Since you have to ALTER columns only one at a time (why, MS, why??), I'd create a new table and INSERT every row into it, converting every column at one...

sp_MSforeachdb is convenient, but it has huge overhead and is probably overkill for what you need.

You can use sp_executesql to return a value from dynamic SQL, albeit that it's not...

TheSQLGuru (12/16/2012)

---

My statement about making "queries" faster is about the ENTIRE process. The act of populating a temp table with an index already on the temp table before data...

If I understand correctly, maybe something like below.

[Btw, as a general performance guideline, make sure the datatype of the time in the table exactly matches your list of times. ...

And don't forget about computed columns. You can make this a computed column, so you don't have to have the generating code in a trigger. Since it's a...

I disagree completely.

I've repeatedly gotten huge performance gains sometimes by indexing temp tables properly. They are tables like any other table, after all. For large tables, I've seen...

You could also use an INNER JOIN to a hard-coded table:

select

case parents.whichParent when 'father' then s.father when mother then s.mother else s.parent_guardian end,

...

CELKO (12/14/2012)

---

But it is fast and portable 😀

On a more serious...

CELKO (12/14/2012)

---

Yeah, it was awful! :crazy:

What I wanted to do was...

Sorry, I didn't have test data so I couldn't pre-test it myself.

If you'll post some data as INSERT statements, I'll be happy to correct the query.

GSquared (12/14/2012)

---

CELKO (12/13/2012)

---

Please read just one little book on SQL and RDBMS. Please, just one little book! This is not a table; it has no key;...

CELKO (12/13/2012)

---

Please read just one little book on SQL and RDBMS. Please, just one little book! This is not a table; it has no key; it...

SELECT object_name, database_name, DATEPART(Week, event_time) AS EventWeek, SUM(sequence_number) AS [#TimesAccessed],

DATEADD(HOUR, DATEDIFF(HOUR, 0, event_time), 0) AS [EventHour]

INTO #EventHour

FROM ...

ScottPletcher (12/12/2012)

---

CELKO (12/11/2012)

---

In COBOL, the display and the internal format of data was the same. Each field had a PICTURE clause to define it. Your question makes sense only in...

It's normal IF the indexing on the data was not optimal, so that you were scanning old data before that has now been removed.

Note that when you DELETE a large...