data_cleansing 0.4.0 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 2dfd8030076d210f31f5993ed3646abbe38934ea
-  data.tar.gz: a077de74aec84cb805e4087e4e6d04c5c8972a73
+  metadata.gz: 720a83a0b486ac0d4ee6de7c7716de4d47ae6331
+  data.tar.gz: 5f2f86c19d4cfb896f124802bef947e6ebbaf1a7
 SHA512:
-  metadata.gz: 9ee068df2896ee7ae1656fc8af6a4baf446b5969972dae873e62833689a3a110e546135a0cd8b37c23847374839aec79b9e12ef8d9c3baf138a1aaf85c00d47d
-  data.tar.gz: dac70b47e27ac5fbf4d96ae12570c071ebb694b69f0f039a798cf6d2a5e28a850845fd03879751eb2fb0eb1b6a11515f8163ee240763d5ac32d4f30d7d7797a2
+  metadata.gz: 006f1106f3f67dad19c979d0384cb85f0c2bdbe8d3150d6c6ff13629b0c24fa1aaad70b0c9e66718d615dc63d5ca4202bf2125f8fd91a0190f4a1601a3a83a3e
+  data.tar.gz: 83a1ec0a5da0c2ec7d0a3475db95e5360613c53d36a4837b33385e1a8309e0cda894ede12d9dd8861c6d6db1ad80f683cdadff34191b99e87a299de897d6074b

data/README.md

@@ -23,7 +23,27 @@ or pull request.
 
 * Supports global cleansing definitions that can be associated with any Ruby,
   Rails, Mongoid, or other model.
-* Supports custom cleansing definitions for a single attribute
+* Supports custom cleansing definitions that can be defined in-line using block.
+* A cleansing block can access the other attributes in the model in determining
+  how to cleanse the current attribute
+* In a cleansing block other can also be modified if necessary
+* Cleansers are executed in the order they are defined. As a result multiple
+  cleansers can be run against the same field and the order is preserved
+* Multiple cleansers can be specified for a list of attributes at the same time
+* Inheritance is supported. The cleansers for parent classes are run before
+  the child's cleansers
+* Cleansers can be called outside of a model instance for cases where fields
+  need to be cleansed before the model is created, or needs to be found
+* Logging of data cleansing with the before and after values for troubleshooting.
+  Depending on the log level all modified fields are logged, or just the ones
+  completely wiped out to nil
+
+## ActiveRecord (ActiveModel) Features
+
+* Passes the value of the attribute before the Rails type cast so that the
+  original text can be cleansed before passing back to rails for type conversion.
+  This is important for numeric and date fields where spaces and control characters
+  can have undesired effects
 
 ## Examples
 
@@ -134,6 +154,62 @@ 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 configuration
+
+When DataCleansing is used in a Rails environment it can be configured using the
+regular Rails configuration mechanisms. For example:
+
+```ruby
+module MyApplication
+  class Application < Rails::Application
+
+   # Data Cleansing Configuration
+
+   # Attributes who's values are to be masked out during logging
+   config.data_cleansing.register_masked_attributes :bank_account_number, :social_security_number
+
+   # Optionally override the default log level
+   #   Set to :trace or :debug to log all fields modified
+   #   Set to :info to log only those fields which were nilled out
+   #   Set to :warn or higher to disable logging of cleansing actions
+   config.data_cleansing.logger.level = :info
+
+   # Register any global cleaners
+   config.data_cleansing.register_cleaner(:strip) {|string| string.strip!}
+
+  end
+end
+```
+
+## Logging
+
+DataCleansing uses SemanticLogger for logging due to it's excellent integration
+with Rails and its ability to log data in it's raw form to Mongo and to files.
+
+If running a Rails application it is recommended to install the gem
+rails_semantic_logger which replaces the default Rails logger. It is however
+possible to configure the semantic_logger gem to use the existing Rails logger
+in a Rails initializer as follows:
+
+```ruby
+SemanticLogger.default_level = Rails.logger.level
+SemanticLogger.add_appender(Rails.logger)
+```
+
+By changing the log level for DataCleansing the type of output for data
+cleansing can be controlled:
+
+* :trace or :debug to log all fields modified
+* :info to log only those fields which were nilled out
+* :warn or higher to disable logging of cleansing actions
+
+To change the log level, either use the Rails configuration approach, or set it
+directly:
+
+```ruby
+DataCleansing.logger.level = :info
+```
+
 ## Notes
 
 Cleaners are called in the order in which they are defined, so subsequent cleaners
@@ -178,7 +254,7 @@ DataCleansing requires the following dependencies
 
 * Ruby V1.8.7, V1.9.3 or V2 and greater
 * Rails V2 or greater for Rails integration ( Only if Rails is being used )
-* Mongoid V2 or greater for Rails integration ( Only if Mongoid is being used )
+* Mongoid V2 or greater for Mongoid integration ( Only if Mongoid is being used )
 
 ## Meta
 

data/lib/data_cleansing/cleanse.rb

@@ -114,11 +114,12 @@ module DataCleansing
       def data_cleansing_execute_cleaners(cleaners)
         return false if cleaners.nil?
 
