algorithm - The Great Tree list recursion program -

i faced interesting problem called great tree-list problem. problem follows :

in ordered binary tree, each node contains single data element , "small" , "large" pointers sub-trees .all nodes in "small" sub-tree less or equal data in parent node. nodes in "large" sub-tree greater parent node. , circular doubly linked list consist of previous , next pointers.

the problem take ordered binary tree , rearrange internal pointers make circular doubly linked list out of it. "small" pointer should play role of "previous" , "large" pointer should play role of "next". list should arranged nodes in increasing order. have write recursive function & return head pointer new list.

the operation should done in o(n) time.

i understand recursion go down tree, how recursively change small , large sub-trees lists, have append lists parent node.

how should approach problem?.. need direction solve problem!.

the idea create method converts tree node containing subtrees (children nodes) loop. , given node has converted children (e.g. after recursive calls came back), create new loop pointing large pointer (next) of largest node smallest node, , small pointer of smallest node largest node.

may not complete, close this:

tree_node {     small     large }  convert(node){     //base case 1, if leaf node     if node.small == null && node.large == null         return (self, self)      //recursively convert children     if node.small !=null         smallest, larger = convert(node.small)     else         smallest = larger = self      if node.large !=null         smaller, largest = convert(node.large)     else         smaller = largest = self      //wrap ends of chain     largest.large = smallest     smallest.small = largest     //wrap mid part     smaller.small = larger     larger.large = small      //return pointers absolute smallest , largest of subtree     return (smallest, largest) }  //actually doing  convert(tree.root)