we should take disabled option not only from `html_options` hash but from
`options` hash too like `build_select` method does it. So

datetime_select("post", "updated_at", { :discard_minute => true }, { :disabled => true })
datetime_select("post", "updated_at", :discard_minute => true , :disabled => true)

both these variants work now

closes #7431

This functionality will be available from gem
`active_record-session_store` instead.

With a value of "nosniff", this prevents Internet Explorer from MIME-sniffing a response away from the declared content-type.

Updated documentation to demonstrate start_hour and end_hour options

Allow data attributes to be set as a first-level option for form_for, so you can write `form_for @record, data: { behavior: 'autosave' }` instead of `form_for @record, html: { data: { behavior: 'autosave' } }` *DHH*

We recommend the use of Unobtrusive JavaScript instead. For example:

  link_to "Greeting", "#", :class => "nav_link"

  $(function() {
    $('.nav_link').click(function() {
      // Some complex code

      return false;
    });
  });

or

  link_to "Greeting", '#', onclick: "alert('Hello world!'); return false", class: "nav_link"

for simple cases.

This reverts commit 3acdd652.

Revert "Merge pull request #7033 from kron4eg/master". Not a a fan at all of what this makes ERB files look like.

This reverts commit 46b8bcee, reversing
changes made to 2f58795e.

Restoring the '%' trim mode for ERb templates, allowing for a leading percent sign on a line to indicate non-inserted Ruby code.

Rails were including 'application.js' to the pack when using
`javascript_include_tag :all` even there's no application.js in the
public directory.

Passing options as the last value in an array doesn't work with form_for.
This reverts commit 6be564c7.

Rather than keep the url options in record_or_hash_or_array, extract it
and reverse merge with options as it may contain important private keys
like `:routing_type`.

Closes #7259

This applies to the following helpers:
`button_to`
`button_tag`
`image_submit_tag`
`link_to`
`submit_tag`

Also fix some wrong formatting.

Related discussion:
https://github.com/rails/rails/commit/ab72040b74f742b6676b2d2a5dd029bfdca25a7a#commitcomment-1525256

Similarly to 65250022, this allows to use routes helpers for mounted
helpers, but this time in ActionView::TestCase

In integration tests, you might want to use helpers from engines that
you mounted in your application. It's not hard to add it by yourself,
but it's unneeded boilerplate. mounted_helpers are now included by
default. That means that given engine mounted like:

    mount Foo::Engine => "/foo", :as => "foo"

you will be able to use paths from this engine in tests this way:

    foo.root_path #=> "/foo"

(closes #6573)

Prior to this patch the existing .force_ssl method handles both defining
the filter and handling the logic for performing the redirect.

With this patch the logic for redirecting to the HTTPS protocol is
separated from the filter logic that determines if a redirect should
occur.  By separating the two levels of behavior, an instance method
for ActionController (i.e. #force_ssl_redirect) is exposed and available
for more granular SSL enforcement.

Cleaned up indentation.

Make possible to use a block in button_to helper if button text is hard
to fit into the name parameter, e.g.:

    <%= button_to [:make_happy, @user] do %>
      Make happy <strong><%= @user.name %></strong>
    <% end %>
    # => "<form method="post" action="/users/1/make_happy" class="button_to">
    #      <div>
    #        <button type="submit">
    #          Make happy <strong>Name</strong>
    #        </button>
    #      </div>
    #    </form>"

When more than one directory for helpers is provided to a controller, it
should preserver the order of directories. Given 2 paths:

    MyController.helpers_paths = ["dir1/helpers", "dir2/helpers"]

helpers from dir1 should be loaded first. Before this commit, all
helpers were mixed and then sorted alphabetically, which essentially
would require to rename helpers to get desired order.

This is a problem especially for engines, where you would like to be
able to predict accurately which engine helpers will load first.

(closes #6496)

This options can be used to not escape the result by default.