data_cleansing 0.2.0 → 0.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: c1895359140ba6c85db9f47c33a3ddb7b8558d6d
-  data.tar.gz: d2f610c71b6c413114d1b30f41aab6f8d9fdd403
+  metadata.gz: f07805d14a1985523defc78b1dead46165975d7e
+  data.tar.gz: 6d64e90d76621a9656a40610fcb5e9373062eb25
 SHA512:
-  metadata.gz: 0503371aefa9478b62345cdbce5fcc88d344c2b080471e38e5c1d2bf0bdb0f7cd5f00b61e182d100b477994c567e5211f662a828ec781ef6955a24a19686aaa1
-  data.tar.gz: 1e48577434bafba21a884156ade715dbdc56d7dcabcb9ecdccbed057c22f2bcccd162e8f5c38b3822dbb03bde927b55a3967f940a550386bb9c8670b53aa9693
+  metadata.gz: f51316bdd3a17e6be5dcebb973883b826dfa38d3956a345a455f9d795cd7859a9f9818386846dc6dd326eed96036b4aec907529911ce4c1903b1415aab40820f
+  data.tar.gz: b33a0fc76fa07efc63fcc74d343ae123fc2b9b7572492e9d1387bff0da4ba1490ab82f066537e50366dcb2f57e74779ca71c24d5c3519383a986faa95c25cc2e

data/README.md

@@ -3,7 +3,7 @@ data_cleansing
 
 Data Cleansing solution for Ruby with additional support for Rails and Mongoid
 
-* http://github.com/ClarityServices/data_cleansing
+* http://github.com/reidmorrison/data_cleansing
 
 ## Introduction
 
@@ -15,7 +15,7 @@ in a model and have the data cleansed transparently when required.
 DataCleansing is a framework that allows any data cleansing to be applied to
 specific attributes or fields. At this time it does not supply the cleaning
 solutions themselves since they are usually straight forward, or so complex
-that they don't tend to be too usefull to others. However, over time built-in
+that they don't tend to be too useful to others. However, over time built-in
 cleansing solutions may be added. Feel free to submit any suggestions via a ticket
 or pull request.
 
@@ -27,39 +27,91 @@ or pull request.
 
 ## Examples
 
+### Simple Example
+```ruby
+require 'data_cleansing'
+
+# Define a global cleaner
+DataCleansing.register_cleaner(:strip) {|string| string.strip!}
+
+class User
+  include DataCleansing::Cleanse
+
+  attr_accessor :first_name, :last_name
+
+  # Strip leading and trialing whitespace from first_name and last_name
+  cleanse :first_name, :last_name, :cleaner => :strip
+end
+
+u = User.new
+u.first_name = '    joe   '
+u.last_name = "\n  black\n"
+puts "Before data cleansing #{u.inspect}"
+# Before data cleansing Before data cleansing #<User:0x007fc9f1081980 @first_name="    joe   ", @last_name="\n  black\n">
+
+u.cleanse_attributes!
+puts "After data cleansing #{u.inspect}"
+# After data cleansing After data cleansing #<User:0x007fc9f1081980 @first_name="joe", @last_name="black">
+```
+
 ### Ruby Example
 
 ```ruby
+require 'data_cleansing'
+
+# Define a global cleaners
+DataCleansing.register_cleaner(:strip) {|string| string.strip!}
+DataCleansing.register_cleaner(:upcase) {|string| string.upcase!}
+
 class User
   include DataCleansing::Cleanse
 
-  attr_accessor :first_name, :last_name, :address1, :address2
+  attr_accessor :first_name, :last_name, :title, :address1, :address2, :gender
 
   # Use a global cleaner
   cleanse :first_name, :last_name, :cleaner => :strip
 
   # Define a once off cleaner
   cleanse :address1, :address2, :cleaner => Proc.new {|string| string.strip!}
-end
 
-# Define a global cleanser
-DataCleansing.register_cleaner(:strip) {|string, params, object| string.strip!}
+  # Use multiple cleaners, and a custom block
+  cleanse :title, :cleaner => [:strip, :upcase, Proc.new {|string| "#{string}." unless string.end_with?('.')}]
+
+  # Change the cleansing rule based on the value of other attributes in that instance of user
+  # The 'title' is retrieved from the current instance of the user
+  cleanse :gender, :cleaner => [
+    :strip,
+    :upcase,
+    Proc.new do |gender|
+      if (gender == "UNKNOWN") && (title == "MR.")
+        "Male"
+      else
+        "Female"
+      end
+    end
+  ]
+end
 
 u = User.new
 u.first_name = '    joe   '
 u.last_name = "\n  black\n"
 u.address1 = "2632 Brown St   \n"
+u.title = "   \nmr   \n"
+u.gender = " Unknown  "
 puts "Before data cleansing #{u.inspect}"
+# Before data cleansing #<User:0x007fdd5a83a8f8 @first_name="    joe   ", @last_name="\n  black\n", @address1="2632 Brown St   \n", @title="   \nmr   \n", @gender=" Unknown  ">
+
 u.cleanse_attributes!
 puts "After data cleansing #{u.inspect}"
+# After data cleansing #<User:0x007fdd5a83a8f8 @first_name="joe", @last_name="black", @address1="2632 Brown St", @title="MR.", @gender="Male">
 ```
 
