Re: Tree (forest) design

From: Robert Stearns <rstearns1241_at_charter.net>
Date: Sun, 29 Feb 2004 12:04:59 -0500
Message-ID: <40421BBB.8080103_at_charter.net>


Neo wrote:

>>I am working on a design problem where I have many (possibly 
>>multinational) customers with tree structured administrative structures 
>>(cust, nation, region, site, building for instance). At the leaf nodes 
>>of the administrative tree, there is a tree organized location system 
>>(building, floor, room, rack, bin for instance). Any customer may be 
>>lacking in any number of levels.

>
>
> Is the below a correct representation of the problem?
> (I realize each levels is optional)
>
> Nation
> Region
> Site
> Bldg (this building is associated with a customer ?)
> Floor
> Room
> Rack
> Bin
> Part
> Etc...
>
> If you would like to post some sample data, I can model it with an
> experimental db. While you won't be able to translates it completely
> to your application, it may be of some value.
>
>
>>There must be the ability to coalesce nodes into one another, as 
>>well as to split a current node into two, moving part of the subtree of 
>>the old node to the new node at any level.

>
>
> The experimental db's GUI should be able do the above in a manner
> similar to cut/paste in File Explorer?

That is the right idea. People (administration, therefore security and privilege) are associated with levels I named bldg and above; inventory is associated with levels bldg and below; In the simplest case a customer might have only the bldg element; everyone and everything assigned to one bldg; a small specialty manufacturer of something like kayaks for instance. All other cases which include the bldg element and consist of contiguous elements on either or both sides will occur. I may not have enough levels for some customer on either side of bldg so the tree must be able to grow deeper, at need, in both directions. This a real requirements set, for a real project, not a school exercise (not even for my PhD dissertation :-). Received on Sun Feb 29 2004 - 18:04:59 CET

Original text of this message