[SPARK-17296] Spark SQL: cross join + two joins = BUG

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.0.0
• Fix Version/s: 2.0.1, 2.1.0
• Component/s: SQL
• Assignee: Herman van Hövell
• Reporter: Furcy Pin
• Created: 29/Aug/16 15:47
• Updated: 08/Dec/16 22:56
• Resolved: 06/Sep/16 22:58

Description

In spark shell :

CREATE TABLE test (col INT) ;
INSERT OVERWRITE TABLE test VALUES (1), (2) ;

SELECT 
COUNT(1)
FROM test T1 
CROSS JOIN test T2
JOIN test T3
ON T3.col = T1.col
JOIN test T4
ON T4.col = T1.col
;

returns :

Error in query: cannot resolve '`T1.col`' given input columns: [col, col]; line 6 pos 12

Apparently, this example is minimal (removing the CROSS or one of the JOIN causes no issue).

Issue Links

• duplicates: SPARK-17384 SQL - Running query with outer join from 1.6 fails
• is duplicated by: SPARK-17253 Left join where ON clause does not reference the right table produces analysis error
• is duplicated by: SPARK-17384 SQL - Running query with outer join from 1.6 fails
• links to: [Github] Pull Request #14867 (hvanhovell)
• links to: [Github] Pull Request #14984 (hvanhovell)