(incubator-paimon) branch master updated: [orc] Optimize ensureSize for orc ColumnVector (#2845)

[lzljs3620320]

This is an automated email from the ASF dual-hosted git repository.

lzljs3620320 pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-paimon.git


The following commit(s) were added to refs/heads/master by this push:
     new f91a0341e [orc] Optimize ensureSize for orc ColumnVector (#2845)
f91a0341e is described below

commit f91a0341e4e3fcb37d5cab2a6902209b7b43d718
Author: tsreaper <tsreape...@gmail.com>
AuthorDate: Fri Feb 2 20:46:22 2024 +0800

    [orc] Optimize ensureSize for orc ColumnVector (#2845)
---
 .../format/orc/writer/RowDataVectorizer.java       | 25 ++++++++++++++++------
 1 file changed, 19 insertions(+), 6 deletions(-)

diff --git 
a/paimon-format/src/main/java/org/apache/paimon/format/orc/writer/RowDataVectorizer.java
 
b/paimon-format/src/main/java/org/apache/paimon/format/orc/writer/RowDataVectorizer.java
index 5349935ca..b05b5e208 100644
--- 
a/paimon-format/src/main/java/org/apache/paimon/format/orc/writer/RowDataVectorizer.java
+++ 
b/paimon-format/src/main/java/org/apache/paimon/format/orc/writer/RowDataVectorizer.java
@@ -194,8 +194,10 @@ public class RowDataVectorizer extends 
Vectorizer<InternalRow> {
         listColumnVector.lengths[rowId] = arrayData.size();
         listColumnVector.offsets[rowId] = listColumnVector.childCount;
         listColumnVector.childCount += listColumnVector.lengths[rowId];
-        listColumnVector.child.ensureSize(
-                listColumnVector.childCount, listColumnVector.offsets[rowId] 
!= 0);
+        ensureSize(
+                listColumnVector.child,
+                listColumnVector.childCount,
+                listColumnVector.offsets[rowId] != 0);
 
         InternalRow convertedRowData = convert(arrayData, 
arrayType.getElementType());
         for (int i = 0; i < arrayData.size(); i++) {
@@ -221,10 +223,14 @@ public class RowDataVectorizer extends 
Vectorizer<InternalRow> {
         mapColumnVector.lengths[rowId] = mapData.size();
         mapColumnVector.offsets[rowId] = mapColumnVector.childCount;
         mapColumnVector.childCount += mapColumnVector.lengths[rowId];
-        mapColumnVector.keys.ensureSize(
-                mapColumnVector.childCount, mapColumnVector.offsets[rowId] != 
0);
-        mapColumnVector.values.ensureSize(
-                mapColumnVector.childCount, mapColumnVector.offsets[rowId] != 
0);
+        ensureSize(
+                mapColumnVector.keys,
+                mapColumnVector.childCount,
+                mapColumnVector.offsets[rowId] != 0);
+        ensureSize(
+                mapColumnVector.values,
+                mapColumnVector.childCount,
+                mapColumnVector.offsets[rowId] != 0);
 
         InternalRow convertedKeyRowData = convert(keyArray, 
mapType.getKeyType());
         InternalRow convertedValueRowData = convert(valueArray, 
mapType.getValueType());
@@ -258,6 +264,13 @@ public class RowDataVectorizer extends 
Vectorizer<InternalRow> {
         }
     }
 
+    private static void ensureSize(ColumnVector cv, int size, boolean 
preserveData) {
+        int currentLength = cv.isNull.length;
+        if (currentLength < size) {
+            cv.ensureSize(Math.max(currentLength * 2, size), preserveData);
+        }
+    }
+
     /**
      * Converting ArrayData to RowData for calling {@link 
RowDataVectorizer#setColumn(int,
      * ColumnVector, DataType, InternalRow, int)} recursively with array.