qbloom_filter 0.1.0 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: c8384171d7c3a67af223b2425c08ec9fa8ccf4c784d658f5951b37642708a384
-  data.tar.gz: a50a3260fcd55e0284e85a1b78f4f34dce3f5ff9bf4a56add2cdebc3aecd1d07
+  metadata.gz: 1a5c8c820e92cec2133acd196b0b42ff9713b5b3446226db27dfa1d8db7ba28f
+  data.tar.gz: f2074640b8228181c145500e39cba8467d7fd7d743c94fecc11c8aefc5a013ba
 SHA512:
-  metadata.gz: 9fe8aeae6469952e99be203cbe4c12cf03317b7eb9499a2bde069cf11c980d3f7746d29fcb2853451ae82ff5d244f87491103bd15d9125e445f44bb63bbabbfb
-  data.tar.gz: 4a43600e93388a498797dcb6706801b105c470602e0449da6bd37faf885ea660f17c3a9a5fcbfede9f1860af9c0cc8fc47a960073ba893e523fca13d4a8e6590
+  metadata.gz: c76fb3bea4b43e1ede2ddfcc6570ace55bd7e3e6a1e5fd3f3b6cb1845f082b8f55716345d40b7202c9d6798e61b84619d0e83a2158c050b966be3451cec83a73
+  data.tar.gz: bb76e4786858a0b0c86fecbd49e6971cccad8b34aaeca4d802940184ca6b38343b39df76da9bdb37fbfe4e9e478260874b400fc2f1ad31b0c3b015cbd3fbc829

data/README.md

@@ -30,7 +30,7 @@ And two parameters can be used to describe the bloom filter:
 bloom_filter = BloomFilter::Filter.new(1000, 0.001)
 ```
 
-#### API
+#### Methods
 __add(value)__ - add item into filter
 
 __includes?(value)__ - check if filter includes the value
@@ -39,6 +39,16 @@ __contains?(value)__ - alias of __includes?(value)__
 
 __count__ - returns number of inserted items
 
+##### Methods for union and intersection several bloom filters
+
+__bit_size__ - returns number of bits in the bit array
+
+__get_bit(position)__ - returns value of a bit(true/false) in the bit array, rises an error if position is out of range of the bit array
+
+__set_bit(position)__ - set a bit to TRUE in the bit array, rises an error if position is out of range of the bit array
+
+__clear_bit(position)__ - set a bit to FALSE in the bit array, rises an error if position is out of range of the bit array
+
 ## Contributing
 
 Bug reports and pull requests are welcome on GitHub at https://github.com/superedriver/bloom_filter

data/lib/bloom_filter.rb

@@ -5,6 +5,8 @@ require 'digest/md5'
 module BloomFilter
   PRIME = 100_000_000_003
   MAX_HASH_PARAM = 1000
+  OUT_OF_RANGE = "Position is out of range"
+
   class Filter
     attr_reader :count
 
@@ -19,20 +21,16 @@ module BloomFilter
 
       #number of hash functions that minimizes the probability of false positives
       @k = (Math.log(2) * (@m / capacity)).ceil
-
-      # a, b params for hash functions
-      @hash_params = []
-      @k.times { @hash_params.push([rand(1000), rand(1000)]) }
     end
 
     def add(value)
       x = get_hash(value)
       was_inserted = true
       @k.times do |i|
-        a, b = @hash_params[i]
+        a, b = get_hash_params(i)
         position = get_position(a, b, x)
-        was_inserted = false unless @bitset[position]
-        @bitset[position] = true
+        was_inserted = false unless self.get_bit(position)
+        self.set_bit(position)
       end
       @count += 1 unless was_inserted
       value
@@ -42,14 +40,30 @@ module BloomFilter
       x = get_hash(value)
       result = true
       @k.times do |i|
-        a, b = @hash_params[i]
-        result = false unless @bitset[get_position(a, b, x)]
+        a, b = get_hash_params(i)
+        result = false unless self.get_bit(get_position(a, b, x))
       end
 
       result
     end
     alias :includes? :contains?
 
+    def get_bit(position)
+      @bitset[position] if is_valid_position(position)
+    end
+
+    def set_bit(position)
+      @bitset[position] = true if is_valid_position(position)
+    end
+
+    def clear_bit(position)
+      @bitset[position] = false if is_valid_position(position)
+    end
+
+    def bit_size
+      @m
+    end
+
     private
 
     def get_position(a, b, val)
@@ -59,5 +73,14 @@ module BloomFilter
     def get_hash(value)
       Digest::MD5.hexdigest(value.to_s).to_i(16)
     end
+
+    def is_valid_position(position)
+      raise OUT_OF_RANGE if position >= @m
+      true
+    end
+
+    def get_hash_params(i)
+      return 2*i + 1, 2*i + 2
+    end
   end
 end

data/lib/bloom_filter/version.rb

@@ -1,3 +1,3 @@
 module BloomFilter
-  VERSION = "0.1.0"
+  VERSION = "0.2.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: qbloom_filter
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.2.0
 platform: ruby
 authors:
 - qaz
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2020-09-27 00:00:00.000000000 Z
+date: 2020-09-28 00:00:00.000000000 Z
 dependencies: []
 description: Simple Bloom Filter
 email: