[SPARK-40819][SQL] Timestamp nanos behaviour regression #38312
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Binary file not shown.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -29,6 +29,7 @@ import org.apache.parquet.schema.Type._ | |
| import org.apache.spark.SparkException | ||
| import org.apache.spark.sql.catalyst.ScalaReflection | ||
| import org.apache.spark.sql.execution.datasources.SchemaColumnConvertNotSupportedException | ||
| import org.apache.spark.sql.functions.desc | ||
| import org.apache.spark.sql.internal.SQLConf | ||
| import org.apache.spark.sql.internal.SQLConf.ParquetOutputTimestampType._ | ||
| import org.apache.spark.sql.test.SharedSparkSession | ||
|
|
@@ -45,15 +46,17 @@ abstract class ParquetSchemaTest extends ParquetTest with SharedSparkSession { | |
| binaryAsString: Boolean, | ||
| int96AsTimestamp: Boolean, | ||
| writeLegacyParquetFormat: Boolean, | ||
| expectedParquetColumn: Option[ParquetColumn] = None, | ||
| nanosAsLong: Boolean = false): Unit = { | ||
| testSchema( | ||
| testName, | ||
| StructType.fromAttributes(ScalaReflection.attributesFor[T]), | ||
| messageType, | ||
| binaryAsString, | ||
| int96AsTimestamp, | ||
| writeLegacyParquetFormat, | ||
| expectedParquetColumn = expectedParquetColumn, | ||
| nanosAsLong = nanosAsLong) | ||
| } | ||
|
|
||
| protected def testParquetToCatalyst( | ||
|
|
@@ -65,12 +68,14 @@ abstract class ParquetSchemaTest extends ParquetTest with SharedSparkSession { | |
| caseSensitive: Boolean = false, | ||
| inferTimestampNTZ: Boolean = true, | ||
| sparkReadSchema: Option[StructType] = None, | ||
| expectedParquetColumn: Option[ParquetColumn] = None, | ||
| nanosAsLong: Boolean = false): Unit = { | ||
| val converter = new ParquetToSparkSchemaConverter( | ||
| assumeBinaryIsString = binaryAsString, | ||
| assumeInt96IsTimestamp = int96AsTimestamp, | ||
| caseSensitive = caseSensitive, | ||
| inferTimestampNTZ = inferTimestampNTZ, | ||
| nanosAsLong = nanosAsLong) | ||
|
|
||
| test(s"sql <= parquet: $testName") { | ||
| val actualParquetColumn = converter.convertParquetColumn( | ||
|
|
@@ -119,7 +124,8 @@ abstract class ParquetSchemaTest extends ParquetTest with SharedSparkSession { | |
| writeLegacyParquetFormat: Boolean, | ||
| outputTimestampType: SQLConf.ParquetOutputTimestampType.Value = | ||
| SQLConf.ParquetOutputTimestampType.INT96, | ||
| expectedParquetColumn: Option[ParquetColumn] = None, | ||
| nanosAsLong: Boolean = false): Unit = { | ||
|
|
||
| testCatalystToParquet( | ||
| testName, | ||
|
|
@@ -134,7 +140,8 @@ abstract class ParquetSchemaTest extends ParquetTest with SharedSparkSession { | |
| parquetSchema, | ||
| binaryAsString, | ||
| int96AsTimestamp, | ||
| expectedParquetColumn = expectedParquetColumn, | ||
| nanosAsLong = nanosAsLong) | ||
| } | ||
|
|
||
| protected def compareParquetColumn(actual: ParquetColumn, expected: ParquetColumn): Unit = { | ||
|
|
@@ -149,7 +156,14 @@ abstract class ParquetSchemaTest extends ParquetTest with SharedSparkSession { | |
| val expectedDesc = expected.descriptor.get | ||
| assert(actualDesc.getMaxRepetitionLevel == expectedDesc.getMaxRepetitionLevel) | ||
| assert(actualDesc.getMaxRepetitionLevel == expectedDesc.getMaxRepetitionLevel) | ||
|
|
||
| actualDesc.getPrimitiveType.getLogicalTypeAnnotation match { | ||
| case timestamp: LogicalTypeAnnotation.TimestampLogicalTypeAnnotation | ||
| if timestamp.getUnit == LogicalTypeAnnotation.TimeUnit.NANOS => | ||
| assert(actual.sparkType == expected.sparkType) | ||
| case _ => | ||
| assert(actualDesc.getPrimitiveType === expectedDesc.getPrimitiveType) | ||
| } | ||
| } | ||
|
|
||
| assert(actual.repetitionLevel == expected.repetitionLevel, "repetition level mismatch: " + | ||
|
|
@@ -197,6 +211,32 @@ abstract class ParquetSchemaTest extends ParquetTest with SharedSparkSession { | |
| } | ||
|
|
||
| class ParquetSchemaInferenceSuite extends ParquetSchemaTest { | ||
| testSchemaInference[Tuple1[Long]]( | ||
|
Contributor
There was a problem hiding this comment. Can this case run passed before Spark 3.2, in my impression, Parquet 1.10.1 used by Spark 3.1 does not support nanos type, does it?
