This project is archived and is in readonly mode.

Weird classes namespace whith cache_classes = false

#5735

Hi,

I've observed a weird behavior of loading classes with namespace. I have a class that is not in a module:

class RedactionControllers < ActionController::Base
end

I can access it normally:

Loading development environment (Rails 3.0.0)
ruby-1.9.2-p0 > RedactionController 
 => RedactionController

But I should not be able to access it in a namespace:

ruby-1.9.2-p0 > Redaction::RedactionController
 => RedactionController

But, if I change config.cache_classes from false to true (in my development environment), I have the expected behavior:

ruby-1.9.2-p0 > Redaction::RedactionController
NameError: uninitialized constant Redaction::RedactionController

This bug has some consequences. In my case, I have this route:

namespace :redaction do
  root :to => "redaction#index"
end

Everything works well in development (with config.cache_classes = false), but it explodes in production (with config.cache_classes = true):

Routing Error
uninitialized constant Redaction::RedactionController

We should fix this difference between development and production to avoid that developers discover some bugs on the production.

Reported by Bruno Michel · September 29th, 2010 @ 08:45 PM

Activity

Bruno Michel
I've addded the failing test case.

September 29th, 2010 @ 09:28 PM
Bruno Michel
For the record, this bug seems related to #5383 AS dependencies resolver broken: not missing constant error

October 1st, 2010 @ 11:32 AM
Bruno Michel
#2548 Model namespace trouble is the same bug.

October 1st, 2010 @ 11:36 AM
Bruno Michel
In my example, Redaction is a module, and the bug is only if the nesting namesapce is a module (not a class). So i've updated my test case to reflect that.

I've tried to fix this bug, but I can't find how to differentiate A::B from module A; B; end in the autoloading stuff.

October 2nd, 2010 @ 04:04 PM

David Trasbo

State changed from new to incomplete
Importance changed from to Low

The patch with the failing test case no longer seems to apply cleanly:

Applying: Failing test case for #5735
fatal: corrupt patch at line 23
Patch failed at 0001 Failing test case for #5735
When you have resolved this problem run "git am --resolved".
If you would prefer to skip this patch, instead run "git am --skip".
To restore the original branch and stop patching run "git am --abort".

Could you please update it, and perhaps provide a fix if you figured it out?

October 8th, 2010 @ 09:43 AM

Bruno Michel
I've updated the patch for the failing test case, but I didn't figure how to fix it without breaking others tests. Seems like it's a complex task, that needs a deep comprehension of loading of classes and modules.
- failing_test_case.patch 950 Bytes
October 10th, 2010 @ 06:44 PM
David Trasbo
- State changed from incomplete to open
October 10th, 2010 @ 08:29 PM
Jeff Kreeftmeijer
- State changed from open to incomplete
- Tag changed from cache_classes, namespace to cache_classes, namespace, patch
Please use the "patch" tag when adding a patch to make sure patched tickets end up in the open patches bin. :)

October 10th, 2010 @ 08:30 PM
Ryan Bigg
- Tag cleared.
Automatic cleanup of spam.

October 11th, 2010 @ 02:59 AM
Ryan Bigg
Automatic cleanup of spam.

October 11th, 2010 @ 02:59 AM
Bruno Michel
Jeff: I haven't use the patch tag on purpose, because it's only a failing test case and I haven't figured how to fix it (and even if it's possible to do that).

October 14th, 2010 @ 08:05 PM
Jeff Kreeftmeijer
- State changed from incomplete to open
- Tag set to cache_classes, namespace
Bruno: I guess I didn't see that was a failing-test-patch. I have no clue why I marked it as incomplete either. My bad, sorry.

October 14th, 2010 @ 08:29 PM
Ryan Bigg
- Tag cleared.
Automatic cleanup of spam.

October 19th, 2010 @ 08:32 AM
Santiago Pastorino
This issue has been automatically marked as stale because it has not been commented on for at least three months.

The resources of the Rails core team are limited, and so we are asking for your help. If you can still reproduce this error on the 3-0-stable branch or on master, please reply with all of the information you have about it and add "[state:open]" to your comment. This will reopen the ticket for review. Likewise, if you feel that this is a very important feature for Rails to include, please reply with your explanation so we can consider it.

Thank you for all your contributions, and we hope you will understand this step to focus our efforts where they are most helpful.

February 2nd, 2011 @ 04:47 PM
Santiago Pastorino
- State changed from open to stale
February 2nd, 2011 @ 04:47 PM
Bruno Michel
- failing_test_case_on_3-0-stable.diff 950 Bytes
February 2nd, 2011 @ 10:11 PM
Bruno Michel
- State changed from stale to open
I've update the failing test case for 3-0-stable. I don't have any idea on how to fix it but I hope someone else will be able to help me on that.

[state:open]

February 2nd, 2011 @ 10:13 PM
rails
This issue has been automatically marked as stale because it has not been commented on for at least three months.

The resources of the Rails core team are limited, and so we are asking for your help. If you can still reproduce this error on the 3-0-stable branch or on master, please reply with all of the information you have about it and add "[state:open]" to your comment. This will reopen the ticket for review. Likewise, if you feel that this is a very important feature for Rails to include, please reply with your explanation so we can consider it.

Thank you for all your contributions, and we hope you will understand this step to focus our efforts where they are most helpful.

May 3rd, 2011 @ 01:00 AM
rails
- State changed from open to stale
May 3rd, 2011 @ 01:00 AM