It's so odd to hear everyone talking about getting "better" and not programming like they were in the '60's or '70's anymore. I wonder what people would say if...

Sorry... had a change of thought... posted those thoughts further below.

Bob Hovious (4/12/2009)

---

Jeff you shoulda been a lawyer :-D. Wasn't it you asking why people needed to do this on another thread? I'm in total agreement that...

GilaMonster (4/8/2009)

---

Steve Jones - Editor (4/7/2009)

---

GilaMonster (4/7/2009)

---

Similar thing here. In my country, woman and IT don't usually go in the same sentence. The number of times I've been 'mistaken' for...

GilaMonster (4/12/2009)

---

Same here. It comes out of my 'education and training' savings.

If you're paying your own way, ignore the conference...

Paul White (4/12/2009)

---

Jeff,

You are so right to point out the data-consistency advantage of the computed column. I guess I could add a constraint or something to simulate that benefit,...

sku370870 (4/12/2009)

---

I now have this ...

SELECT #tblHolidays.HolidayDescription As [Description], #tblHolidays.HolidayDate AS [Holiday Date],

count(case #tblDepartmentHolidays.DepartmentID when 0 then 1 else null end) as [Global],

count(case #tblDepartmentHolidays.DepartmentID when 1 then 1 else...

Heh... ok, Paul. From the way you were talking, I thought you were on to an improvement.

Thanks for your concern. If I'm not mistaken, PIVOT can do multiple aggregations... I just gotta look back in my notes because I don't use PIVOT for all the...

All of that may be true... I just hope they never remove the capability because it is an option. It's kind of like cursors... I generally deplore their usage...

Paul White (4/12/2009)

---

Jeff Moden (4/11/2009)

---

You may want to reconsider ever using a Pivot operator... a properly formed crosstab can be nearly twice as fast...

http://www.sqlservercentral.com/articles/T-SQL/63681/

Performance meaurements are at the bottom of...

Paul White (4/11/2009)

---

Hey Jeff,

It would, but an ordinary column would be my first choice, mainly because persisted computed columns...

Outstanding... thanks for the post and the code, Paul... I'll check it out. That's what I love about forums. Lots of folks with lots of ideas.

Bob Hovious (4/11/2009)

---

See if these are quick enough for you. The use of FOR XML is the fastest concatenation technique I've come across, but this sort of string...

I agree... it would probably be better to not move the data before processing. And, if that's still somehow necessary, I've had great success with just dropping the table...