bloomfilter-rb 2.1.1 → 2.1.2

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: f6473c63de973faec078172e7d3817280d448b8b1d3dd963e44a300a2aff60b2
+  data.tar.gz: c92fedd09bcea80a7ff93bff4ccb7a0bfe87b2e55af2d714b96193a4d5bbc870
+SHA512:
+  metadata.gz: 5bac15299e3a82b183c24095771ff3e302e2f39ffefbeb20185118e37fe1a50170dd4675043761695dd963f3b93764ac16629ccfe22c8b5b71c6b1e14c75e28b
+  data.tar.gz: 86b59a43ae46eac06d8392fb772b0f661531e31c697907775898dd02a67b7c79f020e99b987b6fb5319e7b4636949101e6a5911c24712e8e7a25bbb22dc59ceb

data/.gitignore

@@ -1,4 +1,10 @@
 *.o
 *.bundle
 *.swp
+.rvmrc
+.DS_Store
 ext/Makefile
+lib/cbloomfilter.so
+tmp
+Gemfile.lock
+pkg

data/Gemfile

@@ -1,3 +1,3 @@
 source "http://rubygems.org"
 
-gemspec
+gemspec

data/README.md

@@ -23,26 +23,28 @@ Performance of the Bloom filter depends on a number of variables:
 
 MRI/C implementation which creates an in-memory filter which can be saved and reloaded from disk.
 
-    require 'bloomfilter'
+```ruby
+require 'bloomfilter-rb'
 
-    bf = BloomFilter::Native.new(:size => 100, :hashes => 2, :seed => 1, :bucket => 3, :raise => false)
-    bf.insert("test")
-    bf.include?("test")     # => true
-    bf.include?("blah")     # => false
+bf = BloomFilter::Native.new(:size => 100, :hashes => 2, :seed => 1, :bucket => 3, :raise => false)
+bf.insert("test")
+bf.include?("test")     # => true
+bf.include?("blah")     # => false
 
-    bf.delete("test")
-    bf.include?("test")     # => false
+bf.delete("test")
+bf.include?("test")     # => false
 
-    # Hash with a bloom filter!
-    bf["test2"] = "bar"
-    bf["test2"]             # => true
-    bf["test3"]             # => false
+# Hash with a bloom filter!
+bf["test2"] = "bar"
+bf["test2"]             # => true
+bf["test3"]             # => false
 
-    bf.stats
-    Number of filter bits (m): 10
-    Number of filter elements (n): 2
-    Number of filter hashes (k) : 2
-    Predicted false positive rate = 10.87%
+bf.stats
+# => Number of filter bits (m): 10
+# => Number of filter elements (n): 2
+# => Number of filter hashes (k) : 2
+# => Predicted false positive rate = 10.87%
+```
 
 ***
 
@@ -50,14 +52,16 @@ MRI/C implementation which creates an in-memory filter which can be saved and re
 
 Uses [getbit](http://redis.io/commands/getbit)/[setbit](http://redis.io/commands/setbit) on Redis strings - efficient, fast, can be shared by multiple/concurrent processes.
 
-    bf = BloomFilter::Redis.new
+```ruby
+bf = BloomFilter::Redis.new
 
-    bf.insert('test')
-    bf.include?('test')     # => true
-    bf.include?('blah')     # => false
+bf.insert('test')
+bf.include?('test')     # => true
+bf.include?('blah')     # => false
 
-    bf.delete('test')
-    bf.include?('test')     # => false
+bf.delete('test')
+bf.include?('test')     # => false
+```
 
 ### Memory footprint
 
@@ -67,16 +71,18 @@ Uses [getbit](http://redis.io/commands/getbit)/[setbit](http://redis.io/commands
 
 ***
 
-## Redis-backed counting bloom filter with TTL's
+## Redis-backed counting bloom filter with TTLs
 Uses regular Redis get/set counters to implement a counting filter with optional TTL expiry. Because each "bit" requires its own key in Redis, you do incur a much larger memory overhead.
 
-    bf = BloomFilter::CountingRedis.new(:ttl => 2)
+```ruby
+bf = BloomFilter::CountingRedis.new(:ttl => 2)
 
-    bf.insert('test')
-    bf.include?('test')     # => true
+bf.insert('test')
+bf.include?('test')     # => true
 
-    sleep(2)
-    bf.include?('test')     # => false
+sleep(2)
+bf.include?('test')     # => false
+```
 
 ## Credits
 
@@ -84,4 +90,4 @@ Tatsuya Mori <valdzone@gmail.com> (Original C implementation: http://vald.x0.com
 
 ## License
 
-(MIT License) - Copyright (c) 2011 Ilya Grigorik
+MIT License - Copyright (c) 2011 Ilya Grigorik

data/Rakefile

@@ -1,9 +1,11 @@
+require 'bundler/gem_tasks'
 require 'rake'
 require 'rspec'
 require 'rspec/core/rake_task'
 require 'rake/extensiontask'
-require 'bundler'
 
