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

The MD5 sum of an attachment is only necessary if the model has a corresponding
_fingerprint column. If this column is absent, there is no need to calculate the
MD5, which can be an expensive operation for large files.

Accomplish this by deferring the fingerprint calculation using a block. If the
_fingerprint column is absent, the block is never called, and the calculation
is avoided.

These variables were introduced in:
https://github.com/thoughtbot/paperclip/commit/6ea7c26512434a
but their usage were canceled in:
https://github.com/thoughtbot/paperclip/commit/c740fb171fe

Extract another repository only concerned with locales:
https://github.com/thoughtbot/paperclip-i18n

A user of paperclip might choose to add `paperclip-i18n` as a dependency
for their application, or copy over the related locale file into their
project.

This is a first step in splitting some parts of the codebase into other
projects. Simple and relevant contributions like new translations might
get lost in the shuffle of dozens of PRs awaiting for approval, that
take time to review and merge.

Locales are a simple enough problem to be handled in an external
repository, where tests for consistency across translations run in less
than a second, and PRs can almost trivially be merged in.

the name string is used multiple times in interpolation so storing it reduces object creation

this commit primarily uses frozen strings to reduce object creation during interpolation.
the :basename method now uses File.basename(file, ".*") rather than a Regexp. basename may be called multiple times.

by using the attachment name (symbol) and Class as keys we reduce the number of Strings created before hitting the cache

* Regression fix
* Add specs for intermediate_files var in Paperclip::Attachment

[fixes #1908]

Before the error message was sometimes millions of characters long.

> [paperclip] Content Type Spoof: Filename a.csv
> (application/octet-stream from Headers,
> [#<MIME::Type::Columnar:0x007f9f90f89fa8
> @container=#<MIME::Types:0x007f9f90b09d98 ... snip millions of
> characters of output here ...>], @content_type="text/csv",
> @raw_media_type="text", @raw_sub_type="csv", @simplified="text/csv",
> @i18n_key="text.csv", @media_type="text", @sub_type="csv",
> @extensions=["csv"]>] from Extension), content type discovered from
> file command: application/zip. See documentation to allow this
> combination.

Now becomes:

> [paperclip] Content Type Spoof: Filename a.csv
> (application/octet-stream from Headers,
> ["text/comma-separated-values", "text/csv"] from Extension), content
> type discovered from file command: application/zip. See documentation
> to allow this combination.

[fixes #2017]

Removes note for running paperclip in EOL'd version of Rails (2.3).

Removes `column_type` and `column_options` so that it works well in both
versions.

Backcported from master, to release a patch version with this bugfix.

[closes #1947]

[fixes #1977]

Removes `column_type` and `column_options` so that it works well in both
versions.

[closes #1947]

Turns out it didn't fix test failures like I thought. It only added
more!

This reverts commit b8221c6b.

Both v1 and v2 of the AWS Ruby SDK accept the string literal AES256, no need to differentiate. (thank you, @trevorrowe)

Further changes to Paperclip::Storage::S3 for choosing aws-sdk-v2 before aws-sdk-v1; use object.upload_file rather than object.put; implement v2 expiring_url method

According to the ImageMagick documentation the '@' flag is a special case that
can be used in conjunction with '>' to prevent IM from scaling an image up.
The '<' flag is ignored when using '@'.

http://www.imagemagick.org/Usage/resize/#pixel

This fixes the issue I reported in #1851