# name: test/db-benchmark/groupby.test_slow
# description: Group By benchmark (0.5GB - small dataset) from h2oai db-benchmark (https://github.com/h2oai/db-benchmark)
# group: [db-benchmark]

require httpfs

statement ok
pragma threads=16

# 5% nulls
statement ok
CREATE TABLE x AS SELECT * FROM read_csv_auto('https://github.com/duckdb/duckdb-data/releases/download/v1.0/G1_1e7_1e2_5_0.csv.gz');

# statement ok
# CREATE TABLE x AS SELECT * FROM read_csv_auto('G1_1e7_1e2_5_0.csv.gz');

# q1
statement ok
CREATE TABLE ans AS SELECT id1, sum(v1) AS v1 FROM x GROUP BY id1

query II
SELECT COUNT(*), sum(v1)::varchar AS v1 FROM ans
----
96	28498857

statement ok
DROP TABLE ans

# q2
statement ok
CREATE TABLE ans AS SELECT id1, id2, sum(v1) AS v1 FROM x GROUP BY id1, id2;

query II
SELECT count(*), sum(v1) AS v1 FROM ans;
----
9216	28498857

statement ok
DROP TABLE ans;

# q3
statement ok
CREATE TABLE ans AS SELECT id3, sum(v1) AS v1, avg(v3) AS v3 FROM x GROUP BY id3;

query III
SELECT COUNT(*), sum(v1) AS v1, sum(v3) AS v3 FROM ans;
----
95001	28498857	4749467.631946747

statement ok
DROP TABLE ans;

# q4
statement ok
CREATE TABLE ans AS SELECT id4, avg(v1) AS v1, avg(v2) AS v2, avg(v3) AS v3 FROM x GROUP BY id4;

query IIII
SELECT COUNT(*), sum(v1) AS v1, sum(v2) AS v2, sum(v3) AS v3 FROM ans
----
96	287.9894309270616821	767.8529216923457105	4799.873270453372

statement ok
DROP TABLE ans;

# q5
statement ok
CREATE TABLE ans AS SELECT id6, sum(v1) AS v1, sum(v2) AS v2, sum(v3) AS v3 FROM x GROUP BY id6;

query IIII
SELECT COUNT(*), sum(v1) AS v1, sum(v2) AS v2, sum(v3) AS v3 FROM ans
----
95001	28498857	75988394	474969574.04777884

statement ok
DROP TABLE ans;

# q6
statement ok
CREATE TABLE ans AS SELECT id4, id5, quantile_cont(v3, 0.5) AS median_v3, stddev(v3) AS sd_v3 FROM x GROUP BY id4, id5;

# WARNING: this result might be incorrect
# could not verify using Postgres because of lack of median
query III
SELECT COUNT(*), sum(median_v3) AS median_v3, sum(sd_v3) AS sd_v3 FROM ans
----
9216	460771.216444	266006.904622

statement ok
DROP TABLE ans;

# q7
statement ok
CREATE TABLE ans AS SELECT id3, max(v1)-min(v2) AS range_v1_v2 FROM x GROUP BY id3;

query II
SELECT count(*), sum(range_v1_v2) AS range_v1_v2 FROM ans;
----
95001	379850

statement ok
DROP TABLE ans;

# q8
statement ok
CREATE TABLE ans AS SELECT id6, v3 AS largest2_v3 FROM (SELECT id6, v3, row_number() OVER (PARTITION BY id6 ORDER BY v3 DESC) AS order_v3 FROM x WHERE v3 IS NOT NULL) sub_query WHERE order_v3 <= 2

query II
SELECT count(*), sum(largest2_v3) AS largest2_v3 FROM ans
----
190002	18700554.779631943

statement ok
DROP TABLE ans;

# q9
statement ok
CREATE TABLE ans AS SELECT id2, id4, pow(corr(v1, v2), 2) AS r2 FROM x GROUP BY id2, id4;

query II
SELECT count(*), sum(r2) AS r2 FROM ans
----
9216	9.940515516534346

statement ok
DROP TABLE ans;

# q10
statement ok
CREATE TABLE ans AS SELECT id1, id2, id3, id4, id5, id6, sum(v3) AS v3, count(*) AS count FROM x GROUP BY id1, id2, id3, id4, id5, id6;

query II
SELECT sum(v3) AS v3, sum(count) AS count FROM ans;
----
474969574	10000000

statement ok
DROP TABLE ans