This is an automated email from the ASF dual-hosted git repository.
JingsongLi pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/paimon.git
The following commit(s) were added to refs/heads/master by this push:
new 8d67f8c6f5 [format] Fix positional schema evolution for ORC reads
(#9464)
8d67f8c6f5 is described below
commit 8d67f8c6f5a25d00de4baa510c7cb3a1a7207c34
Author: Arnav Balyan <[email protected]>
AuthorDate: Wed Sep 2 13:50:35 2026 +0530
[format] Fix positional schema evolution for ORC reads (#9464)
---
.../apache/paimon/format/orc/OrcReaderFactory.java | 6 ++
.../orc/OrcPositionalSchemaEvolutionTest.java | 95 ++++++++++++++++++++++
2 files changed, 101 insertions(+)
diff --git
a/paimon-format/src/main/java/org/apache/paimon/format/orc/OrcReaderFactory.java
b/paimon-format/src/main/java/org/apache/paimon/format/orc/OrcReaderFactory.java
index 39abe58541..32daef9c70 100644
---
a/paimon-format/src/main/java/org/apache/paimon/format/orc/OrcReaderFactory.java
+++
b/paimon-format/src/main/java/org/apache/paimon/format/orc/OrcReaderFactory.java
@@ -437,6 +437,7 @@ public class OrcReaderFactory implements
FormatReaderFactory {
getOffsetAndLengthForSplit(splitStart, splitLength,
orcReader.getStripes());
// create ORC row reader configuration
+ boolean forcePositionalEvolution =
OrcConf.FORCE_POSITIONAL_EVOLUTION.getBoolean(conf);
org.apache.orc.Reader.Options options =
new org.apache.orc.Reader.Options()
.schema(schema)
@@ -444,8 +445,13 @@ public class OrcReaderFactory implements
FormatReaderFactory {
.useZeroCopy(OrcConf.USE_ZEROCOPY.getBoolean(conf))
.skipCorruptRecords(OrcConf.SKIP_CORRUPT_DATA.getBoolean(conf))
.tolerateMissingSchema(OrcConf.TOLERATE_MISSING_SCHEMA.getBoolean(conf))
+ .forcePositionalEvolution(forcePositionalEvolution)
.isSchemaEvolutionCaseAware(
OrcConf.IS_SCHEMA_EVOLUTION_CASE_SENSITIVE.getBoolean(conf));
+ if (forcePositionalEvolution) {
+ options.positionalEvolutionLevel(
+ OrcConf.FORCE_POSITIONAL_EVOLUTION_LEVEL.getInt(conf));
+ }
if (!conjunctPredicates.isEmpty() && !deletionVectorsEnabled &&
selection == null) {
// row group filter push down will make row number change
incorrect
// so deletion vectors mode and bitmap index cannot work with
row group push down
diff --git
a/paimon-format/src/test/java/org/apache/paimon/format/orc/OrcPositionalSchemaEvolutionTest.java
b/paimon-format/src/test/java/org/apache/paimon/format/orc/OrcPositionalSchemaEvolutionTest.java
new file mode 100644
index 0000000000..2efb610ef1
--- /dev/null
+++
b/paimon-format/src/test/java/org/apache/paimon/format/orc/OrcPositionalSchemaEvolutionTest.java
@@ -0,0 +1,95 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.format.orc;
+
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.format.FormatReaderContext;
+import org.apache.paimon.fs.Path;
+import org.apache.paimon.fs.local.LocalFileIO;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.RowType;
+
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.hive.ql.exec.vector.LongColumnVector;
+import org.apache.hadoop.hive.ql.exec.vector.VectorizedRowBatch;
+import org.apache.orc.OrcConf;
+import org.apache.orc.OrcFile;
+import org.apache.orc.TypeDescription;
+import org.apache.orc.Writer;
+import org.junit.jupiter.api.Test;
+import org.junit.jupiter.api.io.TempDir;
+
+import java.io.File;
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.List;
+
+import static org.assertj.core.api.Assertions.assertThat;
+
+/** Test positional schema evolution in {@link OrcReaderFactory}. */
+class OrcPositionalSchemaEvolutionTest {
+
+ @TempDir File folder;
+
+ @Test
+ void testPositionalSchemaEvolution() throws Exception {
+ Configuration conf = new Configuration(false);
+ OrcConf.FORCE_POSITIONAL_EVOLUTION.setBoolean(conf, true);
+
+ assertThat(readOrcFile(conf)).containsExactly(42);
+ }
+
+ @Test
+ void testPositionalSchemaEvolutionDisabledByDefault() throws Exception {
+ assertThat(readOrcFile(new
Configuration(false))).containsExactly((Integer) null);
+ }
+
+ private List<Integer> readOrcFile(Configuration conf) throws Exception {
+ Path path = writeOrcFile();
+ RowType readType = RowType.builder().field("id",
DataTypes.INT()).build();
+ OrcReaderFactory factory =
+ new OrcReaderFactory(conf, readType, Collections.emptyList(),
1024, false, false);
+ LocalFileIO fileIO = new LocalFileIO();
+ List<Integer> values = new ArrayList<>();
+ try (RecordReader<InternalRow> reader =
+ factory.createReader(
+ new FormatReaderContext(
+ fileIO, path, fileIO.getFileSize(path), null,
null))) {
+ reader.forEachRemaining(row -> values.add(row.isNullAt(0) ? null :
row.getInt(0)));
+ }
+
+ return values;
+ }
+
+ private Path writeOrcFile() throws Exception {
+ Path path = new Path(folder.getPath(), "legacy.orc");
+ TypeDescription schema =
TypeDescription.fromString("struct<_col0:int>");
+ try (Writer writer =
+ OrcFile.createWriter(
+ new org.apache.hadoop.fs.Path(path.toString()),
+ OrcFile.writerOptions(new
Configuration(false)).setSchema(schema))) {
+ VectorizedRowBatch batch = schema.createRowBatch();
+ ((LongColumnVector) batch.cols[0]).vector[0] = 42;
+ batch.size = 1;
+ writer.addRowBatch(batch);
+ }
+ return path;
+ }
+}
