From fb5a22cb5cb6ea1ceacd89d7fb81bec85b586c37 Mon Sep 17 00:00:00 2001
From: Pavan Lanka <planka@apple.com>
Date: Tue, 5 Oct 2021 10:13:53 -0700
Subject: [PATCH] Fixed #4 - Obtain ORC stripe offsets from writer instead of
 opening the written file for the offsets

---
 .../java/org/apache/iceberg/orc/OrcFileAppender.java  |  7 +++----
 .../org/apache/iceberg/spark/data/TestOrcWrite.java   | 11 +++++++++++
 2 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/orc/src/main/java/org/apache/iceberg/orc/OrcFileAppender.java b/orc/src/main/java/org/apache/iceberg/orc/OrcFileAppender.java
index f407bdcf436c..a72a959ce1ec 100644
--- a/orc/src/main/java/org/apache/iceberg/orc/OrcFileAppender.java
+++ b/orc/src/main/java/org/apache/iceberg/orc/OrcFileAppender.java
@@ -37,7 +37,6 @@
 import org.apache.iceberg.relocated.com.google.common.base.Preconditions;
 import org.apache.iceberg.relocated.com.google.common.collect.Lists;
 import org.apache.orc.OrcFile;
-import org.apache.orc.Reader;
 import org.apache.orc.StripeInformation;
 import org.apache.orc.TypeDescription;
 import org.apache.orc.Writer;
@@ -147,11 +146,11 @@ public long length() {
   @Override
   public List<Long> splitOffsets() {
     Preconditions.checkState(isClosed, "File is not yet closed");
-    try (Reader reader = ORC.newFileReader(file.toInputFile(), conf)) {
-      List<StripeInformation> stripes = reader.getStripes();
+    try {
+      List<StripeInformation> stripes = writer.getStripes();
       return Collections.unmodifiableList(Lists.transform(stripes, StripeInformation::getOffset));
     } catch (IOException e) {
-      throw new RuntimeIOException(e, "Can't close ORC reader %s", file.location());
+      throw new RuntimeIOException(e, "Cannot receive stripe information from writer for %s", file.location());
     }
   }
 
diff --git a/spark/v3.2/spark/src/test/java/org/apache/iceberg/spark/data/TestOrcWrite.java b/spark/v3.2/spark/src/test/java/org/apache/iceberg/spark/data/TestOrcWrite.java
index 1e51a088390e..b9cbedb3de9e 100644
--- a/spark/v3.2/spark/src/test/java/org/apache/iceberg/spark/data/TestOrcWrite.java
+++ b/spark/v3.2/spark/src/test/java/org/apache/iceberg/spark/data/TestOrcWrite.java
@@ -22,11 +22,17 @@
 
 import java.io.File;
 import java.io.IOException;
+import java.util.stream.Collectors;
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.Path;
 import org.apache.iceberg.Files;
 import org.apache.iceberg.Schema;
 import org.apache.iceberg.io.FileAppender;
 import org.apache.iceberg.orc.ORC;
 import org.apache.iceberg.types.Types;
+import org.apache.orc.OrcFile;
+import org.apache.orc.Reader;
+import org.apache.orc.StripeInformation;
 import org.apache.spark.sql.catalyst.InternalRow;
 import org.junit.Assert;
 import org.junit.Rule;
@@ -55,5 +61,10 @@ public void splitOffsets() throws IOException {
     writer.addAll(rows);
     writer.close();
     Assert.assertNotNull("Split offsets not present", writer.splitOffsets());
+    // writer offsets are the same as the ORC reader offsets
+    Reader reader = OrcFile.createReader(new Path(testFile.toURI()), OrcFile.readerOptions(new Configuration()));
+    Assert.assertEquals(reader.getStripes().stream().map(StripeInformation::getOffset).collect(Collectors.toList()),
+        writer.splitOffsets());
+    reader.close();
   }
 }