Rails Ruby on Rails

Comments and changes to this ticket

• You flagged this item as spam.
  

  Postmodern June 20th, 2010 @ 01:07 AM

  Initial patch here: http://github.com/postmodern/rails/commit/7dd0718239c6747e1a6981aed...

• 

  Ryan Bigg June 21st, 2010 @ 04:04 AM

  • State changed from “new” to “needs-more-info”

  You'll need to add tests if you want this in. I'm on the fence about this. The data I load to test things in development is exactly the same data I use in production. What use case is there for having two separate seed files?

• You flagged this item as spam.
  

  Postmodern June 21st, 2010 @ 04:15 AM

  Alright, I will get on writing tests.

  One common scenario for having multiple seeds files, is using a very large and arbitrary data-set for development, in order to test the performance of complex queries.

  The other scenario I recently ran into was designing an app that used a shared production database, where certain tables were populated by other backend python scripts; maintained by another developer. In my testing and development databases, which I controlled, I still needed to load in a canned snapshot of the data from the production database.

  Finally, this would mirror the same file organization used by config/environment.rb and config/environments.

• You flagged this item as spam.
  

  Postmodern June 21st, 2010 @ 04:21 AM

  Where would I write tests against active_record/railties/database.rake? Most of the tests in activerecord appear to be testing the ORM code.

• You flagged this item as spam.
  

  Robert Pankowecki August 9th, 2010 @ 03:42 PM

  • Importance changed from “” to “Low”

  What use case is there for having two separate seed files?

  Keeping seed.rb with data that must be in application to run properly ex. some dictionaries. Keeping data that makes writing tests easier in seeds/test.rb (works like fixtures for me). Keeping data that makes using the application easier in seeds/development.rb ex. few more users, some objects already created, permissions already set and so on...

  I hope that this feature will be implemented :-)

• You flagged this item as spam.
  

  CodeOfficer September 30th, 2010 @ 04:26 AM

  I'd also love to see an environment based solution as there are 2 ways that I tend to use seeds:

  1) seeding my development environment with lots of dummy data to flush out views
  2) seeding production on deploy with data the app needs to run

  I dont use seed files at all in the testing environment.

• 

  David Trasbo October 10th, 2010 @ 06:25 PM

  • State changed from “needs-more-info” to “wontfix”

  My personal opinion: -1

  Why not just use if statements instead?

  if Rails.env.development?
   # ...
  end
  
  if Rails.env.production?
    # ...
  end
  
  # So on, so forth
  

  If scalability is the issue, you could move environment specific seeds into a separate file and require it:

  if Rails.env.development?
    require 'db/seeds/development'
  end
  
  # So on.
  

  Will mark this as won't fix for now. It's not really my decision, but I'll reopen if necessary.

• You flagged this item as spam.
  

  Postmodern October 11th, 2010 @ 12:14 AM

  Hmm, or we could use:

  require File.join('db','seeds',Rails.env)
  

• You flagged this item as spam.
  

  james2m April 4th, 2011 @ 04:58 PM

  @postmodern If it's useful to you, I put together a little gem to do this Seedbank. It lets you have seeds for different environments.