Increase readability of ternary statement in build method

Object#respond_to? returns singletons and thus we inherit that contract.
The implementation of the predicate is good, but the test is only
checking boolean semantics, which in this case is not enough.

The entry is basically copy & paste of the commit message, but the CHANGELOG
has a different purpose than Git history, it just communicates what is new:

* No need to explain why did the bug happen (unless it is truly relevant).

* No need to explain how was the bug fixed.

* Whether the user gives new names to columns does not really matter, use of
  select to cherry-pick a column for example also presented that behaviour.
  Non-selected attributes are the key, either because they were not included
  in the selected list, or because they were but with a different alias.

* In the case of an attribute alias, what you really want to depict is that
  respond_to? returns false for the original attribute name.

Typo Fix in AR CHANGELOG [ci skip]

fix respond_to? for non selected column

fixes #4208

If a query selects only a few columns and gives custom names to
those columns then respond_to? was returning true for the non
selected columns. However calling those non selected columns
raises exception.

    post = Post.select("'title' as post_title").first

In the above case when `post.body` is invoked then an exception is
raised since `body` attribute is not selected. Howevere `respond_to?`
did not behave correctly.

    pos.respond_to?(:body) #=> true

Reason was that Active Record calls `super` to pass the call to
Active Model and all the columns are defined on Active Model.

Fix is to actually check if the data returned from the db contains
the data for column in question.

symbolize_keys => symbolize_keys! on new hash; change sentence

Adding documentation to Renderer

This reverts commit e8727d37, reversing
changes made to d098e1c2.

Reason: it broke the mysql build

autosave_association issue that occurs when table has unique index

fixes bug introduced by  #3329
These are the conditions necessary to reproduce the bug:
- For an association, autosave => true.
- An association record is being destroyed
- A new association record is being created.
- There is a unique index one of the association's fields.
- The record being created has the same value as the record being
destroyed on the indexed field.

Before, the deletion of records was postponed until after all
insertions/saves.  Therefore the new record with the identical value in
the indexed field caused a non-unique value error to be thrown at the database
level.

With this fix, the deletions happen first, before the insertions/saves.
Therefore the record with the duplicate value is gone from the database
before the new record is created, thereby avoiding the non-uniuqe value
error.

Postgresql array columns don't properly escape single quote strings when loading fixtures

"There're" => There are for better readability

Support transactions in Migrator.run

Conflicts:
	activerecord/CHANGELOG.md

Fix freeze applying to cloned objects

Reorganize MessageEncryptor

1) According to OpenSSL's documentation, cipher.random_iv must be called
   after cipher.encrypt and already sets the generated IV on the cipher.

2) OpenSSL::CipherError was moved to OpenSSL::Cipher::CipherError in
   Ruby 1.8.7. Since Rails 4 requires at least Ruby 1.9.3, support for
   the old location can be dropped.

renderer is clearer.

 - A string in an array of strings that has a quote char (') needs to have that quote char escaped if the array is getting wrapped in quote chars.

Otherwise we get extra indent for the <td> elements due to the extra
spaces at the beginning of the lines, like this:

      <tr>
          <td><%= author.name %></td>
          <td><%= link_to 'Show', author %></td>
        <td><%= link_to 'Edit', edit_author_path(author) %></td>

This reverts commit 521035af, reversing
changes made to 222011db.

Reason: This broke the tests

This reverts commit 57583879.

Conflicts:
	railties/lib/rails/generators/app_base.rb

Reason: sprockets 2.9.1 fixes this issue

DB with postgres string array column doesn't load fixtures well

Add support for extracting the port from the :host option and for
removing the subdomain by using nil, false or ''.