SQL Server - conditional aggregation with correlation

Question

Background:

The original case was very simple. Calculate running total per user from highest revenue to lowest:

CREATE TABLE t(Customer INTEGER  NOT NULL         


        

Answer 1

There is an easier solution:

SELECT c.Customer, c."User", c."Revenue",
       1.0 * Revenue/ NULLIF(c2.sum_total, 0) AS percentage,
       1.0 * c2.sum_running / NULLIF(c2.sum_total, 0) AS running_percentage
FROM t c CROSS APPLY
     (SELECT SUM(c2.Revenue) AS sum_total,
             SUM(CASE WHEN c2.Revenue >= x.Revenue THEN c2.Revenue ELSE 0 END) 
                 as sum_running
      FROM t c2 CROSS JOIN
           (SELECT c.REVENUE) x
      WHERE c."User" = c2."User"
     ) c2
ORDER BY "User", Revenue DESC;

I am not sure why or if this limitation is in the SQL '92 standard. I did have it pretty well memorized 20 or so years ago, but I don't recall that particular limitation.

I should note:

• At the time of the SQL 92 standard, lateral joins were not really on the radar. Sybase definitely had no such concept.
• Other databases do have problems with outer references. In particular, they often limit the scoping to one level deep.
• The SQL Standard itself tends to highly political (that is, vendor-driven) rather than driven by actual database user requirements. Well, over time, it does move in the right direction.