imgproxy is a fast and secure standalone server for resizing and converting remote images. The main principles of imgproxy are simplicity, speed, and security. It is a Go application, ready to be installed and used in any Unix environment—also ready to be containerized using Docker.
imgproxy can be used to provide a fast and secure way to get rid of all the image resizing code in your web application (like calling ImageMagick or GraphicsMagick, or using libraries), while also being able to resize everything on the fly on a separate server that only you control. imgproxy is fast, easy to use, and requires zero processing power or storage from the main application. imgproxy is indispensable when handling image resizing of epic proportions, especially when original images are coming from a remote source.
imgproxy.rb is a framework-agnostic Ruby Gem for imgproxy that includes proper support for Ruby on Rails' most popular image attachment options: Active Storage and Shrine.
Important
This readme shows documentation for version 3.x.
See Upgrading imgproxy.rb for the upgrade guide.
Add this to your Gemfile
:
gem "imgproxy"
or install system-wide:
gem install imgproxy
imgproxy.rb uses anyway_config to load configuration, so you can configure it in different ways.
With a separate config file:
# <Rails root>/config/imgproxy.yml
development:
# Full URL to where your imgproxy lives.
endpoint: "http://imgproxy.example.com"
# Hex-encoded signature key and salt
key: "your_key"
salt: "your_salt"
production: ...
test: ...
With a secrets.yml
entry for imgproxy:
# secrets.yml
production:
...
imgproxy:
# Full URL to where your imgproxy lives.
endpoint: "http://imgproxy.example.com"
# Hex-encoded signature key and salt
key: "your_key"
salt: "your_salt"
...
With environment variables:
IMGPROXY_ENDPOINT="http://imgproxy.example.com" \
IMGPROXY_KEY="your_key" \
IMGPROXY_SALT="your_salt" \
rails s
...or right in your application code:
# config/initializers/imgproxy.rb
Imgproxy.configure do |config|
# Full URL to where your imgproxy lives.
config.endpoint = "http://imgproxy.example.com"
# Hex-encoded signature key and salt
config.key = "your_key"
config.salt = "your_salt"
end
endpoint
(IMGPROXY_ENDPOINT
) - Full URL to your imgproxy instance. Default:nil
.key
(IMGPROXY_KEY
) - Hex-encoded signature key. Default:nil
.salt
(IMGPROXY_SALT
) - Hex-encoded signature salt. Default:nil
.raw_key
(IMGPROXY_RAW_KEY
) - Raw (not hex-encoded) signature key. Default:nil
.raw_salt
(IMGPROXY_RAW_SALT
) - Raw (not hex-encoded) signature salt. Default:nil
.signature_size
(IMGPROXY_SIGNATURE_SIZE
) - Signature size. See URL signature section of imgproxy docs. Default: 32.use_short_options
(IMGPROXY_USE_SHORT_OPTIONS
) - Use short processing options names (rs
forresize
,g
forgravity
, etc). Default: true.base64_encode_urls
(IMGPROXY_BASE64_ENCODE_URLS
) - Encode source URLs to base64. Default: false.always_escape_plain_urls
(IMGPROXY_ALWAYS_ESCAPE_PLAIN_URLS
) - Always escape plain source URLs even when ones don't need to be escaped. Default: false.source_url_encryption_key
(IMGPROXY_SOURCE_URL_ENCRYPTION_KEY
) - Hex-encoded source URL encryption key. Default:nil
.raw_source_url_encryption_key
(IMGPROXY_RAW_SOURCE_URL_ENCRYPTION_KEY
) - Raw (not hex-encoded) source URL encryption key. Default:nil
.always_encrypt_source_urls
(IMGPROXY_ALWAYS_ENCRYPT_SOURCE_URLS
) - Always encrypt source URLs. Default: false.use_s3_urls
(IMGPROXY_USE_S3_URLS
) - Uses3://...
source URLs for Active Storage and Shrine attachments stored in Amazon S3. Default: false.use_gcs_urls
(IMGPROXY_USE_GCS_URLS
) - Usegs://...
source URLs for Active Storage and Shrine attachments stored in Google Cloud Storage. Default: false.gcs_bucket
(IMGPROXY_GCS_BUCKET
) - Google Cloud Storage bucket name. Default:nil
.shrine_host
(IMGPROXY_SHRINE_HOST
) - Shrine host for locally stored files.
imgproxy.rb comes with the Active Storage support built-in. It is enabled automagically if you load imgproxy
gem after rails
(basically, just put gem "imgproxy"
after gem "rails"
in your Gemfile
). Otherwise, modify your initializer at config/initializers/imgproxy.rb
:
# config/initializers/imgproxy.rb
Imgproxy.extend_active_storage!
