If you're just looking for the literal PC, then your expression just needs to be "PC" (with the quotes included). SSIS uses double-quotes, so using the snigle-quotes will get you...

Jeff Moden (2/5/2008)

---

It IS a cool solution... Wasn't so good in the performance department on my machine, though... it took 29 minutes to do the million row example on my...

then try the COMMIT? you should at least TRY to commit said transactions.

you'd have to run it twice )once per transaction still open. The second one might take...

Try checking the following:

- under Control Panel, Admin tools, services - double-check to see what the service account is for SQL Agent. That account needs to be able...

Assuming you're using an identity field as your PK, use

DBCC CHECKIDENT('mytable',RESEED,1)

will reset this back to 1

Note that you only need to do that if you use DELETE from mytable. ...

Even if they're perfect...

antonio.collins (2/5/2008)

---

the query that produces our 22K row ranked result set takes 150s. i was looking for some top N wizardry to cajole the query engine to skip producing...

The big question is - how would you correlate the rows without SSIS? How would you relate data from today to data from yesterday, row-by row? Something has...

Think of using the tally table has merit, but I'm thinking your using that hammer on the wrong kind of nail.

How about populating a temp table with a minute by...

Fully agree with GSquared. The correlated sub-query won't scale well, so if your data gets big it will start to choke on performance.

That being said - you want to...

That's certainly one way to do it! Depending on what you're doing - might also be a lookup (for data from another table) or data conversion (string-->numeric for example)

you can get SSMS to script them all by using the Generate scripts option under "all tasks". Simply right-click on the DB name to get started.

I usually am "dipped" in it by this time of day....and it ain't smelling pretty today....

Glad that helps!

One warning though - it seems to do some interesting things...

If the family+depcode generates duplicates, then no- PIVOT will aggregate them out. You'd also have to concatenate the first and last names together, since PIVot will only do one...

CROSS or OUTER APPLY with TVF's is basically a correlated sub-query in disguise. So - not incredibly bad if the sets are small, but still not as good as...