New versions: Dev (3.14) | Latest (3.13) | 3.12 | 3.11 | 3.10 | 3.9 | 3.8 | Old versions: 3.7 | 3.6 | 3.5

CUME_DIST

Applies to ✅ Open Source Edition   ✅ Express Edition   ✅ Professional Edition   ✅ Enterprise Edition

The CUME_DIST() hypothetical set function calculates the cumulative distribution of the hypothetical value, i.e. the relative rank from 1/N to 1 (PERCENT_RANK produces values from 0 to 1)

SELECT 
  cume_dist(0) WITHIN GROUP (ORDER BY ID),
  cume_dist(2) WITHIN GROUP (ORDER BY ID),
  cume_dist(4) WITHIN GROUP (ORDER BY ID)
FROM BOOK
create.select(
         cumeDist(val(0)).withinGroupOrderBy(BOOK.ID),
         cumeDist(val(2)).withinGroupOrderBy(BOOK.ID),
         cumeDist(val(4)).withinGroupOrderBy(BOOK.ID))
      .from(BOOK)

Producing:

+--------------+--------------+--------------+
| cume_dist(0) | cume_dist(2) | cume_dist(4) |
+--------------+--------------+--------------+
|          0.2 |          0.6 |          1.0 |
+--------------+--------------+--------------+

Dialect support

This example using jOOQ:

cumeDist(val(0)).withinGroupOrderBy(BOOK.ID)

Translates to the following dialect specific expressions:

-- CUBRID, INFORMIX, ORACLE, POSTGRES, SQLSERVER, SYBASE
cume_dist(0) WITHIN GROUP (ORDER BY BOOK.ID)

-- ACCESS, ASE, AURORA_MYSQL, AURORA_POSTGRES, COCKROACHDB, DB2, DERBY, FIREBIRD, H2, HANA, HSQLDB, INGRES, MARIADB, MEMSQL, 
-- MYSQL, REDSHIFT, SQLDATAWAREHOUSE, SQLITE, TERADATA, VERTICA
/* UNSUPPORTED */

(These are currently generated with jOOQ 3.14, see #10141)

The jOOQ Logo