From 2c56424858d7832ba147f3d7dfd8ab07a51a38d1 Mon Sep 17 00:00:00 2001
From: Costas Zarifis <costas.zarifis@databricks.com>
Date: Mon, 28 Sep 2026 07:52:20 +0000
Subject: [PATCH] GH-3829: Preserve column path identity when copying row
 groups

---
 .../parquet/hadoop/ParquetFileWriter.java     | 15 +++++---
 .../hadoop/TestParquetWriterAppendBlocks.java | 37 +++++++++++++++++++
 2 files changed, 46 insertions(+), 6 deletions(-)

diff --git a/parquet-hadoop/src/main/java/org/apache/parquet/hadoop/ParquetFileWriter.java b/parquet-hadoop/src/main/java/org/apache/parquet/hadoop/ParquetFileWriter.java
index 82f4577b83..bb6e0e8668 100644
--- a/parquet-hadoop/src/main/java/org/apache/parquet/hadoop/ParquetFileWriter.java
+++ b/parquet-hadoop/src/main/java/org/apache/parquet/hadoop/ParquetFileWriter.java
@@ -36,6 +36,7 @@
 import java.util.Map.Entry;
 import java.util.Optional;
 import java.util.Set;
+import java.util.stream.Collectors;
 import java.util.zip.CRC32;
 import org.apache.hadoop.conf.Configuration;
 import org.apache.hadoop.fs.FSDataInputStream;
@@ -1702,21 +1703,21 @@ public void appendRowGroup(SeekableInputStream from, BlockMetaData rowGroup, boo
     withAbortOnFailure(() -> {
       startBlock(rowGroup.getRowCount());
 
-      Map<String, ColumnChunkMetaData> columnsToCopy = new HashMap<String, ColumnChunkMetaData>();
+      Map<ColumnPath, ColumnChunkMetaData> columnsToCopy = new HashMap<ColumnPath, ColumnChunkMetaData>();
       for (ColumnChunkMetaData chunk : rowGroup.getColumns()) {
-        columnsToCopy.put(chunk.getPath().toDotString(), chunk);
+        columnsToCopy.put(chunk.getPath(), chunk);
       }
 
       List<ColumnChunkMetaData> columnsInOrder = new ArrayList<ColumnChunkMetaData>();
 
       for (ColumnDescriptor descriptor : schema.getColumns()) {
-        String path = ColumnPath.get(descriptor.getPath()).toDotString();
+        ColumnPath path = ColumnPath.get(descriptor.getPath());
         ColumnChunkMetaData chunk = columnsToCopy.remove(path);
         if (chunk != null) {
           columnsInOrder.add(chunk);
         } else {
-          throw new IllegalArgumentException(
-              String.format("Missing column '%s', cannot copy row group: %s", path, rowGroup));
+          throw new IllegalArgumentException(String.format(
+              "Missing column '%s', cannot copy row group: %s", path.toDotString(), rowGroup));
         }
       }
 
@@ -1724,7 +1725,9 @@ public void appendRowGroup(SeekableInputStream from, BlockMetaData rowGroup, boo
       if (!dropColumns && !columnsToCopy.isEmpty()) {
         throw new IllegalArgumentException(String.format(
             "Columns cannot be copied (missing from target schema): %s",
-            String.join(", ", columnsToCopy.keySet())));
+            columnsToCopy.keySet().stream()
+                .map(ColumnPath::toDotString)
+                .collect(Collectors.joining(", "))));
       }
 
       // copy the data for all chunks
diff --git a/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestParquetWriterAppendBlocks.java b/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestParquetWriterAppendBlocks.java
index 3eb6128209..9a9a8e0d8d 100644
--- a/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestParquetWriterAppendBlocks.java
+++ b/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestParquetWriterAppendBlocks.java
@@ -140,6 +140,43 @@ public void testBasicBehavior() throws IOException {
     assertThat(expected).as("All records should be present").isEmpty();
   }
 
+  @Test
+  public void testAppendKeepsCollidingDotStringPathsDistinct() throws IOException {
+    MessageType schema = Types.buildMessage()
+        .required(BINARY)
+        .as(UTF8)
+        .named("a.b")
+        .requiredGroup()
+        .required(BINARY)
+        .as(UTF8)
+        .named("b")
+        .named("a")
+        .named("AppendCollisionTest");
+    SimpleGroupFactory factory = new SimpleGroupFactory(schema);
+    Group expected = factory.newGroup().append("a.b", "top-level");
+    expected.addGroup("a").append("b", "nested");
+
+    Path source = newTemp();
+    try (ParquetWriter<Group> sourceWriter =
+        ExampleParquetWriter.builder(source).withType(schema).build()) {
+      sourceWriter.write(expected);
+    }
+
+    Path copied = newTemp();
+    ParquetFileWriter copyWriter = new ParquetFileWriter(CONF, schema, copied);
+    copyWriter.start();
+    copyWriter.appendFile(CONF, source);
+    copyWriter.end(EMPTY_METADATA);
+
+    try (ParquetReader<Group> reader =
+        ParquetReader.builder(new GroupReadSupport(), copied).build()) {
+      Group actual = reader.read();
+      assertThat(actual.getString("a.b", 0)).isEqualTo("top-level");
+      assertThat(actual.getGroup("a", 0).getString("b", 0)).isEqualTo("nested");
+      assertThat(reader.read()).isNull();
+    }
+  }
+
   /**
    * This test is similar to {@link #testBasicBehavior()} only that it uses static files generated by a previous release
    * (1.11.1). This test is to validate the fix of PARQUET-2027.