 Bundler::GemHelper.install_tasks
+Rake::ExtensionTask.new('cbloomfilter')
 RSpec::Core::RakeTask.new(:spec)
-Rake::ExtensionTask.new('cbloomfilter')
+Rake::Task[:spec].prerequisites << :clean
+Rake::Task[:spec].prerequisites << :compile

data/bloomfilter-rb.gemspec

@@ -8,14 +8,15 @@ Gem::Specification.new do |s|
   s.platform    = Gem::Platform::RUBY
   s.authors     = ["Ilya Grigorik", "Tatsuya Mori"]
   s.email       = ["ilya@igvita.com"]
-  s.homepage    = "http://github.com/igrigorik/bloomfilter"
+  s.homepage    = "http://github.com/igrigorik/bloomfilter-rb"
   s.summary     = "Counting Bloom Filter implemented in Ruby"
   s.description = s.summary
   s.rubyforge_project = "bloomfilter-rb"
 
-  s.add_dependency "redis"
-  s.add_development_dependency "rspec"
+  s.add_development_dependency "redis"
+  s.add_development_dependency "rspec", ">= 3"
   s.add_development_dependency "rake"
+  s.add_development_dependency "rake-compiler" 
 
   s.extensions = ["ext/cbloomfilter/extconf.rb"]
 
@@ -23,4 +24,4 @@ Gem::Specification.new do |s|
   s.test_files    = `git ls-files -- {test,spec,features}/*`.split("\n")
   s.executables   = `git ls-files -- bin/*`.split("\n").map{ |f| File.basename(f) }
   s.require_paths = ["lib"]
-end
+end

data/examples/pure-ruby-bf.rb

@@ -10,8 +10,8 @@ class BloomFilter
   def initialize(max_entries, num_hashes, seed)
     @num_hashes = num_hashes
     @size = max_entries.to_i
-    @bitmap = BitSet.new(@size)
-    @__mask = BitSet.new(@size)
+    @bitmap = Bitset.new(@size)
+    @__mask = Bitset.new(@size)
     @seed = seed
   end
 
@@ -41,7 +41,7 @@ def main
   while line = ARGF.gets
     data = line.chop
 
-    if bf.new_entry?(data)
+    if bf.new?(data)
       num += 1
       bf.insert(data)
     end

data/ext/cbloomfilter/cbloomfilter.c

@@ -19,7 +19,6 @@ struct BloomFilter {
     int k; /* # of hash functions */
     int s; /* # seed of hash functions */
     int r; /* # raise on bucket overflow? */
-    int num_set; /* # of set bits */
     unsigned char *ptr; /* bits data */
     int bytes; /* size of byte data */
 };
@@ -37,8 +36,13 @@ void bucket_unset(struct BloomFilter *bf, int index) {
     if ((c & mask) == 0) {
       // do nothing
     } else {
-        bf->ptr[byte_offset] -= (1 << bit_offset) & ((1 << 8) - 1);
-        bf->ptr[byte_offset + 1] -= ((1 << bit_offset) & ((1 << 16) - 1)) >> 8;
+        // reduce the counter: 11 00 => 10 00 (suppose bf->b is 2)
+        c -= (1 << bit_offset) & ((1 << 8) -1);
+        // shift the bitmap right by 1 bit: 10 00 => 01 00
+        c = (~mask & c) | ((c & mask) >> (bit_offset + 1) << bit_offset);
+
+        bf->ptr[byte_offset] = c & ((1 << 8) - 1);
+        bf->ptr[byte_offset + 1] = (c & ((1 << 16) - 1)) >> 8;
     }
 
 }
@@ -52,8 +56,9 @@ void bucket_set(struct BloomFilter *bf, int index) {
     if ((c & mask) == mask) {
         if (bf->r == 1) rb_raise(rb_eRuntimeError, "bucket got filled up");
     } else {
-        bf->ptr[byte_offset] += (1 << bit_offset) & ((1 << 8) - 1);
-        bf->ptr[byte_offset + 1] += ((1 << bit_offset) & ((1 << 16) - 1)) >> 8;
+        c = c + ((1 << bit_offset) & ((1 << 8) -1)) | c;
+        bf->ptr[byte_offset] = c & ((1 << 8) - 1);
+        bf->ptr[byte_offset + 1] = (c & ((1 << 16) - 1)) >> 8;
     }
 }
 
@@ -127,7 +132,6 @@ static VALUE bf_s_new(int argc, VALUE *argv, VALUE self) {
     bf->k = k;
     bf->s = s;
     bf->r = r;
-    bf->num_set = 0;
 
     bf->bytes = ((m * b) + 15) / 8;
     bf->ptr = ALLOC_N(unsigned char, bf->bytes);
@@ -170,10 +174,22 @@ static VALUE bf_r(VALUE self) {
     return bf->r == 0 ? Qfalse : Qtrue;
 }
 
