bit more cleanup

amogh-jahagirdar · amogh-jahagirdar · commit 110c80fcf735 · 2025-04-30T16:29:34.000-06:00
diff --git a/arrow/src/main/java/org/apache/iceberg/arrow/vectorized/VectorizedArrowReader.java b/arrow/src/main/java/org/apache/iceberg/arrow/vectorized/VectorizedArrowReader.java
@@ -628,12 +628,11 @@ public VectorHolder read(VectorHolder reuse, int numValsToRead) {
       VectorHolder ids = idReader.read(null, numValsToRead);
       BigIntVector vec = allocateBigIntVector(ROW_ID_ARROW_FIELD, numValsToRead);
       ArrowBuf dataBuffer = vec.getDataBuffer();
-      boolean isNullReader = ids.vector() == null;
       ArrowVectorAccessor<?, String, ?, ?> idsAccessor =
-          isNullReader ? null : ArrowVectorAccessors.getVectorAccessor(ids);
+          ids.vector() == null ? null : ArrowVectorAccessors.getVectorAccessor(ids);
       for (int i = 0; i < numValsToRead; i += 1) {
         long bufferOffset = (long) i * Long.BYTES;
-        if (isNullReader || ids.nullabilityHolder().isNullAt(i) == 1) {
+        if (idsAccessor == null || isNull(ids, i)) {
           long rowId = firstRowId + (Long) positions.getObject(i);
           dataBuffer.setLong(bufferOffset, rowId);
         } else {
@@ -688,14 +687,14 @@ public VectorHolder read(VectorHolder reuse, int numValsToRead) {
       BigIntVector vec = allocateBigIntVector(LAST_UPDATED_SEQ, numValsToRead);
       ArrowBuf dataBuffer = vec.getDataBuffer();
       VectorHolder seqNumbers = seqReader.read(null, numValsToRead);
-      ArrowVectorAccessor<?, String, ?, ?> accessor =
+      ArrowVectorAccessor<?, String, ?, ?> seqAccessor =
           seqNumbers.vector() == null ? null : ArrowVectorAccessors.getVectorAccessor(seqNumbers);
       for (int i = 0; i < numValsToRead; i += 1) {
         long bufferOffset = (long) i * Long.BYTES;
-        if (seqNumbers.vector() == null || seqNumbers.nullabilityHolder().isNullAt(i) == 1) {
+        if (seqAccessor == null || isNull(seqNumbers, i)) {
           dataBuffer.setLong(bufferOffset, lastUpdatedSeq);
         } else {
-          long materializedSeqNumber = accessor.getLong(i);
+          long materializedSeqNumber = seqAccessor.getLong(i);
           dataBuffer.setLong(bufferOffset, materializedSeqNumber);
         }
       }
@@ -725,6 +724,10 @@ public void close() {
     }
   }
 
+  private static boolean isNull(VectorHolder holder, int index) {
+    return holder.nullabilityHolder().isNullAt(index) == 1;
+  }
+
   private static BigIntVector allocateBigIntVector(Field field, int valueCount) {
     BigIntVector vector = (BigIntVector) field.createVector(ArrowAllocation.rootAllocator());
     vector.allocateNew(valueCount);
diff --git a/spark/v3.5/spark/src/test/java/org/apache/iceberg/spark/data/GenericsHelpers.java b/spark/v3.5/spark/src/test/java/org/apache/iceberg/spark/data/GenericsHelpers.java
@@ -77,11 +77,11 @@ public static void assertEqualsBatch(
       Iterator<Record> expectedRecords,
       ColumnarBatch batch,
       Map<Integer, Object> idToConstant,
-      Integer numRowsAlreadyRead) {
+      Integer batchFirstRowPos) {
     for (int rowPos = 0; rowPos < batch.numRows(); rowPos++) {
       InternalRow row = batch.getRow(rowPos);
       Record expectedRecord = expectedRecords.next();
-      assertEqualsUnsafe(struct, expectedRecord, row, idToConstant, numRowsAlreadyRead + rowPos);
+      assertEqualsUnsafe(struct, expectedRecord, row, idToConstant, batchFirstRowPos + rowPos);
     }
   }
 
diff --git a/spark/v3.5/spark/src/test/java/org/apache/iceberg/spark/data/TestHelpers.java b/spark/v3.5/spark/src/test/java/org/apache/iceberg/spark/data/TestHelpers.java
@@ -50,7 +50,6 @@
 import org.apache.iceberg.FileContent;
 import org.apache.iceberg.FileScanTask;
 import org.apache.iceberg.ManifestFile;
-import org.apache.iceberg.MetadataColumns;
 import org.apache.iceberg.Schema;
 import org.apache.iceberg.Snapshot;
 import org.apache.iceberg.Table;

Original file line number	Diff line number	Diff line change
`@@ -77,11 +77,11 @@ public static void assertEqualsBatch(`
`77`	`77`	`Iterator<Record> expectedRecords,`
`78`	`78`	`ColumnarBatch batch,`
`79`	`79`	`Map<Integer, Object> idToConstant,`
`80`		`- Integer numRowsAlreadyRead) {`
	`80`	`+ Integer batchFirstRowPos) {`
`81`	`81`	`for (int rowPos = 0; rowPos < batch.numRows(); rowPos++) {`
`82`	`82`	`InternalRow row = batch.getRow(rowPos);`
`83`	`83`	`Record expectedRecord = expectedRecords.next();`
`84`		`- assertEqualsUnsafe(struct, expectedRecord, row, idToConstant, numRowsAlreadyRead + rowPos);`
	`84`	`+ assertEqualsUnsafe(struct, expectedRecord, row, idToConstant, batchFirstRowPos + rowPos);`
`85`	`85`	`}`
`86`	`86`	`}`
`87`	`87`