Core: Cache dropStats in ManifestReader iterator (apache#5836)

alok123t · Nov 9, 2022 · a50a295 · a50a295
1 parent 305e320
commit a50a295
Show file tree

Hide file tree

Showing 2 changed files with 21 additions and 2 deletions.
diff --git a/core/src/main/java/org/apache/iceberg/ManifestReader.java b/core/src/main/java/org/apache/iceberg/ManifestReader.java
@@ -272,8 +272,8 @@ CloseableIterable<ManifestEntry<F>> liveEntries() {
   /** @return an Iterator of DataFile. Makes defensive copies of files before returning */
   @Override
   public CloseableIterator<F> iterator() {
-    return CloseableIterable.transform(liveEntries(), e -> e.file().copy(!dropStats(columns)))
-        .iterator();
+    boolean dropStats = dropStats(columns);
+    return CloseableIterable.transform(liveEntries(), e -> e.file().copy(!dropStats)).iterator();
   }
 
   private static Schema projection(

diff --git a/core/src/test/java/org/apache/iceberg/TestManifestReaderStats.java b/core/src/test/java/org/apache/iceberg/TestManifestReaderStats.java
@@ -147,6 +147,25 @@ public void testReadIteratorWithFilterAndSelectStatsIncludesFullStats() throws I
     }
   }
 
+  @Test
+  public void testReadIteratorWithProjectStats() throws IOException {
+    ManifestFile manifest = writeManifest(1000L, FILE);
+    try (ManifestReader<DataFile> reader =
+        ManifestFiles.read(manifest, FILE_IO)
+            .project(new Schema(ImmutableList.of(DataFile.FILE_PATH, DataFile.VALUE_COUNTS)))) {
+      DataFile entry = reader.iterator().next();
+
+      Assert.assertEquals(FILE_PATH, entry.path());
+      Assert.assertEquals(VALUE_COUNT, entry.valueCounts());
+      Assert.assertNull(entry.columnSizes());
+      Assert.assertNull(entry.nullValueCounts());
+      Assert.assertNull(entry.nanValueCounts());
+      Assert.assertNull(entry.lowerBounds());
+      Assert.assertNull(entry.upperBounds());
+      assertNullRecordCount(entry);
+    }
+  }
+
   @Test
   public void testReadEntriesWithSelectNotProjectStats() throws IOException {
     ManifestFile manifest = writeManifest(1000L, FILE);