-static VALUE bf_num_set(VALUE self) {
+static VALUE bf_s(VALUE self) {
+    struct BloomFilter *bf;
+    Data_Get_Struct(self, struct BloomFilter, bf);
+    return INT2FIX(bf->s);
+}
+
+static VALUE bf_set_bits(VALUE self){
     struct BloomFilter *bf;
+    int i,j,count = 0;
     Data_Get_Struct(self, struct BloomFilter, bf);
-    return INT2FIX(bf->num_set);
+    for (i = 0; i < bf->bytes; i++) {
+        for (j = 0; j < 8; j++) {
+            count += (bf->ptr[i] >> j) & 1;
+        }
+    }
+    return INT2FIX(count);
 }
 
 static VALUE bf_insert(VALUE self, VALUE key) {
@@ -203,21 +219,64 @@ static VALUE bf_insert(VALUE self, VALUE key) {
         bucket_set(bf, index);
     }
 
-    bf->num_set += 1;
     return Qnil;
 }
 
 static VALUE bf_merge(VALUE self, VALUE other) {
     struct BloomFilter *bf, *target;
+    int i;
     Data_Get_Struct(self, struct BloomFilter, bf);
     Data_Get_Struct(other, struct BloomFilter, target);
-    int i;
     for (i = 0; i < bf->bytes; i++) {
         bf->ptr[i] |= target->ptr[i];
     }
     return Qnil;
 }
 
+static VALUE bf_and(VALUE self, VALUE other) {
+    struct BloomFilter *bf, *bf_other, *target;
+    VALUE klass, obj, args[5];
+    int i;
+
+    Data_Get_Struct(self, struct BloomFilter, bf);
+    Data_Get_Struct(other, struct BloomFilter, bf_other);
+    args[0] = INT2FIX(bf->m);
+    args[1] = INT2FIX(bf->k);
+    args[2] = INT2FIX(bf->s);
+    args[3] = INT2FIX(bf->b);
+    args[4] = INT2FIX(bf->r);
+    klass = rb_funcall(self,rb_intern("class"),0);
+    obj = bf_s_new(5,args,klass);
+    Data_Get_Struct(obj, struct BloomFilter, target);
+    for (i = 0; i < bf->bytes; i++){
+        target->ptr[i] = bf->ptr[i] & bf_other->ptr[i];
+    }
+
+    return obj;
+}
+
+static VALUE bf_or(VALUE self, VALUE other) {
+    struct BloomFilter *bf, *bf_other, *target;
+    VALUE klass, obj, args[5];
+    int i;
+
+    Data_Get_Struct(self, struct BloomFilter, bf);
+    Data_Get_Struct(other, struct BloomFilter, bf_other);
+    args[0] = INT2FIX(bf->m);
+    args[1] = INT2FIX(bf->k);
+    args[2] = INT2FIX(bf->s);
+    args[3] = INT2FIX(bf->b);
+    args[4] = INT2FIX(bf->r);
+    klass = rb_funcall(self,rb_intern("class"),0);
+    obj = bf_s_new(5,args,klass);
+    Data_Get_Struct(obj, struct BloomFilter, target);
+    for (i = 0; i < bf->bytes; i++){
+        target->ptr[i] = bf->ptr[i] | bf_other->ptr[i];
+    }
+
+    return obj;
+}
+
 static VALUE bf_delete(VALUE self, VALUE key) {
     int index, seed;
     int i, len, m, k, s;
@@ -245,7 +304,6 @@ static VALUE bf_delete(VALUE self, VALUE key) {
         bucket_unset(bf, index);
     }
 
-    bf->num_set += 1;
     return Qnil;
 }
 
@@ -309,13 +367,11 @@ static VALUE bf_bitmap(VALUE self) {
     struct BloomFilter *bf;
     Data_Get_Struct(self, struct BloomFilter, bf);
 
-    VALUE str = rb_str_new(0, bf->m);
+    VALUE str = rb_str_new(0, bf->bytes);
     unsigned char* ptr = (unsigned char *) RSTRING_PTR(str);
 
-    int i;
-    for (i = 0; i < bf->m; i++)
-        *ptr++ = bucket_get(bf, i);
-
+    memcpy(ptr, bf->ptr, bf->bytes);
+    
     return str;
 }
 
@@ -324,11 +380,7 @@ static VALUE bf_load(VALUE self, VALUE bitmap) {
     Data_Get_Struct(self, struct BloomFilter, bf);
     unsigned char* ptr = (unsigned char *) RSTRING_PTR(bitmap);
 
