RSpec 2.6.0 deprecation warning

[searls]

RSpec 2.6.0 is giving this warning when I run my specs:

*****************************************************************
DEPRECATION WARNING: you are using deprecated behaviour that will
be removed from RSpec 3.

You have set some configuration options after an example group has
already been defined.  In RSpec 3, this will not be allowed.  All
configuration should happen before the first example group is
defined.  The configuration is happening at:

  /path/to/my/project/spec/spec_helper.rb:50
*****************************************************************

This refers to this generated block:

RSpec.configure do |config|
  config.before(:each) do
    database_connections.each do |name|
      plsql(name).savepoint "before_each"
    end
  end
  config.after(:each) do
    # Always perform rollback to savepoint after each test
    database_connections.each do |name|
      plsql(name).rollback_to "before_each"
    end
  end
  config.after(:all) do
    # Always perform rollback after each describe block
    database_connections.each do |name|
      plsql(name).rollback
    end
  end
end

Thoughts?

[rsim]

I do not have such warnings. Can you ensure that you are not requiring spec_helper.rb twice? (Maybe using different path names to this file) Or maybe try to add some debugging message before this code block to see if it is executed once or several times.

[searls]

Indeed, only one of several workstations is seeing the error. I'll look into it in more detail the next time I see it.

On Jul 20, 2011, at 9:53, rsim reply@reply.github.com wrote:

I do not have such warnings. Can you ensure that you are not requiring spec_helper.rb twice? (Maybe using different path names to this file) Or maybe try to add some debugging message before this code block to see if it is executed once or several times.

Reply to this email directly or view it on GitHub: https://github.com/rsim/ruby-plsql-spec/issues/8#issuecomment-1615534

[jgebal]

Is it still valid after recent updates or can we close this one?

[jgebal]

if we're unable to reproduce, and no one complains, can it be closed?

[javornikolov]

I don't see warnings either, using rspec 3.4. I'm closing the issue...