SQL GROUP BY Alias - An Oracle feature that would be great in SQL Server

By: Aaron Bertrand   |   Updated: 2023-12-01   |   Comments   |   Related: > TSQL

Problem

GROUP BY queries can become overly convoluted if your grouping column is a complex expression. Because of the logical processing order of a query, you're often forced to repeat such an expression since its alias can't be used in the GROUP BY clause.

Oracle recently solved this in their 23c release by adding the ability to GROUP BY column_alias. This is such simple but powerful syntax, and I'm hoping we can get SQL Server to follow Oracle's lead.

Solution

Reviewing the Logical Processing Order section of the official documentation:

Since the GROUP BY is processed first, aliases defined in the SELECT list are simply unavailable. This means you can't say, for example:

SELECT TheMonth   = DATEFROMPARTS(YEAR(create_date), MONTH(create_date), 1),
       MonthlyDBs = COUNT(*)
FROM sys.databases
GROUP BY TheMonth;

With this query, the following error is returned:

Msg 207, Level 16, State 1
Invalid column name 'TheMonth'.

To group by the TheMonth column, we need to use one of the following workarounds:

Repeating the expression

SELECT TheMonth   = DATEFROMPARTS(YEAR(create_date), MONTH(create_date), 1),
       MonthlyDBs = COUNT(*)
FROM sys.databases
GROUP BY DATEFROMPARTS(YEAR(create_date), MONTH(create_date), 1);

Using a common table expression (CTE)

WITH cte(TheMonth) AS 
(
  SELECT DATEFROMPARTS(YEAR(create_date), MONTH(create_date), 1)
  FROM sys.databases
)
SELECT TheMonth, MonthlyDBs = COUNT(*)
FROM cte
GROUP BY TheMonth;

Using a subquery

SELECT TheMonth, MonthlyDBs = COUNT(*)
FROM
(
  SELECT DATEFROMPARTS(YEAR(create_date), MONTH(create_date), 1)
  FROM sys.databases
) AS sq(TheMonth)
GROUP BY TheMonth;

Using CROSS APPLY

SELECT TheMonth, MonthlyDBs = COUNT(*)
FROM sys.databases
CROSS APPLY 
(
  SELECT DATEFROMPARTS(YEAR(create_date), MONTH(create_date), 1)
) AS ca(TheMonth)
GROUP BY TheMonth;

Note: In SQL Server 2022 or Azure SQL Database, you could simplify the expression toDATETRUNC(MONTH, create_date). But you still can't avoid one of the above workarounds.

In Oracle (again, starting with 23c), you can simply say (forgetting about SQL Server-specific metadata):

SELECT TRUNC(DateTimeColumnName, 'MONTH') AS TheMonth,
       COUNT(*) AS MonthlyWhatever
FROM TableName
GROUP BY TheMonth;

This form is much more convenient and removes the need to understand logical processing order – including clearing up the often-confusing fact that aliases from SELECT are available in ORDER BY.

Wouldn't it be great to have this syntax support in SQL Server?

On the other hand…

I don't want SQL Server to blindly follow Oracle's lead. Another recent addition was GROUP BY column_position, which allows you to reference a grouping column by its ordinal (which has been in MySQL for some time). Like with ORDER BY column_position, I consider this a ticking time bomb, and I do not want this parity at all.

Next Steps

Vote for this feature suggestion.

Also, see these tips and other resources:

• Learning the SQL GROUP BY Clause
• SQL GROUP BY Queries to Summarize and Rollup Sets of Records
• SQL Server DATEFROMPARTS Function
• New Date and Time Functions in SQL Server 2012
• SELECT (Transact-SQL)

About the author

Aaron Bertrand (@AaronBertrand) is a passionate technologist with industry experience dating back to Classic ASP and SQL Server 6.5. He is editor-in-chief of the performance-related blog, SQLPerformance.com, and also blogs at sqlblog.org.

This author pledges the content of this article is based on professional experience and not AI generated.

View all my tips