Skip to content

[SPARK-11694][SQL] Parquet logical types are not being tested properly #9660

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
HyukjinKwon wants to merge 7 commits into from
Closed

[SPARK-11694][SQL] Parquet logical types are not being tested properly #9660

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
870a37a
[SPARK-11694][SQL] Parquet logical types are not being tested properly
HyukjinKwon Nov 12, 2015
9a5c2a3
[SPARK-11694][SQL] Remove mistakenly added characters.
HyukjinKwon Nov 12, 2015
02f3ef9
[SPARK-11694][SQL] Overload writeMetaFile() and Add all the logical t…
HyukjinKwon Nov 13, 2015
22e1374
[SPARK-11694][SQL] Shorten line less than 100 characters
HyukjinKwon Nov 13, 2015
8e3e470
[SPARK-11694][SQL] Add a import manually
HyukjinKwon Nov 13, 2015
3c7277a
[SPARK-11694][SQL] Add an import for `ParquetMetadata`
HyukjinKwon Nov 13, 2015
ce36be0
[SPARK-11694][SQL] Add an import for `Collections`
HyukjinKwon Nov 13, 2015
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
39 changes: 30 additions & 9 deletions ...re/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetIOSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -91,6 +91,33 @@ class ParquetIOSuite extends QueryTest with ParquetTest with SharedSQLContext {
}
}

test("SPARK-11694 Parquet logical types are not being tested properly") {
val parquetSchema = MessageTypeParser.parseMessageType(
"""message root {
| required int32 a(INT_8);
| required int32 b(INT_16);
| required int32 c(DATE);
| required int32 d(DECIMAL(1,0));
| required int64 e(DECIMAL(10,0));
| required binary f(UTF8);
| required binary g(ENUM);
| required binary h(DECIMAL(32,0));
| required fixed_len_byte_array(32) i(DECIMAL(32,0));
|}
""".stripMargin)
Copy link
Contributor

@liancheng liancheng Nov 12, 2015

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we add all Parquet logical types that Spark supports here for completeness? For example:

fixed_len_byte_array f0(DECIMAL(d, s))
binary f1(UTF8);


val expectedSparkTypes = Seq(ByteType, ShortType, DateType, DecimalType(1, 0),
DecimalType(10, 0), StringType, StringType, DecimalType(32, 0), DecimalType(32, 0))

withTempPath { location =>
val path = new Path(location.getCanonicalPath)
val conf = sparkContext.hadoopConfiguration
writeMetadata(parquetSchema, path, conf)
val sparkTypes = sqlContext.read.parquet(path.toString).schema.map(_.dataType)
assert(sparkTypes === expectedSparkTypes)
}
}

test("string") {
val data = (1 to 4).map(i => Tuple1(i.toString))
// Property spark.sql.parquet.binaryAsString shouldn't affect Parquet files written by Spark SQL
Expand Down Expand Up @@ -350,16 +377,10 @@ class ParquetIOSuite extends QueryTest with ParquetTest with SharedSQLContext {
""".stripMargin)

withTempPath { location =>
val extraMetadata = Collections.singletonMap(
CatalystReadSupport.SPARK_METADATA_KEY, sparkSchema.toString)
val fileMetadata = new FileMetaData(parquetSchema, extraMetadata, "Spark")
val extraMetadata = Map(CatalystReadSupport.SPARK_METADATA_KEY -> sparkSchema.toString)
val path = new Path(location.getCanonicalPath)

ParquetFileWriter.writeMetadataFile(
sparkContext.hadoopConfiguration,
path,
Collections.singletonList(
new Footer(path, new ParquetMetadata(fileMetadata, Collections.emptyList()))))
val conf = sparkContext.hadoopConfiguration
writeMetadata(parquetSchema, path, conf, extraMetadata)

assertResult(sqlContext.read.parquet(path.toString).schema) {
StructType(
Expand Down
17 changes: 17 additions & 0 deletions sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetTest.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,8 @@ package org.apache.spark.sql.execution.datasources.parquet

import java.io.File

import org.apache.parquet.schema.MessageType

import scala.collection.JavaConverters._
import scala.reflect.ClassTag
import scala.reflect.runtime.universe.TypeTag
Expand Down Expand Up @@ -117,6 +119,21 @@ private[sql] trait ParquetTest extends SQLTestUtils {
ParquetFileWriter.writeMetadataFile(configuration, path, Seq(footer).asJava)
}

/**
* This is an overloaded version of `writeMetadata` above to allow writing customized
* Parquet schema.
*/
protected def writeMetadata(
parquetSchema: MessageType, path: Path, configuration: Configuration,
extraMetadata: Map[String, String] = Map.empty[String, String]): Unit = {
val extraMetadataAsJava = extraMetadata.asJava
val createdBy = s"Apache Spark ${org.apache.spark.SPARK_VERSION}"
val fileMetadata = new FileMetaData(parquetSchema, extraMetadataAsJava, createdBy)
val parquetMetadata = new ParquetMetadata(fileMetadata, Seq.empty[BlockMetaData].asJava)
val footer = new Footer(path, parquetMetadata)
ParquetFileWriter.writeMetadataFile(configuration, path, Seq(footer).asJava)
}

protected def readAllFootersWithoutSummaryFiles(
path: Path, configuration: Configuration): Seq[Footer] = {
val fs = path.getFileSystem(configuration)
Expand Down