-    int i;
-    for (i = 0; i < bf->m; i++) {
-      if (*ptr++)
-        bucket_set(bf, i);
-    }
+    memcpy(bf->ptr, ptr, bf->bytes);    
 
     return Qnil;
 }
@@ -340,12 +392,15 @@ void Init_cbloomfilter(void) {
     rb_define_method(cBloomFilter, "k", bf_k, 0);
     rb_define_method(cBloomFilter, "b", bf_b, 0);
     rb_define_method(cBloomFilter, "r", bf_r, 0);
-    rb_define_method(cBloomFilter, "num_set", bf_num_set, 0);
+    rb_define_method(cBloomFilter, "set_bits", bf_set_bits, 0);
+    rb_define_method(cBloomFilter, "s", bf_s, 0);
     rb_define_method(cBloomFilter, "insert", bf_insert, 1);
     rb_define_method(cBloomFilter, "delete", bf_delete, 1);
     rb_define_method(cBloomFilter, "include?", bf_include, -1);
     rb_define_method(cBloomFilter, "clear", bf_clear, 0);
     rb_define_method(cBloomFilter, "merge!", bf_merge, 1);
+    rb_define_method(cBloomFilter, "&", bf_and, 1);
+    rb_define_method(cBloomFilter, "|", bf_or, 1);
 
     rb_define_method(cBloomFilter, "to_s", bf_to_s, 0);
     rb_define_method(cBloomFilter, "bitmap", bf_bitmap, 0);
@@ -353,7 +408,5 @@ void Init_cbloomfilter(void) {
 
     /* functions that have not been implemented, yet */
 
-    //  rb_define_method(cBloomFilter, "&", bf_and, 1);
-    //  rb_define_method(cBloomFilter, "|", bf_or, 1);
     //  rb_define_method(cBloomFilter, "<=>", bf_cmp, 1);
 }

data/lib/bloomfilter-rb.rb

@@ -1,4 +1,3 @@
-require 'redis'
 require 'zlib'
 
 require 'cbloomfilter'
@@ -6,4 +5,4 @@ require 'bloomfilter/filter'
 require 'bloomfilter/native'
 require 'bloomfilter/counting_redis'
 require 'bloomfilter/redis'
-require 'bloomfilter/version'
+require 'bloomfilter/version'

data/lib/bloomfilter.rb

@@ -0,0 +1 @@
+require 'bloomfilter-rb'

data/lib/bloomfilter/counting_redis.rb

@@ -3,14 +3,15 @@ module BloomFilter
 
     def initialize(opts = {})
       @opts = {
-        :size    => 100,
-        :hashes  => 4,
-        :seed    => Time.now.to_i,
-        :bucket  => 3,
-        :ttl => false,
-        :server => {}
+        :identifier => 'rbloom',
+        :size       => 100,
+        :hashes     => 4,
+        :seed       => Time.now.to_i,
+        :bucket     => 3,
+        :ttl        => false,
+        :server     => {}
       }.merge opts
-      @db = ::Redis.new(@opts[:server])
+      @db = @opts.delete(:db) || ::Redis.new(@opts[:server])
     end
 
     def insert(key, ttl=nil)
@@ -25,8 +26,10 @@ module BloomFilter
 
     def delete(key)
       indexes_for(key).each do |idx|
-        if @db.decr(idx).to_i <= 0
+        count = @db.decr(idx).to_i
+        if count <= 0
           @db.del(idx)
+          @db.setbit(idx, 0) if count < 0
         end
       end
     end
@@ -38,7 +41,7 @@ module BloomFilter
     alias :key? :include?
 
     def num_set
-      @db.keys("rbloom:*").size
+      @db.eval("return #redis.call('keys', '#{@opts[:identifier]}:*')")
     end
     alias :size :num_set
 
@@ -52,7 +55,7 @@ module BloomFilter
       def indexes_for(key)
         indexes = []
         @opts[:hashes].times do |i|
-          indexes.push "rbloom:" + (Zlib.crc32("#{key}:#{i+@opts[:seed]}") % @opts[:size]).to_s
+          indexes.push @opts[:identifier] + ":" + (Zlib.crc32("#{key}:#{i+@opts[:seed]}") % @opts[:size]).to_s
         end
 
         indexes

data/lib/bloomfilter/native.rb

@@ -1,7 +1,9 @@
 module BloomFilter
+  BloomFilter::ConfigurationMismatch = Class.new(ArgumentError)
+  
   class Native < Filter
     attr_reader :bf
-
+    
     def initialize(opts = {})
       @opts = {
         :size    => 100,
@@ -33,9 +35,34 @@ module BloomFilter
 
     def delete(key); @bf.delete(key); end
     def clear; @bf.clear; end
-    def size; @bf.num_set; end
+    def size; @bf.set_bits; end
     def merge!(o); @bf.merge!(o.bf); end
 
+    # Returns the number of bits that are set to 1 in the filter.
+    def set_bits
+      @bf.set_bits
+    end
+
+    # Computes the intersection of two Bloom filters.
+    # It assumes that both filters have the same size -
+    # if this is not true +BloomFilter::ConfigurationMismatch+ is raised.
+    def &(o)
+      raise BloomFilter::ConfigurationMismatch.new unless same_parameters?(o)
+      result = self.class.new
+      result.instance_variable_set(:@bf,@bf.&(o.bf))
+      result
+    end
+
+    # Computes the union of two Bloom filters.
+    # It assumes that both filters have the same size -
+    # if this is not true +BloomFilter::ConfigurationMismatch+ is raised.
+    def |(o)
+      raise BloomFilter::ConfigurationMismatch.new unless same_parameters?(o)
+      result = self.class.new
+      result.instance_variable_set(:@bf,@bf.|(o.bf))
+      result
+    end
+
     def bitmap
       @bf.bitmap
     end
@@ -43,8 +70,8 @@ module BloomFilter
     def marshal_load(ary)
       opts, bitmap = *ary
 
