Summarize
The SUMMARIZE command can be used to easily compute a number of aggregates over a table or a query.
The SUMMARIZE command launches a query that computes a number of aggregates over all columns (min, max, approx_unique, avg, std, q25, q50, q75, count), and return these along the column name, column type, and the percentage of NULL values in the column.
Note that the quantiles and percentiles are approximate values.
Usage
To summarize the contents of a table, use SUMMARIZE followed by the table name.
Query
SUMMARIZE TABLE tbl;Result
column_name | column_type | min | max | approx_unique | avg | std | q25 | q50 | q75 | count | null_percentage------------------+-------------+----------+-------+---------------+--------------------+--------------------+------+------+-------+-------+----------------- id | INTEGER | 1 | 7 | 3 | 3.6666666666666665 | 3.055050463303893 | 2 | 3 | 6 | 3 | 0.00 name | VARCHAR | bookmark | other | 3 | NULL | NULL | NULL | NULL | NULL | 3 | 0.00 col1 | DOUBLE | 1 | 9 | 3 | 4.666666666666667 | 4.041451884327381 | 1.75 | 4 | 7.75 | 3 | 0.00 col2 | DOUBLE | 2 | 16 | 3 | 7.666666666666667 | 7.371114795831994 | 2.75 | 5 | 13.25 | 3 | 0.00 some column name | VARCHAR | alpha | gamma | 3 | NULL | NULL | NULL | NULL | NULL | 3 | 0.00 i | INTEGER | 1 | 3 | 3 | 2 | 1 | 1 | 2 | 3 | 3 | 0.00 a | INTEGER | 10 | 12 | 3 | 11 | 1 | 10 | 11 | 12 | 3 | 0.00 b | INTEGER | 20 | 22 | 3 | 21 | 1 | 20 | 21 | 22 | 3 | 0.00 c | INTEGER | 30 | 32 | 3 | 31 | 1 | 30 | 31 | 32 | 3 | 0.00 number | INTEGER | 10 | 20 | 2 | 15 | 7.0710678118654755 | 10 | 15 | 20 | 3 | 33.33 number1 | INTEGER | 100 | 300 | 3 | 200 | 100 | 125 | 200 | 275 | 3 | 0.00 number2 | INTEGER | 1000 | 3000 | 3 | 2000 | 1000 | 1250 | 2000 | 2750 | 3 | 0.00To summarize a query, prepend SUMMARIZE to a query.
Query
SUMMARIZE SELECT * FROM tbl;Example
Below is an example of SUMMARIZE on a sample lineitem table modeled on the TPC-H schema.
Query
CREATE TABLE lineitem AS SELECT * FROM (VALUES (1, 155, 4, 1, 17.0, 21168.23, 0.04, 0.02, 'N', 'O', DATE '1996-03-13', DATE '1996-02-12', DATE '1996-03-22', 'DELIVER IN PERSON', 'TRUCK', 'to beans x-ray carefully'), (1, 67, 9, 2, 36.0, 45983.16, 0.09, 0.06, 'N', 'O', DATE '1996-04-12', DATE '1996-02-28', DATE '1996-04-20', 'TAKE BACK RETURN', 'MAIL', 'slyly bold pinto beans'), (2, 106, 5, 1, 38.0, 44694.46, 0.00, 0.05, 'N', 'O', DATE '1997-01-28', DATE '1997-01-14', DATE '1997-02-02', 'TAKE BACK RETURN', 'RAIL', 'requests against the'), (3, 4, 2, 1, 45.0, 54058.05, 0.06, 0.00, 'R', 'F', DATE '1994-02-02', DATE '1994-01-04', DATE '1994-02-23', 'NONE', 'AIR', 'pending foxes wake'), (3, 19, 8, 2, 49.0, 46796.47, 0.10, 0.00, 'R', 'F', DATE '1993-11-09', DATE '1993-12-20', DATE '1993-11-24', 'TAKE BACK RETURN', 'RAIL', 'sleep quickly bold pinto beans')) AS t(l_orderkey, l_partkey, l_suppkey, l_linenumber, l_quantity, l_extendedprice, l_discount, l_tax, l_returnflag, l_linestatus, l_shipdate, l_commitdate, l_receiptdate, l_shipinstruct, l_shipmode, l_comment);Query
