(parquet-java) branch master updated: GH-3188: Set the global configured column stats enable flag to default (#3189)

[gangwu]

This is an automated email from the ASF dual-hosted git repository.

gangwu pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/parquet-java.git


The following commit(s) were added to refs/heads/master by this push:
     new 00b6bab59 GH-3188: Set the global configured column stats enable flag 
to default (#3189)
00b6bab59 is described below

commit 00b6bab597d55e5fc7af7e19f951794c31d87577
Author: huaxiangsun <huaxiang...@apache.org>
AuthorDate: Fri Apr 11 18:00:49 2025 -0700

    GH-3188: Set the global configured column stats enable flag to default 
(#3189)
---
 .../src/main/java/org/apache/parquet/column/ParquetProperties.java | 1 +
 .../src/test/java/org/apache/parquet/hadoop/TestParquetWriter.java | 7 ++++++-
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git 
a/parquet-column/src/main/java/org/apache/parquet/column/ParquetProperties.java 
b/parquet-column/src/main/java/org/apache/parquet/column/ParquetProperties.java
index 9aaef4b3c..cb5931581 100644
--- 
a/parquet-column/src/main/java/org/apache/parquet/column/ParquetProperties.java
+++ 
b/parquet-column/src/main/java/org/apache/parquet/column/ParquetProperties.java
@@ -708,6 +708,7 @@ public class ParquetProperties {
     }
 
     public Builder withStatisticsEnabled(boolean enabled) {
+      this.statistics.withDefaultValue(enabled);
       this.statisticsEnabled = enabled;
       return this;
     }
diff --git 
a/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestParquetWriter.java 
b/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestParquetWriter.java
index c8e8f71a9..739aa85d2 100644
--- 
a/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestParquetWriter.java
+++ 
b/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestParquetWriter.java
@@ -543,7 +543,7 @@ public class TestParquetWriter {
   }
 
   @Test
-  public void testSizeStatisticsControl() throws Exception {
+  public void testSizeStatisticsAndStatisticsControl() throws Exception {
     MessageType schema = Types.buildMessage()
         .required(BINARY)
         .named("string_field")
@@ -568,6 +568,7 @@ public class TestParquetWriter {
     try (ParquetWriter<Group> writer = ExampleParquetWriter.builder(path)
         .withType(schema)
         .withSizeStatisticsEnabled(false)
+        .withStatisticsEnabled(false) // Disable column statistics globally
         .build()) {
       writer.write(group);
     }
@@ -576,6 +577,7 @@ public class TestParquetWriter {
       // Verify size statistics are disabled globally
       for (BlockMetaData block : reader.getFooter().getBlocks()) {
         for (ColumnChunkMetaData column : block.getColumns()) {
+          assertTrue(column.getStatistics().isEmpty()); // Make sure there is 
no column statistics
           assertNull(column.getSizeStatistics());
         }
       }
@@ -589,6 +591,7 @@ public class TestParquetWriter {
         .withType(schema)
         .withSizeStatisticsEnabled(true) // enable globally
         .withSizeStatisticsEnabled("boolean_field", false) // disable for 
specific column
+        .withStatisticsEnabled("boolean_field", false) // disable column 
statistics
         .build()) {
       writer.write(group);
     }
@@ -599,8 +602,10 @@ public class TestParquetWriter {
         for (ColumnChunkMetaData column : block.getColumns()) {
           if (column.getPath().toDotString().equals("boolean_field")) {
             assertNull(column.getSizeStatistics());
+            assertTrue(column.getStatistics().isEmpty());
           } else {
             assertTrue(column.getSizeStatistics().isValid());
+            assertFalse(column.getStatistics().isEmpty());
           }
         }
       }