Rails Ruby on Rails

Comments and changes to this ticket

• You flagged this item as spam.
  

  Philipp Steinacher February 19th, 2011 @ 11:27 PM

  • Tag set to “development environment, console, reload!”

  It also doesn't work with Rails 3.0.3 and ruby-1.9.2-p136

• You flagged this item as spam.
  

  William Denniss February 20th, 2011 @ 03:36 AM

  I'm not sure if the version of ruby matters, but for the record:

  $ ruby --version ruby 1.8.7 (2009-06-12 patchlevel 174) [universal-darwin10.0]

• 

  Jeff Kreeftmeijer February 28th, 2011 @ 08:17 PM

  • State changed from “new” to “open”
  • Importance changed from “” to “Low”

  I couldn't reproduce this issue on master or 3.0.4:

  # app/models/user.rb
  
  class User < ActiveRecord::Base
    validates :email, :presence => true
    #validates :name, :presence => true
  end
  

  $ script/rails c
  Loading development environment (Rails 3.1.0.beta)
  irb(main):001:0> u = User.new
  => #<User id: nil, name: nil, email: nil, created_at: nil, updated_at: nil>
  irb(main):002:0> u.valid?
  => false
  irb(main):003:0> u.errors
  => #<ActiveModel::Errors:0xb7204b50 @base=#<User id: nil, name: nil, email: nil, created_at: nil, updated_at: nil>, @messages=#<OrderedHash {:email=>["can't be blank"]}>>
  
  # Now, before reloading and without leaving the console, I uncomment the :name validation in the user model.
  
  irb(main):004:0> reload!
  Reloading...
  => true
  irb(main):005:0> u = User.new
  => #<User id: nil, name: nil, email: nil, created_at: nil, updated_at: nil>
  irb(main):006:0> u.valid?
  => false
  irb(main):007:0> u.errors
  => #<ActiveModel::Errors:0xb6d3ed00 @base=#<User id: nil, name: nil, email: nil, created_at: nil, updated_at: nil>, @messages=#<OrderedHash {:email=>["can't be blank"], :name=>["can't be blank"]}>>
  

• 

  Jeff Kreeftmeijer February 28th, 2011 @ 09:01 PM

  • State changed from “new” to “needs-more-info”
  • Assigned user set to “Jeff Kreeftmeijer”

  Sorry, in my last comment I previously confirmed the issue, but I was doing something wrong. I can't reproduce this issue and I updated my last comment.

  Can anyone else reproduce this? If not, I'll mark this one as invalid. :)

• You flagged this item as spam.
  

  William Denniss February 28th, 2011 @ 11:49 PM

  Hi Jeff,

  Thank you for testing this. I'm sorry for not creating a better bug report, it seems that the test case in #3822 does in fact pass. I had reload! problems in 3.0.4, and incorrectly assumed that #3822 had regressed exactly.

  Please see the updated bug report above for a test case that fails in 3.0.4.

  Thanks again for your time.

  Best Regards,
  Will

• 

  Jeff Kreeftmeijer March 1st, 2011 @ 06:45 PM

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

  Thanks Will, I tried to reproduce again:

  # app/models/user.rb
  
  class User < ActiveRecord::Base
    def reloaded?
      false
    end
  end
  

  $ script/rails c
  Loading development environment (Rails 3.1.0.beta)
  irb(main):001:0> u = User.new
  => #<User id: nil, name: nil, email: nil, created_at: nil, updated_at: nil>
  irb(main):002:0> u.reloaded?
  => false
  irb(main):003:0> reload!
  Reloading...
  => true
  irb(main):004:0> u.reloaded?
  => false
  irb(main):005:0> u = User.new
  => #<User id: nil, name: nil, email: nil, created_at: nil, updated_at: nil>
  irb(main):006:0> u.reloaded?
  => true
  

  Again, before calling reload!, I changed the model to make User#reloaded? return true. When I call that method on the existing u object, it still returns false after reloading. When I create a new User object, #reload? returns true.

  This confirms your issue on master, I'll re-open the ticket and I'll have a look at it as soon as I can. :)

• You flagged this item as spam.
  

  William Denniss March 2nd, 2011 @ 03:03 AM

  Thanks for looking into it Jeff! Glad you were able to repro it.

• 

  Jeff Kreeftmeijer March 5th, 2011 @ 12:03 PM

  • State changed from “open” to “wontfix”

  Hi William,

  After talking to Josh about this, we came to this conclusion:

  The u object doesn't get reloaded because it already existed before the model changed and reload! only takes care of the model in this case. I'm marking this one as wontfix, since reloading instances of objects is not reload!'s job, it simply reloads the environment.

  Thanks for reporting! :)