SUMMARIZE lineitem;Result
column_name | column_type | min | max | approx_unique | avg | std | q25 | q50 | q75 | count | null_percentage-----------------+--------------+--------------------+--------------------------+---------------+----------------------+----------------------+------------+------------+------------+-------+----------------- l_orderkey | INTEGER | 1 | 3 | 3 | 2 | 1 | 1 | 2 | 3 | 5 | 0.00 l_partkey | INTEGER | 4 | 155 | 5 | 70.2 | 62.22298610642212 | 15 | 67 | 118 | 5 | 0.00 l_suppkey | INTEGER | 2 | 9 | 5 | 5.6 | 2.8809720581775866 | 4 | 5 | 8 | 5 | 0.00 l_linenumber | INTEGER | 1 | 2 | 2 | 1.4 | 0.5477225575051662 | 1 | 1 | 2 | 5 | 0.00 l_quantity | DECIMAL(3,1) | 17.0 | 49.0 | 5 | 37 | 12.349089035228468 | 31 | 38 | 46 | 5 | 0.00 l_extendedprice | DECIMAL(7,2) | 21168.23 | 54058.05 | 5 | 42540.074 | 12490.322396220607 | 38812 | 45983 | 48612 | 5 | 0.00 l_discount | DECIMAL(3,2) | 0.00 | 0.10 | 5 | 0.057999999999999996 | 0.04024922359499621 | 0 | 0 | 0 | 5 | 0.00 l_tax | DECIMAL(3,2) | 0.00 | 0.06 | 4 | 0.026000000000000002 | 0.027928480087537882 | 0 | 0 | 0 | 5 | 0.00 l_returnflag | VARCHAR | N | R | 2 | NULL | NULL | NULL | NULL | NULL | 5 | 0.00 l_linestatus | VARCHAR | F | O | 2 | NULL | NULL | NULL | NULL | NULL | 5 | 0.00 l_shipdate | DATE | 1993-11-09 | 1997-01-28 | 5 | 1995-07-02 04:48:00 | NULL | 1994-01-12 | 1996-03-13 | 1996-06-24 | 5 | 0.00 l_commitdate | DATE | 1993-12-20 | 1997-01-14 | 5 | 1995-06-17 00:00:00 | NULL | 1993-12-31 | 1996-02-12 | 1996-05-18 | 5 | 0.00 l_receiptdate | DATE | 1993-11-24 | 1997-02-02 | 5 | 1995-07-13 19:12:00 | NULL | 1994-01-31 | 1996-03-22 | 1996-07-01 | 5 | 0.00 l_shipinstruct | VARCHAR | DELIVER IN PERSON | TAKE BACK RETURN | 3 | NULL | NULL | NULL | NULL | NULL | 5 | 0.00 l_shipmode | VARCHAR | AIR | TRUCK | 4 | NULL | NULL | NULL | NULL | NULL | 5 | 0.00 l_comment | VARCHAR | pending foxes wake | to beans x-ray carefully | 5 | NULL | NULL | NULL | NULL | NULL | 5 | 0.00Using SUMMARIZE in a Subquery
SUMMARIZE can be used as a subquery. This allows creating a table from the summary, for example:
Query
CREATE TABLE tbl_summary AS SELECT * FROM (SUMMARIZE TABLE tbl);Summarizing Remote Tables
It is possible to summarize remote tables over HTTP(S) and S3 using the SUMMARIZE TABLE statement. For example:
Query
SUMMARIZE TABLE 'https://example.com/data/data.csv';Result
column_name | column_type | min | max | approx_unique | avg | std | q25 | q50 | q75 | count | null_percentage-------------+-------------+-------+-------+---------------+------+--------------------+------+------+------+-------+----------------- a | BIGINT | 1 | 2 | 2 | 1.5 | 0.7071067811865476 | 1 | 2 | 2 | 2 | 0.00 b | VARCHAR | [2,3] | [5,8] | 2 | NULL | NULL | NULL | NULL | NULL | 2 | 0.00 c | DOUBLE | 4.5 | 9.5 | 2 | 7 | 3.5355339059327378 | 4.5 | 7 | 9.5 | 2 | 0.00