db_id
stringclasses 12
values | query
stringlengths 16
593
| setup_sql
stringclasses 8
values | validation_sql
stringclasses 15
values | question
stringlengths 22
239
| category
stringclasses 5
values |
---|---|---|---|---|---|
hn | SELECT COUNT(*) as domain_count,
SUBSTRING(SPLIT_PART(url, '//', 2), 1, POSITION('/' IN SPLIT_PART(url, '//', 2)) - 1) as domain
FROM hacker_news
WHERE url IS NOT NULL GROUP BY domain ORDER BY domain_count DESC LIMIT 10; | ; | SELECT * FROM ddb_benchmark_result; | what are the top domains being shared on hacker_news? | hard |
laptop | SELECT c.firstname, c.lastname, COUNT(*) AS num_pcs_bought
FROM customers c
JOIN sales s ON c.customer_id = s.customer_id
JOIN pcs p ON s.model = p.model
GROUP BY c.customer_id, c.firstname, c.lastname
ORDER BY num_pcs_bought DESC
LIMIT 1; | ; | SELECT * FROM ddb_benchmark_result; | Who bought the most PCs, print also the users name? | medium |
transactions | select users.id, users.name, sum(transactions.amount) as balance from users join transactions on users.id = transactions.user_id group by users.id, users.name having balance = 0 | ; | SELECT * FROM ddb_benchmark_result; | list the names off account holders who have negative balances | easy |
laptop | SELECT model FROM products WHERE maker = 'B'; | ; | SELECT * FROM ddb_benchmark_result; | List only the model number of all products made by maker B. | easy |
laptop | SELECT model FROM products WHERE maker <> 'B'; | ; | SELECT * FROM ddb_benchmark_result; | List the model numbers of all products not made by maker B. | easy |
laptop | SELECT AVG(speed) FROM pcs WHERE speed >= 3.00 | ; | SELECT * FROM ddb_benchmark_result; | Return the average speed all PCs with speed >= 3.00 | easy |
laptop | SELECT MAX(price) FROM printers WHERE color = 'TRUE' AND type='laser' | ; | SELECT * FROM ddb_benchmark_result; | Return the price of the most expensive color laser printer | medium |
laptop | SELECT MIN(paid) FROM sales WHERE type_of_payment LIKE '%visa%' | ; | SELECT * FROM ddb_benchmark_result; | Return the minimum amount paid by customers who used a visa card (debit or credit) to purchase a product | medium |
laptop | SELECT customer_id FROM customers WHERE firstname LIKE '%e%' OR lastname LIKE '%e%' | ; | SELECT * FROM ddb_benchmark_result; | Find the customer_id of customers who have the letter 'e' either in their first name or in their last name | medium |
laptop | SELECT model, price/0.85 AS 'price (USD)' FROM laptops WHERE ram >= 1024 | ; | SELECT * FROM ddb_benchmark_result; | Assume all prices in the table Laptops are in Euro. List the prices of laptops with at least 1024 ram. You should return the price in USD in a column called 'price (USD)'. Assume that 1 USD = 0.85 EURO. Name the price column 'price (USD)'. | hard |
laptop | SELECT maker FROM products GROUP BY maker HAVING COUNT(maker) > 4; | ; | SELECT * FROM ddb_benchmark_result; | Return a list of makers that make more than four different products. | medium |
laptop | SELECT model FROM laptops WHERE speed > 1.7 ORDER BY speed DESC; | ; | SELECT * FROM ddb_benchmark_result; | List all the laptop model numbers that have a speed greater than 1.7 in descending order of speed. | medium |
laptop | SELECT firstname
FROM sales
JOIN customers ON sales.customer_id = customers.customer_id
GROUP BY firstname
ORDER BY COUNT(firstname); | ; | SELECT * FROM ddb_benchmark_result; | List firstnames of customers in an ascending order based on the number of purchases made by customers with this firstname. | medium |
laptop | SELECT DISTINCT maker FROM products JOIN pcs ON products.model = pcs.model WHERE ram > 1500; | ; | SELECT * FROM ddb_benchmark_result; | List all the makers (with only one entry per maker) who make PCs with RAM greater than 1500. | medium |
laptop | SELECT city, AVG(paid) as 'avg_spend' FROM sales JOIN customers ON sales.customer_id = customers.customer_id GROUP BY city | ; | SELECT * FROM ddb_benchmark_result; | Find the city and the average amount of money spent by customers in each city. Name the column for the amount 'avg_spend' | medium |
laptop | SELECT color, MAX(price) as 'max_price' FROM printers GROUP BY color; | ; | SELECT * FROM ddb_benchmark_result; | Find the maximum price for each color of printer. Name the column for the maximum price 'max_price' | medium |
