#1570 named_scope not preserving proper model in STI - 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.

#1570 ✓resolved

named_scope not preserving proper model in STI

Reported by Paweł Kondzior | December 14th, 2008 @ 03:26 AM | in 2.3.9

Simple example:


class User < ActiveRecord::Base
  named_scope(:simple_test, lambda { puts self.inspect; {} })
end

class Employee < ActiveRecord::Base
end

And the result:

Loading development environment (Rails 2.3.0)

Employee.simple_test(123) User(id: integer, created_at: datetime, updated_at: datetime, active: boolean, type: string, location_id: integer, name: string, email: string) => []

In such case we don't have access to Emloyee class methods, reflection and other from lambda.

Comments and changes to this ticket

Paweł Kondzior December 14th, 2008 @ 03:27 AM


Loading development environment (Rails 2.3.0)
>> Employee.simple_test(123)
User(id: integer, created_at: datetime, updated_at: datetime, active: boolean, type: string, location_id: integer, name: string, email: string)
=> []

You flagged this item as spam.
Paweł Kondzior December 14th, 2008 @ 03:33 AM
```
class Employee < User
end
```
And because it's bit late, i did typo in example code :(
Frederick Cheung December 14th, 2008 @ 10:31 AM
This is not so much named_scope doing something but ruby closures. The block associated with the lambda remembers what self was when that block was made.
You flagged this item as spam.
Paweł Kondzior December 14th, 2008 @ 11:29 AM
Yup, i figured out that. But still, problem exists. Workaround could be redefine simple_test from Employee method, but we do STI to be DRY, this is not DRY. The best solution would be Proc#call_with_binding in Ruby (hard thing!) or rework of scopes, and create some helper method for all lambdas.

For Proc#call_with_binding... well it's funny but we have some similar proposals for 1.9? http://www.ruby-forum.com/topic/...

The second solution would be probably very ugly and no one from core will acceppt such and api for lambdas ? Am i right ?
Frederick Cheung December 14th, 2008 @ 01:15 PM
Or if the @proxy_scope variable from the Scope (which should be the "right" class/scope) were somehow passed to the lambda, but I'm not sure how you'd do that without breaking everyone's existing code.
You flagged this item as spam.
Paweł Kondzior December 14th, 2008 @ 03:16 PM
I hoped that solution will be easy, and i was right :) Active Support gives us Proc#bind that emulates 1.9 instance_exec (instance_eval with arguments)

It was after i post proposal http://www.ruby-forum.com/topic/..., details about instance_exec you can found here http://eigenclass.org/hiki.rb?in...

I thnik DHH and rest of core should lobby ruby-core for Proc#bind native implementation in 1.9, it's better to have it in ruby than in active_support just like Symbol#to_proc. Just use your power that comes from being core of ruby community ;-)

Small patch with testcase is on the way.
You flagged this item as spam.
Paweł Kondzior December 14th, 2008 @ 03:49 PM
And here it is.
- 001_named_scope_preserve_sti_self_context.patch 2.1 KB
You flagged this item as spam.
Paweł Kondzior December 14th, 2008 @ 05:33 PM
Same stuff but here i'm using instance_exec functionality.
- 001_named_scope_perserve_sti_self_context_using_instance_exec.patch 2.1 KB
Pratik March 8th, 2009 @ 12:57 PM
- Assigned user changed from “Jeremy Kemper” to “Pratik”
- Tag changed from “2.3, bug, named_scope, sti” to “2.3, bug, named_scope, patch, sti”
- State changed from “new” to “incomplete”
Patch doesn't apply any more. Also, we should have a test for cases when such a named scope is chained with other named scopes.

Thanks.
Jeremy Kemper May 4th, 2010 @ 06:48 PM
- Milestone changed from “2.x” to “3.x”
[bulk edit]
Neeraj Singh June 23rd, 2010 @ 02:37 AM
- Assigned user changed from “Pratik” to “José Valim”
Attached is test.

Attached is code patch. For the code patch please provide credit to Paweł Kondzior. One of these days I will figure out myself how to do that.

Assigning this ticket to Mr. Valim . He approved #4507 which is similar to this one.
- fix_against_2-3-stable.diff 1 KB
Neeraj Singh June 23rd, 2010 @ 02:38 AM
Here is test.
- test_against_2_3_master.diff 1.5 KB
Repository June 23rd, 2010 @ 08:17 AM
(from [687d7f52c4a1d85666f1efc08ffa182e6dfeba57]) STI should identify itself inside named_scope

[#1570 state:resovled]

Signed-off-by: José Valim jose.valim@gmail.com
http://github.com/rails/rails/commit/687d7f52c4a1d85666f1efc08ffa18...
José Valim June 23rd, 2010 @ 08:18 AM
- Milestone changed from “3.x” to “2.3.9”
- State changed from “incomplete” to “resolved”
You flagged this item as spam.
James Le Cuirot July 4th, 2010 @ 02:08 PM
- Importance changed from “” to “”
This commit causes a massive slow down and memory leak. Please see #5044.