When an application has multiple root entries with different
constraints, the current solution is to use `get '/'`. Example:

**Currently I have to do:**
```ruby
get '/', to: 'portfolio#show', constraints: ->(req) { Hostname.portfolio_site?(req.host) }
get '/', to: 'blog#show',      constraints: ->(req) { Hostname.blog_site?(req.host) }
root 'landing#show'
```

**But I would like to do:**
```ruby
root 'portfolio#show', constraints: ->(req) { Hostname.portfolio_site?(req.host) }
root 'blog#show',      constraints: ->(req) { Hostname.blog_site?(req.host) }
root 'landing#show'
```

Other URL matchers such as `get`, `post`, etc, already allows this, so I
think it's fair that `root` also allow it since it's just a shortcut for
a `get` internally.

Fix cache fetch instrumentation

used predicate methods to avoid is_a? checks

update to ruby-oci8 - 2.2.0

It stopped `ruby 1.8` support, we already switched to `ruby >= 2.2.2`

monitor is require for SneakersAdapter

change `prepare_destination` to public API [ci skip]

With the previous implementation, the block passed to
define_singleton_method, which will live forever as the method body,
captures the parameters (args and block) in its enclosure.

For the current_scope registry, that can include an AR::Relation.

Modify the scope method documentation

Follow DRY principle and remove duplication

Remove unused require

`require 'active_support/core_ext/string/filters'` was added in b3bfa361. However, it is no longer needed from 3ae98181.

Minor cleanup for form helpers guide

remove unused require

Fix subsecond precision supported version

Improve readability in CSRF section of guide [ci skip]

`shellwords` is no longer needed from #20605.

Remove unused `pk_and_sequence_for` in AbstractMysqlAdapter

Fractional Seconds supported version is 5.6.4 or higher.

http://dev.mysql.com/doc/refman/5.6/en/fractional-seconds.html

`prepare_destination` has been used in the template file for the generator,
I think it should be a public API

ref: https://github.com/rails/rails/blob/master/railties/lib/rails/generators/test_unit/generator/templates/generator_test.rb#L8

Adds a paragraph to the documentation of the `ActiveRecord::Scoping::Named.scope` method,
explaining that the method is intended to return an ActiveRecord::Relation object to be
composable with other scopes.

In the case that in the case that `nil` or `false` are returned, the method returns
an `all` relation instead.
This unexpected behaviour is mentioned in #19249 #14256 #21465 and #21882 and wasn't
documented at all. This commit adds this documentation.

Saw this while doing a review of a patch:

* Normalize case and punctuation across comments.
* ascii -> ASCII
* Since I was on it, some blank lines that visually
  add some clarity IMO.

Mounted engine route fix

Change the reporter to just read the option.

Pass output_inline where needed in tests.

Lets us cut the verbose and straight up duplicated setup in 3 tests down to one line.

We'd see the failures and errors reported after the run, which is needless, when we've already
reported them.

Turns:

```

.......................................S....................F

This failed

bin/rails test test/models/bunny_test.rb:14

....

Finished in 0.100886s, 1020.9583 runs/s, 1001.1338 assertions/s.

  2) Failure:
BunnyTest#test_something_failing [/Users/kasperhansen/Documents/code/collection_caching_test/test/models/bunny_test.rb:15]:
This failed

103 runs, 101 assertions, 1 failures, 0 errors, 1 skips

You have skipped tests. Run with --verbose for details.
```

Into:

```

...................S.......................................F

This failed

bin/rails test test/models/bunny_test.rb:14

......................

Finished in 0.069910s, 1473.3225 runs/s, 1444.7143 assertions/s.

103 runs, 101 assertions, 1 failures, 0 errors, 1 skips
```

Added missing specs for not modifying queues when using AJ test helpers

This reverts commit afe67218.

Reason: This change was done by mistake. This option is at the `config`
object not in the `config.assets`.

See
https://github.com/rails/rails/blob/cfb371e96a2789790359ec1db779df490fd253d6/railties/lib/rails/application/configuration.rb#L16

`pk_and_sequence_for` is implemented for PG and MySQL adapters (not
implemented for Sqlite3 adapter). But MySQL adapters are not using
`pk_and_sequence_for` already.

When generating the url for a mounted engine through its proxy, the path should be the sum of three parts:

1. Any `SCRIPT_NAME` request header or the value of `ActionDispatch::Routing::RouteSet#relative_url_root`.
2. A prefix (the engine's mounted path).
3. The path of the named route inside the engine.

Since commit https://github.com/rails/rails/commit/44ff0313c121f528a68b3bd21d6c7a96f313e3d3, this has been broken. Step 2 has been changed to:

2. A prefix (the value of `ActionDispatch::Routing::RouteSet#relative_url_root` + the engine's mounted path).

The value of `ActionDispatch::Routing::RouteSet#relative_url_root` is taken into account in step 1 of the route generation and should be ignored when generating the mounted engine's prefix in step 2.

This commit fixes the regression by having `ActionDispatch::Routing::RouteSet#url_for` check `options[:relative_url_root]` before falling back to `ActionDispatch::Routing::RouteSet#relative_url_root`. The prefix generating code then sets `options[:relative_url_root]` to an empty string. This empty string is used instead of `ActionDispatch::Routing::RouteSet#relative_url_root` and avoids the duplicate `relative_url_root` value in the final result.

This resolves #20920 and resolves #21459

Fix "Esperanto" name

Correct spelling is "Esperanto": 

https://en.wikipedia.org/wiki/Esperanto

Fix typo in docs

Fix pluralization in docs [ci skip]