Re: [Rails] Impossible to get data with ActiveResource

[Colin Law]

On 18 October 2012 23:30, chardy <thiba...@gmail.com> wrote:
> Hi,
>
> I'm trying to work with a RESTful server but I have some problem to setup my
> project.
> I have a model class like this :
>
>> class User < ActiveResource::Base
>>
>> self.site =
>> "http://diufvm31.unifr.ch:8090/CyberCoachServer/resources/users/"
>>
>> end
>
>
> this controller :
>
>> class UsersController < ApplicationController
>>
>> def new
>>
>> end
>>
>> def index
>>
>> @users = User.find(:all)
>>
>>
>> respond_to do |format|
>>
>> format.html # index.html.erb
>>
>> format.json { render :json => @users }
>>
>> format.xml { render :xml => @users}
>>
>> end
>>
>> end
>>
>> end
>>
>>
>
>
> and simply this view :
>
>> <% @users.each do |user| %>
>>
>> <p><%= user.username %></p>
>>
>> <% end %>
>>
>>
>
>
> But I can't retrieve the data, @users is empty.
> Anyone has an idea how to fix this issue?

Is there anything useful in log/development.log?

Colin

>
> Thanks
>
> --
> You received this message because you are subscribed to the Google Groups
> "Ruby on Rails: Talk" group.
> To post to this group, send email to rubyonra...@googlegroups.com.
> To unsubscribe from this group, send email to
> rubyonrails-ta...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/rubyonrails-talk/-/fjonPR37-AQJ.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

[chardy]

Thanks for your help Colin!

My logs : 

Started GET "/users/index" for 127.0.0.1 at Thu Oct 18 23:01:19 +0200 2012

Processing by UsersController#index as HTML

  Rendered users/index.html.erb within layouts/application (0.6ms)

Completed 500 Internal Server Error in 140ms

ActionView::Template::Error (undefined method `each' for nil:NilClass):

    1: <% @users.each do |user| %>

    2: <h1><%= user.username %></h1>

    3: 

    4: <% end %>

  app/views/users/index.html.erb:1:in `_app_views_users_index_html_erb___162613411_2191123680'

I think the problem is the Error 500...? It's strange because if you try the link in the browser it's works... 

Thanks for your help!

[Colin Law]

The 500 error is not the problem, that is just the result of @users being nil.

I think you need to enable logging so that you can see what
activeresource is doing. I have not used activeresource but a quick
google found [1] which looks useful and has a section Logging
Requests. Hopefully you will then see what is going on.

Colin

[1] http://ofps.oreilly.com/titles/9780596521424/activeresource_id59243.html

[chardy]

Yes, good idea, now I can understand where is the problem : 

GET http://diufvm31.unifr.ch:8090/CyberCoachServer/resources/users/users.xml

--> 404 Not Found 32 (129ms)

The server is not configured to respond for .xml (or .json). The content type should be defined in the header...

Is it possible to setup this in ruby...?

Thanks a lot!

[Hassan Schroeder]

On Sun, Oct 21, 2012 at 7:06 AM, chardy <thiba...@gmail.com> wrote:

> GET http://diufvm31.unifr.ch:8090/CyberCoachServer/resources/users/users.xml
> --> 404 Not Found 32 (129ms)
>
> The server is not configured to respond for .xml (or .json).

If this server isn't under your control, shouldn't you have some kind
of documentation on what it does -- what requests it responds to and
in what format?

--
Hassan Schroeder ------------------------ hassan.s...@gmail.com
http://about.me/hassanschroeder
twitter: @hassan

[chardy]

Yes, it's for a project in my university. I know that the we need to specify the format in the header and the server can handle json and xml format.

