be ignored. A regular expression constraint gets overwritten when the
routes.rb file is processed. Changed the overwriting to an ||= instead
of an = assignment.

Override <%== to always behave as literal text rather than toggling based on whether escaping is enabled. Fixes that existing plaintext email templates using <%== unexpectedly flipped to *escaping* HTML when #8235 was merged.

This reverts commit 90c887fa.

[ci skip]

They was extracted from a plugin.

See https://github.com/rails/rails-observers

[Rafael Mendonça França + Steve Klabnik]

Instead of deleting the skip_digest option flag, this changes the method to merely check the
key. This change is because of a discussion in this thread:

https://github.com/rails/rails/pull/8317

This commit also makes #fragment_name_with_digest private due to its
functionality being subsumed by #cache_fragment_name.

It's only possible to inherit from ActiveRecord::Base and not include
it.

This file was using mime_types before load the Mime::Type class.
When trying to register first Mime::Type it load mime_type that loads
mime_types in the end.

Requiring mime_type ensure that we have the class definition and the
mime types

This was forgotten when we did the merge.

Move the routes for each test inside the test method so that
it's easier to see which routes are applicable to which test.

To ensure that each test wasn't invalidated the changes were
done by first removing all of the routes, ensuring that all
of the tests failed and then adding the routes back to each
test one by one. One test for `assert_recognizes` was
removed as it wasn't actually testing the defined routes and
is now tested more thoroughly in routing_assertions_test.rb.

One downside is that the test suite takes about 1s longer
due to having to using `method_missing` for handling the url
helpers as using `include url_helpers` isn't isolated
for each test.

This add support for sending an explicit opt-out of the "Russian-doll"
cache digest feature on a case-by-case basis. This is useful when cache-
expiration needs to be performed manually and it would be otherwise
difficult to know the exact name of a digested cache key.

More information: https://github.com/rails/cache_digests/pull/16

This yields a small bit of performance improvement when building the
defaults from constraints, specially considering that it's rather common
for constraints to be empty.

Also, there's a bit of duplicated code in here that I have to check
before extracting.