[jira] [Created] (SPARK-39302) Spark SQL - wrong field selection in group by

["shaharlewy@gmail.com (Jira)" <ji...@apache.org>]

shaharlewy@gmail.com created SPARK-39302:
--------------------------------------------

             Summary: Spark SQL - wrong field selection in group by
                 Key: SPARK-39302
                 URL: https://issues.apache.org/jira/browse/SPARK-39302
             Project: Spark
          Issue Type: Bug
          Components: SQL
    Affects Versions: 3.1.2
            Reporter: shaharlewy@gmail.com


SQL parser select wrong field to group by.
See following example.
Sample data:

 
{code:java}
spark.sql("""
            select "US" as pv_countryCode
            union all
            select "IL" as pv_countryCode
         """).createOrReplaceTempView("my_test_data")
spark.sql("""
            select -3 as id, "US" as countryCode
         """).createOrReplaceTempView("country_codes_sample")
          {code}
code:
{code:java}
spark.sql("""
            select "US" as pv_countryCode
            union all
            select "IL" as pv_countryCode
         """).createOrReplaceTempView("my_test_data")
spark.sql("""
            select -3 as id, "US" as countryCode
         """).createOrReplaceTempView("country_codes_sample")
          {code}
Error:
{code:java}
org.apache.spark.sql.AnalysisException: expression 'cct.`id`' is neither present in the group by, nor is it an aggregate function. Add to group by or wrap in first() (or first_value) if you don't care which value you get.;
Aggregate [country_id#910], [coalesce(id#886, -3) AS country_id#908, count(1) AS count#909L]
+- Join LeftOuter, (countryCode#887 = country_id#910)
   :- SubqueryAlias BASE
   :  +- Project [pv_countryCode#883 AS country_id#910]
   :     +- SubqueryAlias my_test_data
   :        +- Union false, false
   :           :- Project [US AS pv_countryCode#883]
   :           :  +- OneRowRelation
   :           +- Project [IL AS pv_countryCode#884]
   :              +- OneRowRelation
   +- SubqueryAlias cct
      +- SubqueryAlias country_codes_sample
         +- Project [-3 AS id#886, US AS countryCode#887]
            +- OneRowRelation {code}
I expected Spark to choose the selected country_id(country_id#908) instead of country_id#910

Or at least throw ambiguous exception when grouping by `country_id`.

This lead developers to add `cct.id` into group by which results in unexpected results.
(In case country_id has both null and -3 values)

 



--
This message was sent by Atlassian Jira
(v8.20.7#820007)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org