grosser-algorithms 0.4.0

data/lib/algorithms.rb

@@ -0,0 +1,71 @@
+=begin rdoc
+  The 'Algorithms and Containers' library is an effort to provide a set of commonly used
+  algorithms and containers to Ruby programmers.
+
+  This is a Google Summer of Code 2008 project
+
+  Written by Kanwei Li, mentored by Austin Ziegler
+
+  To avoid typing Algorithms::Containers::xxx to initialize containers, include the Algorithms namespace and Containers module.
+
+    require 'algorithms'
+    include Algorithms
+    include Containers
+
+    tree = RBTreeMap.new
+
+  instead of:
+
+    require 'algorithms'
+
+    tree = Algorithms::Containers::RBTreeMap.new
+
+  Done so far:
+  * Heaps           - Algorithms::Containers::Heap, Algorithms::Containers::MaxHeap, Algorithms::Containers::MinHeap
+  * Priority Queue  - Algorithms::Containers::PriorityQueue
+  * Stack           - Algorithms::Containers::Stack
+  * Queue           - Algorithms::Containers::Queue
+  * Deque           - Algorithms::Containers::Deque, Algorithms::Containers::CDeque (C extension), Algorithms::Containers::RubyDeque
+  * Red-Black Trees - Algorithms::Containers::RBTreeMap, Algorithms::Containers::CRBTreeMap (C extension), Algorithms::Containers::RubyRBTreeMap
+  * Splay Trees     - Algorithms::Containers::SplayTreeMap
+  * Tries           - Algorithms::Containers::Trie
+  * Suffix Array    - Algorithms::Containers::SuffixArray
+  * kd Tree         - Algorithms::Containers::KDTree
+
+  * Search algorithms
+    - Binary Search         - Algorithms::Algorithms::Search.binary_search
+    - Knuth-Morris-Pratt    - Algorithms::Algorithms::Search.kmp_search
+  * Sort algorithms
+    - Bubble sort           - Algorithms::Algorithms::Sort.bubble_sort
+    - Comb sort             - Algorithms::Algorithms::Sort.comb_sort
+    - Selection sort        - Algorithms::Algorithms::Sort.selection_sort
+    - Heapsort              - Algorithms::Algorithms::Sort.heapsort
+    - Insertion sort        - Algorithms::Algorithms::Sort.insertion_sort
+    - Shell sort            - Algorithms::Algorithms::Sort.shell_sort
+    - Quicksort             - Algorithms::Algorithms::Sort.quicksort
+    - Mergesort             - Algorithms::Algorithms::Sort.mergesort
+  * String algorithms
+    - Levenshtein distance  - Algorithms::Algorithms::String.levenshtein_dist
+=end
+
+module Algorithms
+  VERSION = File.read( File.join(File.dirname(__FILE__),'..','VERSION') ).strip
+  
+  module Algorithms; end
+  module Containers; end
+end
+
+require 'algorithms/search'
+require 'algorithms/sort'
+require 'algorithms/string'
+require 'containers/heap'
+require 'containers/stack'
+require 'containers/deque'
+require 'containers/queue'
+require 'containers/priority_queue'
+require 'containers/rb_tree_map'
+require 'containers/splay_tree_map'
+require 'containers/suffix_array'
+require 'containers/trie'
+require 'containers/kd_tree'
+

