Sergey Shelukhin created HIVE-8462:
--------------------------------------
Summary: CBO duplicates columns
Key: HIVE-8462
URL: https://issues.apache.org/jira/browse/HIVE-8462
Project: Hive
Issue Type: Bug
Reporter: Sergey Shelukhin
Assignee: Sergey Shelukhin
Priority: Critical
{noformat}
select *, rank() over(partition by key order by value) as rr from src1
{noformat}
Original plan appears to be incorrect:
{noformat}
HiveProjectRel(key=[$0], value=[$1], (tok_function rank (tok_windowspec
(tok_partitioningspec (tok_distributeby (tok_table_or_col key)) (tok_orderby
(tok_tabsortcolnameasc (tok_table_or_col value))))))=[$5], rr=[$5])
HiveProjectRel(key=[$0], value=[$1], block__offset__inside__file=[$2],
input__file__name=[$3], row__id=[$4], (tok_function rank (tok_windowspec
(tok_partitioningspec (tok_distributeby (tok_table_or_col key)) (tok_orderby
(tok_tabsortcolnameasc (tok_table_or_col value))))))=[rank() OVER (PARTITION BY
$0 ORDER BY $1 ROWS BETWEEN 2147483647 FOLLOWING AND 2147483647 PRECEDING)])
HiveTableScanRel(table=[[default.src1]])
{noformat}
and final AST has {noformat}
TOK_SELEXPR
.
TOK_TABLE_OR_COL
$hdt$_0
(tok_function rank (tok_windowspec (tok_partitioningspec
(tok_distributeby (tok_table_or_col key)) (tok_orderby (tok_tabsortcolnameasc
(tok_table_or_col value))))))
(tok_function rank (tok_windowspec (tok_partitioningspec
(tok_distributeby (tok_table_or_col key)) (tok_orderby (tok_tabsortcolnameasc
(tok_table_or_col value))))))
TOK_SELEXPR
.
TOK_TABLE_OR_COL
$hdt$_0
(tok_function rank (tok_windowspec (tok_partitioningspec
(tok_distributeby (tok_table_or_col key)) (tok_orderby (tok_tabsortcolnameasc
(tok_table_or_col value))))))
rr
{noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
