OK, Another Problem

Thanks for all the assistance with the time issue I was having. It looks like I can make the last code work somehow. I'll have to figure out how to incorporate it in, but I think I can probably manage that.

Here's my next question! I have a field in my source table that contains a range of values, separated by a dash. I need to parse this out into two separate destination table fields. The lengths before and after the dash will vary, so I need to be able to read up to the character, then from the other side of the character. Any suggestions on parsing? Books on line has not provided me with the answer (yet).

For example:

0.1-3.7 needs to be 0.1 (Field1) and 3.7 (Field2)
12.5-27
100-199

Thanks! This board has been extremely helpful for me (the pseudo-DBA).
Dirkcheckout CHARINDEX and PATINDEX either one of these should work for you.|||PATINDEX works perfectly for what I need.

Thanks again!
Dirk

odd order by behaviour

Might be out to lunch, but I can't figure why this is being ordered the way it is:

SELECT '1' as t
UNION ALL
SELECT 'A' as t
union all
SELECT '['as t
ORDER BY t

I'd expect it to be 1, A, [ but instead it's [, 1, A [ is ascii 91 which is greater than both 1 and A, so why does it come first?

KarlWell, I can tell you that it's related to collation. And I can also tell you that this will get you the results you expect:

SELECT t FROM
(
SELECT '1' as t
UNION ALL
SELECT 'A' as t
union all
SELECT '['as t
) AS A
ORDER BY t collate SQL_Latin1_General_Cp437_BIN

But I can't tell you intelligently why that works.

Terri|||some collations would actually place a "y" towards the beginning of the alphabet...

its not always tied to the ascii codesql