PARQUET-2139: set metadata offset to 0 since it is not written at all

apache · Jun 3, 2024 · f7fb556 · f7fb556
1 parent 8b91d6c
commit f7fb556
Showing 1 changed file with 3 additions and 2 deletions.
diff --git a/...et-hadoop/src/main/java/org/apache/parquet/format/converter/ParquetMetadataConverter.java b/...et-hadoop/src/main/java/org/apache/parquet/format/converter/ParquetMetadataConverter.java
@@ -534,8 +534,9 @@ private void addRowGroup(
     int columnOrdinal = -1;
     ByteArrayOutputStream tempOutStream = null;
     for (ColumnChunkMetaData columnMetaData : columns) {
-      ColumnChunk columnChunk =
-          new ColumnChunk(columnMetaData.getFirstDataPageOffset()); // verify this is the right offset
+      // There is no ColumnMetaData written after the chunk data, so set the ColumnChunk
+      // file_offset to 0
+      ColumnChunk columnChunk = new ColumnChunk(0);
       columnChunk.file_path = block.getPath(); // they are in the same file for now
       InternalColumnEncryptionSetup columnSetup = null;
       boolean writeCryptoMetadata = false;