QuestDB supports the following types of joins: INNER
, OUTER
, CROSS
, ASOF
and SPLICE
. FULL
joins are not yet implemented and are on our roadmap. All supported join types can be combined in a single SQL statement; QuestDB SQL’s optimiser determines the best execution order and algorithms.
There are no known limitations on size of tables or sub-queries participating in joins and there are no limitations on number of joins either.
Join syntax
Following data join columns from joined tables are combined in single row. Same name columns originating from different tables will be automatically aliased to create unique column namespace of the result set.
Though it is a best practice to diligently specify join conditions, QuestDB will also analyse WHERE
clause for implicit join condition and will derive transient join conditions where necessary.
:::tip
When tables are joined on column that has the same name in both tables you can use shorthand ON (column)
clause
:::
Implicit joins
It is possible to join two tables using the following syntax:
SELECT *
FROM a, b
WHERE a.id = b.id;
The type of join as well as the column will be inferred from the where clause, and may end up being either INNER
or CROSS
join. For the example above, the equivalent explicit statement would be:
SELECT *
FROM a
JOIN b ON (id);
(INNER) JOIN
Overview
(INNER) JOIN
is used to return rows from 2 tables where the records on the compared column have matching values in both tables
Examples
The following query will return the movieId and the average rating from table ratings
. It will also add a column for the title
from table movies
. The corresponding title will be identified based on the movieId
in the ratings
table matching an id
in the movies
table.
SELECT movieId a, title, avg(rating)
FROM ratings
INNER JOIN (select movieId id, title from movies)
ON ratings.movieId = id;
By default JOIN
is interpreted as INNER JOIN
. Therefore INNER
does not need to be specified.
SELECT movieId a, title, avg(rating)
FROM ratings
JOIN (select movieId id, title from movies)
ON ratings.movieId = id;
OUTER JOIN
Overview
OUTER JOIN
will return all records from the LEFT table, and if matched, the records of the RIGHT table. When there is no match for the RIGHT table, it will return NULL
values in right table fields
Examples
General syntax is as follows:
SELECT tab1.colA, tab2.colB
FROM table1 tab1
OUTER JOIN table2 tab2
ON tab1.colA = tab2.colB;
OUTER JOIN
query can also be used to select all rows in left table that do not exist in right table.
SELECT tab1.colA, tab2.colB
FROM table1 tab1
OUTER JOIN table2 tab2
ON tab1.colA = tab2.colB
WHERE tab2.colB = NULL;
CROSS JOIN
Overview
CROSS JOIN
will return the cartesian product of the two tables being joined. It can be used to a table with all possible combinations.
:::note
CROSS JOIN
does not have ON
clause.
:::
Example
The following will return all possible combinations of starters and deserts
SELECT *
FROM starters
CROSS JOIN deserts;
ASOF JOIN
Overview
ASOF
joins are used on time series data to join two tables based on timestamp where timestamps do not exactly match. For a given record at a given timestamp, it will return the corresponding record in the other table at the closest timestamp prior to the timestamp in the first table.
:::note
To be able to leverage ASOF JOIN
, both joined table must have a designated timestamp
column. To designate a column as timestamp
, please refer to the CREATE TABLE section.
:::
ASOF
join is performed on tables or result sets that are ordered by time. When table is created as ordered by time order of records is enforced and timestamp column name is in table metadata. ASOF
join will use timestamp column from metadata.
Example
Consider the following tables.
ts | ask |
---|---|
2019-10-17T00:00:00.000000Z | 100 |
2019-10-17T00:00:00.200000Z | 101 |
2019-10-17T00:00:00.400000Z | 102 |
ts | bid |
---|---|
2019-10-17T00:00:00.100000Z | 101 |
2019-10-17T00:00:00.300000Z | 102 |
2019-10-17T00:00:00.500000Z | 103 |
Therefore the following query:
SELECT bids.ts timebid, bid, ask
FROM bids
ASOF JOIN asks;
Will return the following:
timebid | bid | ask |
---|---|---|
2019-10-17T00:00:00.100000Z | 101 | 100 |
2019-10-17T00:00:00.300000Z | 102 | 101 |
2019-10-17T00:00:00.500000Z | 103 | 102 |
:::note
There is no ASKS
at timestamp 2019-10-17T00:00:00.100000Z
. The ASOF JOIN
will look for the value in the BIDS
table that has the closest timestamp inferior or equal to the target timestamp.
:::
In case tables do not have designated timestamp column, but data is in chronological order, timestamp columns can be specified at runtime:
SELECT bids.ts timebid, bid, ask
FROM (bids timestamp(ts))
ASOF JOIN (asks timestamp (ts));
:::caution
ASOF
join does not check timestamp order, if data is not in chronological order join result is non-deterministic
:::
Above query assumes that there is only one instrument in BIDS
and ASKS
tables and therefore does not use the optional ON
clause.
If both tables store data for multiple instruments ON
clause will allow you to find bids for asks with matching instrument value.
SELECT *
FROM asks
ASOF JOIN bids ON (instrument);
SPLICE JOIN
Overview
SPLICE JOIN
is a full ASOF JOIN
. It will return all the records from both tables. For each record from left table splice join will find prevailing record from right table and for each record from right table - prevailing record from left table.
Examples
Considering the following tables.
ts | ask |
---|---|
2019-10-17T00:00:00.000000Z | 100 |
2019-10-17T00:00:00.200000Z | 101 |
2019-10-17T00:00:00.400000Z | 102 |
ts | bid |
---|---|
2019-10-17T00:00:00.100000Z | 101 |
2019-10-17T00:00:00.300000Z | 102 |
2019-10-17T00:00:00.500000Z | 103 |
This query:
SELECT bids.ts timebid, bid, ask
FROM bids
SPLICE JOIN asks;
Will return the following results
timebid | bid | ask |
---|---|---|
null | null | 100 |
2019-10-17T00:00:00.100000Z | 101 | 100 |
2019-10-17T00:00:00.100000Z | 101 | 101 |
2019-10-17T00:00:00.300000Z | 102 | 101 |
2019-10-17T00:00:00.300000Z | 102 | 102 |
2019-10-17T00:00:00.500000Z | 103 | 102 |
Note that the above query does not use the optional ON
clause. In case you need additional filtering on the two tables, you can use the ON
clause as follows:
SELECT ts timebid, instrument bidInstrument, bid, ask
FROM bids
SPLICE JOIN
(
SELECT ts timesask, instrument askInstrument, ask ask
FROM asks
)
ON bidInstrument=askInstrument;