`regexp_extract_all` function

Applies to: Databricks SQL Databricks Runtime

Extracts all of the strings in str that match the regexp expression and correspond to the regex group index.

Syntax

regexp_extract_all(str, regexp [, idx] )

str: A STRING expression to be matched.
regexp: A STRING expression with a matching pattern.
idx: An optional integral number expression greater or equal 0 with default 1.

An ARRAY<STRING>.

The regexp string must be a Java regular expression.

When using literals, use raw-literal (r prefix) to avoid escape character pre-processing.

regexp may contain multiple groups. idx indicates which regex group to extract. An idx of 0 means match the entire regular expression.

SQL
> SELECT regexp_extract_all('100-200, 300-400', '(\\d+)-(\\d+)', 1);
 [100, 300]