Viewing 15 posts - 47,386 through 47,400 (of 59,095 total)
Simon_L (11/5/2008)
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 9:12 pm
I agree with Matt... dynamic, supposedly "smart" triggers whose code is exactly the same for every table is "Death by SQL" and deserves a bucket load of high speed pork...
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 9:00 pm
Perhaps this will help...
http://www.sqlservercentral.com/articles/T-SQL/63681/
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 8:54 pm
AND, if you do happen to need or want to plan on more than 2 billion characters of notes for each row, then I agree with some of the others......
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 8:53 pm
I just wouldn't bother with all of that... we're not talking about in row TEXT columns. We're talking about NVARCHAR(MAX). Just add the Notes column to each table......
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 8:50 pm
I don't see anything in the code that requires dynamic SQL... why the requirement to use it? I won't even ask why you think you need a cursor... 😛
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 8:45 pm
Both 2000 and 2005 do it like Gail said... both make a copy of the table with the new column added, then it moves the data, drops the old table,...
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 8:42 pm
Just use BCP to run it into a temporary staging table with the correct switches set to offload any bad rows into an error file. You should set the...
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 8:39 pm
... and please see the following article for some potential problems you can avoid with some other methods of concatenation...
http://www.sqlservercentral.com/articles/Test+Data/61572/
Jerry's method does not have such problems.
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 8:07 pm
1
... and is TempDB large enough for the task and is auto-shrink on TempDB turned off?
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 8:04 pm
Sorry... not enough info. Please refer to the link in my signature for quicker/better answers.
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 8:02 pm
If the NVARCHAR column has no multi-lingual entries (the main purpose for NVARCHAR), the a simple SELECT CAST(columnname AS VARCHAR(x)) where x is the desired maximum length should work. ...
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 7:56 pm
Probably because it's relatively slower than a Cross-Tab and not backwards compatible to SQL Server 2000 and may not be migratable to other RDBMS's.
So... without any real data to test...
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 7:50 pm
That won't be a recurrsive query... it'll be a simple join between the two tables joined on ID.
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 7:44 pm
Probably not what you want to hear, but I'm thinking that the OCR software, well... sucks and you may have to find some better software.
--Jeff Moden
Change is inevitable... Change for the better is not.
November 5, 2008 at 7:39 pm
Viewing 15 posts - 47,386 through 47,400 (of 59,095 total)