data/lib/algorithms/search.rb

@@ -0,0 +1,86 @@
+=begin rdoc
+    This module implements search algorithms. Documentation is provided for each algorithm.
+    
+=end
+module Algorithms
+  module Algorithms::Search
+    # Binary Search: This search finds an item in log(n) time provided that the container is already sorted.
+    # The method returns the item if it is found, or nil if it is not. If there are duplicates, the first one
+    # found is returned, and this is not guaranteed to be the smallest or largest item.
+    #
+    # Complexity: O(lg N)
+    # 
+    #   Algorithms::Search.binary_search([1, 2, 3], 1) #=> 1
+    #   Algorithms::Search.binary_search([1, 2, 3], 4) #=> nil
+    def self.binary_search(container, item)
+      return nil if item.nil?
+      low = 0
+      high = container.size - 1
+      while low <= high
+        mid = (low + high) / 2
+        val = container[mid]
+        if val > item
+          high = mid - 1
+        elsif val < item
+          low = mid + 1
+        else
+          return val
+        end
+      end
+      nil
+    end
+  
+    # Knuth-Morris-Pratt Algorithm substring search algorithm: Efficiently finds the starting position of a 
+    # substring in a string. The algorithm calculates the best position to resume searching from if a failure
+    # occurs.
+    # 
+    # The method returns the index of the starting position in the string where the substring is found. If there
+    # is no match, nil is returned.
+    #
+    # Complexity: O(n + k), where n is the length of the string and k is the length of the substring.
+    #
+    #   Algorithms::Search.kmp_search("ABC ABCDAB ABCDABCDABDE", "ABCDABD") #=> 15
+    #   Algorithms::Search.kmp_search("ABC ABCDAB ABCDABCDABDE", "ABCDEF") #=> nil
+    def self.kmp_search(string, substring)
+      return nil if string.nil? or substring.nil?
+    
+      # create failure function table
+      pos = 2
+      cnd = 0
+      failure_table = [-1, 0]
+      while pos < substring.length
+        if substring[pos - 1] == substring[cnd]
+          failure_table[pos] = cnd + 1
+          pos += 1
+          cnd += 1
+        elsif cnd > 0
+          cnd = failure_table[cnd]
+        else
+          failure_table[pos] = 0
+          pos += 1
+        end
+      end
+
+      m = i = 0
+      while m + i < string.length
+        if substring[i] == string[m + i]
+          i += 1
+          return m if i == substring.length
+        else
+          m = m + i - failure_table[i]
+          i = failure_table[i] if i > 0
+        end
+      end
+      return nil
+    end
+  
+    # Allows kmp_search to be called as an instance method in classes that include the Search module.
+    #
+    #   class String; include Algorithms::Search; end
+    #   "ABC ABCDAB ABCDABCDABDE".kmp_search("ABCDABD") #=> 15
+    def kmp_search(substring)
+      Algorithms::Search.kmp_search(self, substring)
+    end
+  
+  end
+end

