In case of any doubt, I should clarify that you only need to do this for typing strings into the query window (or batch script, etc. - whenever using...

This is the best I have come up with so far. The first version works for a single row, the second is a version which processes a whole rowset. The code assumes...

When you specify a string literal (i.e. a readable string value, in quotes), any embedded single quotes are doubled (i.e. into a pair of adjacent single quotes, not into double-quotes)....

Sorry I misspelt your name Sergiy. Still I'm afraid your dark mutterings about procedural programmers (not me) don't cut any ice unless backed with some sort of substantive criticism.

Gald to be of help. The need for complex or difficult constraints (e.g. the kind that need triggers) can often (not always!) be avoided by improved normalisation of your database...

sys.partitions.rows (bigint) - Approximate number of rows in this partition.

Note approx, but then given we are finding the height of all tables in the DB, this sounds like more of an...

You can create the temp table statically with just an IDENITITY column, say, and then add columns using an ALTER TABLE statement in dynamic SQL. the ability to do this is...

True for v8 and earlier. In v9, MS have actually managed to keep the row count (and the good old rowmodctr) updated automatically. How clever!

I have to add that all this talk of 1nf and not passing CSVs really only arises because of the shortcomings of TSQL - not being able to pass tables...

I came up with this, more as an exercise than a real-world solution:

 

declare

@csv varchar(8000

Yes, a trigger  I think - but a different db design would be better. It looks like this field is meant to track some sort of progress. It might be...

What is the revlevance of your post, Sergei?

Or

set

nocount on set ansi_warnings off

yes the information schema views as well as having an irritatingly long schema name are still quite basic and designed more like reports than useful programming tools. the sys schema has lots...

yes they have a spurious entry with indid = 0. indid 1 is always clustered.

I don't know what the forward-compatible, MS-sanctioned version of the code is though.