privacy 0.1.3 → 0.1.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 8be253d2d26c29d1d4f5145621bafad66858146e1e2e1e311f2a97b3a5f56479
-  data.tar.gz: b20aaaa204b7b1bb814c6a42bb32105b2236641d381048941dae452f9db69362
+  metadata.gz: e388db32406558d4970fab62d76fb4bf8e3a6cccbddf3cfd6f5a47b0a3af160d
+  data.tar.gz: 8358c360c06bb0386c768e283ffb123f14030b81bc96f79f4c3cb4dfec51f2f9
 SHA512:
-  metadata.gz: 1ec29124dac0e2b84f2bfbb0a2625f9441e43f51f4416c62b606cbf5b91a5d6cda846caf629b3b6cfbbb1989e71ec2321479d37b557423cba9995a95e5808184
-  data.tar.gz: 49802f38af22e86afb99ab8c628f2f367415116bcc4301e9757fc14d7f54ed339338e3c4356dffbab4b3c672b625cf59d1e52c0ce65c9c1eb8ccb44189f0c560
+  metadata.gz: 8e8a6eadc68c458d508189c35ec34bdfa245d18159755c075cace122029a5c899f589134fc4156618120c8ace5d49a7ca26f770b6125695b5a6a91956cc0454a
+  data.tar.gz: 8a05cd62e6c983aafe69c9dec2a546629f9f2126c2ab25f78fbbe29dbfef6bd2825eae08abcac149fa28be7b64858cfc49eed3302a30c27c176ef2fffa20ca97

data/README.md

@@ -32,6 +32,8 @@ To install this gem onto your local machine, run `bundle exec rake install`. To
 
 To release a new version, change version number and `rake release`.
 
+To work on the code, use `bundle exec privacy process`.
+
 ## Contributing
 
 Bug reports and pull requests are welcome on GitHub at https://github.com/[USERNAME]/privacy. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [code of conduct](https://github.com/[USERNAME]/privacy/blob/master/CODE_OF_CONDUCT.md).

