Secondly it seemed from the method that the intent was to test a case
where layout was declared in a symbol and the method named mention in
layout returns nil.

That case is already covered with class     class WithSymbolReturningNil .

Also the case of SymbolReturningString is covered with the class
WithSymbol.

If you set a js_compressor, it means that you want shit compressed -- we dont need a second option to turn that on

Update CHANGELOG to mention the new SafeBuffer change

Make escape_javascript happy to handle SafeBuffers

* see GH#1553
* allow for the fact that gsub on SafeBuffer does not pass match variables $1, $2 etc to a block

Make test meaningful

If before_filter list was being called mistakenly then
the test should fail. However test will not fail because
second filter is assigning new values to @list. To truly
test that first before_filter is not called when it should
not be called then @list should not assigned value unconditionally.

This patch will make the test fail if first filter is called.

Brought back alternative convention for namespaced models in i18n

Do not hardcode :id as it was ignored by mass assignment filter

As a result test was failing on Oracle where ids are assigned by default from 1000

test name should reflect the name of the method

remove unused class from test

that is actually being called

Fix ActionPack tests on `master`

Don't operate on the safebuffer, operate on a string and convert.

This was introduced in [2684f17a] 💣

* Fix the problem on manipulating on the `ActiveSupport::SafeBuffer`
* Make sure that we run `escape_javascript` on the `String`, to avoid unexpected behavior.

Adapt [823aa223] to make sure we perform cloning before manipulation only on `OutputBuffer`.

After the fragment rendering, `Builder` returns the `String` object instead of `ActionView::OutputBuffer`. Somehow the same procedure which was in [823aa223] does not play nice with the String, and result in the fragment got lost.

Prefer 'each' over 'for in' syntax.

Fragment caching needs to operate on the pure output, not the safebuffer.

safebuffer.

Also make sure that the versions of those methods which modify a string in place such as gsub! can't be called on safe buffers at all.

Conflicts:

	activesupport/test/safe_buffer_test.rb

Allow polymorphic has_one to work when the association is set before the owner has been saved. Fixes #1524.

When you add a record to a polymorphic has_one, you should be able to access the owner from the associated record

Fix typo

Fixes #1489