[SPARK-7196] decimal precision lost when loading DataFrame from JDBC

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.3.1
• Fix Version/s: 1.3.2, 1.4.0
• Component/s: SQL
• Assignee: L. C. Hsieh
• Reporter: Ken Geis
• Created: 28/Apr/15 16:19
• Updated: 25/Aug/15 12:22
• Resolved: 30/Apr/15 22:15

Description

I have a decimal database field that is defined as 10.2 (i.e. ##########.##). When I load it into Spark via sqlContext.jdbc(..), the type of the corresponding field in the DataFrame is DecimalType, with precisionInfo None. Because of that loss of precision information, SPARK-4176 is triggered when I try to .saveAsTable(..).

Issue Links

• relates to: SPARK-4176 Support decimals with precision > 18 in Parquet
• links to: [Github] Pull Request #5777 (viirya)