[SPARK-42683] Automatically rename metadata columns that conflict with data schema columns

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 3.3.2, 3.4.0
• Fix Version/s: 3.5.0
• Component/s: Spark Core
• Target Version/s: 3.5.0
• Assignee: Ryan Johnson
• Reporter: Ryan Johnson
• Created: 06/Mar/23 14:17
• Updated: 30/Mar/23 09:10
• Resolved: 30/Mar/23 09:09

Description

Today, if a datasource already has a column called `_metadata`, queries cannot access the file-source metadata column that normally carries that name. We can address this conflict with two changes to metadata column handling:

1. Automatically rename any metadata column whose name conflicts with a data schema column
2. Add a facility to reliably find metadata columns by their original/logical name, even if they were renamed.

Issue Links

• links to: [Github] Pull Request #40300 (ryan-johnson-databricks)