Re: pointers on representing tree in db?

   Albert   NULL     1000.00   1   12
   Bert     Albert    900.00   2    3
   Chuck    Albert    900.00   4   11
   Donna    Chuck     800.00   5    6
   Eddie    Chuck     700.00   7    8
   Fred     Chuck     600.00   9   10
  

   which would look like this as a directed graph:   

              Albert (1,12)
              /        \
            /            \
      Bert (2,3)    Chuck (4,11)
                     /    |   \
                   /      |     \
                 /        |       \
               /          |         \
          Donna (5,6)  Eddie (7,8)  Fred (9,10)
  

  This (without the lft and rgt columns) is called the adjacency list   model, after the graph theory technique of the same name; the pairs of   nodes are adjacent to each other. The problem with the adjacency list   model is that the boss and employee columns are the same kind of thing   (i.e. names of personnel), and therefore should be shown in only one   column in a normalized table. To prove that this is not normalized,   assume that "Chuck" changes his name to "Charles"; you have to change   his name in both columns and several places. The defining   characteristic of a normalized table is that you have one fact, one   place, one time.   

  To show a tree as nested sets, replace the nodes with ovals, then nest   subordinate ovals inside each other. The root will be the largest oval   and will contain every other node. The leaf nodes will be the   innermost ovals with nothing else inside them and the nesting will show   the hierarchical relationship. The rgt and lft columns (I cannot use   the reserved words LEFT and RIGHT in SQL) are what shows the nesting.   

  If that mental model does not work, then imagine a little worm crawling   anti-clockwise along the tree. Every time he gets to the left or right   side of a node, he numbers it. The worm stops when he gets all the way   around the tree and back to the top.   

  This is a natural way to model a parts explosion, since a final   assembly is made of physically nested assemblies that final break down   into separate parts.   

  At this point, the boss column is both redundant and denormalized, so   it can be dropped. Also, note that the tree structure can be kept in   one table and all the information about a node can be put in a second   table and they can be joined on employee number for queries.   

  To convert the graph into a nested sets model think of a little worm   crawling along the tree. The worm starts at the top, the root, makes a   complete trip around the tree. When he comes to a node, he puts a   number in the cell on the side that he is visiting and increments his   counter. Each node will get two numbers, one of the right side and one   for the left. Computer Science majors will recognize this as a odified   preorder tree traversal algorithm. Finally, drop the unneeded   ersonnel.boss column which used to represent the edges of a graph.   

  This has some predictable results that we can use for building   queries. The root is always (left = 1, right = 2 * (SELECT COUNT(*)   FROM TreeTable)); leaf nodes always have (left + 1 = right); subtrees   are defined by the BETWEEN predicate; etc. Here are two common queries   which can be used to build others:   

1. An employee and all their Supervisors, no matter how deep the tree.

   SELECT P2.*
     FROM Personnel AS P1, Personnel AS P2     WHERE P1.lft BETWEEN P2.lft AND P2.rgt       AND P1.emp = :myemployee;   

  2. The employee and all subordinates. There is a nice symmetry here.   

   SELECT P2.*
     FROM Personnel AS P1, Personnel AS P2     WHERE P1.lft BETWEEN P2.lft AND P2.rgt       AND P2.emp = :myemployee;   

  3. Add a GROUP BY and aggregate functions to these basic queries and   you have hierarchical reports. For example, the total salaries which   each employee controls:   

   SELECT P2.emp, SUM(P1.salary)
     FROM Personnel AS P1, Personnel AS P2     WHERE P1.lft BETWEEN P2.lft AND P2.rgt     GROUP BY P2.emp;