who | select country_name, max(pm25_concentration) as worst_pm25_for_country
from ambient_air_quality
group by country_name
order by worst_pm25_for_country desc
limit 1 | ; | SELECT * FROM ddb_benchmark_result; | Find the country with the worst single reading of air quality (highest PM 2.5 value). Show the PM 2.5 value as well. | medium |
who | select country_name, avg(pm25_concentration) as worst_avg_pm25_for_country
from ambient_air_quality
group by country_name
order by worst_avg_pm25_for_country desc
limit 1 | ; | SELECT * FROM ddb_benchmark_result; | Find the country with the worst average air quality (highest PM 2.5 value). Show the PM 2.5 value as well. | medium |
who | select distinct country_name from ambient_air_quality order by country_name | ; | SELECT * FROM ddb_benchmark_result; | Find all countries for which WHO air quality data is available. Sort alphabetically. | medium |
who | select year, avg(pm25_concentration) from ambient_air_quality
where country_name = 'Singapore'
group by year
order by year | ; | SELECT * FROM ddb_benchmark_result; | Find Singapore air quality defined as PM2.5 concentration over time | medium |
nyc | SELECT COLUMNS('^trip_') FROM rideshare LIMIT 10; | ; | SELECT * FROM ddb_benchmark_result; | select only the column names from the rideshare table that start with trip_ and return the first 10 values | duckdb |
nyc | SELECT * FROM rideshare USING SAMPLE 1%; | ; | SELECT * FROM ddb_benchmark_result; | select a 1% sample from the nyc.rideshare table | duckdb |
laptop | SELECT * EXCLUDE (customer_id) FROM customers;
| ; | SELECT * FROM ddb_benchmark_result; | select all columns from the customer table, except customer_id | duckdb |
nyc | SUMMARIZE rideshare; | ; | SELECT * FROM ddb_benchmark_result; | show summary statistics of the rideshare table | duckdb |
none | SELECT * FROM read_csv_auto(
'https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv') | ; | SELECT * FROM ddb_benchmark_result; | read a CSV from https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv | duckdb |
none | COPY (SELECT * FROM read_csv_auto(
'https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv'))
TO 'titanic.parquet' (FORMAT 'parquet'); | ; | SELECT * FROM 'titanic.parquet' | read a CSV from https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv and convert it to a parquet file called "titanic" | duckdb |
none | CREATE TABLE titanic AS (SELECT * FROM read_csv_auto(
'https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv')) | ; | SELECT * FROM titanic; | create a table called "titanic" from CSV file https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv | duckdb |
none | PRAGMA default_null_order='NULLS LAST'; | ; | SELECT current_setting('default_null_order'); | configure duckdb to put null values last when sorting | duckdb |
none | CREATE TABLE IF NOT EXISTS products (
maker varchar(10),
model varchar(10),
type varchar(10)); | ; | SELECT * FROM products; | create a table about products, that contains a maker, model and type column | ddl |
product | INSERT INTO products (maker, model, type)
VALUES
('A', '1001', 'pc'); | ; | SELECT * FROM products; | add a row with values for model "1001" of type "pc", from maker "A" to products table | ddl |
none | CALL pragma_version();
| ; | SELECT * FROM ddb_benchmark_result; | get current version of duckdb | duckdb |
nyc | PRAGMA table_info('rideshare'); | ; | SELECT * FROM ddb_benchmark_result; | list all columns in table nyc.rideshare | duckdb |
nyc | PRAGMA show_tables; | ; | SELECT * FROM ddb_benchmark_result; | show all tables in the curent database | duckdb |
laptop | SELECT customer_id, model, sum(paid) FROM sales GROUP BY ALL | ; | SELECT * FROM ddb_benchmark_result; | how much did each customer spend per model type? | easy |
nyc | SELECT Max(datediff('minute', tpep_pickup_datetime, tpep_dropoff_datetime)) from nyc.taxi | ; | SELECT * FROM ddb_benchmark_result; | What was the longest taxi ride in minutes? | hard |
who | with per_region as (
select avg(pm10_concentration) as avg_pm10, who_region from ambient_air_quality group by who_region
), max_region as (
select who_region from per_region where avg_pm10 = (select max(avg_pm10) from per_region)
), min_city_value_in_max_region as (
