3.1

gucci watches

gucci watches

Ugh, typo in comments. How lame is that? Here's the fix.

Santiago, yes it is. (Just checked). Somehow (not sure) it is because of: https://rails.lighthouseapp.com/projects/8994/tickets/6290-patch-ex...

And because of...

2kan have you tried it with the latest master code?. Is still an issue?

[state:open]

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 limit...

Applied and pushed. Thanks.

By the way, the simplistic example given in the original ticket above isn't the motivating use case. That would be using an association or scope to find a recor...

OK, added last! too. (Aaron, you were right!)

Josh, can you add last! as well to the patch and I'll apply it

Pushed to master

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 limit...

Merged to master. Closing.

Today the company to build a website for a job so important that can not simply say yes or no to it. Website has become the media and the leading business tool ...

+1, I encountered this in 3.0.5 as well.

Oops, I've found that my last uploaded patch wasn't good enough (I've uploaded the wrong diff, sorry). Here is the correct one.

+1 This would be a nice change to have, to finalize the migration to html_safe in Rails

I am using non-restful routes, which is perhaps why the catch-all doesn't work for me

match ':controller(/:action(/:id))', :constraints => {:id => /.*/}

Oops, I removed José. Sorry.

A catch-all route at the end of your routes my seem like a simple fix, but it has a major drawback:
If your application relies on engines that extend your app w...

Are you using :action as the dynamic segment name? If so Rails will try to find the action in the controller. The following path works for me:

match '*path', :...

Just to add,
match '/:anything', :to => "home#routing_error", :constraints => { :anything => /./ } or
match "", :to => "home#routing_error"

Don't seem to work w...

This is a huge bug and I feel it's priority should be raised.

It's not acceptable for a broken link in a production app to result in a blank screen, this is why...

I've started a discussion on core with suggestion to change scope syntax. I think the syntax change is certainly a better way of resolving this issue than a hac...

Hi Santiago, is the patch ok?

Akira's patch fixes the issue for me, but the attached test fails because Pet.joins(:toys).where(:toys => {:name => 'Bone'}).count returns 0.

Also, I'm not a gr...

Fixed by https://github.com/rails/rails/commit/94031ef422357e5802c965a9e8365...

Err, that was for the "ERROR: Could not find a valid gem 'pkg/rails-3.1.0.beta.gem' (>= 0) in any repository" problem. Sorry.

I had the exact same issue. Running

rake all:install

resolved the problem.

Any more info is needed with this patch?

Hey Diego, I don't think that's a good idea. We already have a j helper to escape json.
The best is to deprecate the helper for 3-0-stable and remove it from ma...

This is also breaking my mailer templates, since the original method expects the second argument to be a hash and the one in MailHelper expects a fixnum. (I sta...

+1 on bringing it into 3-0-stable. I'm sitting at 3.0.4 due to this bug in 3.0.5 (and having used simple_format in mailer templates).

Further, since this is only a private refactoring, there is no public or even semi-public interface to worry about breaking.

Any reason this can't be in 3-0-stable? There is an additional regression in that this method now raises on nil, whereas it used to return a blank string.

Applied.

Hey Santiago,

There's a patch in a pull request at GitHub. If I can get a sec I'll try to look into how this sort of thig is tested if at all

Can you provide a test case or a patch?

Yep, tried it again, definitely still an issue. Found another one though--will open in a new ticket.

Just tried it again on master, this is still an issue for me.

Jack - you're right, require 'date' is all that is needed to fix this.

Course, as soon as I click submit and try it for real, busted again, as well as a different issue. Will look at it further, but either way, more eyes should tak...

I think this has been inadvertently solved by recent commits. I've seen the issue (had it just happen with a slightly old local install of 3.1.beta) but after g...