Thanks for the tip, @brendon!
https://github.com/thoughtbot/paperclip/pull/2331#issuecomment-262910522

Allows `s3_host_alias` to be set to CDNs which support S3 Bucket prefix.

[fixes #2287]

Bug introduced in 1c7d7f66

Per https://github.com/thoughtbot/paperclip/commit/1c7d7f66ceea27cfe604e3a13c2eb87df217d13a#commitcomment-18765681 request

[Amazon S3 Transfer
Acceleration](http://docs.aws.amazon.com/AmazonS3/latest/dev/transfer-acceleration.html)
enables fast, easy, and secure transfers of files over long distances
between your client and an S3 bucket. And
[aws-sdk](https://github.com/aws/aws-sdk-ruby) has allowed us to enable
this feature in version 2.3.0 (more detail is
[here](https://github.com/aws/aws-sdk-ruby/pull/1163)).

You can now pass a configuration option (:use_accelerate_endpoint) when
config Paperclip with S3. You can enable the [Amazon S3 Transfer
Acceleration](http://docs.aws.amazon.com/AmazonS3/latest/dev/transfer-acceleration.html)
by set this option:
```:use_accelerate_endpoint => true```

[fixes #2291]

* correct S3 specs to use the correct keys in the s3_credentials hash
* S3 tests work with a region that requires s3_domain_url

[fixes #1985]

Adapters now accept an options parameter, that currently specifies
the type of hash digest to use.  The default value remains MD5, but
can be specified to be any OpenSSL-supported digest.  The specs are
modified to reflect that.

The task just reassigns all of the attachments, thereby regenerating
their fingerprints.

[closes #2155]

* Add default `content_type_detector` to `UploadedFileAdapter`

Default `s3_protocol` used to be `http` when `:s3_permissions` are
`:public_read` (default), and `https` when `:s3_permissions` are
anything else.

With an empty String as default, if the page is served over HTTPS,
attachment URL will be HTTPS, and if the page is served over HTTP,
attachment url will be HTTP.

`public-read` is an authorization concept, independent form the
encryption of the HTTP connection used to read the file. As such, the
one shouldn't define the other.

[fixes #2038]

If `only_process` list is not empty, but it doesn't contain `:original`
style, original file hasn't been reprocessed and it's not needed to
rewrite/reupload it.

[fixes #1993]
[fixes #2046]
[fixes #1804]

There's a subtle difference between what `\Z` and `\z` consider the "end
of string" which is that the uppercase version allows a single trailing
newline:

```
/\Afoo\Z/.match("foo\n")

/\Afoo\Z/.match("foo\n\n")

