Question

The same question has been asked for MySQL here, anyway that syntax doesn’t work in SQL Server.

I paste the sample from that question (by simplifying it) here because it explains what I need very well.

DECLARE @t1 integer
SET @t1=0;
-- MyTable is a simple Table with 2 fields "MyID" and "MyValue"
insert into MyTable
SELECT  @t1 := @t1+1, --this "should" work in MySQL
  MyAnotherValue
FROM    AnotherTable

Is there a way to achieve the same in SQL Server (without using a cursor)?

Note: this is a to be run once query, it is a maintenance query, so race conditions and locks are not an issue.

Answer 1

This will work, even if you run it more than once:

insert into MyTable(MyId, MyValue)
  select (select isnull(max(MyId), 0) from MyTable) + -- avoid duplicated keys, if you repeat this insert again
         row_number() over(order by MyAnotherValue),
         MyAnotherValue
    from AnotherTable

The isnull() function covers the case when MyTable is empty