 ### Rails Example
 
-To encrypt a field in a Mongoid document, just add ":encrypted => true" at the end
-of the field specifier. The field name must currently begin with "encrypted_"
-
 ```ruby
+# Define a global cleanser
+DataCleansing.register_cleaner(:strip) {|string| string.strip!}
+
 # 'users' table has the following columns :first_name, :last_name, :address1, :address2
 class User < ActiveRecord::Base
   include DataCleansing::Cleanse
@@ -74,9 +126,6 @@ class User < ActiveRecord::Base
   before_validation :cleanse_attributes!
 end
 
-# Define a global cleanser
-DataCleansing.register_cleaner(:strip) {|string, params, object| string.strip!}
-
 # Create a User instance
 u = User.new(:first_name => '    joe   ', :last_name => "\n  black\n", :address1 => "2632 Brown St   \n")
 puts "Before data cleansing #{u.attributes.inspect}"
@@ -85,6 +134,12 @@ puts "After data cleansing #{u.attributes.inspect}"
 u.save!
 ```
 
+## Notes
+
+Cleaners are called in the order in which they are defined, so subsequent cleaners
+can assume that the previous cleaners have run and can therefore access or even
+modify previously cleaned attributes
+
 ## Installation
 
 ### Add to an existing Rails project
@@ -120,9 +175,9 @@ tries to convert it to an integer or float.
 Meta
 ----
 
-* Code: `git clone git://github.com/ClarityServices/data_cleansing.git`
-* Home: <https://github.com/ClarityServices/data_cleansing>
-* Issues: <http://github.com/ClarityServices/data_cleansing/issues>
+* Code: `git clone git://github.com/reidmorrison/data_cleansing.git`
+* Home: <https://github.com/reidmorrison/data_cleansing>
+* Issues: <http://github.com/reidmorrison/data_cleansing/issues>
 * Gems: <http://rubygems.org/gems/data_cleansing>
 
 This project uses [Semantic Versioning](http://semver.org/).
@@ -135,7 +190,7 @@ Reid Morrison :: reidmo@gmail.com :: @reidmorrison
 License
 -------
 
-Copyright 2013 Clarity Services, Inc.
+Copyright 2013 Reid Morrison, Clarity Services, Inc.
 
 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.

data/lib/data_cleansing/cleanse.rb

@@ -1,16 +1,17 @@
 module DataCleansing
   # Mix-in to add cleaner
   module Cleanse
-    CleanerStruct = Struct.new(:cleaner, :attributes, :params)
+    CleanerStruct = Struct.new(:cleaners, :attributes, :params)
 
     module ClassMethods
       # Define how to cleanse one or more attributes
       def cleanse(*args)
         last = args.last
         params = (last.is_a?(Hash) && last.instance_of?(Hash)) ? args.pop.dup : {}
-        cleaner = params.delete(:cleaner)
-        raise(ArgumentError, "Mandatory :cleaner parameter is missing: #{params.inspect}") unless cleaner
-        (@cleaners ||= ThreadSafe::Array.new) << CleanerStruct.new(cleaner, args, params)
+        cleaners = Array(params.delete(:cleaner))
+        raise(ArgumentError, "Mandatory :cleaner parameter is missing: #{params.inspect}") unless cleaners
+        (@cleaners ||= ThreadSafe::Array.new) << CleanerStruct.new(cleaners, args, params)
+        nil
       end
 
       def cleaners
@@ -23,7 +24,6 @@ module DataCleansing
       def cleanse_attributes!
         self.class.cleaners.each do |cleaner_struct|
           params  = cleaner_struct.params
-          cleaner = cleaner_struct.cleaner
           attrs = cleaner_struct.attributes
 
