dwain.c (2/22/2013)

---

mister.magoo (2/22/2013)

---

ChrisM@Work (2/22/2013)

---

L' Eomot Inversé (2/22/2013)

---

mister.magoo (2/22/2013)

---

And here is the code (with notes removed for brevity) and the changes underlined.

Sorry if I have made a mistake, but I think...

Not always true 🙂 It really depends upon...

aochss (1/16/2013)

---

The performance of this is amazing, as is the thought behind the code. These kind of posts really make you think and learn a whole new set of...

No worries 🙂 Your way of explaining was rather good for me to analyze how better my explanation could have been 😉

Nice. I would have thought it the same way 🙂 But I think the self join can be omitted

;WITH Tally(N) AS (SELECT ROW_NUMBER() OVER(ORDER BY (SELECT NULL))

...

yasubmj (10/15/2012)

---

Hmmm... so do you mean that if the other 2 columns (Message, DateTimeStamp) were a part of another index on the table... and were not a part of the...

To me this is fine and the optimizer is doing it what seems to be an efficient way. In first case, the optimizer knows that it has to update the...

SQL Kiwi (9/17/2012)

---

In any case, the assumption must be that TRY_* will only be used where necessary, and (relatively) slow as the intrinsic functions may be, it will likely out-perform...

Very nicely thought indeed. Great 😎 Not to mention how tricky it is to understand NOT LIKE %[NOT DIGITS]% 😉 Also the reasons why ISNUMERIC fails (do not work as...

SQL Kiwi (9/11/2012)

---

Not for me, no 🙂

The source code...

peter-757102 (9/11/2012)

---

Having to specify collations everywhere would make the function cumbersome to use. It might even throw off the optimizer in not using indexes due to collation differences.

An example...

m.t.cleary (9/11/2012)

---

peter-757102 (9/11/2012)

---

The recent discussion has hit an interesting topic, namely that of collations. While I can subscribe to the idea that it is ok for the delimiter being compared...

m.t.cleary (9/11/2012)

---

Jeff,

Usman’s N1 splitter is in http://www.sqlservercentral.com/Forums/FindPost1303895.aspx

On the subject of benchmark frameworks, I have some sympathy for storing the output. Splitters are often used where you are going...

SQL Kiwi (9/11/2012)

---

m.t.cleary (9/11/2012)

---

I agree that the CLR has additional and desirable features, but if we're comparing apples we need to make sure the functionality we are comparing is the...

ASFSDJG (9/7/2012)

---

CREATE NONCLUSTERED INDEX [IX_FIS_Pub_Author_Publication_ID] ON [dbo].[FIS_Publication_Author]

(

[Publication_ID] ASC,

[IsDeleted] ASC

)

INCLUDE ( [Author_ID],

[Seq_No]) WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, SORT_IN_TEMPDB = OFF, IGNORE_DUP_KEY = OFF, DROP_EXISTING = OFF,...