-      @bf = Native.new(opts)
-      @bf.bf.load(bitmap) if !bitmap.nil?
+      initialize(opts)
+      @bf.load(bitmap) if !bitmap.nil?
     end
 
     def marshal_dump
@@ -61,5 +88,14 @@ module BloomFilter
       end
     end
 
+    protected
+
+    # Returns true if parameters of the +o+ther filter are
+    # the same.
+    def same_parameters?(o)
+      @bf.m == o.bf.m && @bf.k == o.bf.k &&
+        @bf.s == o.bf.s && @bf.b == o.bf.b
+    end
+
   end
 end

data/lib/bloomfilter/redis.rb

@@ -10,7 +10,7 @@ module BloomFilter
         :eager  => false,
         :server => {}
       }.merge opts
-      @db = ::Redis.new(@opts[:server])
+      @db = @opts.delete(:db) || ::Redis.new(@opts[:server])
 
       if @opts[:eager]
         @db.setbit @opts[:namespace], @opts[:size]+1, 1
@@ -45,11 +45,7 @@ module BloomFilter
     alias :key? :include?
 
     def delete(key)
-      @db.pipelined do
-        indexes_for(key) do |idx|
-          @db.setbit @opts[:namespace], idx, 0
-        end
-      end
+      warn "Deletes are disabled on non-counting filter, see: https://github.com/igrigorik/bloomfilter-rb/issues/37. This method will be deprecated in a future release."
     end
 
     def clear

data/lib/bloomfilter/version.rb

@@ -1,3 +1,3 @@
 module BloomFilter
-  VERSION = "2.1.1"
+  VERSION = "2.1.2"
 end

data/spec/counting_redis_spec.rb

@@ -1,52 +1,61 @@
 require 'helper'
 
 describe BloomFilter::CountingRedis do
-  include BloomFilter
 
-  context "use Redis for storage" do
-    it "should store data in Redis" do
-      bf = CountingRedis.new
+  it "should connect to remote redis server" do
+    expect { BloomFilter::CountingRedis.new }.not_to raise_error
+  end
 
-      bf.insert(:abcd)
-      bf.insert('test')
-      bf.include?('test').should be_true
-      bf.key?('test').should be_true
+  it "should allow redis client instance to be passed in" do
+    redis_client = double("Redis")
+    bf = BloomFilter::CountingRedis.new(:db => redis_client)
+    expect(bf.instance_variable_get(:@db)).to be(redis_client)
+  end
 
-      bf.include?('test', 'test2').should be_false
-      bf.include?('test', 'abcd').should be_true
+  context "a default CountingRedis instance" do
+    before do
+      # clear all redis databases
+      subject.instance_variable_get(:@db).flushall
     end
+    
+    it "should store data in Redis" do
+      subject.insert(:abcd)
+      subject.insert('test')
 
-    it "should accept a TTL value for a key" do
-      bf = CountingRedis.new(:ttl => 1)
-
-      bf.insert('test')
-      bf.include?('test').should be_true
+      expect(subject.include?('test')).to be true
+      expect(subject.key?('test')).to be true
 
-      sleep(2)
-      bf.include?('test').should be_false
+      expect(subject.include?('test', 'test2')).to be false
+      expect(subject.include?('test', 'abcd')).to be true
+      expect(subject.include?('test', 'abcd', 'nada')).to be false
     end
 
     it "should delete keys from Redis" do
-      bf = CountingRedis.new
-
-      bf.insert('test')
-      bf.include?('test').should be_true
+      subject.insert('test')
+      expect(subject.include?('test')).to be true
 
-      bf.delete('test')
-      bf.include?('test').should be_false
+      subject.delete('test')
+      expect(subject.include?('test')).to be false
     end
 
     it "should output current stats" do
-      bf = CountingRedis.new
-      bf.clear
-
-      bf.insert('test')
-      bf.size.should == 4
-      lambda { bf.stats }.should_not raise_error
+      subject.insert('test')
+      expect(subject.size).to eq(4)
+      expect { subject.stats }.not_to raise_error
     end
+  end
 
