We are setting a limit unconditionally just below,
which overrides any existing one anyway.

Handle `Rack::QueryParser` errors in `ActionDispatch::ExceptionWrapper`

Give importance to rails command [ci skip]

Correct link for model generator field types

The most complete list of field types appears in the `SchemaStatements` docs rather than the `TableDefinition` docs.

The change to link to `SchemaStatements` means that the explanation for the `index` parameter is no longer available on the linked-to page. A brief explanation of the `index` parameter is added directly in the guide to make up for this.

Wrap module around lazy load hooks

This test was added to protect the test suite from our mistakes but now
it is failing because bundler does add duplicated libs in the load path
by design (if the repository has more than one gem)

Rack [recently](https://github.com/rack/rack/commit/7e7a3890449b5cf5b86929c79373506e5f1909fb)
moved the namespace of its `ParameterTypeError` and `InvalidParameterError`
errors. Whilst an alias for the old name was added, the logic in
`ActionDispatch::ExceptionWrapper` was still broken by this change, since it
relies on the class name.

This PR updates `ActionDispatch::ExceptionWrapper` to handle the Rack 2.0
namespaced errors correctly. We no longer need to worry about the old names,
since Rails specifies Rack ~> 2.0.

[ci skip] Update redirecting links in guides

Speed up RDoc generation when it's already been done once

rails -> Rails [ci skip]

Raise ArgumentError for bad strptime arguments

Fix for issue https://github.com/rails/rails/issues/25784

Prior to this commit the lazy_load_hooks.rb file contained important lazy load
hooks. Since [7c90d91c](https://github.com/rails/rails/commit/7c90d91c3c43bdbba25d38589aed0e2940af3bc8) the [documentation](http://api.rubyonrails.org/files/activesupport/lib/active_support/lazy_load_hooks_rb.html) did not display
the comments in this file as the docs for load hooks.

This commit wraps the code within this file in a module so we can display the
documentation for `ActiveSupport` load hooks. By extending `ActiveSupport` with
this module, all the methods within it should still be accessible through
`ActiveSupport`.

Remove duplicate test and fix a typo in the test

- Tests for dup'ing params was separately added in a separate file in
  https://github.com/rails/rails/pull/25735.

Stop changes to a dupped `ActionController::Parameters` mutating the original

The current implementation serializes zero-length durations incorrectly (it serializes as `"-P"`), and cannot un-serialize itself:

```
[1] pry(main)> ActiveSupport::Duration.parse(0.minutes.iso8601)
ActiveSupport::Duration::ISO8601Parser::ParsingError: Invalid ISO 8601 duration: "-P" is empty duration
from /Users/rando/.gem/ruby/2.3.1/gems/activesupport-5.0.0/lib/active_support/duration/iso8601_parser.rb:96:in `raise_parsing_error'
```

Postgres empty intervals are serialized as `"PT0S"`, which is also parseable by the Duration deserializer, so I've modified the `ISO8601Serializer` to do the same.

Additionally, the `#normalize` function returned a negative sign if `parts` was blank (all zero). Even though this fix does not rely on the sign, I've gone ahead and corrected that, too, in case a future refactoring of `#serialize` uses it.

Fix Active Record test failure

Caused at a45363a2.

[ci skip] add class level documentation to ActiveModel::Type::Boolean

When `group` is used in combination with any calculation method, the
resulting hash uses the grouping expression as the key. Currently we're
incorrectly always favoring the type reported by the query, instead of
the type known by the class. This causes differing behavior depending on
whether the adaptor actually gives proper types with the query or not.
After this change, the behavior will be the same on all adaptors -- we
see if we know the type from the class, fall back to the type from the
query, and finally fall back to the identity type.

Fixes #25595

add documentation of the behaviors of type coercion at the class level

Only include files that contain changes since the last generation
of the API documentation.

Thus, only modified files are re-generated instead of the whole API.

The `created.rid` file won't contain the whole list of files anymore
if we generate several times the API but we don't really care about it,
only the generation date and time is important.

It is possible to fall back to the previous behavior by defining the ALL
environment variable running the `rake rdoc` task.

Update class_attribute docs

Add tests for 1xx, 204 and 304 responses to response_test.rb

remove unnecessary send

`class_variable_get` is public from Ruby 1.9.
Ref: http://ruby-doc.org/core-2.2.0/Module.html#method-i-class_variable_get

Cleanup CHANGELOGs [ci skip]

Fix AJ tests on ruby 2.4 being caused since classes are unified for Integer

Fix AJ tests on ruby 2.4 being causes since classes are unified for Integer, and we create test name nased on arg class. Append ar as well to the test name

- Cleanup Active Record CHANGELOG.
- Add missing CHANGELOG for https://github.com/rails/rails/pull/25688.
- Clarify that assets requests logging is suppressed.

In response_test.rb, we haven't had a test to make sure that

1) these responses don't have a message-body as described in RFC7231[1]

2) 1xx and 204 responses must not have a Content-Length header field
   as described in RFC7230-section3.3.2[2]

[1] https://tools.ietf.org/html/rfc7231
[2] https://tools.ietf.org/html/rfc7230#section-3.3.2

Even though our implementation doesn't allow users to send
a Content-Length header field in a 304 response, sending the
header field is valid as mentioned in RFC7230-section3.3.2[2].
So I've decided not to test whether or not a 304 response has
the header.

The citation from the section is as follows;
```
A server MAY send a Content-Length header field in a 304 (Not
Modified) response to a conditional GET request (Section 4.1 of
[RFC7232]); a server MUST NOT send Content-Length in such a response
unless its field-value equals the decimal number of octets that would
have been sent in the payload body of a 200 (OK) response to the same
request.
```