Northern Monkey (8/3/2016)

---

I have some dim tables for find and replace for colours and sizes as I have multiple columns in a table that could hold the value see below

,FilterColour=CASEWHEN(SELECT...

Exactly, unless they're the same. I did that to reduce noise...

Out

I'm sorry, I didn't realized that the id values had gaps.

Here's a script that might help you and others for future references.

CREATE TABLE [dbo].[instrument_ref](

[instrument_id] [int] NOT NULL,

[description] [nvarchar](100)...

Your sample data doesn't seem to correspond with your expected results. Are you missing tables?

Use Cross tabs instead. I'm not sure if there's only one date per month so I included a range, but you can change the condition as needed.

SELECT SUM( CASE WHEN...

Please

Going further on J Livingston's formula.

DECLARE @str VARCHAR(100) = 'Discontinued Rx # 0011122657 via Prescription File'

SELECT SUBSTRING(@str,19,10) AS FixedLength_FixedPos,

SUBSTRING(@str,PATINDEX('%[0-9]%',@str),10) AS FixedLength_VarPos,

SUBSTRING(@str,19,PATINDEX('%[0-9] %',@str)-18)...

Rod at work (8/2/2016)

---

Eric M Russell (8/1/2016)

---

Maybe local user SQL Server groups should organize van pools for out-of-town SQL Saturday events. Not only would it help mitigate the expense of...

Grumpy DBA (8/2/2016)

---

Phil Parkin (8/2/2016)

---

I just read a post on another site which suggested using a 'carrot' as a column delimiter, rather than a comma.

Took me a moment to realise...

Quick tought.

Test for performance instead of trying to guess.

The only way to avoid implicit conversions is to change the data types in the tables or making them explicit which isn't an improvement over implicit ones.

You could try Eirikur's...

Sorry, I wasn't born yet.

chlorine

Running profiler against the server then navigating through the object explorer details tab, generates different dynamic sql calls.

AFAIK, that would mean that there's no stored procedure and those are ad-hoc...