Delete Duplicates
by Isai Damier, Android Engineer @ Google

#=======================================================================
# Author: Isai Damier
# Title: Singly Linked List
# Project: geekviewpoint
# Package: datastructure
#
# Description: A LinkedList is a data structure that allows access
#   to a collection of data using pointers/references. While an
#   array can also be defined as above, LinkedLists and arrays differ
#   in how they are stored in memory and in the operations they
#   allow. Unlike an array that must be stored in a block of memory,
#   the nodes of a LinkedList can be stored anywhere because each
#   node has a reference to the node that succeeds it. Because the
#   nodes are stored so loosely, inserting nodes into a LinkedList
#   is easy; whereas in an array, all the succeeding elements must
#   be shifted. Of course, insertion also means changing the size of
#   the array, which means creating the entire array anew.
#
#   Perhaps the greatest beauty of LinkedList is that it allows
#   accessing an entire sequence of nodes using only one variable:
#   a reference to the first node in the sequence.
#
#   Countless operations can be performed on LinkedLists. Following
#   are a few, ranging from the common to the very interesting.
#=======================================================================
  #=====================================================================
  # Time Complexity of Solution:
  #   O(n).
  #
  # Description: Delete all duplicates from this list, such that
  #   iabcdeafgdc becomes iabcdefg
  #
  # Technical Details: The easiest way to remove duplicates from
  #   a collection tends to be with a hashmap. Here is a process:
  #   0] traverse the list, sending data to a linked hashmap, which
  #      discards duplicates. (Do deleteFromHead if acceptable)
  #   1] empty the linked hashmap into a new linked list.
  #   Time Complexity is O(n) Space Complexity is O(n)
  #
  # Note: I code to python 2.7 because that's what most people are
  #  using and it's harmless for fundamentals. But you don't need
  #  ordereddict for your linked-hashmap if you have python 3.x
  #=====================================================================
 
 import collections
class SinglyLinkedList( object ):

  def __init__( self ):
    self.head , self.tail = None, None

  def deleteDuplicates( self ) : # min, max, avg, sum, size
    noDups = collections.OrderedDict()
    t = self.deleteFromHead()
    while ( None != t ): # if(!noDups.containsKey(t.data)): not necessary
      noDups[ t.data] = 1
      t = self.deleteFromHead()

    # reconstruct linkedlist
    for k in noDups.keys():
      self.addToTail( k )

class Node( object ):

  def __init__( self, data, next = None ):
    self.data = data
    self.next = next
import unittest
from algorithms.SinglyLinkedList import SinglyLinkedList
import random

class Test( unittest.TestCase ):
  #=====================================================================
  # Test of deleteDuplicates method, of class SinglyLinkedList.
  #=====================================================================
  def testDeleteDuplicates( self ):
    tape = [2, 19, 11, 5, 17, 11, 1, 17, 2, 11]
    linkedList = SinglyLinkedList()
    for i in range( len( tape ) ):
      linkedList.addToTail( tape[i] )

    exp = [2, 19, 11, 5, 17, 1]
    linkedList.deleteDuplicates()
    self.assertEquals( exp, linkedList.toArray() )