select min(pm10_concentration) from ambient_air_quality where who_region in (from max_region)
), min_city_in_max_region as (
select city from ambient_air_quality where pm10_concentration in (from min_city_value_in_max_region) and who_region in (from max_region)
)
from min_city_in_max_region | ; | SELECT * FROM ddb_benchmark_result; | What is the city with the lowest pm10 concentration in the region with the highest average pm10 concentration? | hard |
hn | SELECT *, regexp_extract(text, '([a-z0-9_\.-]+)@([\da-z\.-]+)\.([a-z\.]{2,63})',0) email from hacker_news where email[:4]='test' | ; | SELECT * FROM ddb_benchmark_result; | Get all posts on hn that contain an email address starting with test. Return all original columns, plus a new column containing the email address. | hard |
json | SELECT employee.id, employee.first_name FROM employee_json | ; | SELECT * FROM ddb_benchmark_result; | Extract id and first_name properties as individual columns from the employee struct | duckdb |
who | SELECT who_region[1]::INT as region, * EXCLUDE (who_region) FROM who.ambient_air_quality | ; | SELECT * FROM ddb_benchmark_result; | count quality measurements per region. Make sure to return the region code (first char of who_region) as integer and sort by region. | duckdb |
flightinfo | SELECT seat.seat_number FROM seat
JOIN direct_flight ON direct_flight.flight_number = seat.flight_number
JOIN airport AS departure_airport ON departure_airport.iata_code = direct_flight.departure_airport_iata_code
JOIN airport AS arriving_airport ON arriving_airport.iata_code = direct_flight.arriving_airport_iata_code
JOIN city AS departure_city ON departure_city.city_zipcode = departure_airport.city_zip_code
JOIN city AS arriving_city ON arriving_city.city_zipcode = arriving_airport.city_zip_code
WHERE departure_city.city_name = 'Bruxelles' AND arriving_city.city_name = 'Newark'; | ; | SELECT * FROM ddb_benchmark_result; | Which seats were available on the flight from Bruxelles to Newark? | hard |
laptop | COPY customers FROM 'customers_12_12_2023.csv'; | COPY customers TO 'customers_12_12_2023.csv'; | SELECT * FROM customers; | copy content of csv file customers_12_12_2023.csv into customers table | duckdb |
laptop | COPY customers FROM 'customers_12_12_2023.csv' (DELIMITER '\t'); | COPY customers TO 'customers_12_12_2023.csv' (FORMAT CSV, DELIMITER '\t'); | SELECT * FROM customers; | copy content of csv file costomers_12_12_2023.csv into customers table with tab separator | duckdb |
laptop | COPY customers FROM 'customers_partitioned/city=Amsterdam/*.parquet'; | COPY customers TO 'customers_partitioned' (FORMAT PARQUET, PARTITION_BY (city), OVERWRITE_OR_IGNORE True); | SELECT * FROM customers;; | copy any parquet files from 'customers_partitioned/city=Amsterdam/' into customers table | duckdb |
laptop | COPY customers(customer_id) FROM 'customers_customer_ids.csv'; | COPY customers(customer_id) TO 'customers_customer_ids.csv'; | SELECT * FROM customers; | copy only the customer_id column from the customers_customer_ids.csv into the customers tables | duckdb |
laptop | CREATE TABLE test_tbl AS SELECT * FROM read_json_auto('test.json'); | COPY customers TO 'test.json'
| SELECT * FROM test_tbl; | read json file from test.json and create new table from it called 'test_tbl' | duckdb |
laptop | SELECT * FROM read_csv_auto('test.csv'); | COPY customers TO 'test.csv'; | SELECT * FROM ddb_benchmark_result; | read csv from test.csv | duckdb |
laptop | SELECT * FROM read_csv_auto('test.csv', columns={'customer_id': 'VARCHAR', 'firstname': 'VARCHAR', 'lastname': 'VARCHAR'}); | COPY customers(customer_id, firstname, lastname) TO 'test.csv'; | SELECT * FROM ddb_benchmark_result; | read csv from test.csv with predefined column and types - customer_id: string, firstname: string, lastname: string | duckdb |
laptop | SELECT * EXCLUDE (ram, hd) FROM pcs; | ; | SELECT * FROM ddb_benchmark_result; | select all columns from pcs table except for ram and hd | duckdb |
laptop | SELECT COLUMNS('name$') FROM customers; | ; | SELECT * FROM ddb_benchmark_result; | select all columns ending with 'name' from customers table | duckdb |
laptop | SELECT LENGTH(COLUMNS('name$')) FROM customers | ; | SELECT * FROM ddb_benchmark_result; | for each column ending with 'name' in the customers table, compute the string length | duckdb |
