OLAP-bewerkingen: GROUPING SETS

Data-driven beslissingen nemen met SQL

Irene Ortner

Data Scientist at Applied Statistics

Overzicht van OLAP-operators in SQL

Uitbreidingen in SQL voor OLAP-bewerkingen

GROUP BY CUBE
GROUP BY ROLLUP
GROUP BY GROUPING SETS

Tabel renting_extended

De eerste rijen van de tabel renting_extended:

| renting_id | country  | genre  | rating |
|------------|----------|--------|--------|
| 2          | Belgium  | Drama  | 10     |
| 32         | Belgium  | Drama  | 10     |
| 203        | Austria  | Drama  | 6      |
| 292        | Austria  | Comedy | 8      |
| 363        | Belgium  | Drama  | 7      |
| .......... | ........ | ...... | ...... |

GROUP BY GROUPING SETS

Voorbeeld van een query met de operator GROUPING SETS:

SELECT country, 
       genre, 
       COUNT(*)
FROM rentings_extended
GROUP BY GROUPING SETS ((country, genre), (country), (genre), ());

Kolomnamen tussen haakjes vormen één aggregatieniveau.
GROUP BY GROUPING SETS geeft een UNION over meerdere GROUP BY-queries.

GROUPING SETS en GROUP BY-queries

SELECT country, 
       genre, 
       COUNT(*)
FROM renting_extended
GROUP BY GROUPING SETS (country, genre);

Tel verhuur per unieke combinatie van land en genre.
Expressie in GROUPING SETS: (country, genre)

SELECT country, 
       genre, 
       COUNT(*)
FROM renting_extended
GROUP BY country, genre;

| country | genre  | count  |
|---------|--------|--------|
| Austria | Comedy | 2      |
| Belgium | Drama  | 15     |
| Austria | Drama  | 4      |
| Belgium | Comedy | 1      |

GROUPING SETS en GROUP BY-queries

SELECT country, COUNT(*)
FROM renting_extended
GROUP BY GROUPING SETS (country);

Tel filmverhuur per land.
Expressie in GROUPING SETS: (country)

SELECT country, COUNT(*)
FROM renting_extended
GROUP BY country;

| country | count |
|---------|-------|
| Austria | 16    |
| Belgium | 6     |

GROUPING SETS en GROUP BY-queries

SELECT genre, COUNT(*)
FROM renting_extended
GROUP BY GROUPING SETS (genre);

Tel filmverhuur per genre.
Expressie in GROUPING SETS: (genre)

SELECT genre, COUNT(*)
FROM renting_extended
GROUP BY genre;

| country | count |
|---------|-------|
| Comedy  | 3     |
| Drama   | 19    |

GROUPING SETS en GROUP BY-queries

SELECT COUNT(*)
FROM renting_extended
GROUP BY GROUPING SETS ();

Totale aggregatie: tel alle verhuur.
Expressie in GROUPING SETS: ()

SELECT COUNT(*)
FROM renting_extended;

| count |
|-------|
| 22    |

Notatie voor GROUP BY GROUPING SETS

GROUP BY GROUPING SETS (...)

SELECT country, genre, COUNT(*)
FROM renting_extended
GROUP BY GROUPING SETS ((country, genre), (country), (genre), ());

UNION van 4 eerdere queries.
Bundel alle pivot-tabelinfo in één query.
Equivalent aan GROUP BY CUBE (country, genre).

Resultaat met GROUPING SETS-operator

SELECT country, genre, COUNT(*)
FROM renting_extended
GROUP BY GROUPING SETS ((country, genre), (country), (genre), ());

| country | genre  |count|
|---------|--------|-----|
| NULL    | NULL   | 22  |
| Austria | Comedy | 2   |
| Belgium | Drama  | 15  |
| Austria | Drama  | 4   |
| Belgium | Comedy | 1   |
| Belgium | NULL   | 16  |
| Austria | NULL   | 6   |
| NULL    | Comedy | 3   |
| NULL    | Drama  | 19  |

Aantal verhuur en gemiddelde rating berekenen

Combineer alleen geselecteerde aggregaties:
- land en genre
- genre

Gebruik aantal verhuur en gemiddelde beoordeling voor aggregatie.

SELECT country, 
     genre, 
     COUNT(*), 
     AVG(rating) AS avg_rating
FROM renting_extended
GROUP BY GROUPING SETS ((country, genre), (genre));

Aantal verhuur en gemiddelde rating berekenen

SELECT country, genre, COUNT(*), AVG(rating) AS avg_rating
FROM renting_extended
GROUP BY GROUPING SETS ((country, genre), (genre));

| country | genre  | count| avg_rating |
|---------|--------|------|------------|
| Austria | Comedy | 2    | 8.00       |
| Belgium | Drama  | 15   | 9.17       |
| Austria | Drama  | 4    | 6.00       |
| Belgium | Comedy | 1    | NULL       |
| NULL    | Comedy | 3    | 8.00       |
| NULL    | Drama  | 19   | 8.38       |

Laten we oefenen!

Data-driven beslissingen nemen met SQL