data/lib/algorithms/sort.rb

@@ -0,0 +1,243 @@
+require 'containers/heap' # for heapsort
+
+=begin rdoc
+    This module implements sorting algorithms. Documentation is provided for each algorithm.
+    
+=end
+module Algorithms
+  module Algorithms
+    module Sort
+  
+      # Bubble sort: A very naive sort that keeps swapping elements until the container is sorted.
+      # Requirements: Needs to be able to compare elements with <=>, and the [] []= methods should
+      # be implemented for the container.
+      # Time Complexity: О(n^2)
+      # Space Complexity: О(n) total, O(1) auxiliary
+      # Stable: Yes
+      # 
+      #   Algorithms::Sort.bubble_sort [5, 4, 3, 1, 2] => [1, 2, 3, 4, 5]
+      def self.bubble_sort(container)
+        loop do
+          swapped = false
+          (container.size-1).times do |i|
+            if (container[i] <=> container[i+1]) == 1
+              container[i], container[i+1] = container[i+1], container[i] # Swap
+              swapped = true
+            end
+          end
+          break unless swapped
+        end
+        container
+      end
+  
+      # Comb sort: A variation on bubble sort that dramatically improves performance.
+      # Source: http://yagni.com/combsort/
+      # Requirements: Needs to be able to compare elements with <=>, and the [] []= methods should
+      # be implemented for the container.
+      # Time Complexity: О(n^2)
+      # Space Complexity: О(n) total, O(1) auxiliary
+      # Stable: Yes
+      # 
+      #   Algorithms::Sort.comb_sort [5, 4, 3, 1, 2] => [1, 2, 3, 4, 5]
+      def self.comb_sort(container)
+        container
+        gap = container.size
+        loop do
+          gap = gap * 10/13
+          gap = 11 if gap == 9 || gap == 10
+          gap = 1 if gap < 1
+          swapped = false
+          (container.size - gap).times do |i|
+            if (container[i] <=> container[i + gap]) == 1
+              container[i], container[i+gap] = container[i+gap], container[i] # Swap
+              swapped = true
+            end
+          end
+          break if !swapped && gap == 1
+        end
+        container
+      end
+  
+      # Selection sort: A naive sort that goes through the container and selects the smallest element,
+      # putting it at the beginning. Repeat until the end is reached.
+      # Requirements: Needs to be able to compare elements with <=>, and the [] []= methods should
+      # be implemented for the container.
+      # Time Complexity: О(n^2)
+      # Space Complexity: О(n) total, O(1) auxiliary
+      # Stable: Yes
+      # 
+      #   Algorithms::Sort.selection_sort [5, 4, 3, 1, 2] => [1, 2, 3, 4, 5]
+      def self.selection_sort(container)
+        0.upto(container.size-1) do |i|
+          min = i
+          (i+1).upto(container.size-1) do |j|
+            min = j if (container[j] <=> container[min]) == -1
+          end
+          container[i], container[min] = container[min], container[i] # Swap
+        end
+        container
+      end
+  
+      # Heap sort: Uses a heap (implemented by the Containers module) to sort the collection.
+      # Requirements: Needs to be able to compare elements with <=>
+      # Time Complexity: О(n^2)
+      # Space Complexity: О(n) total, O(1) auxiliary
+      # Stable: Yes
+      # 
+      #   Algorithms::Sort.heapsort [5, 4, 3, 1, 2] => [1, 2, 3, 4, 5]
+      def self.heapsort(container)
+        heap = ::Algorithms::Containers::Heap.new(container)
+        ary = []
+        ary << heap.pop until heap.empty?
+        ary
+      end
+  
+      # Insertion sort: Elements are inserted sequentially into the right position.
+      # Requirements: Needs to be able to compare elements with <=>, and the [] []= methods should
+      # be implemented for the container.
+      # Time Complexity: О(n^2)
+      # Space Complexity: О(n) total, O(1) auxiliary
+      # Stable: Yes
+      # 
+      #   Algorithms::Sort.insertion_sort [5, 4, 3, 1, 2] => [1, 2, 3, 4, 5]
+      def self.insertion_sort(container)
+        return container if container.size < 2
+        (1..container.size-1).each do |i|
+          value = container[i]
+          j = i-1
+          while j >= 0 and container[j] > value do
+            container[j+1] = container[j]
+            j = j-1
+          end
