picguard 1.0.1 → 1.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 97758b22222ef4f51b940030b23d6bdc5f434aac
-  data.tar.gz: 452acbd8bdc5bfbca225e79659eab662e1a9da48
+  metadata.gz: a98abe2c1e9ffc8a0a0619f1a29197e0aa43175f
+  data.tar.gz: 68409340239c5e44662d5694c0036ac726e1302e
 SHA512:
-  metadata.gz: 6d9ffb3845ac4ae51f51bd688ca61387bde74a994d994cefcd9a521a5ca664fdbd1a79d881c87472775a041b72efec88a56532d725868dee0692bb7547ae473f
-  data.tar.gz: d398879a6b644e0f80279d668b8b244545ae1f2f0da201cfe5e05a50c4a533c73b887ca6728fd8592ba86ca6b3c7bb580d5af8e332ce89b3ec4b1da4031a9a6f
+  metadata.gz: 36b340420006eb8367bc1168d7e4977792538789243481993ecef496365ea53006ea73cc4914baea88a8ad30c3ae5011836c91467501e23ccc32711d39f9e29e
+  data.tar.gz: 0a52d1234658f3224ae7255715e626aace926a8d4f93d9d764b10d2d97010acb9857c7c4fdbcb94abb60a3d9655d9d7b7ef5ed151cb220dbcd3ff91213cd0e1f

data/README.md

@@ -6,9 +6,27 @@ Picguard guards your application by filtering out the pictures showing inappropr
 It uses Google Vision API to process the pictures that are being uploaded via your application.
 
 # Why use Picguard?
-Imagine a business application that allows user to upload a photo when creating his profile. The user is free to upload any kind of picture (including pictures showing violent scenes or adult content). The most popular solution for that is implementing a feature that allows other users to report the inappropriate content - which means you rely on your users when it comes to the safety of your application - not so great. Another scenario would be: what if for a profile picture an application should only allow the picture showing a human face? The soultion would be often the same as in the first example.
+Imagine a business application that allows user to upload a photo when creating his profile. The user is free to upload any kind of picture (including pictures showing violent scenes or adult content). The most popular solution for that is implementing a feature that allows other users to report the inappropriate content - which means you rely on your users when it comes to the safety of your application - not so great. Another scenario would be: what if for a profile picture an application should only allow the picture showing a human face? The solution would be often the same as in the first example.
 Picguard lets you configure your preferences (globally or separately for each model) for image filtering and gives you a clean solution to validate the picture before it's saved.
 
+## Requirements
+
+[ImageMagick](http://www.imagemagick.org) must be installed and Picguard must have access to it.
+
+If you're on Mac OS X, you'll want to run the following with Homebrew:
+
+```bash
+brew install imagemagick
+```
+
+If you're on Ubuntu (or any Debian base Linux distribution), you'll want to run the following with apt-get:
+
+```bash
+sudo apt-get install imagemagick
+```
+
+After the installation you should be able to run `which convert` in your terminal and you should see for example `/usr/local/bin/convert` as a response.
+
 ## Installation
 
 Add Picguard to your Gemfile:
@@ -34,9 +52,9 @@ end
 
 ##### Let's go through all of the attributes you need to set up in the config:
 
-`google_api_key` is the secret key that you use to authenticate to the Google Vision API, you can generate one that is connected to your Google Project. More about generating the key can be found [here](https://cloud.google.com/vision/docs/getting-started#setting_up_an_api_key). If you are completely new to the Google Cloud Platform you should probably start [here](https://cloud.google.com/vision/docs/getting-started).
-`threshold_adult` and `threshold_violence` are the thresholds for the adult and violent content. This is the highest value that you consider acceptable, everything above this level will be categorised as unsafe. For all of the likelihood levels please check [this piece of documentation](https://cloud.google.com/vision/reference/rest/v1/images/annotate#Likelihood).
-`threshold_face` is the threshold for face recognition. Google responds with a float value from 0 to 1 that reflects how sure the Google API is when it comes to face recognition. Only the picture with values above your threshold will be categorised as the ones containing human face.
+`google_api_key` is the secret key that you use to authenticate to the Google Vision API, you can generate one that is connected to your Google Project. More about generating the key can be found [here](https://cloud.google.com/vision/docs/getting-started#setting_up_an_api_key). If you are completely new to the Google Cloud Platform you should probably start [here](https://cloud.google.com/vision/docs/getting-started).   
+`threshold_adult` and `threshold_violence` are the thresholds for the adult and violent content. This is the highest value that you consider acceptable, everything above this level will be categorised as unsafe. For all of the likelihood levels please check [this piece of documentation](https://cloud.google.com/vision/reference/rest/v1/images/annotate#Likelihood).   
+`threshold_face` is the threshold for face recognition. Google responds with a float value from 0 to 1 that reflects how sure the Google API is when it comes to face recognition. Only the pictures with values above your threshold will be categorised as the ones showing human face.   
 
