decidim-cdtb 0.1.3 → 0.1.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: d908d1cc4021da9869dd10cacb1fe06b360301d4e13f1a70a55fdc90a2712b6a
-  data.tar.gz: e7a0c808803bbbd6dc75e5bdfaee7b3acfb1311c085f17eb41a9de32fc7795fa
+  metadata.gz: 515d740942f3089a3cb906fab890191dc4b19f97c98ccff17fac5aac2aa83d9e
+  data.tar.gz: 557fa8a269b25d0ac8b15c522d16072600c0b5a0d39fee2250cf9c80299486ab
 SHA512:
-  metadata.gz: 20ddde6eddf59ab12b7ca8d6f489692a04183331f0c1ef496b199b4350cdd103d2b23ca94329199d36f2a9e7222ea0c3269da25648ec0acb47bdde6b38666b92
-  data.tar.gz: 2dd41a8a1f79ffd4506c6b18d58393d3a88d8a021e0acf05cfbd4e4ac1dcc22250294c8de29d7168c2fce29bf2d2df71c0b9905414a3df1e28991aa665be8d36
+  metadata.gz: 519c7234a2cff3adba2732e07e4fb9c2aa111fb0803ac8cac8180a9683c501676bb2f0aa96059f1a5cedd063db98471f70e1321d33be0e8db4a9947341df3b6a
+  data.tar.gz: be0c056d4b91a543a528628b4f820fd06a7913899c12f3fa4e5db646ddbd1599970fcba41d67d583472b53abe2eef447e8b0d4aede8d83222617e3f3d80f9df8

data/.rubocop.yml

@@ -11,7 +11,7 @@ AllCops:
 
 Metrics/BlockLength:
   Enabled: true
-  Max: 30
+  Max: 50
   Exclude:
     - lib/tasks/*.rake
 

data/CHANGELOG.md

@@ -1,5 +1,9 @@
 ## [Unreleased]
 
+## [0.1.4] - 2024-01-30 (Peus grans com gegants)
+
+- Add users spam detector task
+
 ## [0.1.3] - 2023-06-23 (Tan iguals com especials)
 
 - Validate migrations task

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    decidim-cdtb (0.1.3)
+    decidim-cdtb (0.1.4)
       decidim (>= 0.26.2)
       rails (>= 6)
       ruby-progressbar
@@ -775,4 +775,4 @@ DEPENDENCIES
   sqlite3
 
 BUNDLED WITH
-   2.3.6
+   2.4.22

data/README.md

@@ -76,6 +76,28 @@ To migrate from S3 to local storage, the identified steps will be:
     `bin/rake cache:clear`
 5. Restart the Rails server
 
+### Detect spam
+
+To detect spam in Decidim.
+
+#### Detect spam users
+Detects users susceptible of being spammers. It can run on all organizations or be scoped to a single organization by passing the organization ID as the rake task parameter.
+
+This rake task export a .csv with a list of all the searched users. A column indicates if each user is suspicious of being a spammer or not.
+The columns in the CSV are: "ID, "Is suspicious?", "Name", "Email", "Nickname", "Personal URL", "About"
+
+Examples:
+`bin/rake cdtb:spam:users[org_id]` --> find users in organization with an id.
+`bin/rake cdtb:spam:users` --> find all users in all organizations.
+
+To set custom words in the rake, you can override it with an initalizer:
+
+```
+Decidim::Cdtb.configure do |config|
+  config.spam_words = ENV["CDTB_SPAM_WORDS"]&.split(",")
+end
+```
+
 ### Upgrades:
 
 #### Upgrade modules

data/lib/decidim/cdtb/engine.rb

@@ -10,6 +10,10 @@ module Decidim
         # Workaround for https://stackoverflow.com/questions/72970170/upgrading-to-rails-6-1-6-1-causes-psychdisallowedclass-tried-to-load-unspecif
         Rails.application.config.active_record.use_yaml_unsafe_load = true
       end
+
+      config.after_initialize do
+        Decidim::Cdtb.config.spam_regexp = Regexp.union(Decidim::Cdtb.config.spam_words)
+      end
     end
   end
 end

