Conflicts:
	railties/guides/source/active_record_querying.textile

This works just fine, is less code, and reduces the risk of someone implementing a SQL injection vulnerability.

Fix trouble using :subdomain in development environment when using numeric addresses.

    See-also pull request #3561 from 3-1-stable

    Otherwise the following occurs:

    TypeError: can't convert nil into String
        /Users/bfolkens/dev/bfolkens-rails-core/actionpack/lib/action_dispatch/http/url.rb:75:in host_or_subdomain_and_domain'
        /Users/bfolkens/dev/bfolkens-rails-core/actionpack/lib/action_dispatch/http/url.rb:37:in url_for'
        /Users/bfolkens/dev/bfolkens-rails-core/actionpack/lib/action_dispatch/routing/url_for.rb:147:in test_subdomain_may_be_accepted_with_numeric_host'
        /Users/bfolkens/dev/bfolkens-rails-core/activesupport/lib/active_support/testing/setup_and_teardown.rb:67:in run'
        /Users/bfolkens/dev/bfolkens-rails-core/activesupport/lib/active_support/callbacks.rb:426:in send'
        /Users/bfolkens/dev/bfolkens-rails-core/activesupport/lib/active_support/callbacks.rb:81:in run'

Should be checking if file exists or not.

I found this bug when running rake test:uncommitted 
on a newly generated rails app which don't have
test file for application_controller.
Can see detail here #3461

Warnings removed.

CHANGELOG to CHANGELOG.md

Doc patch, DrbStore no more exists in rails

remove_method tests and changes

Remove redundant blank line at the bottom of generated controller code

The output in Travis is a bit different. The SQLite
documentation (http://www.sqlite.org/eqp.html) warns
output may change dramatically between releases. I
do not want to mock the result set because I want a
real EXPLAIN to happen. I prefer a test that may fail
in future releases than a test that may give false
positives in future releases.

Removed argument throwing warnings when running tests

The last argument to DateTime.new/civil is the Julian day number for
when the calendar reform occured. DateTime ignores it if it's 0, and
sets it to the default of DateTime::ITALY instead, so we could just
leave it out. This also removes the warnings while running the
ActiveSupport tests (there still are some for the setup though).

Fix postgres adapter to handle spaces between schemas

Make pluralize follow singularize's logic of uncountability check

The bug is that an error would be raised while attempting to convert a
template from one encoding to another.

Please see http://redmine.ruby-lang.org/issues/5564 for more details.

The workaround is to load all conversions into memory ahead of time,
and will only happen if the ruby version is *exactly* 1.9.3p0. The
hope is obviously that the underlying problem will be resolved in
the next patchlevel release of 1.9.3.

This reverts commit 10773f94.

Refactored pluralize and singularize into a common method