+          container[j+1] = value
+        end
+        container
+      end
+  
+      # Shell sort: Similar approach as insertion sort but slightly better.
+      # Requirements: Needs to be able to compare elements with <=>, and the [] []= methods should
+      # be implemented for the container.
+      # Time Complexity: О(n^2)
+      # Space Complexity: О(n) total, O(1) auxiliary
+      # Stable: Yes
+      # 
+      #   Algorithms::Sort.shell_sort [5, 4, 3, 1, 2] => [1, 2, 3, 4, 5]
+      def self.shell_sort(container)
+        increment = container.size/2
+        while increment > 0 do
+          (increment..container.size-1).each do |i|
+            temp = container[i]
+            j = i
+            while j >= increment && container[j - increment] > temp do
+              container[j] = container[j-increment]
+              j -= increment
+            end
+            container[j] = temp
+          end
+          increment = (increment == 2 ? 1 : (increment / 2.2).round)
+        end
+        container
+      end
+  
+      # Quicksort: A divide-and-conquer sort that recursively partitions a container until it is sorted.
+      # Requirements: Container should implement #pop and include the Enumerable module.
+      # Time Complexity: О(n log n) average, O(n^2) worst-case
+      # Space Complexity: О(n) auxiliary
+      # Stable: No
+      # 
+      #   Algorithms::Sort.quicksort [5, 4, 3, 1, 2] => [1, 2, 3, 4, 5]
+      # def self.quicksort(container)
+      #   return [] if container.empty?
+      #   
+      #   x, *xs = container
+      # 
+      #   quicksort(xs.select { |i| i <  x }) + [x] + quicksort(xs.select { |i| i >= x })
+      # end
+  
+      def self.partition(data, left, right)
+        pivot = data[front]
+        left += 1
+
+        while left <= right do
+          if data[frontUnknown] < pivot
+            back += 1
+            data[frontUnknown], data[back] = data[back], data[frontUnknown] # Swap
+          end
+
+          frontUnknown += 1
+        end
+
+        data[front], data[back] = data[back], data[front] # Swap
+        back
+      end
+
+
+      # def self.quicksort(container, left = 0, right = container.size - 1)
+      #   if left < right 
+      #     middle = partition(container, left, right)
+      #     quicksort(container, left, middle - 1)
+      #     quicksort(container, middle + 1, right)
+      #   end
+      # end
+  
+      def self.quicksort(container)
+        bottom, top = [], []
+        top[0] = 0
+        bottom[0] = container.size
+        i = 0
+        while i >= 0 do
+          l = top[i]
+          r = bottom[i] - 1;
+          if l < r
+            pivot = container[l]
+            while l < r do
+              r -= 1 while (container[r] >= pivot  && l < r)
+              if (l < r)
+                container[l] = container[r]
+                l += 1
+              end
+              l += 1 while (container[l] <= pivot  && l < r)
+              if (l < r)
+                container[r] = container[l]
+                r -= 1
+              end
+            end
+            container[l] = pivot
+            top[i+1] = l + 1
+            bottom[i+1] = bottom[i]
+            bottom[i] = l
+            i += 1
+          else
+            i -= 1
+          end
+        end
+        container    
+      end
+
+      # Mergesort: A stable divide-and-conquer sort that sorts small chunks of the container and then merges them together.
+      # Returns an array of the sorted elements.
+      # Requirements: Container should implement []
+      # Time Complexity: О(n log n) average and worst-case
+      # Space Complexity: О(n) auxiliary
+      # Stable: Yes
+      # 
+      #   Algorithms::Sort.mergesort [5, 4, 3, 1, 2] => [1, 2, 3, 4, 5]
+      def self.mergesort(container)
+        return container if container.size <= 1
+        mid   = container.size / 2
+        left  = container[0...mid]
+        right = container[mid...container.size]
+        merge(mergesort(left), mergesort(right))
+      end
+
+      def self.merge(left, right)
+        sorted = []
+        until left.empty? or right.empty?
+          left.first <= right.first ? sorted << left.shift : sorted << right.shift
+        end
+        sorted + left + right
+      end
+
+    end
+  end
+end

