Help! SQL & recursion.

Hi,

Can anyone help with this little dilemma:

FILE1 <->> FILE2 <->>FILE3 <->> ....

FILE1 has reference ID uniq, autoincrement field, PREVID=0 FILE2 has reference PREVID = ID from FILE1 FILE3 has reference PREVID = ID from FILE2 etc.

In my case FILE1, FILE2,FILE3, etc are referred to the same physical datafile.
Number of levels is not limited or many.

1. I would like to delete all childrens by specifying only ID of their parent node;
2. Copy subtree with ID to another subtree, don't forget that new ID must be autoincremented and UNIQ;
3. I would like to go through the whole tree from ID down;
4. Find the head/s for ID. Head means no parents. As explained before the relation File1 to File2 is 1 to many, but my case it is many to many, so ID record can have many head nodes.

I would really appreciate any help.

Thanks
Evgeny Received on Sat May 30 1998 - 00:00:00 CDT