CSCI 311 -- Program One - Spring 2009

AVL Trees Plus

Dr. Melody Stapleton   

PLEASE NOTE: SINCE YOUR TA HAD ALREADY WRITTEN HIS SOLUTION AND GENERATED TEST CASES WITH A STANDING FOR “INSERT” AND “I” STANDING FOR INORDER TRAVERSAL, I AM SWITCHING MY PROGRAM INPUT FILE CODE TO MATCH HIS.  SORRY FOR ANY INCONVENIENCE, BUT THIS SHOULD BE A SIMPLE CHANGE IN YOUR CODE!

You are to implement the operations of an AVL tree using C++ classes.  You are to structure your main program to accept and interpret file input with the following format.  Listed below are the possibilities for the different categories of commands:

A    INTEGER_KEY    for Insert the integer to the AVL tree and do any necessary rebalancing

D    INTEGER_KEY    for Delete the integer from the AVL tree and do any necessary rebalancing

S    INTEGER_KEY    for Search the AVL tree for the integer key and for a successful search, return the pointer to the node in the tree

                                     where the key is stored.   For unsuccessful search, return a NULL pointer

P                                    for Preorder Traversal

I                                    for Inorder Traversal

R                                    for Postorder Traversal

B                                    for a report that gives the number of subtrees that are left heavy, right heavy and balanced (see more explanation below).  You will also print out                                                                       within this report the current cumulative statistic for deletions, as described below.

Special criteria: The commands may come in any order.  Although insertions to the tree are relatively straightforward, deletions of parent nodes that have two children may result in ambiguities.  You are to write two different versions of a portion of your deletion function, each version reflecting a different approach to deletion and you are going to gather statistics to measure which version is more effective at balancing the tree with the least amount of “effort”.

Deletion algorithm cases:

1.     If a parent node having two children is deleted, it is to be replaced by the deeper (as in depth in the tree) of the inorder predecessor versus the inorder successor.  If the inorder predecessor and the successor are the same depth, use the inorder predecessor.

2.      If a parent node having two children is deleted, it is to be replaced by the more shallow (as in depth in the tree) of the inorder predecessor versus the inorder successor.  If the inorder predecessor and the successor are the same depth, use the inorder predecessor.

With each deletion within each version you will collect the statistic for the number of rotations that were necessary to put the tree back into balanced form as an AVL tree.  You will accumulate this statistic over the entire run of the program with that version of deletion with each run against an input file.

For the command "B" in the input file, you will print out a report.  In this report you will consider each node in the tree.  For each node in the tree you will assess whether the the subtree rooted at that node is left heavy, right heavy, or balanced.  Each time the command "B" is issued you will generate a report of how many subtrees total in the current tree are left heavy, right heavy or balanced.  Your instructor will provide numerous examples.

Send your output for this program to a file.  For inserts it is an error to insert a duplicate value into the tree, be sure to notify the user of such an attempt.  Tell the user if an operation is successful, as well.  Be sure to give appropriate 'error' messages back to the user of your program.  For example, if asked to Delete a value from the tree that was not found in the tree, you should reply that the value was not found in the tree and therefore, not deleted.  For the Search operation, if the item that is being sought is found, print out a message that says you found it and what its value is.  If the item is not found, print out a message to that effect.  Of course, output the results of a traversal to this file, as well.

A good strategy for developing your program: Write pseudocode for your program that includes:

  1. A complete AVL tree class declaration and any other data structure declarations - this to become your .h file
  2. pseudocode for your main program - this to become your avlmain.cpp file
  3. pseudocode for each function definition - this to become your documentation for your avl.cpp file

Take this link for a sample input file to run your program against.  Your TA will provide final input files which are test cases to run your program against.  Note: Your program should be *bullet-proof*, debugged and able to run correctly against *any file*!