[SPARK-20921] While reading from oracle database, it converts to wrong type.

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Duplicate
• Affects Version/s: 2.1.0
• Component/s: SQL
• Assignee: Unassigned
• Reporter: vivek dixit
• Created: 30/May/17 10:12
• Updated: 15/Sep/17 08:14
• Resolved: 15/Sep/17 08:14

Description

While reading data from oracle using jdbc, it converts integer column to boolean. if it has the column size as one which is wrong.

https://github.com/apache/spark/commit/39a2b2ea74d420caa37019e3684f65b3a6fcb388#diff-edf9e48ce837c5f2a504428d9aa67970R46 this condition in oracledjdbcdialect is doing it. This has caused our data load process to fail.

Issue Links

• Is contained by: SPARK-20427 Issue with Spark interpreting Oracle datatype NUMBER
• links to: [Github] Pull Request #18195 (wangyum)