by using a series of tables and mapping parent IDs to child IDs in a many to many relationship.
And then using JOIN queries pull back the children of a parent ID.
One schema could have all the component items in a monolithic table and then use a table that mapped component ID to component ID with columns of parentID and childID.
another schema could have tables of components that can only be parents with a table of child components
You also will have to consider the child components that can also have children and so on.
There are a few ways of planning it.
one would be to start with the individual components and design the DB schemas backwards.
which should lead to a more efficient structure
downside would be more complex queries.
at the opposite end you could start off with the whole and break it into assemblies, then sub assemblies and so on.
This would very likely lead to less efficient tables with more redundant data. The queries would be less complex but there would be more round trips to the DB.
All in all not a trivial task.
Hopefully you are not trying to do this in Access 
__________________
Chris. ->> Links are advertising NOT optimising!! <<-
A foolish consistency is the hobgoblin of little minds
Thought for today:- I SEO the only industry where all the cowboys are Indians?
|