data/lib/algorithms/string.rb

@@ -0,0 +1,8 @@
+=begin rdoc
+    This module implements string algorithms. Documentation is provided for each algorithm.
+    
+=end
+
+begin
+  require 'CString'
+end

data/lib/containers/deque.rb

@@ -0,0 +1,176 @@
+=begin rdoc
+    A Deque is a container that allows items to be added and removed from both the front and back,
+    acting as a combination of a Stack and Queue.
+
+    This implementation uses a doubly-linked list, guaranteeing O(1) complexity for all operations.
+=end
+module Algorithms
+  module Containers
+    class RubyDeque
+      include Enumerable
+  
+      Node = Struct.new(:left, :right, :obj)
+  
+      # Create a new Deque. Takes an optional array argument to initialize the Deque.
+      #
+      #   d = Containers::Deque.new([1, 2, 3])
+      #   d.front #=> 1
+      #   d.back #=> 3
+      def initialize(ary=[])
+        @front = nil
+        @back = nil
+        @size = 0
+        ary.to_a.each { |obj| push_back(obj) }
+      end
+  
+      # Returns true if the Deque is empty, false otherwise.
+      def empty?
+        @size == 0
+      end
+  
+      # Removes all the objects in the Deque.
+      def clear
+        @front = @back = nil
+        @size = 0
+      end
+  
+      # Return the number of items in the Deque.
+      #
+      #   d = Containers::Deque.new([1, 2, 3])
+      #   d.size #=> 3
+      def size
+        @size
+      end
+      alias_method :length, :size
+  
+      # Returns the object at the front of the Deque but does not remove it.
+      #
+      #   d = Containers::Deque.new
+      #   d.push_front(1)
+      #   d.push_front(2)
+      #   d.front #=> 2
+      def front
+        @front && @front.obj
+      end
+  
+      # Returns the object at the back of the Deque but does not remove it.
+      #
+      #   d = Containers::Deque.new
+      #   d.push_front(1)
+      #   d.push_front(2)
+      #   d.back #=> 1
+      def back
+        @back && @back.obj
+      end
+  
+      # Adds an object at the front of the Deque.
+      #
+      #   d = Containers::Deque.new([1, 2, 3])
+      #   d.push_front(0)
+      #   d.pop_front #=> 0
+      def push_front(obj)
+        node = Node.new(nil, nil, obj)
+        if @front
+          node.right = @front
+          @front.left = node
+          @front = node
+        else
+          @front = @back = node
+        end
+        @size += 1
+        obj
+      end
+  
+      # Adds an object at the back of the Deque.
+      #
+      #   d = Containers::Deque.new([1, 2, 3])
+      #   d.push_back(4)
+      #   d.pop_back #=> 4
+      def push_back(obj)
+        node = Node.new(nil, nil, obj)
+        if @back
+          node.left = @back
+          @back.right = node
+          @back = node
+        else
+          @front = @back = node
+        end
+        @size += 1
+        obj
+      end
+  
+      # Returns the object at the front of the Deque and removes it.
+      #
+      #   d = Containers::Deque.new
+      #   d.push_front(1)
+      #   d.push_front(2)
+      #   d.pop_front #=> 2
+      #   d.size #=> 1
+      def pop_front
+        return nil unless @front
+        node = @front
+        if @size == 1
+          clear
+          return node.obj
+        else
+          @front.right.left = nil
+          @front = @front.right
+        end
+        @size -= 1
+        node.obj
+      end
+  
+      # Returns the object at the back of the Deque and removes it.
+      #
+      #   d = Containers::Deque.new
+      #   d.push_front(1)
+      #   d.push_front(2)
+      #   d.pop_back #=> 1
+      #   d.size #=> 1
+      def pop_back
+        return nil unless @back
+        node = @back
+        if @size == 1
+          clear
+          return node.obj
+        else
+          @back.left.right = nil
+          @back = @back.left
+        end
+        @size -= 1
+        node.obj
+      end
+  
+      # Iterate over the Deque in FIFO order.
+      def each_forward
+        return unless @front
+        node = @front
+        while node
+          yield node.obj
+          node = node.right
+        end
+      end
+      alias_method :each, :each_forward
+  
+      # Iterate over the Deque in LIFO order.
+      def each_backward
+        return unless @back
+        node = @back
+        while node
+          yield node.obj
+          node = node.left
+        end
+      end
+      alias_method :reverse_each, :each_backward
+
+    end
+  end
+  
+  begin
+    require 'CDeque'
+    Containers::Deque = Containers::CDeque
+  rescue LoadError # C Version could not be found, try ruby version
+    Containers::Deque = Containers::RubyDeque
+  end
+
+end