Skip to content

[SPARK-53291][SQL] Fix nullability for value column #52043

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
cashmand wants to merge 2 commits into from
Closed

[SPARK-53291][SQL] Fix nullability for value column #52043

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
60668aa
Fix
cashmand Aug 15, 2025
012463c
More tests
cashmand Aug 15, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
More tests
  • Loading branch information
@cashmand
cashmand committed Aug 15, 2025
commit 012463cddaa08305013f88d6f46e24a719196a44
19 changes: 19 additions & 0 deletions sql/core/src/test/scala/org/apache/spark/sql/VariantWriteShreddingSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -203,6 +203,8 @@ class VariantWriteShreddingSuite extends SparkFunSuite with ExpressionEvalHelper
testWithSchema(obj, t, Row(obj.getMetadata, untypedValue(obj), null))
}

testWithSchema(obj, VariantType, Row(obj.getMetadata, untypedValue(obj)))

// Happy path
testWithSchema(obj, StructType.fromDDL("a int, b string"),
Row(obj.getMetadata, null, Row(Row(null, 1), Row(null, "hello"))))
Expand All @@ -228,6 +230,11 @@ class VariantWriteShreddingSuite extends SparkFunSuite with ExpressionEvalHelper
testWithSchema(obj, ArrayType(StructType.fromDDL("a int, b string")),
Row(obj.getMetadata, untypedValue(obj), null))

// Shred with no typed_value in field schema
testWithSchema(obj, StructType.fromDDL("a variant, b variant"),
Row(obj.getMetadata, null,
Row(Row(untypedValue("1")), Row(untypedValue("\"hello\"")))))

// Similar to the case above where "b" was not in the shredding schema, but with the unshredded
// value being an object. Check that the copied value has correct dictionary IDs.
val obj2 = parseJson("""{"a": 1, "b": {"c": "hello"}}""")
Expand All @@ -248,6 +255,9 @@ class VariantWriteShreddingSuite extends SparkFunSuite with ExpressionEvalHelper
StructType.fromDDL("a int, b string")).foreach { t =>
testWithSchema(arr, t, Row(arr.getMetadata, untypedValue(arr), null))
}

testWithSchema(arr, VariantType, Row(arr.getMetadata, untypedValue(arr)))

// First element is shredded
testWithSchema(arr, ArrayType(StructType.fromDDL("a int, b string")),
Row(arr.getMetadata, null, Array(
Expand All @@ -272,6 +282,15 @@ class VariantWriteShreddingSuite extends SparkFunSuite with ExpressionEvalHelper
Row(null, 2),
Row(null, 3)
)))

// No typed_value in element schema
testWithSchema(arr, ArrayType(VariantType),
Row(arr.getMetadata, null, Array(
Row(untypedValue("""{"a": 1, "b": "hello"}""")),
Row(untypedValue("2")),
Row(untypedValue("null")),
Row(untypedValue("4"))
)))
}

}