#4714 Rails 2.3.8: breaks Selenium test - Ruby on Rails

Jeff Kreeftmeijer

Could you show us the code and explain how to reproduce it?

May 27th, 2010 @ 05:28 PM

TMorgan99

Not much to add, really.
The app is available from pragprog,

http://media.pragprog.com/titles/achbd/code/achbd-code.tgz
unzip the folder 'automated_browser/01'

Then, I have added a Gemfile to specify my gems ...

# A sample Gemfile
# Not *used* by Rails 2.3, but used by 'bundle install' to prep the gems manually
source :gemcutter

gem "rails", '2.3.5'
gem "sqlite3-ruby", :require => 'sqlite3'

gem 'rake'
gem 'rspec'
gem 'rspec-rails'
gem 'cucumber'
gem 'cucumber-rails'
gem 'database_cleaner'
gem 'nokogiri'
gem 'webrat'

gem 'mongrel'
gem 'selenium-client'

$ bundle install
$ rake db:migrate db:test:prepare  cucumber

Test runs ok.

Now change gemfile to rails '2.3.8'
Change RAILS_GEM_VERSION in ./config/environment.rb

$ bundle install
$ rake cucumber

Test fails on timeout

May 27th, 2010 @ 06:26 PM

TMorgan99

I have traced this a bit further.
I used a cucumber enabled git bisect search to find the point of failure.

It appears that the problem is the upgrading to rack 1.1.0
If I reverse this patch

patch -p1 -R <  0001-rack-dependency-bumped-to-110-tests-passing.patch

my test runs again.
But I am not sure why.

What were the tests that were passing here, so I could add my failing test to them?

May 31st, 2010 @ 11:30 PM

TMorgan99

State changed from new to resolved
Assigned user set to josh

git bisect search finds this commit in rails...

74206aeff2b6a960612a820ec47e91ef9b778262 is first bad commit
commit 74206aeff2b6a960612a820ec47e91ef9b778262
Author: Rizwan Reza <rizwanreza@gmail.com>
Date:   Sun May 16 23:40:17 2010 +0430

    Rack dependency bumped to 1.1.0, tests passing. [#3558 state:resolved]

    Signed-off-by: José Valim <jose.valim@gmail.com>

further git bisect searching within rack leads to this ...

commit 8f836f406ca10274c6465e17c2b5646257a8412b
Author: Eric Wong <normalperson@yhbt.net>
Date:   Thu Dec 10 21:34:17 2009 -0600

avoid HeaderHash#to_hash in middlewares

Since HeaderHash objects are valid header responses, avoid
converting the headers to Hash objects only to have it
reconverted back to HeaderHash in the next middleware.

Signed-off-by: Joshua Peek &lt;josh@joshpeek.com&gt;

I have opened a ticket on Rack

http://rack.lighthouseapp.com/projects/22435-rack/tickets/99-rack-1...

I suppose the tests in rack were insufficient?

June 1st, 2010 @ 12:59 AM

TMorgan99

oops - it seemed that I have inadvertently resolved this issue.
If I reverse the commit in rack, I can run my test successfully.

It appear that subsequent middlewares were not advised as to this change, and will fail if they were relying on
the .to_hash call

June 1st, 2010 @ 11:00 PM

Jeremy Kemper

State changed from resolved to open

June 22nd, 2010 @ 08:14 PM

Santiago Pastorino

Importance changed from to Low

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 @ 05:01 PM

Santiago Pastorino

State changed from open to stale

February 2nd, 2011 @ 05:02 PM

bingbing

omega watch

March 26th, 2011 @ 01:38 AM

Rails 2.3.8: breaks Selenium test

Activity