-    it "should connect to remote redis server" do
-      lambda { CountingRedis.new }.should_not raise_error
+  context "a TTL 1 instance" do
+    subject { BloomFilter::CountingRedis.new(:ttl => 1) }
+
+    it "should accept a TTL value for a key" do
+      subject.instance_variable_get(:@db).flushall
+
+      subject.insert('test')
+      expect(subject.include?('test')).to be true
+
+      sleep(2)
+      expect(subject.include?('test')).to be false
     end
   end
-end
+end

data/spec/helper.rb

@@ -1,2 +1,3 @@
+require 'redis'
 require 'bundler/setup'
-require 'bloomfilter-rb'
+require 'bloomfilter-rb'

data/spec/native_spec.rb

@@ -1,58 +1,116 @@
 require 'helper'
 
 describe BloomFilter::Native do
-  include BloomFilter
 
   it "should clear" do
-    bf = Native.new(:size => 100, :hashes => 2, :seed => 1, :bucket => 3, :raise => false)
+    bf = BloomFilter::Native.new(:size => 100, :hashes => 2, :seed => 1, :bucket => 3, :raise => false)
     bf.insert("test")
-    bf.include?("test").should be_true
+    expect(bf.include?("test")).to be true
     bf.clear
-    bf.include?("test").should be_false
+    expect(bf.include?("test")).to be false
   end
 
   it "should merge" do
-    bf1 = Native.new(:size => 100, :hashes => 2, :seed => 1, :bucket => 3, :raise => false)
-    bf2 = Native.new(:size => 100, :hashes => 2, :seed => 1, :bucket => 3, :raise => false)
+    bf1 = BloomFilter::Native.new(:size => 100, :hashes => 2, :seed => 1, :bucket => 3, :raise => false)
+    bf2 = BloomFilter::Native.new(:size => 100, :hashes => 2, :seed => 1, :bucket => 3, :raise => false)
     bf2.insert("test")
-    bf1.include?("test").should be_false
+    expect(bf1.include?("test")).to be false
     bf1.merge!(bf2)
-    bf1.include?("test").should be_true
-    bf2.include?("test").should be_true
+    expect(bf1.include?("test")).to be true
+    expect(bf2.include?("test")).to be true
   end
 
   context "behave like a bloomfilter" do
-    it "should test set memerbship" do
-      bf = Native.new(:size => 100, :hashes => 2, :seed => 1, :bucket => 3, :raise => false)
+    it "should test set membership" do
+      bf = BloomFilter::Native.new(:size => 100, :hashes => 2, :seed => 1, :bucket => 3, :raise => false)
       bf.insert("test")
       bf.insert("test1")
 
-      bf.include?("test").should be_true
-      bf.include?("abcd").should be_false
-      bf.include?("test", "test1").should be_true
+      expect(bf.include?("test")).to be true
+      expect(bf.include?("abcd")).to be false
+      expect(bf.include?("test", "test1")).to be true
     end
 
     it "should work with any object's to_s" do
-      bf = Native.new
-      bf.insert(:test)
-      bf.insert(:test1)
-      bf.insert(12345)
-
-      bf.include?("test").should be_true
-      bf.include?("abcd").should be_false
-      bf.include?("test", "test1", '12345').should be_true
+      subject.insert(:test)
+      subject.insert(:test1)
+      subject.insert(12345)
+
+      expect(subject.include?("test")).to be true
+      expect(subject.include?("abcd")).to be false
+      expect(subject.include?("test", "test1", '12345')).to be true
+    end
+
+    it "should return the number of bits set to 1" do
+      bf = BloomFilter::Native.new(:hashes => 4)
+      bf.insert("test")
+      expect(bf.set_bits).to be == 4
+      bf.delete("test")
+      expect(bf.set_bits).to be == 0
+
+      bf = BloomFilter::Native.new(:hashes => 1)
+      bf.insert("test")
+      expect(bf.set_bits).to be == 1
+    end
+
+    it "should return intersection with other filter" do
+      bf1 = BloomFilter::Native.new(:seed => 1)
+      bf1.insert("test")
+      bf1.insert("test1")
+
+      bf2 = BloomFilter::Native.new(:seed => 1)
+      bf2.insert("test")
+      bf2.insert("test2")
+
+      bf3 = bf1 & bf2
+      expect(bf3.include?("test")).to be true
+      expect(bf3.include?("test1")).to be false
+      expect(bf3.include?("test2")).to be false
+    end
+
+    it "should raise an exception when intersection is to be computed for incompatible filters" do
+      bf1 = BloomFilter::Native.new(:size => 10)
+      bf1.insert("test")
+
+      bf2 = BloomFilter::Native.new(:size => 20)
+      bf2.insert("test")
+
+      expect { bf1 & bf2 }.to raise_error(BloomFilter::ConfigurationMismatch)
+    end
+
+    it "should return union with other filter" do
+      bf1 = BloomFilter::Native.new(:seed => 1)
+      bf1.insert("test")
+      bf1.insert("test1")
+
+      bf2 = BloomFilter::Native.new(:seed => 1)
+      bf2.insert("test")
+      bf2.insert("test2")
+
+      bf3 = bf1 | bf2
+      expect(bf3.include?("test")).to be true
+      expect(bf3.include?("test1")).to be true
+      expect(bf3.include?("test2")).to be true
+    end
+
+    it "should raise an exception when union is to be computed for incompatible filters" do
+      bf1 = BloomFilter::Native.new(:size => 10)
+      bf1.insert("test")
+
+      bf2 = BloomFilter::Native.new(:size => 20)
+      bf2.insert("test")
+
+      expect {bf1 | bf2}.to raise_error(BloomFilter::ConfigurationMismatch)
     end
   end
 
   context "behave like counting bloom filter" do
     it "should delete / decrement keys" do
-      bf = Native.new
+      subject.insert("test")
+      expect(subject.include?("test")).to be true
 
