despamilator 2.1 → 2.1.1

data/History.txt

@@ -67,6 +67,21 @@
 
 * Minor bug fix in shouting filter.
 
-=== IN PROGRESS
+=== 2.1 2011-09-11
 
-* Removed the "matched_by" method. Replaced by "matches".
+* Significant improvement in filtering accuracy. Added...
+ * Suspect punctuation detection.
+ * URL obfuscation detection.
+ * Price detection (dollars only).
+ * Long domain name detection.
+ * Spammy TLD detection.
+
+* Added URL negation for many filters to reduce false positives.
+
+* Deprecated the "matched_by" method. Replaced by "matches".
+
+* Internal refactor to make filters stateless.
+
+=== 2.1.1 2011-09-24
+
+* Fixed syntax error fo Ruby 1.8.

data/README.rdoc

@@ -23,11 +23,11 @@ Using Despamilator:
   dspam = Despamilator.new('some text with an <h2> tag qthhg')
 
   dspam.score #=> the total score for this string (1 is considered high)
-  dspam.matched_by #=> array of matching filters
-  first_match = dspam.matched_by.first #=> first matching filter
-  first_match.name #=> some string with the name of the filter
-  first_match.description #=> some string to describe
-  first_match.score #=> the individual score assigned by this filter
+  dspam.matches #=> array of matching filters
+  first_match = dspam.matches.first #=> first matching filter
+  first_match[:filter].name #=> some string with the name of the filter
+  first_match[:filter].description #=> some string to describe
+  first_match[:score] #=> the individual score assigned by this matching filter
 
 == FILTERING:
 
@@ -49,15 +49,15 @@ They should always supply the following methods:
 
  * name #=> the name of your filter.
  * description #=> what your filter will look for.
- * parse(text) #=> the method that will be called when parsing. A copy of the message is passed in.
+ * parse(subject) #=> the method that will be called when parsing. A copy of the message is passed in.
 
-Along side the above, the following methods are made available to each filter:
+The subject of the detection (including text) is passed into the "parse" method. It provides
+the following methods...
 
- * append_score= #=> method to append a score to the text if there are matches in your parser.
- * matched? #=> whether or not any filter has so far detected something suspect
- * score #=> the current score assigned to the text
+ * text #=> The text of the message. This is immutable but you can alter a duplicate (using "dup").
+ * register_match! #=> Method that receives the instance of the filter and the assigned score if matched.
 
-Take a look at the "naughty_q" code and tests in "spec/filters/naughty_q.rb".
+Take a look at the "prices" code and tests in "spec/filters/prices.rb".
 
 ==== Example Filter:
 
@@ -68,7 +68,7 @@ lib/despamilator/filter/detect_letter_a.rb:
 
   module DespamilatorFilter
 
-    class DetectLetterA < Despamilator::FilterBase
+    class DetectLetterA < Despamilator::Filter
 
       def name
         'Detecting the letter A'
@@ -78,10 +78,10 @@ lib/despamilator/filter/detect_letter_a.rb:
         'Detects the letter "a" in a string for no reason other than a demo'
       end
 
-      def parse text
-        if text.downcase.scan(/a/)
-        # add 0.1 to the score of the text
-        self.append_score = 0.1
+      def parse subject
+        if subject.text.downcase.scan(/a/)
+          # add 0.1 to the score of the text
+          subject.register_match!({:score => 0.1, :filter => self})
       end
     end
   end
@@ -90,8 +90,7 @@ As previously stated, ensure you put a spec test together as well!
 
 == REQUIREMENTS:
 
-* hoe
-* rspec
+* domainatrix
 
 
 == INSTALL:

data/lib/despamilator.rb

@@ -15,8 +15,8 @@ require 'ostruct'
 #
 #  dspam = Despamilator.new('some text with an <h2> tag qthhg')
 #
-#  dspam.score #=> the total score for this string (1 is normally my threshold)
-#  dspam.matched_by #=> array of matching filters
+#  dspam.score #=> the total score for this string (1 is normally my threshold).
+#  dspam.matches #=> array of hashes containing matching filters and their score.
 
 class Despamilator
 

data/lib/despamilator/filter/obfuscated_urls.rb

@@ -15,7 +15,7 @@ module DespamilatorFilter
       count += find_space_separated_characters text
 
       # weird maths below is due to some issue with ruby 1.9.2 multiplying floats by 3 (?!)
-      subject.register_match!({score: (4.0 * count) / 10, filter: self}) if count > 0
+      subject.register_match!({:score => (4.0 * count) / 10, :filter => self}) if count > 0
     end
 
     private

data/lib/despamilator/filter/spammy_tlds.rb

@@ -13,7 +13,7 @@ module DespamilatorFilter
     end
 
     def parse subject
-      matches = subject.text.count(/\w{5,}\.(info|biz)\b/)
+      matches = subject.text.count(/\w{5,}\.(info|biz|xxx)\b/)
       subject.register_match!({:score => 0.05 * matches, :filter => self}) if matches > 0
     end
 

data/lib/despamilator/version.rb

@@ -1,3 +1,3 @@
 class Despamilator
-  VERSION = 2.1
+  VERSION = '2.1.1'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: despamilator
 version: !ruby/object:Gem::Version
-  version: '2.1'
+  version: 2.1.1
   prerelease: 
 platform: ruby
 authors:
@@ -9,11 +9,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2011-09-11 00:00:00.000000000Z
+date: 2011-09-24 00:00:00.000000000Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: domainatrix
-  requirement: &70312466224620 !ruby/object:Gem::Requirement
+  requirement: &70225887914340 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,7 +21,7 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *70312466224620
+  version_requirements: *70225887914340
 description: ! 'Despamilator is a plugin based spam detector designed for use on your
   web forms borne out of two annoyances: Spam being submitted in my web forms and
   CAPTCHAS being intrusive. Despamilator will apply some commonly used heuristics
@@ -74,7 +74,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ! '>='
     - !ruby/object:Gem::Version
-      version: 1.3.6
+      version: 1.3.5
 requirements: []
 rubyforge_project: despamilator
 rubygems_version: 1.8.6