stringfu 0.0.1 → 0.0.2

data/README.markdown

@@ -0,0 +1,3 @@
+# StringFu
+
+StringFu are some simple methods to manipulate strings in order for them to be cleaned up for Natural Language Processing (NLP).

data/lib/stringfu/version.rb

@@ -1,3 +1,3 @@
 module Stringfu
-  VERSION = "0.0.1"
+  VERSION = "0.0.2"
 end

data/lib/stringfu.rb

@@ -1,5 +1,66 @@
 require "stringfu/version"
 
-module Stringfu
-  # Your code goes here...
+class String
+  include StringFu
+end
+
+module StringFu
+  def punc_strip
+    self.gsub(/[!@#\$%^&*;:,<.>?\/|+=\[\]"{}()_-]/,'').gsub(/^'/,'')
+  end
+
+  def punc_strip!
+    replace(self.gsub(/[!@#\$%^&*;:,<.>?\/|+=\[\]"{}()_-]/,'').gsub(/^'/,''))
+  end
+
+  def punc_sub pattern=' '
+    self.sub(/[!@#\$%^&*;:,<.>?\/|+=\[\]"{}()_-]/,pattern).sub(/^'/,pattern)
+  end
+
+  def punc_sub! pattern=' '
+    replace(self.sub(/[!@#\$%^&*;:,<.>?\/|+=\[\]"{}()_-]/,pattern).sub(/^'/,pattern))
+  end
+
+  def punc_gsub pattern=' '
+    self.gsub(/[!@#\$%^&*;:,<.>?\/|+=\[\]"{}()_-]/,pattern).gsub(/^'/,pattern)
+  end
+
+  def punc_gsub! pattern=' '
+    replace(self.gsub(/[!@#\$%^&*;:,<.>?\/|+=\[\]"{}()_-]/,pattern).gsub(/^'/,pattern))
+  end
+
+  def ngrams *args
+    wordphrase = self.split
+
+    if args.size == 0 
+      args = [1..wordphrase.length]
+    end
+  
+    # Inspired by http://www.rubyquiz.com/quiz4.html
+    args = args.map { |arg| Array(arg) }.flatten.uniq.sort
+    args = args.select { |arg| arg <= wordphrase.length and arg > 0 }  
+
+    args.each do |num| 
+      prefix = latinfy num
+      self.instance_variable_set "@#{prefix}grams", []
+      rounds = wordphrase.length - num + 1
+      rounds.times do |iter| 
+        self.instance_variable_get("@#{prefix}grams") << wordphrase[iter...(iter+num)].join(" ")
+      end
+      # Calls :attr_accessor to add new instance variable
+      self.class.__send__(:attr_accessor, "#{prefix}grams".to_sym)
+      self.instance_variable_get("@#{prefix}grams")
+    end
+    args
+  end
+
+  private
+  def latinfy num
+    prefix = {1 => "uni", 2 => "bi", 3 => "tri"}
+    if prefix.has_key? num 
+      prefix[num]
+    else
+      "_#{num}"
+    end
+  end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: stringfu
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.0.2
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2011-09-19 00:00:00.000000000Z
+date: 2011-09-20 00:00:00.000000000Z
 dependencies: []
 description: Extending String with methods to stem, count, and clean words
 email:
@@ -20,6 +20,7 @@ extra_rdoc_files: []
 files:
 - .gitignore
 - Gemfile
+- README.markdown
 - Rakefile
 - lib/stringfu.rb
 - lib/stringfu/version.rb