Viewing 15 posts - 751 through 765 (of 1,554 total)
Well, let me disagree with the absolutness of that statement. It depends. You can very well put table access inside UDF's, sometimes it's even the best thing to do - BUT... everyone...
November 16, 2005 at 1:56 am
I'm in a picky mood today
Don't use this old legacy syntax:
update a set a.MyNewColumn = b.MyOldColumn from table1 a ,...
November 16, 2005 at 1:31 am
I don't quite see why that should be a problem... Could you elaborate on your particular situation?
/Kenneth
November 15, 2005 at 8:57 am
No, not directly - identity doesn't take a variable for it's seed.
This doesn't work...
declare @i int
set @i = 100
create table #x( id int not null identity(@i,1), myVal...
November 14, 2005 at 9:16 am
..or use LIKE / NOT LIKE with wildcards if ISNUMERIC() doesn't cut it.
Using wildcards will also provide some control that ISNUMERIC() doesn't give you, should it be needed.
BOL has...
November 14, 2005 at 6:19 am
Before declarative constraints (eg FK's) came to SQL Server, triggers was used to enforce referential integrity. If that is your purpose, then start by looking at CREATE TRIGGER' in BOL...
November 14, 2005 at 6:14 am
This would work well, IF..... all FK's are implemented as declared foreign keys (ie that it actually exists a foreign key constraint).
Unfortunately there's nothing that prevents a column to have...
November 11, 2005 at 7:26 am
David, what I ment is just that some types of functions implements a 'cursory behaviour' under the hood when used in a query. In effect what you get is a...
November 10, 2005 at 8:31 am
Yes, looks like it. The suffix generated looks just like a random number in hex.
Never given a second thought if there was any special function that did that.
November 10, 2005 at 8:26 am
Well, I doubt it.
I'm guessing that you're getting scans all the way, because you have functions on the dates and also negations in the wehere clause, all things that invalidates...
November 10, 2005 at 8:23 am
Have no clue. Why do you want to know? If you need a ready-baked randomfunction, why not NEWID() ..?
/Kenneth
November 10, 2005 at 7:52 am
Yes, you're on the right track. But, as you've noted, it's not 100%.
Though, what you have is about as close you can get - for the rest you need to...
November 10, 2005 at 6:16 am
Your example suggests that the objids are in descending order.
Why don't you use David's example and check with that yourself? After all, you know your data best.
Using a function this...
November 10, 2005 at 6:06 am
Well, here's bad news for you - there is no reliable way inside SQL Server to see all object dependencies with certainty. (sysdepends et all does not work reliably)
If you...
November 10, 2005 at 5:57 am
Not sure I completely grasp your actual problem here, but.. why do you think you need dynamic SQL just to move rows from one table to another..?
/Kenneth
November 10, 2005 at 5:49 am
Viewing 15 posts - 751 through 765 (of 1,554 total)