Now, to add imgproxy processing to your image attachments, just use the imgproxy_url
method:
user.avatar.imgproxy_url(width: 500, height: 400, resizing_type: :fill)
This method will return a URL to your user's avatar, resized to fill 500x400px on the fly.
If you're a happy user of imgproxy Pro, you may find useful it's Getting an image info feature. imgproxy.rb allows you to easily generate info URLs for your images:
user.avatar.imgproxy_info_url(detect_objects: true, palette: 128)
This method will return a URL to the JSON with the requested info about your user's avatar.
If you have configured both your imgproxy server and Active Storage to work with Amazon S3, you can use use_s3_urls
config option (or IMGPROXY_USE_S3_URLS
env variable) to make imgproxy.rb use short s3://...
source URLs instead of long ones generated by Rails.
You can also enable gs://...
URLs usage for the files stored in Google Cloud Storage with use_gcs_urls
and gcs_bucket
config options (or IMGPROXY_USE_GCS_URLS
and IMGPROXY_GCS_BUCKET
env variables).
Important
You need to explicitly provide GCS bucket name since Active Storage "hides" the GCS config.
You can also use imgproxy.rb's built-in Shrine support. It is enabled automagically if you load imgproxy
gem after shrine
(basically, just put gem "imgproxy"
after gem "shrine"
in your Gemfile
). Otherwise, modify your initializer at config/initializers/imgproxy.rb
:
# config/initializers/imgproxy.rb
Imgproxy.extend_shrine!
Now you can use imgproxy_url
method of Shrine::UploadedFile
:
user.avatar.imgproxy_url(width: 500, height: 400, resizing_type: :fill)
This method will return a URL to your user's avatar, resized to fill 500x400px on the fly.
If you're a happy user of imgproxy Pro, you may find useful it's Getting an image info feature. imgproxy.rb allows you to easily generate info URLs for your images:
user.avatar.imgproxy_info_url(detect_objects: true, palette: 128)
This method will return a URL to the JSON with the requested info about your user's avatar.
Important
If you use Shrine::Storage::FileSystem
as storage, uploaded file URLs won't include the hostname, so imgproxy server won't be able to access them. To fix this, use shrine_host
config.
Alternatively, you can launch your imgproxy server with the IMGPROXY_BASE_URL
setting:
IMGPROXY_BASE_URL="http://your-host.test" imgproxy
If you have configured both your imgproxy server and Shrine to work with Amazon S3, you can use use_s3_urls
config option (or IMGPROXY_USE_S3_URLS
env variable) to make imgproxy.rb use short s3://...
source URLs instead of long ones generated by Shrine.
If you use another gem for your attachment operations, you like to keep things minimal or Rails-free, or if you want to generate imgproxy URLs for pictures that did not originate from your application, you can use the Imgproxy.url_for
method:
Imgproxy.url_for(
"http://images.example.com/images/image.jpg",
width: 500,
height: 400,
resizing_type: :fill
)
This method will return a URL to the image, resized to fill 500x400px on the fly.
If you're a happy user of imgproxy Pro, you may find useful it's Getting an image info feature. imgproxy.rb allows you to easily generate info URLs for your images:
Imgproxy.info_url_for(
"http://images.example.com/images/image.jpg",
detect_objects: true,
palette: 128
)
This method will return a URL to the JSON with the requested info about the image.
You can reuse processing options by using Imgproxy::Builder
:
builder = Imgproxy::UrlBuilders::Processing.new(
width: 500,
height: 400,
resizing_type: :fill,
sharpen: 0.5
)
builder.url_for("http://images.example.com/images/image1.jpg")
builder.url_for("http://images.example.com/images/image2.jpg")
info_builder = Imgproxy::UrlBuilders::Info.new(
detect_objects: true,
palette: 128
)
info_builder.url_for("http://images.example.com/images/image1.jpg")
info_builder.url_for("http://images.example.com/images/image2.jpg")
base64_encode_url
— per-call redefinition ofbase64_encode_urls
config.escape_plain_url
— per-call redefinition ofalways_escape_plain_urls
config.use_short_options
— per-call redefinition ofuse_short_options
config.encrypt_source_url
- (pro) per-call redefinition ofalways_encrypt_source_urls
config.source_url_encryption_iv
- (pro) an initialization vector (IV) to be used for the source URL encryption if encryption is needed. If not specified, a random IV is used.
See Supported processing options for the supported processing options list and their arguments.
See Supported info options for the supported info options list and their arguments.
