Jeff Moden (5/7/2008)

---

Carl Federl (5/7/2008)

---

Another great article ! Thanks

The first reference for a tally table that I can recall is in "Guide to Sybase and SQL Server" by C....

Hi Jeff,

My preference is actually the CLR function I coded up a while back. It's range is the (-) to (+) limit of the bigint type and it takes no...

Great article.

I used an "Integers" table for years, but starting with 2005 found that CLR functions to stream out integers is better (faster and takes virtually no space in the...

Ed Sanford (2/18/2008)

---

Just wanted to add that my main purpose for wanting to use this Array logic is to be able to pass selection lists to a query.

......

dmbaker (2/18/2008)

---

Eric Wilson (2/18/2008)

---

With due respect, I'm afraid you missed my point. One (major) goal of the Relational Model was to be as uncluttered as possible. To that end, relations...

dmbaker (2/18/2008)

---

Eric Wilson (2/14/2008)

---

Regarding the overall article's topic:

The only possible reason to want an ARRAY type is because of assumptions of how the data will be *physically* organized. But we...

By the way, regarding an earlier comment that you cannot Index table variables...

1) True, technically, and another example of the failure to separate logical and physical issues. (Tables should behave...

Regarding the overall article's topic:

We DO NOT NEED ARRAYS!

To use XML and Table Variables to try to "emulate" arrays is like grabbing your neighbor's Ferrari to try and emulate a...

This just shows how people will overly complicate a set-oriented or SQL based solution because they are trying to fit it into a world they are more comfortable with.

And this...

Idiot me!

Somehow this posted to the "whole author's stuff"-

Meant it just for the recent article on constants. NOT the author in general.

Will try to delete, but in case not,...

Sergiy, too funny. Thanks for the laugh. I think you made one good point, and missed about 3 of my points. Oh well.

For me, I've *only* found XML to be...

Other than about an order-of-magnitude performance hit (due to saving to disk), not too much.

But they are also not true variables and don't have rigorous scoping rules, etc. Global temp-tables...

One thing no-one seems to point out:

Why not ASK MICROSOFT and other DBMS vendors to build in Table-Vars as fully fledged things?! Why are we stuck with XML work-arounds (which...

Good reminder of work-arounds to us NOT HAVING A RELATIONAL database product. Passing a table into an operation should have been solved in System R originally and part of every...

Oh my heavens what crap!

WARNING TO NEWBIES: Please do not be dissuaded by the seeming "balance" of comments supporting bit-masking as a "useful" technique.

POINT #1: You cannot do bitmasking in...