On 7/26/07, Sébastien Lorion <[EMAIL PROTECTED]> wrote: > I am in the process of putting the final touch to a tree structure object > model and implementation, but I am still not decided on an important point. > > Considering that operations on the tree are centralized in the Tree class, > ie you write tree.AddNode(child, parent) and not node.AddChild(child), what > would be the better way to store node relationships in your opinion ? > > First option: > > Each node keep a pointer to its parent and store its children locally. > > Second option: > > The tree contains a hash table associating each node to (1) its parent and > (2) its children. (2) could be optional, but I think the convenience and > access speed more than make up for the loss in memory. Each node will then > need a pointer to the tree. > > I already made an implementation for both options and have > drawn some conclusions, but I will not disclose them as to not > influence the comments. > > Also, if you think centralizing operations in the Tree class is not a > good ideas for some reasons, feel free to let me know. > > Thanks for your advise! >
I'll ask the obvious questions - what is it that you're doing that requires the tree? What kind of tree is that? What are the operations you perform on it except adding nodes? Cheers, Stoyan =================================== This list is hosted by DevelopMentor® http://www.develop.com View archives and manage your subscription(s) at http://discuss.develop.com