laptop | SELECT * REPLACE (upper(city) AS city) FROM customers; | ; | SELECT * FROM ddb_benchmark_result; | get all columns from customer table, and make all city names uppercase | duckdb |
laptop | EXPLAIN SELECT * FROM customers | ; | SELECT * FROM ddb_benchmark_result; | show query plan for query: SELECT * from customers | duckdb |
laptop | SELECT ascii(lastname) FROM customers; | ; | SELECT * FROM ddb_benchmark_result; | get the first character of the firstname column and cast it to an INT | duckdb |
laptop | SELECT model, speed::INTEGER FROM laptops; | ; | SELECT * FROM ddb_benchmark_result; | get laptop name and speed, return the speed as integer | duckdb |
laptop_array | SELECT phone_numbers[1] FROM customers; | ; | SELECT * FROM ddb_benchmark_result; | get the first phone number of each customer | duckdb |
laptop_array | INSERT INTO customers(customer_id, phone_numbers) VALUES (5, ['12312323', '23123344']); | ; | SELECT * FROM customers; | insert two phone numbers to customer with id 5 [\"12312323\", and '23123344'] | duckdb |
laptop | ALTER TABLE customers ADD COLUMN phone_numbers VARCHAR[]; | ; | DESCRIBE customers; | how to add a new column phone_numbers to the customers table, with array type varchar | duckdb |
laptop | SELECT firstname[1] FROM customers; | ; | SELECT * FROM ddb_benchmark_result; | get the first letter of the customers firstname | duckdb |
laptop_array | SELECT phone_numbers[:2] FROM customers; | ; | SELECT * FROM ddb_benchmark_result; | get the first two phone numbers from the phone numbers array of each customer | duckdb |
laptop | SELECT {'a':1, 'b':2, 'c':3}; | ; | SELECT * FROM ddb_benchmark_result; | create a struct with keys a, b, c and values 1,2,3 | duckdb |
laptop | SELECT [1,2,3];
| ; | SELECT * FROM ddb_benchmark_result; | create array with values 1,2,3 | duckdb |
laptop | CREATE TABLE test (embeddings FLOAT[100]); | ; | DESCRIBE test; | create table test with a fix-sized array column with 100 dimenions, called embeddings | duckdb |
laptop | CREATE TABLE test (person STRUCT(name VARCHAR, id INTEGER)); | ; | DESCRIBE test; | create table test with a struct column called person with properties name and id | duckdb |
laptop_struct | SELECT person.name, person.id FROM test; | ; | SELECT * FROM ddb_benchmark_result; | get persons name and persons id from the test table. | duckdb |
laptop | UPDATE customers SET email = NULL; | ; | SELECT email FROM customers; | remove all values from email column in customers table | duckdb |
laptop_json | ALTER TABLE customers ALTER COLUMN email SET DATA TYPE VARCHAR; | ; | DESCRIBE customers; | make customer email of type VARCHAR | duckdb |
laptop_json | INSERT INTO customers (customer_id, email) VALUES (5,'{"from": "[email protected]", "to": "[email protected]"}'); | ; | SELECT * FROM customers; | insert json into customer email for customer id 5: {'from': '[email protected]', 'to': '[email protected]'} | duckdb |
laptop_json | SELECT customers.email->>'from' FROM customers; | ; | SELECT * FROM ddb_benchmark_result; | get 'from' field from customer email json | duckdb |
laptop | SUMMARIZE customers; | ; | SELECT * FROM ddb_benchmark_result; | summarize the customer table | duckdb |
laptop | SELECT * FROM customers USING SAMPLE 10% (reservoir); | ; | SELECT count(*) FROM ddb_benchmark_result; | sample 10% from the customers table using reservoir sampling | duckdb |
laptop | SET threads = 10; | ; | SELECT current_setting('threads'); | set number of threads to 10 | duckdb |
laptop | SET memory_limit = '20G';
| ; | SELECT current_setting('memory_limit'); | set memory limit to 20 gigabyte | duckdb |
laptop | SELECT * EXCLUDE (price), avg(price) FROM laptops GROUP BY ALL; | ; | SELECT * FROM ddb_benchmark_result; | show the average price of laptop and group by the remaining columns | duckdb |
laptop | SELECT * FROM laptops WHERE price > 1000 ORDER BY ALL;
| ; | SELECT * FROM ddb_benchmark_result; | show all laptops with price above 1000 and order by all columns | duckdb |
laptop | ATTACH 'who.ddb'; | ; | SHOW DATABASES; | attach database file who.ddb | duckdb |
No dataset card yet
New: Create and edit this dataset card directly on the website!
Contribute a Dataset Card- Downloads last month
- 2