/\Afoo\z/.match("foo\n")
```

* There is no need to pass options argument to UrlGenerator
* Houndci compliance

I recently implemented a custom processor that used the Exifr gem to
extract EXIF information from images uploaded to a Paperclip attachment.
Exifr's processor uses readbyte to parse the EXIF header, so it hit
errors when Paperclip's File-like object didn't have one. There's also a
test for this delegation.

I've also tidied up the README documentation for custom processors,
hopefully to be more clear than before. There was some duplicated
content between the "Post Processing" section and the "Custom Attachment
Processing" section, and those sections were separated in the file. I've
dedicated the "Post Processing" section to Paperclip's built in
thumbnailing processors, and made "Custom Attachment Processing" section
solely about writing & using your own custom processors. This should be
a more understandable progression, as built-in functionality is
discussed first & separately from extending Paperclip.k

- The uri io adapter now seeks for the content-disposition header
  if this is pressent the value filename is taken instead of the
  last path segment for the resource file name
- Fix style comments
- Applied the Tute Costa refactor to URI Adapter.
- Added entry to  the NEWS file.
- Removed editor tracking file
- Fix test cases

AWS no longer accepts `:aes256` as an encryption value and will raise
`Aws::S3::Errors::InvalidArgument` with the message `The encryption method
specified is not supported`. It should instead be `"AES256"`.

This commit makes no implementation change, but for those that use
source code and tests as documentation, it should help prevent
confusion.

* Add `read_timeout` option to `UriAdapter#download_content` method
* Default `read_timeout` is `nil`
* Update NEWS

* replaced if-not-blank with if-present

an instance of code was testing presence via `!a.blank?`. this is now
replaced with a more readable version: `a.present?`

* replaced 2 if-not-blank checks with if-present checks

* temporarily restrict travis

* reverted travis changes

* readded whitespace

Note: this is most likely not the way we want to test our matchers, but it
demonstrates the current issue.

Excessive logging can fill up disk space and become a denial of service attack, see https://cwe.mitre.org/data/definitions/779.html

This is part of the effort started at
https://github.com/thoughtbot/paperclip/issues/2199 to ensure the test
suite is green locally.

The attachment processing spec was failing since the setup required done
in a before block, namely a call to `rebuild_class`, was running
restricted to one of two contexts in the test file.
This was leaving the `Dummy` class in a bad state and affected other tests.

This PR moves the `before` block to run for the both of the contexts.

This is part of the effort started at
https://github.com/thoughtbot/paperclip/issues/2199 to ensure the test
suite is green locally.

The validators spec was failing since the setup required was missing a
call to 'rebuild_class' and leaving the Dummy class in a bad state.

This PR adds the call to the before block setup.

This is part of the effort started at https://github.com/thoughtbot/paperclip/issues/2199)
to ensure the test suite is green locally.

This spec was failing on obtaining a S3 instance with an `nil` region
key. Since `Aws::Resource` does a regex match against the region key
passed in, we would fail on `NoMethodError`.

This corrects the issue by setting a region key to an empty string in
the test setup.

[fixes #1932]

Currently paperclip fails with `NoMethodError` on `nil` when file is not
present. Since `nil` is proper response from fog on missing files, and
this method already returns `false` if any error happen (see line 178),
returning `false` here would be properly handled on client code.

[closes #2173]

Adds failing spec to command line dup check

[fixes #1670]
[fixes #2169]

`Paperclip::Storage::Fog#host_name_for_directory` assumes a String-like
object is set and doesn't check if `@options[:fog_directory]` is
callable, while `Paperclip::Storage::Fog#directory` does. This extracts
a new method with the condition and refactors the other two methods to
call it.

Fixes #2018, #2093.

This configuration constant was useful for supporting different versions
of AWS SDK. We don't any longer, and we can set the value directly.

[fixes #2148]

We don't support AWS SDK v1 any longer.

Drops Rails 3 branches.

[fixes #2101]

Rails 3 is not supported anymore. We can safely remove mentions and
snippets for it.

Drop unused variable definition (`options`).

While working on this branch, Kito found that this test fails due to:

https://github.com/rails/rails/commit/6ec8ba16d85d5feaccb993c9756c1edcbbf0ba13#diff-fdcf8b65b5fb954372c6fe1ddf284c78R76

We are not yet sure if it's a bug in paperclip or in Rails itself. With
current `ActiveModel::Errors` implementation the following happens:

```
record.errors # => @messages = {}
record.errors.include?(:foo) # => false
record.errors # => @messages = { :foo => [] }
```

Which bit us in:
https://github.com/thoughtbot/paperclip/blob/69f18375333234b6f395300266e2612936bd242e/lib/paperclip/validators/attachment_file_name_validator.rb#L23

Another related Rails commit:
https://github.com/rails/rails/commit/b97035df64f5b2f912425c4a7fcb6e6bb3ddab8d

I worked around the issue by changing what we assert in this spec. I am
still not sure that this is a bug in current Rails master.

cc @kitop for review

Ref: https://github.com/rails/rails/commit/481e49c64f790e46f4aff3ed539ed227d2eb46cb

`silence_stream` was deprecated in Rails 4.2 and removed in Rails 5
because it was not thread safe. Paperclip is not running the tests in
threaded mode, it's safe to add the method back in a test support
module.

`bundle exec rails -v` was taking seconds every time, making the test
suite run very slow. This changes the code to compare the version
against `ActiveRecord::VERSION`, which is going to be the same one as
`rails -v`, making it much, much faster.

Simplifies condition: we no longer support Rails < 4.2, so we test for
`>= "5.0"` or fallback to `4.2` branch.

- Also bump Travis Ruby 2.2.x version to 2.2.4

We found that uploading large files to S3 would result in a socket error
("connection reset by peer") occasionally and lately much more
consistently. In researching this I saw that many people got this error
when uploading too large of a file without multipart chunking. I would
have assumed fog did this automatically but the default chunk size may
be too high. In order to address this I wanted to drop the chunk size to
100MB.

Rather than hard-code this I opted to expose a `fog_option`
configuration option that lets me pass any additional options I want to
the fog's `#create` call. This is similar to the `fog_attributes` option
implemented in CarrierWave which [addresses the same
problem](http://stackoverflow.com/a/11867978/201911).

We've been running this now for a week in production and it seems to
resolve the issue.

https://github.com/thoughtbot/paperclip/pull/2135