Rake task to drop and reseed database in Rails

When working on a Rails app, you might sometimes need to drop the local database and start fresh with data loaded from db/seeds.rb, what do you do?

Short answer: use rake db:reset. This drops the database, then loads the schema with rake db:schema:load and then seeds the data with rake db:seed. This is what you should be using for the vast majority of cases.

However, during very early stages of development, I often edit the existing migration files instead of creating new ones. This is useful when I’m still figuring out the schema, and often need to add fields to existing models.

In such cases, we need to run db:migrate instead of db:schema:load when we reset the database. For this, we create a new task, db:reseed.

# lib/tasks/db.rake
namespace :db do
  desc 'Drop, create, migrate then seed the development database'
  task reseed: [ 'db:drop', 'db:create', 'db:migrate', 'db:seed' ] do
    puts 'Reseeding completed.'
  end
end

This describes a db:reseed task which is dependent on the 4 other tasks. The dependencies get executed first and then the given block. The block is optional, and since it only prints a message here, I usually remove it.

The reason db:reset uses db:schema:load and not db:migrate is that the former is much faster. It loads the schema that we’ve already generated from db/schema.rb instead of going through all the migrations again. But when we change existing migration files, we need to run the migrations again to update the schema file.

Editing migration files is not a great idea, especially when you’re working in teams, but I find this works fine in the first few days of development when you don’t yet have a public version of the app in production. Once you’ve got the app in production, you should definitely not be editing migration files.