#5579 Rails3 - accepts_nested_attributes_for reject_if option does not work. - Ruby on Rails

Type	To find
responsible:me	tickets assigned to you
tagged:"@high"	tickets tagged @high
milestone:next	tickets in the upcoming milestone
state:invalid	tickets with the state invalid
created:"last week"	tickets created last week
sort:number, importance, updated	tickets sorted by #, importance or updated
Combine keywords for powerful searching.
Use advanced searching »

This project is archived and is in readonly mode.

#5579 ✓resolved

Rails3 - accepts_nested_attributes_for reject_if option does not work.

Reported by Robert Pankowecki | September 8th, 2010 @ 09:35 AM | in 3.0.2


class Project < ActiveRecord::Base
  has_many :tasks
  accepts_nested_attributes_for :tasks, :reject_if => proc { |attributes| true } # Always reject : just to prove the case.
end

class Task < ActiveRecord::Base
  belongs_to :project
end

p = Project.create!(:name => "project")
p.tasks.create!(:name => "1")
p.tasks.create!(:name => "2")

p.reload

p.attributes = {
  "tasks_attributes" => {
    "0" => {"name"=>"new name", "_destroy"=>"0", "id"=> p.tasks.first.id},
  }
}

p.tasks
 => [#<Task id: 1, project_id: 1, name: "new name", created_at: "2010-09-08 08:26:49", updated_at: "2010-09-08 08:26:49">, #<Task id: 2, project_id: 1, name: "2", created_at: "2010-09-08 08:26:49", updated_at: "2010-09-08 08:26:49">]

The task name is changed but it should not be due to reject_if proc.

This is on Rails 3.0.0

Comments and changes to this ticket

Mark Mulder September 8th, 2010 @ 11:38 AM

Hmm, it's behaviour seems weird although I might be missing something as well.

I tried your example and editing a tasks attributes seems to work, adding a new one does not:

ruby-1.9.2-p0 > p = Project.create!(:name => 'project')
 => #<Project id: 3, name: "project", created_at: "2010-09-08 10:35:31", updated_at: "2010-09-08 10:35:31"> 
ruby-1.9.2-p0 > p.tasks.create!(:name => 'task 1')
 => #<Task id: 4, name: "task 1", project_id: 3, created_at: "2010-09-08 10:35:39", updated_at: "2010-09-08 10:35:39"> 
ruby-1.9.2-p0 > p.tasks_attributes = [ { :id => 4, :name => 'changed'} ]
 => [{:id=>4, :name=>"changed"}] 
ruby-1.9.2-p0 > p.save
 => true 
ruby-1.9.2-p0 > p.tasks
 => [#<Task id: 4, name: "changed", project_id: 3, created_at: "2010-09-08 10:35:39", updated_at: "2010-09-08 10:36:06">] 
ruby-1.9.2-p0 > p.tasks_attributes = [ { :name => 'new task' } ]
 => [{:id=>5, :name=>"new task"}] 
ruby-1.9.2-p0 > p.save
 => true 
ruby-1.9.2-p0 > p.tasks
 => [#<Task id: 4, name: "changed", project_id: 3, created_at: "2010-09-08 10:35:39", updated_at: "2010-09-08 10:36:06">] 
ruby-1.9.2-p0 >

Neeraj Singh September 8th, 2010 @ 04:12 PM
- State changed from “new” to “open”
- Assigned user set to “Neeraj Singh”
- Tag changed from “accepts_nested_attributes_for, reject_if” to “accepts_nested_attributes_for, patch, reject_if”
- Importance changed from “” to “Low”
@Mark you are right. While creating a new record reject_if is being respected but while updating it was being ignored.

Attached is a patch with test.

Please try this patch and see if it works.
- 5579_fix_for_edge.diff 3.2 KB
Neeraj Singh September 14th, 2010 @ 05:34 AM
Can someone verify my patch and provide comments. Thanks.
You flagged this item as spam.
Lake September 15th, 2010 @ 07:23 AM
Hi Neeraj,

Your patch works well for me. I ran the tests and they pass.

I also ran your patch against the OP code and the patch rejects the name change on the task, just as it is supposed to.

Thanks

+1
Neeraj Singh September 15th, 2010 @ 04:49 PM
@Lake thanks for verifying.
Neeraj Singh September 19th, 2010 @ 02:34 PM
- Milestone set to “3.x”
- Assigned user changed from “Neeraj Singh” to “José Valim”
Neeraj Singh September 24th, 2010 @ 01:49 AM
- Assigned user changed from “José Valim” to “Aaron Patterson”
Assigning it to Mr. Patterson as per advice from Santiago.
Neeraj Singh September 24th, 2010 @ 01:49 AM
- Milestone cleared.
Repository September 24th, 2010 @ 12:11 PM
- State changed from “open” to “resolved”
(from [097240f60215b866d24aebd02cc4159bdc6e7451]) reject_id option should be respected while using nested_attributes

[#5579 state:resolved]

Signed-off-by: José Valim jose.valim@gmail.com
http://github.com/rails/rails/commit/097240f60215b866d24aebd02cc415...
Jeremy Kemper October 15th, 2010 @ 11:02 PM
- Milestone set to “3.0.2”
[bulk edit]
You flagged this item as spam.
Andrea Franceschini December 26th, 2010 @ 01:06 PM
Hi there. Though I'm using rails-3.0.3 (and of course activerecord-3.0.3) I'm facing this very same issue, exactly same behaviour. FTR, here's my sources:

app/models/invoice.rb

class Invoice < ActiveRecord::Base
```
attr_accessible :tax, :discount, :modifier, :paid, :issue_date, :payment_date, :description, :items_attributes
has_many :items, :class_name => 'InvoiceItem', :dependent => :destroy
accepts_nested_attributes_for :items, :reject_if => proc { |a| a[:description].blank? }, :allow_destroy => true
```
end

app/models/invoice_item.rb

class InvoiceItem < ActiveRecord::Base
```
attr_accessible :price, :discount, :description
belongs_to :invoice
```
end

I could verify that the :reject_if is honoured when creating a new Invoice, while it's not when updating an existing one.
You flagged this item as spam.
Dave Myron January 10th, 2011 @ 12:14 PM
The proc (or symbol to a method) given to an accept_nested_attributes_for method for a has_one association never runs (for me, on Rails 3.0.3). Did anyone give that one a test?
You flagged this item as spam.
Robert Pankowecki January 16th, 2011 @ 07:10 PM
@Dave - could you write a failing test case ?