Guosmilesmile commented on code in PR #15906:
URL: https://github.com/apache/iceberg/pull/15906#discussion_r3101717696


##########
data/src/test/java/org/apache/iceberg/data/BaseFormatModelTests.java:
##########
@@ -628,6 +644,239 @@ void 
testReaderBuilderRecordsPerBatchNotSupported(FileFormat fileFormat) throws
         .isInstanceOf(UnsupportedOperationException.class);
   }
 
+  @ParameterizedTest
+  @FieldSource("FILE_FORMATS")
+  void testDataWriterMetricsCollection(FileFormat fileFormat) throws 
IOException {
+    DataGenerator dataGenerator = new DataGenerators.DefaultSchema();
+    Schema schema = dataGenerator.schema();
+
+    List<Record> genericRecords = dataGenerator.generateRecords();
+    DataFile dataFile = writeGenericRecords(fileFormat, schema, 
genericRecords);
+
+    assertThat(dataFile).isNotNull();
+    assertThat(dataFile.recordCount()).isEqualTo(genericRecords.size());
+
+    assertValueAndNullCounts(
+        fileFormat, schema, genericRecords, dataFile.valueCounts(), 
dataFile.nullValueCounts());
+    assertBounds(
+        fileFormat, schema, genericRecords, dataFile.lowerBounds(), 
dataFile.upperBounds());
+
+    assertThat(dataFile.columnSizes()).isNotNull().isNotEmpty();
+  }
+
+  @ParameterizedTest
+  @FieldSource("FILE_FORMATS")
+  void testDataWriterMetricsWithNoneMode(FileFormat fileFormat) throws 
IOException {
+    DataGenerator dataGenerator = new DataGenerators.DefaultSchema();
+    Schema schema = dataGenerator.schema();
+
+    TestTables.TestTable table =
+        TestTables.create(
+            tableDir,
+            "test",
+            schema,
+            PartitionSpec.unpartitioned(),
+            3,
+            ImmutableMap.of(TableProperties.DEFAULT_WRITE_METRICS_MODE, 
"none"));
+
+    MetricsConfig noneConfig = MetricsConfig.forTable(table);
+    List<Record> genericRecords = dataGenerator.generateRecords();
+    DataFile dataFile = writeGenericRecords(fileFormat, schema, 
genericRecords, noneConfig);
+
+    assertThat(dataFile).isNotNull();
+    assertThat(dataFile.recordCount()).isEqualTo(genericRecords.size());
+
+    assertCountsNull(fileFormat, schema, dataFile.valueCounts(), 
dataFile.nullValueCounts());
+    assertBoundsNull(fileFormat, schema, dataFile.lowerBounds(), 
dataFile.upperBounds());
+  }
+
+  @ParameterizedTest
+  @FieldSource("FILE_FORMATS")
+  void testDataWriterMetricsWithCountsMode(FileFormat fileFormat) throws 
IOException {
+    DataGenerator dataGenerator = new DataGenerators.DefaultSchema();
+    Schema schema = dataGenerator.schema();
+    TestTables.TestTable table =
+        TestTables.create(
+            tableDir,
+            "test",
+            schema,
+            PartitionSpec.unpartitioned(),
+            3,
+            ImmutableMap.of(TableProperties.DEFAULT_WRITE_METRICS_MODE, 
"counts"));
+
+    MetricsConfig countsConfig = MetricsConfig.forTable(table);
+
+    List<Record> genericRecords = dataGenerator.generateRecords();
+    DataFile dataFile = writeGenericRecords(fileFormat, schema, 
genericRecords, countsConfig);
+
+    assertThat(dataFile).isNotNull();
+    assertThat(dataFile.recordCount()).isEqualTo(genericRecords.size());
+
+    // In the counts mode, valueCounts and nullValueCounts should be present, 
while lowerBounds and
+    // upperBounds should be null.
+    assertValueAndNullCounts(
+        fileFormat, schema, genericRecords, dataFile.valueCounts(), 
dataFile.nullValueCounts());
+    assertBoundsNull(fileFormat, schema, dataFile.lowerBounds(), 
dataFile.upperBounds());
+  }
+
+  @ParameterizedTest
+  @FieldSource("FILE_FORMATS")
+  void testEqualityDeleteWriterMetricsCollection(FileFormat fileFormat) throws 
IOException {
+    DataGenerator dataGenerator = new DataGenerators.DefaultSchema();
+    Schema schema = dataGenerator.schema();
+    FileWriterBuilder<EqualityDeleteWriter<Record>, Object> writerBuilder =
+        FormatModelRegistry.equalityDeleteWriteBuilder(fileFormat, 
Record.class, encryptedFile);
+
+    EqualityDeleteWriter<Record> writer =
+        writerBuilder
+            .schema(schema)
+            .spec(PartitionSpec.unpartitioned())
+            .equalityFieldIds(1)
+            .build();
+
+    List<Record> genericRecords = dataGenerator.generateRecords();
+
+    try (writer) {
+      genericRecords.forEach(writer::write);
+    }
+
+    DeleteFile deleteFile = writer.toDeleteFile();
+
+    assertThat(deleteFile).isNotNull();
+    assertThat(deleteFile.recordCount()).isEqualTo(genericRecords.size());
+
+    assertValueAndNullCounts(
+        fileFormat, schema, genericRecords, deleteFile.valueCounts(), 
deleteFile.nullValueCounts());
+    assertBounds(
+        fileFormat, schema, genericRecords, deleteFile.lowerBounds(), 
deleteFile.upperBounds());
+  }
+
+  @ParameterizedTest
+  @FieldSource("FILE_FORMATS")
+  void testPositionDeleteWriterMetricsSingleFile(FileFormat fileFormat) throws 
IOException {
+    // Single file reference: counts are removed but bounds are preserved.
+    List<PositionDelete<T>> deletes =
+        ImmutableList.of(
+            PositionDelete.<T>create().set("d-file-1.parquet", 0L),
+            PositionDelete.<T>create().set("d-file-1.parquet", 5L),
+            PositionDelete.<T>create().set("d-file-1.parquet", 3L));
+
+    writePositionDeletesAndAssertMetrics(fileFormat, deletes, true);
+  }
+
+  @ParameterizedTest
+  @FieldSource("FILE_FORMATS")
+  void testPositionDeleteWriterMetricsMultipleFiles(FileFormat fileFormat) 
throws IOException {
+    // Multiple file references: both counts and bounds are removed.
+    List<PositionDelete<T>> deletes =
+        ImmutableList.of(
+            PositionDelete.<T>create().set("d-file-1.parquet", 0L),
+            PositionDelete.<T>create().set("d-file-1.parquet", 5L),
+            PositionDelete.<T>create().set("d-file-2.parquet", 3L));
+
+    writePositionDeletesAndAssertMetrics(fileFormat, deletes, false);
+  }
+
+  private void writePositionDeletesAndAssertMetrics(
+      FileFormat fileFormat, List<PositionDelete<T>> deletes, boolean 
singleFileReference)

Review Comment:
   Ok, Change to checkBounds.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to