-        # Capture all changes to attributes if the log level is high enough
-        changes = {} if DataCleansing.logger.send("#{DataCleansing.cleansing_log_level}?")
+        # Capture all changes to attributes if the log level is :info or greater
+        changes = {} if DataCleansing.logger.info?
+        # Capture all modified fields if log_level is :debug or :trace
+        verbose = DataCleansing.logger.debug?
 
-        #logger.send(self.class.data_cleansing_log_level, "Cleansed Attributes", changes) if changes
-        DataCleansing.logger.send("benchmark_#{DataCleansing.cleansing_log_level}","Cleansed Attributes", :payload => changes) do
+        DataCleansing.logger.benchmark_info("cleanse_attributes!", :payload => changes) do
           cleaners.each do |cleaner_struct|
             params = cleaner_struct.params
             attrs  = cleaner_struct.attributes
@@ -177,10 +178,12 @@ module DataCleansing
                     if previous = changes[attr.to_sym]
                       previous[:after] = new_value
                     else
-                      changes[attr.to_sym] = {
-                        :before => masked ? :masked : value,
-                        :after  => new_value
-                      }
+                      if new_value.nil? || verbose
+                        changes[attr.to_sym] = {
+                          :before => masked ? :masked : value,
+                          :after  => new_value
+                        }
+                      end
                     end
                   end
                 end

data/lib/data_cleansing/data_cleansing.rb

@@ -4,7 +4,6 @@ module DataCleansing
   # Global Data Cleansers
   @@global_cleaners     = ThreadSafe::Hash.new
   @@masked_attributes   = ThreadSafe::Array.new
-  @@cleansing_log_level = :info
 
   # Register a new cleaner
   # Replaces any existing cleaner with the same name
@@ -28,14 +27,4 @@ module DataCleansing
     @@masked_attributes.freeze
   end
 
-  # Set the log_level at which to log cleansing activities at
-  def self.cleansing_log_level
-    @@cleansing_log_level
-  end
-
-  # Set the log_level at which to log cleansing activities at
-  def self.cleansing_log_level=(log_level)
-    @@cleansing_log_level = log_level
-  end
-
 end

data/lib/data_cleansing/railtie.rb

@@ -6,9 +6,25 @@ module RubySkynet #:nodoc:
     # @example Set up configuration in the Rails app.
     #   module MyApplication
     #     class Application < Rails::Application
-    #       config.data_cleansing.masked_attributes :bank_account_number, :social_security_number
-    #       config.data_cleansing.cleansing_log_level :info
+    #
+    #       # Data Cleansing Configuration
+    #
+    #       # By default logging is enabled of data cleansing actions
+    #       # Set to false to disable
+    #       config.data_cleansing.logging_enabled = true
+    #
+    #       # Attributes who's values are to be masked out during logging
+    #       config.data_cleansing.register_masked_attributes :bank_account_number, :social_security_number
+    #
+    #       # Optionally override the default log level
+    #       #   Set to :trace or :debug to log all fields modified
+    #       #   Set to :info to log only those fields which were nilled out
+    #       #   Set to :warn or higher to disable logging of cleansing actions
+    #       config.data_cleansing.logger.level = :info
+    #
+    #       # Register any global cleaners
     #       config.data_cleansing.register_cleaner(:strip) {|string| string.strip!}
+    #
     #     end
     #   end
     config.data_cleansing = ::DataCleansing

data/lib/data_cleansing/version.rb

@@ -1,3 +1,3 @@
 module DataCleansing
-  VERSION = "0.4.0"
+  VERSION = "0.5.0"
 end

data/test/active_record_test.rb

@@ -39,9 +39,9 @@ end
 # Define a global cleaner
 DataCleansing.register_cleaner(:strip) {|string, params, object| string.strip!}
 
-# Log data cleansing result at :info level
-# Set to a level below the current log level to disable
-DataCleansing.cleansing_log_level = :info
+# Log data cleansing result
+# Set to :warn or higher to disable
+DataCleansing.logger.level = :debug
 
 # Set the Global list of fields to be masked
 DataCleansing.register_masked_attributes :ssn, :bank_account_number

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: data_cleansing
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.5.0
 platform: ruby
 authors:
 - Reid Morrison
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-07-25 00:00:00.000000000 Z
+date: 2013-07-26 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: thread_safe