discourse/app/models/screened_email.rb

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

89 lines
2.5 KiB
Ruby
Raw Normal View History

# frozen_string_literal: true
# A ScreenedEmail record represents an email address that is being watched,
# typically when creating a new User account. If the email of the signup form
# (or some other form) matches a ScreenedEmail record, an action can be
# performed based on the action_type.
class ScreenedEmail < ActiveRecord::Base
include ScreeningModel
default_action :block
validates :email, presence: true, uniqueness: true
2014-07-14 10:16:24 -04:00
before_save :downcase_email
def downcase_email
self.email = email.downcase
end
def self.canonical(email)
name, domain = email.split("@", 2)
name = name.gsub(/\+.*/, "")
name = name.gsub(".", "") if %w[gmail.com googlemail.com].include?(domain.downcase)
"#{name}@#{domain}".downcase
end
def self.block(email, opts = {})
email = canonical(email)
find_by_email(email) || create!(opts.slice(:action_type, :ip_address).merge(email: email))
end
def self.should_block?(email)
email = canonical(email)
screened_emails = ScreenedEmail.order(created_at: :desc).limit(100)
distances = {}
2014-07-14 10:16:24 -04:00
screened_emails.each do |se|
distances[se.email] = levenshtein(se.email.downcase, email.downcase)
end
max_distance = SiteSetting.levenshtein_distance_spammer_emails
screened_email =
screened_emails
.select { |se| distances[se.email] <= max_distance }
.sort { |se| distances[se.email] }
.first
screened_email.record_match! if screened_email
screened_email.try(:action_type) == actions[:block]
end
def self.levenshtein(first, second)
matrix = [(0..first.length).to_a]
(1..second.length).each { |j| matrix << [j] + [0] * (first.length) }
(1..second.length).each do |i|
(1..first.length).each do |j|
if first[j - 1] == second[i - 1]
matrix[i][j] = matrix[i - 1][j - 1]
else
matrix[i][j] = [matrix[i - 1][j], matrix[i][j - 1], matrix[i - 1][j - 1]].min + 1
end
end
end
matrix.last.last
end
end
2013-08-27 20:42:58 -04:00
# == Schema Information
#
# Table name: screened_emails
#
# id :integer not null, primary key
2019-01-11 14:29:56 -05:00
# email :string not null
2013-08-27 20:42:58 -04:00
# action_type :integer not null
# match_count :integer default(0), not null
# last_match_at :datetime
# created_at :datetime not null
# updated_at :datetime not null
2013-12-05 01:40:35 -05:00
# ip_address :inet
2013-08-27 20:42:58 -04:00
#
# Indexes
#
# index_screened_emails_on_email (email) UNIQUE
# index_screened_emails_on_last_match_at (last_match_at)
2013-08-27 20:42:58 -04:00
#