Some of the processing and info options like crop
or gravity
may have multiple arguments, and you can define these arguments multiple ways:
First and the most readable way is to use a Hash
with named arguments:
Imgproxy.url_for(
"http://images.example.com/images/image.jpg",
crop: {
width: 500,
height: 600,
gravity: {
type: :nowe,
x_offset: 10,
y_offset: 5
}
}
)
# => .../c:500:600:nowe:10:5/...
You can find argument names on the Supported processing options and Supported info options pages.
You can use named arguments even if the option is not supported by the gem. In this case the arguments won't be reordered nor formatted, so you should provide them in the same order and right the same way they should appear in the URL:
Imgproxy.url_for(
"http://images.example.com/images/image.jpg",
unsupported: {
arg1: 1,
nested1: {
arg2: 2,
nested2: {
arg3: 3
}
}
}
)
# => .../unsupported:1:2:3/...
The arguments of the complex options can be provided as an array of formatted values:
Imgproxy.url_for(
"http://images.example.com/images/image.jpg",
crop: [500, 600, :nowe, 10, 5],
trim: [10, "aabbcc", 1. 1]
)
# => .../c:500:600:nowe:10:5/t:10:aabbcc:1:1/...
If a complex option has a single required argument, and you don't want to use the optional ones, you can just use the required argument value:
Imgproxy.url_for(
"http://images.example.com/images/image.jpg",
gravity: :nowe,
trim: 10
)
# => .../g:nowe/t:10/...
Some of the processing options like watermark_url
or style
require their arguments to be base64-encoded. Good news is that imgproxy.rb will encode them for you:
Imgproxy.url_for(
"http://images.example.com/images/image.jpg",
watermark_url: "http://example.com/watermark.jpg",
style: "color: rgba(255, 255, 255, .5)"
)
# => .../wmu:aHR0cDovL2V4YW1wbGUuY29tL3dhdGVybWFyay5qcGc/st:Y29sb3I6IHJnYmEoMjU1LCAyNTUsIDI1NSwgLjUp/...
By default, Imgproxy.url_for
accepts only String
and URI
as the source URL, but you can extend that behavior by using URL adapters.
Tip
imgproxy.rb provides built-in adapters for Active Storage and Shrine that are automatically added when Active Storage or Shrine support is enabled.
URL adapter is a simple class that implements applicable?
and url
methods. See the example below:
class MyItemAdapter
# `applicable?` checks if the adapter can extract
# source URL from the provided object
def applicable?(item)
item.is_a? MyItem
end
# `url` extracts source URL from the provided object
def url(item)
item.image_url
end
end
# ...
Imgproxy.configure do |config|
config.url_adapters.add MyItemAdapter.new
end
Note
Imgproxy
will use the first applicable URL adapter. If you need to add your adapter to the beginning of the list, use the prepend
method instead of add
.
If you use more than one instance of imgproxy and they have different endpoints and key/salt configurations you can specify them in services
option.
Imgproxy.configure do |config|
config.endpoint = "https://main.imgproxy.com/"
config.service(:pro) do |pro|
pro.endpoint = "https://pro.imgproxy.com/"
pro.key = ENV["IMGPROXY_PRO_KEY"]
pro.salt = ENV["IMGPROXY_PRO_SALT"]
pro.source_url_encryption_key = ENV["IMGPROXY_PRO_ENCRYPTION_KEY"]
pro.always_encrypt_source_urls = true
end
end
Or via YAML config:
endpoint: "https://main.imgproxy.com/"
services:
pro:
endpoint: "https://pro.imgproxy.com/"
key: <%= ENV["IMGPROXY_PRO_KEY"] %>
salt: <%= ENV["IMGPROXY_PRO_SALT"] %>
source_url_encryption_key: ENV["IMGPROXY_PRO_ENCRYPTION_KEY"]
always_encrypt_source_urls: true
If you don't specify key
, salt
, endpoint
, signature_size
, source_url_encryption_key
, or always_encrypt_source_urls
, they are inherited from the global configuration.
Pass the service
option to url_for
and info_url_for
:
Imgproxy.url_for(image, service: :pro)
Imgproxy.info_url_for(image, service: :pro)
Bug reports and pull requests are welcome on GitHub at https://github.com/imgproxy/imgproxy.rb.
If you are having any problems with image processing of imgproxy itself, be sure to visit https://github.com/imgproxy/imgproxy first and check out the docs at https://docs.imgproxy.net/.
The gem is available as open source under the terms of the MIT License.
To report a security vulnerability, please contact us at [email protected]. We will coordinate the fix and disclosure.