Viewing 15 posts - 481 through 495 (of 582 total)
If you want to use TSQL only, you would need to use either a linked server or a remote server with a lot of stored procs with many parameters. I...
Tim Wilkinson
"If it doesn't work in practice, you're using the wrong theory"
- Immanuel Kant
Do you allow them to specify parentheses as well? If not, they are likely to get some unexpected results if they mix AND with OR - especially since AND is...
Tim Wilkinson
"If it doesn't work in practice, you're using the wrong theory"
- Immanuel Kant
Answer 1 is correct if you want a necessary (but not sufficient) condition for a BLU - the non clustered index must not be a covering index.
Answer 2 is correct if you...
Tim Wilkinson
"If it doesn't work in practice, you're using the wrong theory"
- Immanuel Kant
These are a bit dirty and make use of implicit conversions but they are probably pretty efficient:
create function dbo.dateportion (@d as datetime)
returns datetime
as
begin
return floor(cast(@d as float))
end
create function dbo.timeportion (@d as...
Tim Wilkinson
"If it doesn't work in practice, you're using the wrong theory"
- Immanuel Kant
So OS files on your database server aren't backed up?
Tim Wilkinson
"If it doesn't work in practice, you're using the wrong theory"
- Immanuel Kant
If you haven't got SQL backups (inc. tran log backups as in previous post) to cover the period immediately before the incident (but surely you must have???), your only hope...
Tim Wilkinson
"If it doesn't work in practice, you're using the wrong theory"
- Immanuel Kant
Previous poster's point is a good one, arising from the problem of searching on a derived value. Given that the data presumably has a high cardinality ( many different values),...
Tim Wilkinson
"If it doesn't work in practice, you're using the wrong theory"
- Immanuel Kant
The syntax quoted from BOL shows that the first three params are mandatory: they aren't contained in square brackets.
Tim Wilkinson
"If it doesn't work in practice, you're using the wrong theory"
- Immanuel Kant
Phone nos should be all different (at least within each group) in example data.
Tim Wilkinson
"If it doesn't work in practice, you're using the wrong theory"
- Immanuel Kant
Repeated data in NoelD's example:
Table_Pos_Dep_Phone
AssocID DepID PosID fk_PhoneID
1D1P1228374 2D1P1232527 3D1P2234222 4D1P2123423 5D1P2123423 6D2P1123423 7D2P1123423 8D2P3123423 9D2P3123423 10D2P3123423
I'd suggest something like:
tblDept: DeptID (PK) Description e.g. 'Finance' ...
tblJob: JobID (PK) Description e.g. 'Clerical Assistant' ...
tblPost: PostID (PK) DeptID (FK) JobID (FK) Description (non-normal data for illustration only) e.g. 'Finance Clerical Asst...
Tim Wilkinson
"If it doesn't work in practice, you're using the wrong theory"
- Immanuel Kant
Viewing 15 posts - 481 through 495 (of 582 total)