bloomfilter 0.1.0 → 0.1.1

data/History.txt

@@ -1,4 +1,2 @@
-== 0.0.1 2007-09-05
-
-* 1 major enhancement:
-  * Initial release
+== 0.1.1 2007-09-05
+  Some files were missing from the gem manifest.

data/Manifest.txt

@@ -5,9 +5,12 @@ README.txt
 Rakefile
 lib/bloomfilter.rb
 lib/bloomfilter/version.rb
+lib/bloomfilter/bloomfilter.rb
+lib/bloomfilter/external_bloom_filter.rb
 scripts/txt2html
 setup.rb
 test/test_bloomfilter.rb
+test/test_external_bloom_filter.rb
 test/test_helper.rb
 website/index.html
 website/index.txt

data/lib/bloomfilter/bloomfilter.rb

@@ -0,0 +1,87 @@
+require 'digest/sha1'
+require "rubygems"
+require "inline"
+
+class FasterBitField
+  attr_reader :size
+  
+  ELEMENT_WIDTH = 8
+  
+  def initialize(size, field = nil)
+    @size = size
+    @field = field || "\000" * (((size - 1) / ELEMENT_WIDTH) + 1)
+  end
+  
+  def set(position)
+    set_c(@field, position)
+    #@field[position / ELEMENT_WIDTH] |= (1 << (position % ELEMENT_WIDTH))
+  end
+  
+  # Read a bit (1/0)
+  def [](position)
+    get_c(@field, position)
+    #@field[position / ELEMENT_WIDTH] & (1 << (position % ELEMENT_WIDTH)) > 0
+  end
+  
+  inline do |builder|
+    builder.c <<-code
+      void set_c(char *bits, int position) {
+        bits[position / 8] |= (1 << (position % 8) );
+      }
+    code
+    
+    builder.c <<-code
+      VALUE get_c(char *bits, int position) {
+        int flag = bits[position / 8] & (1 << ( position % 8 ));
+        if( flag > 0){
+          return Qtrue;
+        }
+        else
+          return Qfalse;
+      }
+    code
+  end
+  
+  def to_s
+    @field
+  end
+end
+
+class BloomFilter
+  def initialize(num_bits, field = nil)
+    if field
+      @bits = FasterBitField.new(num_bits, field)
+    else
+      @bits = FasterBitField.new(num_bits)
+    end
+    
+    @num_bits = num_bits
+  end
+  
+  def hash_value(item)
+    Digest::SHA1.hexdigest(item).scan(/.{8}/).map{|str| str.to_i(16)}
+  end
+  
+  def add(item)
+    hash_value(item).each{|bit_position| @bits.set(bit_position % @num_bits)}
+  end
+  
+  def include?(item)
+    hash_value(item).each{|bit_position| return false unless @bits[bit_position % @num_bits]}
+    true
+  end
+  
+  def save(path)
+    File.open(path, "w") do |file|
+      file.write @bits.to_s
+    end
+  end
+  
+  def self.load(path)
+    field = ""
+    File.open(path, "r") do |file|
+      field = file.read
+    end
+    BloomFilter.new(field.size, field)
+  end
+end

data/lib/bloomfilter/external_bloom_filter.rb

@@ -0,0 +1,71 @@
+require 'digest/sha1'
+
+class ExternalBitField
+  attr_reader :size
+  
+  ELEMENT_WIDTH = 8
+  
+  def self.create(path, size)
+    File.open(path, "w") do |file|
+      file.write "\000" * (((size - 1) / ELEMENT_WIDTH) + 1)
+    end
+    ExternalBitField.new(path)
+  end
+  
+  def initialize(path)
+    @size = File.size(path) * ELEMENT_WIDTH
+    @field = File.new(path, "r+")
+  end
+  
+  def set(position)
+    #debugger
+    # seek to the position in the file where we'll be making changes
+    @field.seek(position / ELEMENT_WIDTH)
+
+    # read the old value, converted to an integer
+    old_val = @field.read(1)[0]
+    
+    # seek back to our spot
+    @field.seek(position / ELEMENT_WIDTH)
+    
+    # write the new value, as a character again this time
+    @field.write( (old_val | 1 << (position % ELEMENT_WIDTH)).chr )
+  end
+  
+  # Read a bit (1/0)
+  def [](position)
+    # seek to the appropriate byte position
+    @field.seek(position / ELEMENT_WIDTH)
+    
+    # read the value
+    val = @field.read(1)[0]
+
+    # figure out if our bit is flipped or not
+    (val & (1 << (position % ELEMENT_WIDTH))) > 0
+  end
+  
+end
+
+class ExternalBloomFilter
+  def self.create(path, size)
+    ExternalBitField.create(path, size)
+    ExternalBloomFilter.new(path)
+  end
+    
+  def initialize(path)
+    @bits = ExternalBitField.new(path)
+  end
+  
+  def hash_value(item)
+    Digest::SHA1.hexdigest(item).scan(/.{8}/).map{|str| str.to_i(16)}
+  end
+  
+  def add(item)
+    hash_value(item).each{|bit_position| @bits.set(bit_position % @bits.size)}
+  end
+  
+  def include?(item)
+    hash_value(item).each{|bit_position| return false unless @bits[bit_position % @bits.size]}
+    true
+  end
+end

data/lib/bloomfilter/version.rb

@@ -2,7 +2,7 @@ module Bloomfilter #:nodoc:
   module VERSION #:nodoc:
     MAJOR = 0
     MINOR = 1
-    TINY  = 0
+    TINY  = 1
 
     STRING = [MAJOR, MINOR, TINY].join('.')
   end

data/website/index.html

@@ -33,7 +33,7 @@
     <h1>bloomfilter</h1>
     <div id="version" class="clickable" onclick='document.location = "http://rubyforge.org/projects/bloomfilter"; return false'>
       <p>Get Version</p>
-      <a href="http://rubyforge.org/projects/bloomfilter" class="numbers">0.1.0</a>
+      <a href="http://rubyforge.org/projects/bloomfilter" class="numbers">0.1.1</a>
     </div>
     <h1>&#x2192; &#8216;bloomfilter&#8217;</h1>
 

metadata

@@ -3,7 +3,7 @@ rubygems_version: 0.9.4
 specification_version: 1
 name: bloomfilter
 version: !ruby/object:Gem::Version 
-  version: 0.1.0
+  version: 0.1.1
 date: 2007-09-05 00:00:00 -07:00
 summary: Two Bloom filter implementations.
 require_paths: 
@@ -36,9 +36,12 @@ files:
 - Rakefile
 - lib/bloomfilter.rb
 - lib/bloomfilter/version.rb
+- lib/bloomfilter/bloomfilter.rb
+- lib/bloomfilter/external_bloom_filter.rb
 - scripts/txt2html
 - setup.rb
 - test/test_bloomfilter.rb
+- test/test_external_bloom_filter.rb
 - test/test_helper.rb
 - website/index.html
 - website/index.txt