[SPARK-23173] from_json can produce nulls for fields which are marked as non-nullable

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.2.1
• Fix Version/s: 2.3.1, 2.4.0
• Component/s: SQL
• Labels: release-notes
• Assignee: Michał Świtakowski
• Reporter: Herman van Hövell
• Created: 22/Jan/18 01:43
• Updated: 12/Dec/22 18:11
• Resolved: 10/Mar/18 14:05

Description

The from_json function uses a schema to convert a string into a Spark SQL struct. This schema can contain non-nullable fields. The underlying JsonToStructs expression does not check if a resulting struct respects the nullability of the schema. This leads to very weird problems in consuming expressions. In our case parquet writing would produce an illegal parquet file.

There are roughly solutions here:

1. Assume that each field in schema passed to from_json is nullable, and ignore the nullability information set in the passed schema.
2. Validate the object during runtime, and fail execution if the data is null where we are not expecting this.
   I currently am slightly in favor of option 1, since this is the more performant option and a lot easier to do.

WDYT? cc rxin marmbrus hyukjin.kwon brkyvz

Issue Links

• is related to: SPARK-10848 Applied JSON Schema Works for json RDD but not when loading json file
• relates to: SPARK-17763 JacksonParser silently parses null as 0 when the field is not nullable
• links to: [Github] Pull Request #20694 (mswit-databricks)