<img src=“https://travis-ci.org/mavenlink/my_obfuscate.png”>
You want to develop against real production data, but you don’t want to violate your users’ privacy. Enter MyObfuscate: standalone Ruby code for the selective rewriting of SQL dumps in order to protect user privacy. It supports MySQL, Postgres, and SQL Server.
(sudo) gem install my_obfuscate
Make an obfuscator.rb script:
#!/usr/bin/env ruby require "rubygems" require "my_obfuscate" obfuscator = MyObfuscate.new({ :people => { :email => { :type => :email, :skip_regexes => [/^[\w\.\_]+@my_company\.com$/i] }, :ethnicity => :keep, :crypted_password => { :type => :fixed, :string => "SOME_FIXED_PASSWORD_FOR_EASE_OF_DEBUGGING" }, :salt => { :type => :fixed, :string => "SOME_THING" }, :remember_token => :null, :remember_token_expires_at => :null, :age => { :type => :null, :unless => lambda { |person| person[:email] == "hello@example.com" } }, :photo_file_name => :null, :photo_content_type => :null, :photo_file_size => :null, :photo_updated_at => :null, :postal_code => { :type => :fixed, :string => "94109", :unless => lambda {|person| person[:postal_code] == "12345"} }, :name => :name, :full_address => :address, :bio => { :type => :lorem, :number => 4 }, :relationship_status => { :type => :fixed, :one_of => ["Single", "Divorced", "Married", "Engaged", "In a Relationship"] }, :has_children => { :type => :integer, :between => 0..1 }, }, :invites => :truncate, :invite_requests => :truncate, :tags => :keep, :relationships => { :account_id => :keep, :code => { :type => :string, :length => 8, :chars => MyObfuscate::USERNAME_CHARS } } }) obfuscator.fail_on_unspecified_columns = true # if you want it to require every column in the table to be in the above definition obfuscator.globally_kept_columns = %w[id created_at updated_at] # if you set fail_on_unspecified_columns, you may want this as well # If you'd like to also validate against your schema.rb file to make sure all fields and tables are present, see https://gist.github.com/cantino/5376e73b0ad806dc4da4 obfuscator.obfuscate(STDIN, STDOUT)
And to get an obfuscated dump:
mysqldump -c --add-drop-table --hex-blob -u user -ppassword database | ruby obfuscator.rb > obfuscated_dump.sql
Note that the -c option on mysqldump is required to use my_obfuscator. Additionally, the default behavior of mysqldump is to output special characters. This may cause trouble, so you can request hex-encoded blob content with –hex-blob. If you get MySQL errors due to very long lines, try some combination of –max_allowed_packet=128M, –single-transaction, –skip-extended-insert, and –quick.
By default the database type is assumed to be MySQL, but you can use the builtin SQL Server support by specifying:
obfuscator.database_type = :sql_server obfuscator.database_type = :postgres
If using Postgres, use pg_dump to get a dump:
pg_dump database | ruby obfuscator.rb > obfuscated_dump.sql
Available types include: email, string, lorem, name, first_name, last_name, address, street_address, secondary_address, city, state, zip_code, phone, company, ipv4, ipv6, url, integer, fixed, null, and keep.
If you don’t want to type all those table names and column names into your obfuscator.rb script, you can use my_obfuscate to do some of that work for you. It can consume your database dump file and create a “scaffold” for the script. To run my_obfuscate in this mode, start with an “empty” scaffolder.rb script as follows:
#!/usr/bin/env ruby require "rubygems" require "my_obfuscate" obfuscator = MyObfuscate.new({}) obfuscator.scaffold(STDIN, STDOUT)
Then feed in your database dump:
mysqldump -c --hex-blob -u user -ppassword database | ruby scaffolder.rb > obfuscator_scaffold.rb_snippet pg_dump database | ruby scaffolder.rb > obfuscator_scaffold.rb_snippet
The output will be a series of configuration statements of the form:
:table_name => {
:column1_name => :keep # scaffold
:column2_name => :keep # scaffold
... etc.
Scaffolding also works if you have a partial configuration. If your configuration is missing some tables or some columns, a call to ‘scaffold’ will pass through the configuration that exists and augment it with scaffolding for the missing tables or columns.
-
Support for Postgres. Thanks @samuelreh!
-
Support for SQL Server
-
:unless and :if now support :nil as a shorthand for a Proc that checks for nil
-
:name, :lorem, and :address are all now supported types. You can pass :number to :lorem to specify how many sentences to generate. The default is one.
-
{ :type => :whatever }is now optional when no additional options are needed. Just use:whatever. -
Warnings are thrown when an unknown column type or table is encountered. Use
:keepin both cases. -
{ :type => :fixed, :string => Proc { |row| ... } }is now available.
-
Fork the project.
-
Make your feature addition or bug fix.
-
Add tests for it. This is important so I don’t break it in a future version unintentionally.
-
Commit, do not mess with rakefile, version, or history. (If you want to have your own version, that is fine but bump version in a commit by itself I can ignore when I pull)
-
Send me a pull request. Bonus points for topic branches.
Thanks to Honk for the original gem, Iteration Labs for prior maintenance work, and Pivotal Labs for patches and updates!
This work is provided under the MIT License. See the included LICENSE file.
The included English word frequency list used for generating random text is provided under the Creative Commons – Attribution / ShareAlike 3.0 license by invokeit.wordpress.com/frequency-word-lists/