-      bf.insert("test")
-      bf.include?("test").should be_true
-
-      bf.delete("test")
-      bf.include?("test").should be_false
+      subject.delete("test")
+      expect(subject.include?("test")).to be false
     end
   end
 
@@ -60,20 +118,32 @@ describe BloomFilter::Native do
     after(:each) { File.unlink('bf.out') }
 
     it "should marshall the bloomfilter" do
-      bf = Native.new
-      lambda { bf.save('bf.out') }.should_not raise_error
+      bf = BloomFilter::Native.new
+      expect { bf.save('bf.out') }.not_to raise_error
     end
 
     it "should load marshalled bloomfilter" do
-      bf = Native.new
-      bf.insert('foo')
-      bf.insert('bar')
-      bf.save('bf.out')
-
-      bf = Native.load('bf.out')
-      bf.include?('foo').should be_true
-      bf.include?('bar').should be_true
-      bf.include?('baz').should be_false
+      subject.insert('foo')
+      subject.insert('bar')
+      subject.save('bf.out')
+
+      bf2 = BloomFilter::Native.load('bf.out')
+      expect(bf2.include?('foo')).to be true
+      expect(bf2.include?('bar')).to be true
+      expect(bf2.include?('baz')).to be false
+
+      expect(subject.send(:same_parameters?, bf2)).to be true
     end
+
+    it "should serialize to a file size proporational its bucket size" do
+      fs_size = 0
+      8.times do |i|
+        bf = BloomFilter::Native.new(size: 10_000, bucket: i+1)
+        bf.save('bf.out')
+        prev_size, fs_size = fs_size, File.size('bf.out')
+        expect(prev_size).to be < fs_size
+      end
+    end
+
   end
-end
+end

data/spec/redis_spec.rb

@@ -1,54 +1,62 @@
 require 'helper'
 
 describe BloomFilter::Redis do
-  include BloomFilter
 
   context "use Redis bitstring for storage" do
-    let(:bf) { Redis.new }
+    before do
+      # clear all redis databases
+      subject.instance_variable_get(:@db).flushall
+    end
 
     it "should store data in Redis" do
-      bf.insert(:abcd)
-      bf.insert('test')
-      bf.include?('test').should be_true
-      bf.key?('test').should be_true
+      subject.insert(:abcd)
+      subject.insert('test')
+      expect(subject.include?('test')).to be true
+      expect(subject.key?('test')).to be true
 
-      bf.include?('test', 'test2').should be_false
-      bf.include?('test', 'abcd').should be_true
+      expect(subject.include?('test', 'test2')).to be false
+      expect(subject.include?('test', 'abcd')).to be true
     end
 
-    it "should delete keys from Redis" do
-      bf.insert('test')
-      bf.include?('test').should be_true
+    it "should not delete keys from Redis" do
+      subject.insert('test')
+      expect(subject.include?('test')).to be true
 
-      bf.delete('test')
-      bf.include?('test').should be_false
+      subject.delete('test')
+      expect(subject.include?('test')).to be true
     end
 
     it "should clear Redis filter" do
-      bf.insert('test')
-      bf.include?('test').should be_true
+      subject.insert('test')
+      expect(subject.include?('test')).to be true
 
-      bf.clear
-      bf.include?('test').should be_false
+      subject.clear
+      expect(subject.include?('test')).to be false
     end
 
     it "should output current stats" do
-      bf.clear
-      bf.insert('test')
-      lambda { bf.stats }.should_not raise_error
+      subject.clear
+      subject.insert('test')
+      expect { subject.stats }.not_to raise_error
     end
 
     it "should connect to remote redis server" do
-      lambda { Redis.new }.should_not raise_error
+      expect { BloomFilter::Redis.new }.not_to raise_error
+    end
+
+    it "should allow redis client instance to be passed in" do
+      redis_client = double("Redis")
+      bf = BloomFilter::Redis.new(:db => redis_client)
+      expect(bf.instance_variable_get(:@db)).to be redis_client
     end
 
     it "should allow namespaced BloomFilters" do
-      bf1 = Redis.new(:namespace => :a)
-      bf2 = Redis.new(:namespace => :b)
+      bf1 = BloomFilter::Redis.new(:namespace => :a)
+      bf2 = BloomFilter::Redis.new(:namespace => :b)
 
       bf1.insert('test')
-      bf1.include?('test').should be_true
-      bf2.include?('test').should be_false
+      expect(bf1.include?('test')).to be true
+      expect(bf2.include?('test')).to be false
     end
   end
-end
+end

