× Heads up!

Aqua Data Studio / nhilam

Follow
IDE for Relational Databases
×
anilkalshetti reported 2015-10-23T13:48:58Z  · tariqrahiman last modified 2015-12-04T07:01:05Z

"Stored As" field mapping improvement


Priority Low
Complexity Unknown
Component Hive
Version 17.0

Aqua Data Studio 17.0.0-rc-22
Build #: 46570
Built on: 2015-Oct-22 03:35:55 PM     
Database: Hive 1.2.1
Hortonworks HDP 2.3.0 - Hive 1.2.1
Drivers Link- Hive 1.2.1
[ACID Transactions enabled] 
Connection Type: HiveServer 2
Distribution: Hortonworks

Issue: "Stored As" field mapping improvement
This issue is related to Issue #13791

1] For Storage Format- TEXTFILE  
a]  Currently ADS uses, below mapping:

InputFormat:        org.apache.hadoop.mapred.TextInputFormat, 
OutputFormat:     org.apache.hadoop.hive.ql.io.IgnoreKeyTextOutputFormat


b] Mapping should be

InputFormat:       org.apache.hadoop.mapred.TextInputFormat,
OutputFormat:    org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat


Reason:  OutputFormat:     " org.apache.hadoop.hive.ql.io.IgnoreKeyTextOutputFormat "    is deprecated.
This is not updated in Hive confluence Language manual . 

Found correct information from Hive Java Docs [Deprecated Classes]- Ref.Link1 [Hive 0.10]   Link2 [Hive 1.2.1]
Link 
Summarizes to below text

[  - org.apache.hadoop.hive.ql.io.IgnoreKeyTextOutputFormat
use HiveIgnoreKeyTextOutputFormat instead  ]
 
@Dev
Solution: in OutPutFormat  instead "IgnoreKeyTextOutputFormat" use HiveIgnoreKeyTextOutputFormat
 
- Currently this issue is affecting below scenario.
i] Create Table with FileFormat SEQUENCEFILE
ii] Alter Table and change FileFormat to TEXTFILE
ii] Check Table Property - It's not detecting Stored As Format
-Please refer screenshot1.png
 
 
2] For Storage Format- SEQUENCEFILE
a]
SEQUENCEFILE

i] Create Table with FileFormat SEQUENCEFILE
ii] In Table Properties-Storage Tab, observe Stored As- Input and Output Format
 
InputFormat:          org.apache.hadoop.mapred.SequenceFileInputFormat,
OutputFormat:       org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
 
b]  TEXTFILE to SEQUENCEFILE
iii] Create a Table with FileFormat TEXTFILE          
ii] Alter Table and change FileFormat to SEQUENCEFILE
ii]  In Table Properties-Storage Tab, observe Stored As- Input and Output Format
 
InputFormat:org.apache.hadoop.mapred.SequenceFileInputFormat,
OutputFormat:org.apache.hadoop.mapred.SequenceFileOutputFormat
 

iv] Output of  command [desc extended testseq] also matches with above Input and Output format, still Stored As field is not reflected properly.
As this Input and Output format is mapped to SEQUENCEFILE, it should be displayed in Stored As field.
Please refer screenshot2.png

3 attachments

Issue #13794

Closed
Fixed
Resolved 2015-10-26T15:29:59Z
 
 
Completion
No due date
Fixed Build ADS 18.0.0-dev-9 & ADS 17.0.0-rc-24
No time estimate

About AquaClusters Privacy Policy Support Version - 19.0.2-4 AquaFold, Inc Copyright © 2007-2017