Viewing 15 posts - 196 through 210 (of 548 total)
Jeff,
When you deal with two different but similar problems, things tend to get confusing.
Problem 1: SQL Server 2000 and the replace function's reliability
Problem 2: SQL Server 2005 and Paul's performance...
November 21, 2009 at 12:38 pm
Jeff,
By the way, once you have attached the collate qualifier to OriginalString the others are not necessary.
In any case, in my opinion this is a bug in SQL Server 2000...
November 21, 2009 at 11:02 am
Jeff Moden (11/21/2009)
Michael Meierruth (11/21/2009)
November 21, 2009 at 8:34 am
Paul White (11/20/2009)
November 21, 2009 at 3:44 am
C# Screw (11/20/2009)
Hicollation details :
Test database default SQL_Latin1_General_CP1_CI_AS
TempDB default Latin1_General_CI_AS
Master database default Latin1_General_CI_AS
Cheers
Screw
Are on SQL Server 2000? If so, Jeff's code should fail with this. Please try Jeff's...
November 20, 2009 at 5:40 am
Finally found a solution to this collate problem.
In Jeff's code if you change the line
LTRIM(RTRIM(OriginalString))
to
LTRIM(RTRIM(OriginalString collate your_database_collation_name))
the code starts to work. You must specify the collation...
November 20, 2009 at 5:32 am
Regarding collation, what I seem to be discovering is that any SQL Server 2000 database whose collation is different from the instance default will cause Jeff's code NOT to work.
I...
November 20, 2009 at 4:35 am
C#Screw
There are more or less 17 different means of 'Nube', one of which is 'new'. So this is what I'm assuming. You might want to look up the others -...
November 20, 2009 at 4:31 am
C#Screw,
I use your method. But knowing that this not always reliable, I generally run a test several times until it 'settles down' to a pretty constant value. IMO it's the...
November 20, 2009 at 1:44 am
Regarding this issue of collate, I have also run into a situation where, based on the database I'm in, Jeff's sample code from the article simply doesn't work, i.e. the...
November 19, 2009 at 11:13 pm
Paul White (11/19/2009)
Just to draw breath and summarize a bit:The fastest CLR solutions appear to be almost 20x faster than the fastest T-SQL implementation...
How did you make this calculation if...
November 19, 2009 at 4:43 am
I noticed that 'vliet' in a post earlier today seems to know about my binary approach calling it the "65, 33, 17, 9, 5, 3 an 2 spaces nested REPLACE...
November 18, 2009 at 2:42 pm
Hello C# Screw,
I'm surprised to see me below a CLR implementation.
You might want to put in this line which uses hard coded string constants instead of the replicate function.
I might...
November 18, 2009 at 7:16 am
Earlier on in this post I mentioned the use of a binary approach. I finally found some time to try this out.
declare @s-2 varchar(8000)
set @s-2 =...
November 18, 2009 at 6:19 am
Viewing 15 posts - 196 through 210 (of 548 total)