metadata

@@ -1,66 +1,83 @@
---- !ruby/object:Gem::Specification 
+--- !ruby/object:Gem::Specification
 name: bloomfilter-rb
-version: !ruby/object:Gem::Version 
-  prerelease: 
-  version: 2.1.1
+version: !ruby/object:Gem::Version
+  version: 2.1.2
 platform: ruby
-authors: 
+authors:
 - Ilya Grigorik
 - Tatsuya Mori
-autorequire: 
+autorequire:
 bindir: bin
 cert_chain: []
-
-date: 2011-03-31 00:00:00 -04:00
-default_executable: 
-dependencies: 
-- !ruby/object:Gem::Dependency 
+date: 2021-07-10 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
   name: redis
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
   prerelease: false
-  requirement: &id001 !ruby/object:Gem::Requirement 
-    none: false
-    requirements: 
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
     - - ">="
-      - !ruby/object:Gem::Version 
-        version: "0"
-  type: :runtime
-  version_requirements: *id001
-- !ruby/object:Gem::Dependency 
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
   name: rspec
-  prerelease: false
-  requirement: &id002 !ruby/object:Gem::Requirement 
-    none: false
-    requirements: 
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
     - - ">="
-      - !ruby/object:Gem::Version 
-        version: "0"
+      - !ruby/object:Gem::Version
+        version: '3'
   type: :development
-  version_requirements: *id002
-- !ruby/object:Gem::Dependency 
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '3'
+- !ruby/object:Gem::Dependency
   name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
   prerelease: false
-  requirement: &id003 !ruby/object:Gem::Requirement 
-    none: false
-    requirements: 
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: rake-compiler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
     - - ">="
-      - !ruby/object:Gem::Version 
-        version: "0"
+      - !ruby/object:Gem::Version
+        version: '0'
   type: :development
-  version_requirements: *id003
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 description: Counting Bloom Filter implemented in Ruby
-email: 
+email:
 - ilya@igvita.com
 executables: []
-
-extensions: 
+extensions:
 - ext/cbloomfilter/extconf.rb
 extra_rdoc_files: []
-
-files: 
-- .gitignore
-- .rspec
+files:
+- ".gitignore"
+- ".rspec"
 - Gemfile
-- Gemfile.lock
 - README.md
 - Rakefile
 - benchmark/redis-bm.rb
@@ -74,6 +91,7 @@ files:
 - ext/cbloomfilter/crc32.h
 - ext/cbloomfilter/extconf.rb
 - lib/bloomfilter-rb.rb
+- lib/bloomfilter.rb
 - lib/bloomfilter/counting_redis.rb
 - lib/bloomfilter/filter.rb
 - lib/bloomfilter/native.rb
@@ -83,35 +101,29 @@ files:
 - spec/helper.rb
 - spec/native_spec.rb
 - spec/redis_spec.rb
-has_rdoc: true
-homepage: http://github.com/igrigorik/bloomfilter
+homepage: http://github.com/igrigorik/bloomfilter-rb
 licenses: []
-
-post_install_message: 
+metadata: {}
+post_install_message:
 rdoc_options: []
-
-require_paths: 
+require_paths:
 - lib
-required_ruby_version: !ruby/object:Gem::Requirement 
-  none: false
-  requirements: 
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
   - - ">="
-    - !ruby/object:Gem::Version 
-      version: "0"
-required_rubygems_version: !ruby/object:Gem::Requirement 
-  none: false
-  requirements: 
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
   - - ">="
-    - !ruby/object:Gem::Version 
-      version: "0"
+    - !ruby/object:Gem::Version
+      version: '0'
 requirements: []
-
-rubyforge_project: bloomfilter-rb
-rubygems_version: 1.6.2
-signing_key: 
-specification_version: 3
+rubygems_version: 3.0.3
+signing_key:
+specification_version: 4
 summary: Counting Bloom Filter implemented in Ruby
-test_files: 
+test_files:
 - spec/counting_redis_spec.rb
 - spec/helper.rb
 - spec/native_spec.rb

data/Gemfile.lock

@@ -1,28 +0,0 @@
-PATH
-  remote: .
-  specs:
-    bloomfilter-rb (2.0.0)
-      redis
-
-GEM
-  remote: http://rubygems.org/
-  specs:
-    diff-lcs (1.1.2)
-    rake (0.8.7)
-    redis (2.2.0)
-    rspec (2.5.0)
-      rspec-core (~> 2.5.0)
-      rspec-expectations (~> 2.5.0)
-      rspec-mocks (~> 2.5.0)
-    rspec-core (2.5.1)
-    rspec-expectations (2.5.0)
-      diff-lcs (~> 1.1.2)
-    rspec-mocks (2.5.0)
-
-PLATFORMS
-  ruby
-
-DEPENDENCIES
-  bloomfilter-rb!
-  rake
-  rspec