Here is the documentation :

Read only resources:

- SportsResource / - Get list of all sports [GET, public]

- UsersResource /users/ - Get list of all users [GET, public]

- PartnershipsResource /partnerships/ - Get list of all partnerships [GET, public]

- SportResource /sports/{sportname} - Get sport entity and its list of subscriptions [GET, public]

"Public" write resources:

- UserResource /users/{username} - Get user entity [GET, public*] - Create new user [PUT, public] - Update user profile [PUT, private] - Delete user [DELETE, private]

Private write resources:

- PartnershipResource /partnerships/{username1};{username2}/ - Get partnership entity [GET, public*] - Propose (or confirm** proposed) partnership [PUT, private] - Delete (or decline** proposed) partnership [DELETE, private]

- SubscriptionResource /users/{username}/{sportname} - Get subscription entity [GET, public*] /partnerships/{username1};{username2}/{sportname}/ - Subscribe (as user or partnership) to a sport [PUT, private] - Update subscription [PUT, private] - Submit new entry [POST, private] - Delete subscription [DELETE, private]

- EntryResource /[uri of subscription]/{entityid} - Get entry entity [GET, public*] - Update entry data [PUT, private] - Delete entry [DELETE, private]

* = Read accessability depends on the privacy settings for the respective entities ** = A partnership is created/deleted in the following way:

• PUT by user1 proposes a new partnership. PUT by user2 confirmes the proposal and the partnership becomes "operational"

• DELETE on an operational partnership results in a proposed partnership. DELETE on a proposed partnership deletes it.

  
  

[Colin Law]

On 21 October 2012 15:06, chardy <thiba...@gmail.com> wrote:
> Yes, good idea, now I can understand where is the problem :
>
> GET http://diufvm31.unifr.ch:8090/CyberCoachServer/resources/users/users.xml

Is the url supposed to be resources/users/users.xml? Perhaps it
should just be resources/users.xml.

Colin

[Hassan Schroeder]

On Sun, Oct 21, 2012 at 7:44 AM, chardy <thiba...@gmail.com> wrote:

> - UsersResource /users/ - Get list of all users [GET, public]

Not much for "documentation" :-/

http://diufvm31.unifr.ch:8090/CyberCoachServer/resources/users

returns an HTML page with a paginated listing of users, but that's
hardly what you want for ActiveResource.

If no more info on this server is available, I'd probably poke at it a bit
with wget or curl to see if I could get a useful response.

Good luck!

[chardy]

Yes, sorry I made a mistake it's only users.xml but the problem is the same...

Is it possible to retrieve the data without the extension .xml but just specify the type in the header?

I have not found anything on the web...

[chardy]

So, I tried to overwrite the methode collection_path :

def collection_path(prefix_options = {}, query_options = nil)
        check_prefix_options(prefix_options)
        prefix_options, query_options = split_options(prefix_options) if query_options.nil?
        "#{prefix(prefix_options)}#{collection_name}.#{format.extension}#{query_string(query_options)}"
      end
by
def collection_path(prefix_options = {}, query_options = nil)
        check_prefix_options(prefix_options)
        prefix_options, query_options = split_options(prefix_options) if query_options.nil?
        "#{prefix(prefix_options)}#{collection_name}#{query_string(query_options)}"
      end

And now it's OK :) 
GET http://diufvm31.unifr.ch:8090/CyberCoachServer/resources/users
--> 200 OK 1260 (1735.7ms)

...But :(
I have an other error now : 
NoMethodError (undefined method `collect!' for #<Hash:0x1030aeed0>):
  app/controllers/users_controller.rb:6:in `index'

Anyone has an idea

Thanks a lot for your support! 

[Colin Law]

What do you think that error might mean?

Colin

[chardy]

It's finally seems to work :)

My ActiveResource model looks like this :

class User < ActiveResource::Base

  class << self

    def element_path(id, prefix_options = {}, query_options = nil)

      check_prefix_options(prefix_options)

      prefix_options, query_options = split_options(prefix_options) if query_options.nil?

      "#{prefix(prefix_options)}#{collection_name}/#{URI.parser.escape id.to_s}#{query_string(query_options)}"

    end

    def collection_path(prefix_options = {}, query_options = nil)

      check_prefix_options(prefix_options)

      prefix_options, query_options = split_options(prefix_options) if query_options.nil?

      "#{prefix(prefix_options)}#{collection_name}#{query_string(query_options)}"

    end

    def instantiate_collection(collection, prefix_options = {})

      collection = collection[element_name.pluralize] if collection.instance_of?(Hash)

      collection.collect! { |record| instantiate_record(record, prefix_options) }

    end

  end

  self.site = "http://diufvm31.unifr.ch:8090/CyberCoachServer/resources/"

end

I founded how to solve the last error here : https://github.com/rails/rails/issues/2318#issuecomment-3555973

Thanks a lot for your support! :)