 ## Validations
 To validate the content of the picture simply add the following validation to your model:
@@ -50,14 +68,16 @@ To validate the content of the picture simply add the following validation to yo
 ```
 ###### where
 
-`image` is the name of the model's attribute that should be validated
-`guard` is the name of the picguard validator
-`safe_search` and `face_detection` are the flags reflecting what should be validated for given model
-`method_name` is the name of the `image`'s attribute method that returns image file path
+`image` is the name of the model's attribute that should be validated   
+`guard` is the name of the picguard validator  
+`safe_search` and `face_detection` are the flags reflecting what should be validated for given model   
+`method_name` value is the name of the `image`'s attribute method that returns image file path     
+
+*NOTE: Sometimes to return file path it's necessary to chain methods such as `avatar.tempfile.path`. In such case you need to pass an array of symbols to the `method_name` attribute (e.g. `[:tempfile, :path])`. For popular file-uploading gems, Picguard gives you a `tool` attribute that is used interchangeably with the `method_name`. More examples below.*   
 
 ## Sample validations for popular file-uploading gems
 
-### CarrierWave
+### [CarrierWave](https://github.com/carrierwaveuploader/carrierwave)
 
 ```ruby
 class User < ActiveRecord::Base
@@ -66,7 +86,40 @@ class User < ActiveRecord::Base
   validates :avatar, guard: {
     safe_search: true,
     face_detection: true,
-    method_name: :path
+    tool: :carrierwave
+  }
+end
+```
+
+### [Paperclip](https://github.com/thoughtbot/paperclip)
+
+```ruby
+class User < ActiveRecord::Base
+  has_attached_file :avatar,
+    styles: { medium: '300x300>', thumb: '100x100>' },
+    default_url: "/images/:style/missing.jpg"
+
+  validates_attachment_content_type :avatar, content_type: /\Aimage\/.*\Z/
+
+  validates :avatar, guard: {
+    safe_search: false,
+    face_detection: true,
+    threshold_face: 0.5,
+    tool: :paperclip
+  }
+end
+```
+
+### [Dragonfly](https://github.com/markevans/dragonfly)
+
+```ruby
+class User < ActiveRecord::Base
+  dragonfly_accessor :avatar
+
+  validates :avatar, presence: true, guard: {
+    safe_search: true,
+    threshold_adult: 'VERY_LIKELY',
+    tool: :dragonfly
   }
 end
 ```

data/lib/guard_validator.rb

@@ -2,6 +2,12 @@ require 'picguard'
 require 'active_model'
 
 class GuardValidator < ActiveModel::EachValidator
+  METHOD_NAMES = {
+    carrierwave:  :path,
+    paperclip:    :staged_path,
+    dragonfly:    [:tempfile, :path]
+  }.freeze
+  private_constant :METHOD_NAMES
 
   def validate_each(record, attribute, value)
     image_path = fetch_image_path(record, attribute)
@@ -12,10 +18,14 @@ class GuardValidator < ActiveModel::EachValidator
   private
 
   def fetch_image_path(record, attribute)
-    arr = [attribute].push(*Array(options[:method_name]))
+    arr = [attribute].push(*Array(fetch_method_names))
     arr.inject(record, :public_send)
   end
 
+  def fetch_method_names
+    options[:tool].present? ? METHOD_NAMES.fetch(options[:tool]) : options[:method_name]
+  end
+
   def valid?(image_path)
     return false unless path_exists?(image_path)
     result = Picguard.analyze(

data/lib/picguard/version.rb

@@ -1,3 +1,3 @@
 module Picguard
-  VERSION = "1.0.1"
+  VERSION = "1.1.0"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: picguard
 version: !ruby/object:Gem::Version
-  version: 1.0.1
+  version: 1.1.0
 platform: ruby
 authors:
 - Szymon Baranowski
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-03-02 00:00:00.000000000 Z
+date: 2016-03-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: google-api-client