[SPARK-20888] Document HiveCaseSensitiveInferenceMode.INFER_AND_SAVE in Spark SQL 2.1 to 2.2 migration notes

• Type: Task
• Status: Resolved
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: 2.2.0
• Fix Version/s: 2.2.0
• Component/s: Documentation, SQL
• Target Version/s: 2.2.0
• Assignee: Michael MacFadden
• Reporter: Michael MacFadden
• Created: 25/May/17 17:24
• Updated: 19/Oct/17 19:09
• Resolved: 26/May/17 01:27

Description

Spark 2.2 introduces a new default value for the SQL configuration setting, `spark.sql.hive.caseSensitiveInferenceMode`. The new default value is `HiveCaseSensitiveInferenceMode.INFER_AND_SAVE`. As this enables a behavior that differs from Spark 2.1, I believe this should be noted in the Spark SQL migration notes from 2.1 to 2.2.

Issue Links

• is related to: SPARK-22306 INFER_AND_SAVE overwrites important metadata in Parquet Metastore table
• links to: [Github] Pull Request #18112 (mallman)