Suppose I have a table like this:
| subject | flag |
|---|---|
| this is a test | 2 |
subject is of type text, and flag is of type int. I would like to transform this table to something like this in Postgres:
| token | flag |
|---|---|
| this | 2 |
| is | 2 |
| a | 2 |
| test | 2 |
Is there an easy way to do this?
Use a LATERAL join - with string_to_table() in Postgres 14+.
Minimal form:
SELECT token, flag
FROM tbl, string_to_table(subject, ' ') token
WHERE flag = 2;
The comma in the FROM list is (almost) equivalent to CROSS JOIN, LATERAL is automatically assumed for set-returning functions (SRF) in the FROM list. Why "almost"? See:
The alias "token" for the derived table is also assumed as column alias for a single anonymous column, and we assumed distinct column names across the query. Equivalent, more verbose and less error-prone:
SELECT s.token, t.flag
FROM tbl t
CROSS JOIN LATERAL string_to_table(subject, ' ') AS s(token)
WHERE t.flag = 2;
Or move the SRF to the SELECT list, which is allowed in Postgres (but not in standard SQL), to (almost) the same effect:
SELECT string_to_table(subject, ' ') AS token, flag
FROM tbl
WHERE flag = 2;
The last one seems acceptable since SRF in the SELECT list have been sanitized in Postgres 10. See:
If string_to_table() does not return any rows (empty or null subject), the (implicit) join eliminates the row from the result. Use LEFT JOIN ... ON true to keep qualifying rows from tbl. See:
We could also use regexp_split_to_table(), but that's slower. Regular expressions are powerful but expensive. See:
In Postgres 13 or older use unnest(string_to_array(subject, ' ')) instead of string_to_table(subject, ' ').
LATERAL join or with the unnest() function. How would you express this as a lateral join?
SELECT * FROM tbl t CROSS JOIN LATERAL unnest(string_to_array(t.subject, ' ')) AS s(token); Ample explanation in the linked answers.… , … syntax to be a simple cross join.cross join lateral automatically.I think it's not necessary to use a join, just the unnest() function in conjunction with string_to_array() should do it:
SELECT unnest(string_to_array(subject, ' ')) as "token", flag FROM test;
token | flag
-------+-------
this | 2
is | 2
a | 2
test | 2
Using regex split to table function including lateral join,
SELECT s.token, flag
FROM tbl t, regexp_split_to_table(t.subject, ' ') s(token)
WHERE flag = 2;
Refer to https://www.postgresql.org/docs/9.3/functions-string.html for the function details