data/lib/decidim/cdtb/spam/user_spam_detector.rb

@@ -0,0 +1,86 @@
+# frozen_string_literal: true
+
+require "csv"
+
+module Decidim
+  module Cdtb
+    module Spam
+      # Detect spam behavior in users
+      #
+      class UserSpamDetector < ::Decidim::Cdtb::Task
+        # rubocop:disable Style/RedundantRegexpEscape
+        URL_REGEX = %r{(https?:\/\/(?:www\.|(?!www))[a-zA-Z0-9][a-zA-Z0-9-]+[a-zA-Z0-9]\.[^\s]{2,}|
+        www\.[a-zA-Z0-9][a-zA-Z0-9-]+[a-zA-Z0-9]\.[^\s]{2,}|https?:\/\/(?:www\.|
+        (?!www))[a-zA-Z0-9]+\.[^\s]{2,}|www\.[a-zA-Z0-9]+\.[^\s]{2,})}.freeze
+        # rubocop:enable Style/RedundantRegexpEscape
+
+        def initialize(organization = nil)
+          @organization = organization
+          progress_bar = { title: "Decidim::User" }
+          super("SPAM DETECTOR", progress_bar: progress_bar)
+        end
+
+        def prepare_execution(_ctx)
+          @users = if @organization.present?
+                     Decidim::User.where(organization: @organization)
+                   else
+                     Decidim::User.all
+                   end
+
+          @num_users = @users.count
+          log_task_info("Checking #{@num_users} users...")
+        end
+
+        def total_items
+          @num_users
+        end
+
+        def do_execution(context)
+          progress_bar = context[:progress_bar]
+
+          CSV.open("spam_users.csv", "w") do |csv|
+            csv_headers = ["ID", "Is suspicious?", "Name", "Email", "Nickname", "Personal URL", "About"]
+            csv << csv_headers
+
+            @users.find_each do |user|
+              suspicious = "NO"
+
+              if spam_user?(user)
+                suspicious = "YES"
+                @num_applied+= 1
+              end
+
+              csv << [user.id, suspicious, user.name, user.email, user.nickname, user.personal_url, user.about]
+
+              progress_bar.increment
+            end
+          end
+        end
+
+        def end_execution(_ctx)
+          if @num_applied.positive?
+            log_task_step("#{@num_applied} suspicious users")
+            log_task_step("Suspicious users list exported to spam_users.csv")
+          else
+            log_task_step("There are not suspicious users!!")
+          end
+        end
+
+        def spam_user?(user)
+          has_spam_word?(user) || has_spam_url?(user)
+        end
+
+        private
+
+        def has_spam_word?(user)
+          [user.name, user.about, user.nickname,
+           user.personal_url, user.about].compact.join("||").match?(Decidim::Cdtb.config.spam_regexp)
+        end
+
+        def has_spam_url?(user)
+          !!(user&.about =~ URL_REGEX || user.name =~ URL_REGEX)
+        end
+      end
+    end
+  end
+end

data/lib/decidim/cdtb/tasks.rb

@@ -4,6 +4,7 @@ require "decidim/cdtb/tasks_utils"
 require "decidim/cdtb/task"
 require "decidim/cdtb/fixes/nickname_fixer"
 require "decidim/cdtb/multitenants/org_by_host_like"
+require "decidim/cdtb/spam/user_spam_detector"
 require "decidim/cdtb/storage/local_sharding"
 require "decidim/cdtb/storage/set_local_on_blobs"
 require "decidim/cdtb/upgrades/validate_migrations_task"

data/lib/decidim/cdtb/version.rb

@@ -2,7 +2,7 @@
 
 module Decidim
   module Cdtb
-    VERSION = "0.1.3"
+    VERSION = "0.1.4"
     DECIDIM_MIN_VERSION = ">= 0.26.2"
   end
 end

data/lib/decidim/cdtb.rb

@@ -5,7 +5,14 @@ require_relative "cdtb/engine"
 require_relative "cdtb/tasks"
 
 module Decidim
+  # Cdtb configuration
   module Cdtb
+    include ActiveSupport::Configurable
+
     class Error < StandardError; end
+
+    config_accessor :spam_words do
+      %w[viagra sex game free crypto crack xxx luck girls vip download]
+    end
   end
 end

data/lib/tasks/spam.rake

@@ -0,0 +1,13 @@
+# frozen_string_literal: true
+
+namespace :cdtb do
+  namespace :spam do
+    desc "Show a list with users suspected of spam"
+    task :users, %i[org_id] => :environment do |_task, args|
+      organization = args.org_id.present? ? Decidim::Organization.find(args.org_id) : nil
+
+      detector = ::Decidim::Cdtb::Spam::UserSpamDetector.new(organization)
+      detector.execute!
+    end
+  end
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: decidim-cdtb
 version: !ruby/object:Gem::Version
-  version: 0.1.3
+  version: 0.1.4
 platform: ruby
 authors:
 - Oliver Valls
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2023-07-10 00:00:00.000000000 Z
+date: 2024-02-07 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: decidim
@@ -105,6 +105,7 @@ files:
 - lib/decidim/cdtb/engine.rb
 - lib/decidim/cdtb/fixes/nickname_fixer.rb
 - lib/decidim/cdtb/multitenants/org_by_host_like.rb
+- lib/decidim/cdtb/spam/user_spam_detector.rb
 - lib/decidim/cdtb/storage/local_sharding.rb
 - lib/decidim/cdtb/storage/set_local_on_blobs.rb
 - lib/decidim/cdtb/task.rb
@@ -120,6 +121,7 @@ files:
 - lib/tasks/anonymize.rake
 - lib/tasks/cdtb.rake
 - lib/tasks/multitenants.rake
+- lib/tasks/spam.rake
 - lib/tasks/storage.rake
 - lib/tasks/upgrade.rake
 - sig/decidim/cdtb.rbs