|
Comments
|
|
very clear and concise
|
|
Thad Moorer on
10/14/2009
Excellent advanced followup to foriegn key initial videos. Also provides some insight into how to use the concept in place of some of the practices, such as coding and triggers, in order to make your solutions more efficient and effective.
|
|
Don Weigend on
10/14/2009
Very good explanation
|
|
|
Clear example showing powerful feature.
|
|
luther smith on
10/14/2009
Good info
|
|
|
Great work in showing a clear alternative to triggers. Constraints might be willing to be used instead of triggers, if for example, you are using replication and don't want to have your own triggers.
|
|
|
excellent, plain, simple example. clearly spoken word - very nice!
|
|
|
Good One
|
|
|
Thank you and keep creating more!
|
|
|
Caution! Use Cascading Updates on FK's at your own risk.. I've seen lazy DBA's use them and really screw table(s) up bad!!!! You better know what you are doing before using the CASCADE option in a foreign key...
|
|
Emine Dekkar on
10/17/2009
did not know about the use of defaultfor all fk while deleting the pk...nice feature for BI ..thx
|
|
Cheryl Pacenka on
10/19/2009
very good
|
|
|
good
|
|
Steve Culshaw on
10/23/2009
Excellent tip
|
|
|
Love this! :)
|
|
Susan Lammi on
10/30/2009
great practical explaination.
|
|
Steve Harris on
11/18/2009
excellent presentation
|
|
|
Excellent content. I've been working with SQL Server for a number of years and never new of this feature.
|
|
|
Excellent .. clear concise explanation. thanks a lot...
|
|
|
Not so good !
|
|
|
Great video!!! Always good to get back to the basics.
|
|
|
It's worth stressing that using the "CASCADE" is a very risky option for the DELETE option. UPDATE is a bit less risky, but make sure you understand the consequences of the CASCADE before implementing it. Frequently, this is done by DBAs who want to maintain data integrity without realizing what that action will do.
When I first began working with an RDBMS, I applied foreign keys and cascading updates/deletes. These problems tend to pop up on data requiring a level of denormalization (such as a customer's address at the time of an order; the customer's address may change in the future and should not be retroactively applied to a previous order).
Cascading an update of an address to a customer's order could be disastrous (or that could be the intended action).
|