avoid freezing the headers until the web server has actually read data
from the body proxy.  Once the webserver has read data, then we should
throw an error if someone tries to set a header

when streaming responses, we need to make sure the cookie jar is written
to the headers before returning up the stack. This commit introduces a
new method on the response object that writes the cookie jar to the
headers as the response is committed.  The middleware and test framework
will not write the cookie headers if the response has already been
committed.

fixes #14352

detect the type of controller we're testing and return the right type of
response based on that controller.  This allows us to stop doing the
weird sleep thing.

Commit 4f2cd3e9 introduced a bug by reordering the call to
`@controller.recycle!` above the call to `build_request_uri`. The
impact of this was that the `@_url_options` cache ends up not being
reset between building a request URI (occurring within the test
controller) and the firing of the actual request.

We encountered this bug because we had the following setup:

  class MinimumReproducibleController < ActionController::Base
    before_filter { @param = 'param' }

    def index
      render text: url_for(params)
    end

    def default_url_options
      { custom_opt: @param }
    end
  end

  def test_index
    get :index # builds url, then fires actual request
  end

The first step in  `get :index` in the test suite would populate the
@_url_options cache. The subsequent call to `url_for` inside of the
controller action would then utilize the uncleared cache, thus never
calling the now-updated default_url_options.

This commit fixes this bug calling recycle! twice, and removes a call
to set response_body, which should no longer be needed since we're
recycling the request object explicitly.

The request attributes filtered_parameters, filtered_env and filtered_path
are memoized for performance reasons. However this can cause unusual
behavior in tests where there are multiple calls to get, post, etc.

Fixes #13803.

This reverts commit 0303c232.

Conflicts:
	actionpack/lib/action_controller/test_case.rb

REASON: It will conflict with a lot of test cases. Better to call
`process` directly since this is a very uncommon HTTP method.

Fixes #10638.

This worked in Rails 3.2, but was a regression in 4.0.0.beta1

previously when a partial was placed inside a directory
(eg. '/dir/_partial'), `assert_template` did not replace
the '_' prefix when looking through rendered tempaltes,
which resulted in an error.

I modified it to replace both, the leading '_' and the last '_'
after a '/'.

When `assert_template` is used with the :locals option, and the
partial was not rendered, a method_missing error was raised.
This changes first checks, if the partial actually was rendered
and raises a descriptive error.

Bring back the same API we have with Request::Session.

All ActionPack and Railties tests are passing. Closes #8891.

[Carlos Antonio da Silva + Santiago Pastorino]

This reverts commit 637a7d9d, reversing
changes made to 5937bd02.

For instance, it prevents false positive in this case:

    file = nil
    get :index
    assert_template("#{file}")

Store FlashHashes in the session as plain hashes rather than custom objects with unstable class names and instance variables.

Refactor FlashHash to take values for its ivars in the constructor, to pretty up FlashHash.from_session_value.

Remove stale comment on FlashHash: it is no longer Marshaled in the session so we can change its implementation.

Remove blank lines I introduced in controller/test_case.rb.

Unit tests for FlashHash#to_session_value.

Put in a compatibility layer to accept FlashHash serializations from Rails 3.0+.

Test that Rails 3.2 session flashes are correctly converted to the new format.

Remove code path for processing Rails 3.0 FlashHashes since they can no longer deserialize.

Fix session['flash'] deletion condition: it will never be empty?, it will either be nil or a hash with 'discard' and 'flashes' keys.

This reverts commit 36376560, reversing
changes made to 3148ed9a.

Conflicts:
	actionpack/lib/action_dispatch/middleware/flash.rb

Reason: it broke Sam's CI

https://github.com/rails/rails/pull/8017#issuecomment-10210655

Store FlashHashes in the session as plain hashes rather than custom objects with unstable class names and instance variables.

Refactor FlashHash to take values for its ivars in the constructor, to pretty up FlashHash.from_session_value.

Remove stale comment on FlashHash: it is no longer Marshaled in the session so we can change its implementation.

Remove blank lines I introduced in controller/test_case.rb.

Unit tests for FlashHash#to_session_value.

Put in a compatibility layer to accept FlashHash serializations from Rails 3.0+.

Test that Rails 3.2 session flashes are correctly converted to the new format.

Remove code path for processing Rails 3.0 FlashHashes since they can no longer deserialize.

It's sometimes hard to quickly find where deprecated call was performed, especially in case of migrating between Rails versions. So this is an attempt to improve the call stack part of the warning message by providing caller explicitly.

this refactoring extracts the semi complex data structure of rendered locals
per view into into a separate class

the documentation on #assert_template states that the :locals option is
only available in view test cases:

    # In a view test case, you can also assert that specific locals are passed
    # to partials:

I added a warning when it's passed in an inapropriate context to prevent
a NoMethodError.

Given Im rendering an template `/layout/hello.html.erb`, assert_template was
passing with any string that matches. This behavior allowed false passing like:

	assert_template "layout"
	assert_template "out/hello"

Now the passing possibilities are:

	assert_template "layout/hello"
	assert_template "hello"

fixing assert_template bug when template matches expected, but not ends with

Cherry Pick Merge: Fixes issue #3849 assert_template false positive

taking redundant test off

prevening incorrect assert_template when rendering with repeated names in path

updating CHANGELOG with bugfix: assert_template false passing