mbleigh / seed-fu

To seed the production database, sometimes we need to execute Rails or plug-in code in addition to creating a whole bunch of instances and assigning values to their variables.

For example, I use awesome_nested_set (http://github.com/collectiveidea/awesome_nested_set/tree/master). I would like to be able to do something in my code like this:

Thing.seed do |s|
  s.id = "2"
  s.save
  s.move_to_child_of(1)
end

I'm using postgres for my database. When seed records are created the primary key sequence should be updated - see http://railspikes.com/2009/3/6/duplicate-key-violates-unique-constraint for an explanation.

Currently I do rake db:seed in my test helper to make sure the test database is populated with the necessary static data. It would be great to be able to call SeedFu.seed or something instead.

The new (bare bones) seed data feature included with Rails 2.3.4 also uses a rake task of db:seed. This conflicts with the task used by seed_fu.

Suggest that the task be renamed db:seed_fu.

If I use the following code to make a model read-only:

class Permission < ActiveRecord::Base
    def readonly?
        return true
    end
end

then its data doesn't seem to be able to be seeded (ActiveRecord::ReadOnlyRecord is thrown).

I suppose it makes sense really, but I wondered if there's a way to make Seed Fu override this setting, as seeding is the one time I actually want my read-only models to be writeable?

seed_fu_tasks.rake line 40:

  puts "\n== Seed from #{pretty_name} " + ("=" * (60 - (17 + File.split(file).last.length)))

errors out with a "negative argument" if the base filename (result of File.split(file).last.length) is over a certain length (43 characters).

The 1.2.1 gem doesn't contain ./tasks or any of it's files. It looks like it's missing from the gemspec.

SeedFu::Writer::Seed#add_seed calls chunk_this_seed? but this method is not defined.