Contributor
Author
There was a problem hiding this comment. This particular case doesn’t pass, neither does similar tests for
Contributor
There was a problem hiding this comment. Then is this really a regression?
Contributor
Author
There was a problem hiding this comment. So I've been looking further into it, it's because the message is different between In where as So in Spark 3.1.0 you end up hitting this block with returns a where as since 3.2 you hit https://github.com/apache/spark/blob/branch-3.2/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetSchemaConverter.scala#L174 because a case for
Contributor
There was a problem hiding this comment. @cloud-fan moving Parquet from |
||
| "timestamp nanos", | ||
| """ | ||
| |message root { | ||
| | required int64 _1 (TIMESTAMP(NANOS,true)); | ||
| |} | ||
| """.stripMargin, | ||
| binaryAsString = false, | ||
| int96AsTimestamp = true, | ||
| writeLegacyParquetFormat = true, | ||
| expectedParquetColumn = Some( | ||
| ParquetColumn( | ||
| sparkType = StructType.fromAttributes( | ||
| ScalaReflection.attributesFor[Tuple1[Long]]), | ||
| descriptor = None, | ||
| repetitionLevel = 0, | ||
| definitionLevel = 0, | ||
| required = false, | ||
| path = Seq(), | ||
| children = Seq( | ||
| primitiveParquetColumn(LongType, PrimitiveTypeName.INT64, Repetition.REQUIRED, | ||
| 0, 0, Seq("_1"), logicalTypeAnnotation = Some(LogicalTypeAnnotation.intType(64, false))) | ||
| ))), | ||
| nanosAsLong = true | ||
| ) | ||
|
|
||
| testSchemaInference[(Boolean, Int, Long, Float, Double, Array[Byte])]( | ||
| "basic types", | ||
| """ | ||
|
|
@@ -1027,6 +1067,24 @@ class ParquetSchemaSuite extends ParquetSchemaTest { | |
| } | ||
| } | ||
|
|
||
| test("SPARK-40819: parquet file with TIMESTAMP(NANOS, true) (with nanosAsLong=true)") { | ||
| val tsAttribute = "birthday" | ||
| withSQLConf(SQLConf.LEGACY_PARQUET_NANOS_AS_LONG.key -> "true") { | ||
| val testDataPath = testFile("test-data/timestamp-nanos.parquet") | ||
| val data = spark.read.parquet(testDataPath).select(tsAttribute) | ||
| assert(data.schema.fields.head.dataType == LongType) | ||
| assert(data.orderBy(desc(tsAttribute)).take(1).head.getAs[Long](0) == 1668537129123534758L) | ||
| } | ||
| } | ||
|
|
||
| test("SPARK-40819: parquet file with TIMESTAMP(NANOS, true) (with default nanosAsLong=false)") { | ||
| val testDataPath = testFile("test-data/timestamp-nanos.parquet") | ||
| val e = intercept[org.apache.spark.SparkException] { | ||
| spark.read.parquet(testDataPath).collect() | ||
| } | ||
| assert(e.getMessage.contains("Illegal Parquet type: INT64 (TIMESTAMP(NANOS,true)).")) | ||
| } | ||
|
|
||
| // ======================================================= | ||
| // Tests for converting Parquet LIST to Catalyst ArrayType | ||
| // ======================================================= | ||
|
|
||
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@awdavidson I realised that we already released Spark 3.2.3.
Can you make a PR to fix this to 3.2.4?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sure