[
https://issues.apache.org/jira/browse/HIVE-12328?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14988307#comment-14988307
]
Thejas M Nair commented on HIVE-12328:
--------------------------------------
But there seems like a bigger problem here, than just improper error message.
The column code is there in both LHS and RHS. It should have complained about
ambigous column in this case. (fyi, i was trying to use the "join .. using "
syntax, but used " on " instead ).
The join expression looks broken, the "keys" under " Map Join Operator" is
empty -
{code}
0: jdbc:hive2://localhost:10000/default> explain select * from sample_07 a join
sample_07 b on (code);
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--+
|
Explain
|
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--+
| Plan not optimized by CBO.
|
|
|
| Vertex dependency in root stage
|
| Map 1 <- Map 2 (BROADCAST_EDGE)
|
|
|
| Stage-0
|
| Fetch Operator
|
| limit:-1
|
| Stage-1
|
| Map 1
|
| File Output Operator [FS_84]
|
| compressed:false
|
| Statistics:Num rows: 243 Data size: 50660 Basic stats: COMPLETE
Column stats: NONE
|
|
table:{"serde:":"org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe","input
format:":"org.apache.hadoop.mapred.TextInputFormat","output
format:":"org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat"} |
| Select Operator [SEL_83]
|
|
outputColumnNames:["_col0","_col1","_col2","_col3","_col4","_col5","_col6","_col7"]
|
| Statistics:Num rows: 243 Data size: 50660 Basic stats:
COMPLETE Column stats: NONE
|
| Map Join Operator [MAPJOIN_86]
|
| | condition map:[{"":"Inner Join 0 to 1"}]
|
| | keys:{}
|
| |
outputColumnNames:["_col0","_col1","_col2","_col3","_col7","_col8","_col9","_col10"]
|
| | Statistics:Num rows: 243 Data size: 50660 Basic stats:
COMPLETE Column stats: NONE
|
| |<-Map 2 [BROADCAST_EDGE]
|
| | Reduce Output Operator [RS_81]
|
| | sort order:
|
| | Statistics:Num rows: 110 Data size: 22923 Basic stats:
COMPLETE Column stats: NONE
|
| | value expressions:code (type: string), description
(type: string), total_emp (type: int), salary (type: int)
|
| | Filter Operator [FIL_85]
|
| | predicate:code (type: string)
|
| | Statistics:Num rows: 110 Data size: 22923 Basic
stats: COMPLETE Column stats: NONE
|
| | TableScan [TS_78]
|
| | alias:b
|
| | Statistics:Num rows: 221 Data size: 46055 Basic
stats: COMPLETE Column stats: NONE
|
| |<-TableScan [TS_77]
|
| alias:a
|
| Statistics:Num rows: 221 Data size: 46055 Basic stats:
COMPLETE Column stats: NONE
|
|
|
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+--+
{code}
> Join On clause needs a semantic check to verify expression is boolean
> ----------------------------------------------------------------------
>
> Key: HIVE-12328
> URL: https://issues.apache.org/jira/browse/HIVE-12328
> Project: Hive
> Issue Type: Bug
> Components: Query Planning
> Affects Versions: 1.0.0, 1.2.1
> Reporter: Thejas M Nair
> Assignee: Pengcheng Xiong
>
> SQL join query fails at query runtime with a poor error message if the
> expression in the on clause of join is not a boolean.
> Hive should give a proper error message at runtime.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
