp data type. Although there is no requirement to add these columns to any particular table, this is yet another example of Rails making it easy for a common convention to be implemented easily and consistently.

Options for creating tables

We can pass a hash of options as a second parameter to ...