Merge pull request #5256 from consuldemocracy/remove-pdf-metadata

ENS: Strip metadata from attachments

Gemfile

@@ -20,6 +20,7 @@ gem "dalli", "~> 3.2.6"
 gem "delayed_job_active_record", "~> 4.1.7"
 gem "devise", "~> 4.9.2"
 gem "devise-security", "~> 0.18.0"
+gem "exiftool_vendored", "~> 12.60.0"
 gem "file_validators", "~> 3.0.0"
 gem "font-awesome-sass", "~> 5.15.1" # Remember to update vendor/assets/images/fontawesome when updating this gem
 gem "foundation-rails", "~> 6.6.2.0"
@@ -85,6 +86,7 @@ group :test do
   gem "capybara", "~> 3.39.2"
   gem "capybara-webmock", "~> 0.7.0"
   gem "email_spec", "~> 2.2.2"
+  gem "pdf-reader"
   gem "rspec-rails", "~> 5.1.2"
   gem "selenium-webdriver", "~> 4.13.1"
   gem "simplecov", "~> 0.22.0", require: false

Gemfile.lock

@@ -6,6 +6,7 @@ GEM
 GEM
   remote: https://rubygems.org/
   specs:
+    Ascii85 (1.1.0)
     actioncable (6.1.7.6)
       actionpack (= 6.1.7.6)
       activesupport (= 6.1.7.6)
@@ -70,6 +71,7 @@ GEM
     acts_as_votable (0.14.0)
     addressable (2.8.5)
       public_suffix (>= 2.0.2, < 6.0)
+    afm (0.2.2)
     ahoy_matey (4.2.1)
       activesupport (>= 5.2)
       device_detector
@@ -209,6 +211,10 @@ GEM
       multi_json (>= 1.3)
       rake
     execjs (2.8.1)
+    exiftool (1.2.4)
+      json
+    exiftool_vendored (12.60.0)
+      exiftool (>= 0.7.0)
     factory_bot (6.2.0)
       activesupport (>= 5.0.0)
     factory_bot_rails (6.2.0)
@@ -255,6 +261,7 @@ GEM
     gyoku (1.4.0)
       builder (>= 2.1.2)
       rexml (~> 3.0)
+    hashery (2.1.2)
     hashie (5.0.0)
     highline (2.0.3)
     htmlentities (4.3.4)
@@ -408,6 +415,12 @@ GEM
     parser (3.2.2.3)
       ast (~> 2.4.1)
       racc
+    pdf-reader (2.11.0)
+      Ascii85 (~> 1.0)
+      afm (~> 0.2.1)
+      hashery (~> 2.0)
+      ruby-rc4
+      ttfunk
     pg (1.4.3)
     pg_search (2.3.6)
       activerecord (>= 5.2)
@@ -545,6 +558,7 @@ GEM
       rubocop-capybara (~> 2.17)
       rubocop-factory_bot (~> 2.22)
     ruby-progressbar (1.13.0)
+    ruby-rc4 (0.1.5)
     ruby-vips (2.1.4)
       ffi (~> 1.12)
     ruby2_keywords (0.0.5)
@@ -626,6 +640,7 @@ GEM
     tilt (2.0.10)
     timeout (0.4.1)
     tomlrb (2.0.3)
+    ttfunk (1.7.0)
     turbolinks (5.2.1)
       turbolinks-source (~> 5.2)
     turbolinks-source (5.2.0)
@@ -705,6 +720,7 @@ DEPENDENCIES
   devise-security (~> 0.18.0)
   email_spec (~> 2.2.2)
   erb_lint (~> 0.5.0)
+  exiftool_vendored (~> 12.60.0)
   factory_bot_rails (~> 6.2.0)
   faker (~> 3.2.1)
   file_validators (~> 3.0.0)
@@ -734,6 +750,7 @@ DEPENDENCIES
   omniauth-rails_csrf_protection (~> 1.0.1)
   omniauth-twitter (~> 1.4.0)
   paranoia (~> 2.6.2)
+  pdf-reader
   pg (~> 1.4.3)
   pg_search (~> 2.3.6)
   pronto (~> 0.11.1)

app/models/document.rb

@@ -9,6 +9,8 @@ class Document < ApplicationRecord
   validates :documentable_id, presence: true,         if: -> { persisted? }
   validates :documentable_type, presence: true,       if: -> { persisted? }
 
+  before_save :remove_metadata
+
   scope :admin, -> { where(admin: true) }
 
   def self.humanized_accepted_content_types
@@ -36,4 +38,13 @@ class Document < ApplicationRecord
     def documentable_class
       association_class
     end
+
+    def remove_metadata
+      return unless attachment.attached?
+
+      attachment_path = ActiveStorage::Blob.service.path_for(attachment.key)
+      Exiftool.new(attachment_path, "-all:all=")
+    rescue Exiftool::ExiftoolNotInstalled, Exiftool::NoSuchFile
+      nil
+    end
 end

spec/system/documents_spec.rb

@@ -0,0 +1,27 @@
+require "rails_helper"
+
+describe "Documents" do
+  describe "Metadata" do
+    scenario "download document without metadata" do
+      login_as(create(:user))
+      visit new_proposal_path
+
+      fill_in "Proposal title", with: "debate"
+      fill_in "Proposal summary", with: "In summary, what we want is..."
+      fill_in "Full name of the person submitting the proposal", with: "Isabel Garcia"
+      documentable_attach_new_file(file_fixture("logo_with_metadata.pdf"))
+      check "I agree to the Privacy Policy and the Terms and conditions of use"
+
+      click_button "Create proposal"
+
+      io = URI.parse("#{app_host}#{polymorphic_path(Document.last.attachment)}").open
+      reader = PDF::Reader.new(io)
+
+      expect(reader.info[:Keywords]).not_to eq "Test Metadata"
+      expect(reader.info[:Author]).not_to eq "Test Developer"
+      expect(reader.info[:Title]).not_to eq "logo_with_metadata.pdf"
+      expect(reader.info[:Producer]).not_to eq "Test Producer"
+      expect(reader.info).to eq({})
+    end
+  end
+end