That was ...

... interesting:rolleyes:

And certainly not views that do [font="Courier New"]SELECT *[/font]

I couldn't even remember what an SPN was, but I figured -R should be RESET, and -D should be DELETE. That left -L for LIST. Good thing the tool writers...

ronmoses (6/25/2012)

---

I see no problem with the wording. Can a table variable have indexes defined, yes or no? Yes, it can. That's all that was asked -...

There are no ways to DEFINE indexes on table variables. Indexes may or may not be created by the implementation when you define constraints, but that is an implementation detail.

Great question!

It highlights something which could be considered a bug. It's certainly not something you would expect from reading the documentation on CREATE TYPE.

Being someone who actually used sp_addtype "back...

This question is just completely broken. There's a big difference between "preferred" and "valid". And to claim that varbinary(max) isn't a LOB is a bit daft.

While I find the question slightly pointless, I have to say I'm more than a little surprised that so many got it wrong. The question made it perfectly clear that...

Richard M Karpel (12/16/2010)

---

Now, I read the section highlighted in BOL, and I expected to get the not defined error message, but I would need to look further and see...

I don't want to come off as mean, but your problem is that you fail to communicate what your problem is.

If your object is in fact the Java "Date" class,...

Excellent question!

I got it wrong on the strength of some tests I ran that gave me the 'Column name or number of supplied values does not match table definition.' error....

SQL Server does...

Truly a good question because it provoked a lot of thought as to which techniques are possible, valid and practical.

I got it "right" (yay!), but after reading the discussion I...

This is not a particularly very well designed question. What is a comment and what is the definition? I would say the comments inside an object are part of the...

Great question.

But the version should have been specified. In SQL 2000 you can see the execution plan even for encrypted procs. Although that would probably have been a dead giveaway...