           # Special case to include :all fields
@@ -62,14 +62,16 @@ module DataCleansing
 
             # No need to clean if attribute is nil
             unless value.nil?
-              new_value = if cleaner.is_a?(Proc)
-                cleaner.call(value, params)
-              else
-                if c = DataCleansing.cleaner(cleaner)
-                  c.call(value, params)
-                else
-                  raise "No cleaner defined for #{cleaner.to_sym}"
-                end
+              # Allow multiple cleaners to be defined and only set the new value
+              # once all cleaners have run
+              new_value = value
+              cleaner_struct.cleaners.each do |cleaner|
+                # Cleaner itself could be a custom Proc, otherwise do a global lookup for it
+                proc = cleaner.is_a?(Proc) ? cleaner : DataCleansing.cleaner(cleaner.to_sym)
+                raise "No cleaner defined for #{cleaner.inspect}" unless proc
+
+                # Call the cleaner proc within the scope (binding) of this object
+                new_value = instance_exec(new_value, params, &proc)
               end
               # Update value if it has changed
               send("#{attr.to_sym}=".to_sym, new_value) if new_value != value
@@ -77,6 +79,7 @@ module DataCleansing
 
           end
         end
+        nil
       end
     end
 

data/lib/data_cleansing/version.rb

@@ -1,3 +1,3 @@
 module DataCleansing
-  VERSION = "0.2.0"
+  VERSION = "0.3.0"
 end

data/test/ruby_test.rb

@@ -21,6 +21,38 @@ class RubyUser
   cleanse :address1, :address2, :cleaner => Proc.new {|string| "<< #{string.strip!} >>"}
 end
 
+# Another global cleaner, used by RubyUser2
+DataCleansing.register_cleaner(:upcase) {|string| string.upcase!}
+
+class RubyUser2
+  include DataCleansing::Cleanse
+
+  attr_accessor :first_name, :last_name, :title, :address1, :address2, :gender
+
+  # Use a global cleaner
+  cleanse :first_name, :last_name, :cleaner => :strip
+
+  # Define a once off cleaner
+  cleanse :address1, :address2, :cleaner => Proc.new {|string| string.strip!}
+
+  # Use multiple cleaners, and a custom block
+  cleanse :title, :cleaner => [:strip, :upcase, Proc.new {|string| "#{string}." unless string.end_with?('.')}]
+
+  # Change the cleansing rule based on the value of other attributes in that instance of user
+  # The 'title' is retrieved from the current instance of the user
+  cleanse :gender, :cleaner => [
+    :strip,
+    :upcase,
+    Proc.new do |gender|
+      if (gender == "UNKNOWN") && (title == "MR.")
+        "Male"
+      else
+        "Female"
+      end
+    end
+  ]
+end
+
 class RubyTest < Test::Unit::TestCase
   context "Ruby Models" do
 
@@ -54,5 +86,33 @@ class RubyTest < Test::Unit::TestCase
       end
     end
 
+    context "with ruby user2" do
+      setup do
+        @user = RubyUser2.new
+        @user.first_name = '    joe   '
+        @user.last_name  = "\n  black\n"
+        @user.address1   = "2632 Brown St   \n"
+        @user.title      = "   \nmr   \n"
+        @user.gender     = " Unknown  "
+      end
+
+      should 'cleanse_attributes!' do
+        @user.cleanse_attributes!
+        assert_equal 'joe', @user.first_name
+        assert_equal 'black', @user.last_name
+        assert_equal '2632 Brown St', @user.address1
+      end
+
+      should 'cleanse_attributes! with multiple cleaners' do
+        @user.cleanse_attributes!
+        assert_equal 'MR.', @user.title
+      end
+
+      should 'cleanse_attributes! referencing other attributes' do
+        @user.cleanse_attributes!
+        assert_equal 'Male', @user.gender
+      end
+    end
+
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: data_cleansing
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.3.0
 platform: ruby
 authors:
 - Reid Morrison
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-07-11 00:00:00.000000000 Z
+date: 2013-07-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: thread_safe