similar to: Some JSON doc patches

Just a quick pointer to some ActiveSupport patches hoping for reviews: Puts Array#to_query and Object#to_query in the right places in ActiveSupport - http://dev.rubyonrails.org/ticket/10395 Preserve user-specified content type for XML and JSON rendering (currently it''s overridden and set to Mime::XML/Mimi::JSON - http://dev.rubyonrails.org/ticket/10388 Oh and a tiny patch for

I''d really like to see just 2 more functional changes to the JSON serialization/encoding before we get to Rails 2.0. First, and I think this is really crucial because without it all the efforts to output JSON from ActiveRecord objects would have been half in vain, is to allow us to do this in controllers: @authors = Author.find(:all) render :json => @authors.to_json(:only =>

Please check out this ticket (which would explain almost everything): http://dev.rubyonrails.org/ticket/8762 So basically, Rails to_json methods produces invalid JSON out of the box as it doesn''t quote all hash keys. This is valid JSON: { "id": 6589, "code": "SIN", "name": "Singapore" } This is not (Rails'' JSON encoders

I''m really glad the JSON changes got in, many thanks to all involved, committers and contributors alike :). But there''s still one feature that''s missing to achieve feature parity between the JSON and XML encoding/conversion. And that is the abililty to say in your controller: @authors = Author.find(:all) render :json => @authors.to_json(:only => :name)

Just some quick cleanup and documentation. Remove obsolete ActiveSupport::JSON::Variable - http://dev.rubyonrails.org/ticket/9969 Add documentation for Hash#to_json and Enumerable#to_json - http://dev.rubyonrails.org/ticket/9970 Please review :) Thanks! Cheers, Chu Yeow --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google

Sorry for attaching patches in emails to the mailing list, but I''m still unable to login to Devjavu for Backgroundrb (I''m able to login just fine for Merb''s Devjavu and submitted a ticket successfully). Anyway, I encountered a bug when trying to #ask_status of a worker before it''s run (or rather, before its result_hash has been set to something). Not sure why

Hi again everyone, I''ve 2 tiny doc patches mostly involving grammar fixes (and errant whitespace elimination). My English is not perfect so any review would be very welcome. The patches: http://dev.rubyonrails.org/ticket/9819 http://dev.rubyonrails.org/ticket/10062 Thanks! Cheers, Chu Yeow --~--~---------~--~----~------------~-------~--~----~ You received this message because you

Hi guys, Could a core committer review and apply http://dev.rubyonrails.org/ticket/10602? This would get rid of a failing test with the latest version of mocha (0.5.6). Thanks much! Cheers, Chu Yeow --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Core" group. To post to this group, send

I like the merb-style exception pages where there''re links to open the files listed in the stack trace in TextMate and the source around each line a lot so I stole the idea (and the code!) and made a patch for Rails: http://dev.rubyonrails.org/ticket/10401 Here''s how it looks like in Merb: http://yehudakatz.com/wp-content/uploads/stacktrace.gif What do you guys think?

Hi devs, I was trying to open a new ticket with a minor doc patch at the BackgroundDrb Trac (http://backgroundrb.devjavu.com/) and it seems like only admins have permission to do so. Is this intended? I read in the readme that patches should be mailed to this list, but I much prefer to use Trac since there''s already one set up. Any reason Trac is not open to registered Devjavu users?

Hi all, I just submitted a patch that would allow you to pass a proc to ActionController::Base.asset_host rather than just using a %d wildcard (e.g. assets%d.example.com). Basically this frees you from the hard-coded assumption that you have 4 hosts numbered 0-3 and also allows you to do smarter things depending on the asset (e.g. images from a particular domain, stylesheets from another). The

Hi guys, I''ve fixed a pretty annoying bug in Action caching that is detailed here: https://rails.lighthouseapp.com/projects/8994/tickets/6480-action-cache-filter-code-incorrectly-renders-nil-response-body-if-response-is-non-cacheable Would appreciate testers or a direct commit if it checks out :) Cheers, Chu Yeow -- You received this message because you are subscribed to the Google

Hi all, I was looking at this ticket: http://dev.rubyonrails.org/ticket/6265 and it got me thinking: is there any reason we''re not using the tasty ActiveSupport::JSON encoder to assemble options for the various helper methods in PrototypeHelper and ScriptaculousHelper? At the moment, they use options_for_javascript() in JavaScriptHelper, which is a half-hearted JSON generator that

Fix for the bad (failing) test here: http://dev.rubyonrails.org/ticket/10664 On Jan 2, 5:33 pm, alexey.verkhov...@gmail.com wrote: > The build failed. > > CHANGES > ------- > Revision 8522 committed by bitsweat on 2008-01-02 09:03:56 > db:sessions:clear task uses session_table_name method. Closes #10631 [Cheah Chu Yeow] > > M /trunk/railties/lib/tasks/databases.rake >

http://dev.rubyonrails.org/ticket/9990 and reiterated here: The current ActiveSupport JSON decoder uses YAML.load to parse json documents, and YAML has different escaping rules from JSON. In particular, in JSON, forward-slashes must be escaped (I''m not sure why, it isn''t neccessary for Javascript, but thats the spec.) This presents problems for URLs in ActiveSupport JSON:

Hello Rails people, I submitted a patch recently to get rid of an annoying exception raised when an AR object is created - http://dev.rubyonrails.org/ticket/10556 , had some people review it, got some pluses... What am I supposed to do now? Can anybody advise or just review the patch and take care of the issue? Thanks! -- Best regards, Yuri Leikind

I''ve the following situation (which actually works in Rails 2.3.8 and this is a result of the migration of our app to rails3): I have a partial which is rendered a few times (imagine tr''s in a table). On the clientside there''s a script that does a json request for a specific row (jQuery + ajax dataType: ''json'' and type: ''POST'') In

Hi, In short: I can do following: user.to_xml(:root => "xml_user" ) But not following: user.to_json(:root => "json_user") JSON serializer for AR doesn''t allow us to specify "root" option unlike XML serializer. JSON serializer always uses model name as the root if class attribute include_root_in_json is set to true. I''ve added a patch which

Hiya, So, what was the impetus behind having ParamsParser create Date and DateTime objects out of application/json requests? The project I''m on expects parameters to be received as standard types (bools, numbers, strings, arrays, and hashes). Rails shouldn''t try to convert objects that are explicitly passed as these types; it should be up to the application to do that.

Do I need something like this: eval(''('' + this.header(''X-JSON'') + '')'') (taken from Cody Swann, http://ajaxian.com/archives/prototype-extension-dynamic-script-pattern-support ). I''m guessing my problem is simpler, and I''m leaving something small out related to the header though. Thanks for any help -Mark 1. evalJSON: