I have the following rake task defined in my lib/tasks
folder:
namespace :db do
namespace :test do
task :prepare => :environment do
Rake::Task["db:seed"].invoke
end
end
end
Now, what this does is seed the test DB when I run rake db:test:prepare
. I do this because I have some basic records that must exist in order for the app to function, so they're not optional and can't really be mocked.
Separately, I have a model that uses S3 for asset storage in development and production, but I don't want it to use S3 for testing. I have set up a method in the model that changes the storage path from S3 to local if Rails.env.test?
However, this isn't working. I was wondering if the rake task was aware of what environment it was being called from, and it turns out it is NOT. I put this at the top of my seeds.rb file:
puts "Environment Check: Rails Environment = #{Rails.env}"
Sure enough, when the task runs this prints: Environment Check: Rails Environment = development
So, how can I redo this rake task so that when it's seeding the test DB it knows that it's seeding the test DB??
I was having this problem too; in my db/seeds.rb
file I have a block that creates user accounts in the development environment, but they were also being created when preparing the test environment to run rake
for RSpec or Cucumber testing, which resulted in a wall of red.
Updated: I've found that the best way to specify the environment for rake tasks is to specify the environment within the task, above all statements that need the environment to be set. So in this case:
Rails.env = 'test'
Rake::Task["db:seed"].invoke
does the job.