data/lib/privacy/person.rb

@@ -0,0 +1,133 @@
+class Privacy::Person
+  attr_reader :identifier, :row, :index, :columns, :data, :combinations, :initials
+
+  COLUMNS_TO_ANONYMIZE = [
+    'Nom _identifier_',
+    'Prénom _identifier_',
+    'Date de naissance _identifier_',
+    'Lieu naissance Pays _identifier_',
+    'Lieu naissance Ville _identifier_',
+    'Lieu naissance Département _identifier_',
+    'Sexe _identifier_',
+    'Nationalité _identifier_',
+    'Adresse _identifier_ [Date]',
+    'Adresse _identifier_ [Département]',
+    'Adresse _identifier_ [Ville]',
+    'Adresse _identifier_ [Pays]',
+    'Profession _identifier_',
+    'Date profession _identifier_'
+  ]
+
+  INITIALS_SEPARATORS = [
+    '',
+    '.'
+  ]
+
+  def initialize(identifier)
+    @identifier = identifier
+    puts "Identify #{identifier}"
+  end
+
+  def identify_columns(row)
+    @columns = {}
+    row.each do |key, value|
+      COLUMNS_TO_ANONYMIZE.each do |column|
+        name = column_with_identifier column
+        if value == name
+          letter = key.gsub '1', ''
+          @columns[letter] = name
+          puts "  #{name} on column #{letter}"
+        end
+      end
+    end
+  end
+
+  def anonymize(row, index)
+    @row = row
+    @index = index + 1 # Excel is 1-indexed
+    anonymize_personal_data
+    build_combinations
+    remove_combinations
+  end
+
+  protected
+
+  def anonymize_personal_data
+    @data = {}
+    puts
+    puts "Anonymize row #{index}"
+    puts "  #{identifier}"
+    @columns.each do |letter, column|
+      cell = "#{letter}#{index}"
+      value = row[cell]
+      next if value.nil?
+      @data[column] = value
+      puts "      #{value} (#{column})"
+      row[cell] = "[#{column}]"
+    end
+  end
+
+  def build_combinations
+    puts 'build_combinations'
+    @combinations = []
+    first_name_column = column_with_identifier 'Prénom _identifier_'
+    last_name_column = column_with_identifier 'Nom _identifier_'
+    first_name = data[first_name_column]
+    last_name = data[last_name_column]
+    if !first_name.nil? && !last_name.nil?
+      initials = []
+      first_name.split(' ').each { |l| initials << l[0].upcase }
+      last_name.split(' ').each { |l| initials << l[0].upcase }
+      @combinations << "#{first_name} #{last_name}"
+      @combinations << " #{first_name} #{last_name}"
+      @combinations << "#{first_name} #{last_name} "
+      @combinations << "#{last_name} #{first_name}"
+      @combinations << " #{last_name} #{first_name}"
+      @combinations << "#{last_name} #{first_name} "
+      @combinations += initials_combinations(initials)
+    end
+    puts "Combinations: #{@combinations.join(', ')}"
+  end
+
+  def column_with_identifier(column)
+    column.gsub '_identifier_', identifier
+  end
+
+  def initials_combinations(initials)
+    puts "Initials for #{initials}"
+    combinations = []
+    initials.each do |letter|
+      initials.shift
+      recursive_combinations = initials_combinations initials
+      INITIALS_SEPARATORS.each do |separator|
+        initial_with_separator = "#{letter}#{separator}"
+        if recursive_combinations.none?
+          combinations << "#{initial_with_separator}"
+          combinations << "#{initial_with_separator} "
+          combinations << " #{initial_with_separator}"
+        else
+          recursive_combinations.each do |c|
+            combinations << "#{initial_with_separator}#{c}"
+            combinations << " #{initial_with_separator}#{c}"
+            combinations << "#{initial_with_separator} #{c}"
+            combinations << "#{initial_with_separator} #{c} "
+            combinations << "#{initial_with_separator}#{c} "
+          end
+        end
+      end
+    end
+    combinations.uniq
+  end
+
+  def remove_combinations
+    row.each do |key, value|
+      anonymized_value = value
+      if value.is_a? String
+        combinations.each do |token|
+          anonymized_value = anonymized_value.gsub(token, "[#{identifier}]")
+        end
+      end
+      row[key] = anonymized_value
+    end
+  end
+end

data/lib/privacy/processor.rb

@@ -1,14 +1,21 @@
 require 'creek'
 require 'write_xlsx'
+require 'privacy/person'
 
 class Privacy::Processor
   attr_reader :file
 
   DIRECTORY = 'processed'
+  PRIVATE_DATA = {
+    'Nom ppmec': '[Nom]',
+    'Prénom ppmec': '[Prénom]',
+    'Nom ppvic': '[Nom]',
+    'Prénom ppvic': '[Prénom]'
+  }
 
   def initialize(file)
     @file = file
-    process
+    process unless @file.start_with? '~'
   end
 
   protected
@@ -25,7 +32,17 @@ class Privacy::Processor
   end
 
   def filter_data
-    sheet.rows.each do |row|
+    @ppmec = Privacy::Person.new 'ppmec'
+    @ppmec.identify_columns sheet.rows.first
+    @ppvic = Privacy::Person.new 'ppvic'
+    @ppvic.identify_columns sheet.rows.first
+    sheet.rows.first.each do |key, value|
+      worksheet.write key, value
+    end
+    sheet.rows.each_with_index do |row, index|
+      next if index.zero?
+      @ppmec.anonymize row, index
+      @ppvic.anonymize row, index
       row.each do |key, value|
         worksheet.write key, value
       end
@@ -59,7 +76,7 @@ class Privacy::Processor
   end
 
   def data
-    @data ||= Creek::Book.new path
+    @data ||= Creek::Book.new path, with_headers: true
   end
 
   def sheet

data/lib/privacy/version.rb

@@ -1,3 +1,3 @@
 module Privacy
-  VERSION = "0.1.3"
+  VERSION = "0.1.4"
 end

data/privacy.gemspec

@@ -1,31 +1,31 @@
 require_relative 'lib/privacy/version'
 
