Yibo Dong created SPARK-60125:
---------------------------------
Summary: regexp_instr ignores the specified regex group index
Key: SPARK-60125
URL: https://issues.apache.org/jira/browse/SPARK-60125
Project: Spark
Issue Type: Bug
Components: SQL
Affects Versions: 5.0.0
Environment: Spark: 5.0.0-SNAPSHOT
Commit: 38422aefaf599ffb4cfa579d98898d95b1f02081
Java: OpenJDK 17.0.17
OS: macOS 15.7.4
Reporter: Yibo Dong
h2. What happened
{{regexp_instr}} returns the start position of the full regex match even when a
different capture group is selected using the third argument. This produces an
incorrect result when the selected group starts after the full match.
h2. How to reproduce
Run the following SQL using {{spark-sql}}:
{code:sql}
SELECT regexp_instr('abc123', '([a-z]+)([0-9]+)', 2) AS position;
{code}
Save the SQL as {{repro.sql}} and run:
{code:bash}
spark-sql --master 'local[2]'
--conf spark.ui.enabled=false
-f repro.sql
{code}
h2. Expected result
The second capture group matches {{123}}, which starts at the fourth character
of {{abc123}}. The result should be:
{code:text}
position
--------
4
{code}
h2. Actual result
The query returns the start position of the full match:
{code:text}
position
--------
1
{code}
h2. Additional information
Spark's ocumentation defines the third argument, {{idx}}, as the matched regex
group index.
The implementation appears to use the start position of the full match without
considering the selected group index.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
