Tuesday, June 17, 2003, 8:30:04 PM, you wrote:
AC> I'm still interested to know how the "build a key out of a hash of the
AC> node and its depth in the tree" (or whatever it is... I delete these
AC> posts as I read them, so forget the details) copes with moving a node to
AC> elsewhere in the tree?
Pretty easily...
Say I have a model like this (which I do use), similar to Matt's abc
type tree except it allows 999 nodes instead of 26 at the expense of
larger keys.
Name Serial
Cars 001
Ford 001001
Jaguar 001001001
Volvo 001001002
Lincoln 001001003
GM 001002
Buick 001002001
Cadillac 001002003
Escalade 001002003001
DeVille 001002003002
Say GM sells Cadillac to Ford, I need to move the 001002003* tree
below 001001. So it's one simple UPDATE statement using the LIKE
keyword to update everything that begins with 001002003 to 001001 +
(next unused tree node). This model does end up updating a lot of rows
with very large trees, so I have my doubts as to how far it would
scale, but it's extremely simple and fast enough for every tree I've
done.
The actual SQL would end up looking something like this...
UPDATE Categories
SET Serial = '001001004' + SubString(Serial, (Len('001001004') + 1), Len(Serial) )
WHERE Category_ID LIKE '001002003%'
--
mailto:[EMAIL PROTECTED]
----------------------------------------------------------
You are subscribed to cfcdev. To unsubscribe, send an email
to [EMAIL PROTECTED] with the word 'unsubscribe cfcdev'
in the message of the email.
CFCDev is run by CFCZone (www.cfczone.org) and supported
by Mindtool, Corporation (www.mindtool.com).