-Gem::Specification.new do |spec|
-  spec.name          = "privacy"
-  spec.version       = Privacy::VERSION
-  spec.authors       = ["Arnaud Levy"]
-  spec.email         = ["contact@arnaudlevy.com"]
+Gem::Specification.new do |s|
+  s.name          = "privacy"
+  s.version       = Privacy::VERSION
+  s.authors       = ["Arnaud Levy"]
+  s.email         = ["contact@arnaudlevy.com"]
 
-  spec.summary       = "Privacy removes personal data from a xls file"
-  spec.description   = "Remove first name and last name from xls files, replace them with an anonymous placeholder"
-  spec.homepage      = "https://github.com/arnaudlevy/privacy"
-  spec.license       = "MIT"
-  spec.required_ruby_version = Gem::Requirement.new(">= 2.3.0")
+  s.summary       = "Privacy removes personal data from a xls file"
+  s.description   = "Remove first name and last name from xls files, replace them with an anonymous placeholder"
+  s.homepage      = "https://github.com/arnaudlevy/privacy"
+  s.license       = "MIT"
+  s.required_ruby_version = Gem::Requirement.new(">= 2.3.0")
 
-  # spec.metadata["allowed_push_host"] = "TODO: Set to 'http://mygemserver.com'"
+  # s.metadata["allowed_push_host"] = "TODO: Set to 'http://mygemserver.com'"
 
-  spec.metadata["homepage_uri"] = spec.homepage
-  spec.metadata["source_code_uri"] = "https://github.com/arnaudlevy/privacy"
+  s.metadata["homepage_uri"] = s.homepage
+  s.metadata["source_code_uri"] = "https://github.com/arnaudlevy/privacy"
 
   # Specify which files should be added to the gem when it is released.
   # The `git ls-files -z` loads the files in the RubyGem that have been added into git.
-  spec.files         = Dir.chdir(File.expand_path('..', __FILE__)) do
-    `git ls-files -z`.split("\x0").reject { |f| f.match(%r{^(test|spec|features)/}) }
+  s.files         = Dir.chdir(File.expand_path('..', __FILE__)) do
+    `git ls-files -z`.split("\x0").reject { |f| f.match(%r{^(test|s|features)/}) }
   end
-  spec.bindir        = "exe"
-  spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
-  spec.require_paths = ["lib"]
-  spec.add_dependency "thor"
-  spec.add_dependency "creek"
-  spec.add_dependency "write_xlsx"
+  s.bindir        = "exe"
+  s.executables   = s.files.grep(%r{^exe/}) { |f| File.basename(f) }
+  s.require_paths = ["lib"]
+  s.add_dependency "thor"
+  s.add_dependency "creek"
+  s.add_dependency "write_xlsx"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: privacy
 version: !ruby/object:Gem::Version
-  version: 0.1.3
+  version: 0.1.4
 platform: ruby
 authors:
 - Arnaud Levy
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2020-05-28 00:00:00.000000000 Z
+date: 2020-06-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: thor
@@ -61,6 +61,7 @@ executables:
 extensions: []
 extra_rdoc_files: []
 files:
+- ".DS_Store"
 - ".gitignore"
 - CODE_OF_CONDUCT.md
 - Gemfile
@@ -72,6 +73,7 @@ files:
 - bin/setup
 - exe/privacy
 - lib/privacy.rb
+- lib/privacy/person.rb
 - lib/privacy/processor.rb
 - lib/privacy/version.rb
 - privacy.gemspec