sunchao commented on a change in pull request #34611:
URL: https://github.com/apache/spark/pull/34611#discussion_r753433365
##########
File path:
sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/VectorizedPlainValuesReader.java
##########
@@ -53,19 +53,46 @@ public void skip() {
throw new UnsupportedOperationException();
}
+ private void updateCurrentByte() {
+ try {
+ currentByte = (byte) in.read();
+ } catch (IOException e) {
+ throw new ParquetDecodingException("Failed to read a byte", e);
+ }
+ }
+
@Override
public final void readBooleans(int total, WritableColumnVector c, int rowId)
{
- // TODO: properly vectorize this
- for (int i = 0; i < total; i++) {
- c.putBoolean(rowId + i, readBoolean());
+ int i = 0;
+ if (bitOffset > 0) {
+ i = Math.min(8 - bitOffset, total);
+ c.putBooleans(rowId, i, currentByte, bitOffset);
+ bitOffset = (bitOffset + i) & 7;
+ }
+ for (; i + 7 < total; i += 8) {
+ updateCurrentByte();
+ c.putBooleans(rowId + i, currentByte);
+ }
+ if (i < total) {
+ updateCurrentByte();
+ bitOffset = total - i;
+ c.putBooleans(rowId + i, bitOffset, currentByte, 0);
}
}
@Override
public final void skipBooleans(int total) {
- // TODO: properly vectorize this
- for (int i = 0; i < total; i++) {
- readBoolean();
+ int skipBytes = (bitOffset + total) / 8 - 1;
+ bitOffset = (bitOffset + total) & 7;
+ if (skipBytes >= 0) {
+ try {
+ in.skipFully(skipBytes);
+ if (bitOffset > 0) {
+ currentByte = (byte) in.read();
Review comment:
nit: can we call `updateCurrentByte`?
##########
File path:
sql/core/src/main/java/org/apache/spark/sql/execution/vectorized/WritableColumnVector.java
##########
@@ -201,6 +202,25 @@ public WritableColumnVector reserveDictionaryIds(int
capacity) {
*/
public abstract void putBooleans(int rowId, int count, boolean value);
+ /**
+ * Sets bits from [src[srcIndex], src[srcIndex + count]) to [rowId, rowId +
count)
+ * src must contain bit-packed 8 Booleans in the byte.
+ */
+ public void putBooleans(int rowId, int count, byte src, int srcIndex) {
+ byte8[0] = (byte)(src & 1);
+ byte8[1] = (byte)(src >>> 1 & 1);
+ byte8[2] = (byte)(src >>> 2 & 1);
+ byte8[3] = (byte)(src >>> 3 & 1);
+ byte8[4] = (byte)(src >>> 4 & 1);
+ byte8[5] = (byte)(src >>> 5 & 1);
+ byte8[6] = (byte)(src >>> 6 & 1);
+ byte8[7] = (byte)(src >>> 7 & 1);
+ putBytes(rowId, count, byte8, srcIndex);
+ }
+ public void putBooleans(int rowId, byte src) {
Review comment:
I wonder for `OffHeapColumnVector` we can directly operate on the `data`
too using `Platform.putByte`
##########
File path:
sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/VectorizedPlainValuesReader.java
##########
@@ -53,19 +53,46 @@ public void skip() {
throw new UnsupportedOperationException();
}
+ private void updateCurrentByte() {
+ try {
+ currentByte = (byte) in.read();
+ } catch (IOException e) {
+ throw new ParquetDecodingException("Failed to read a byte", e);
+ }
+ }
+
@Override
public final void readBooleans(int total, WritableColumnVector c, int rowId)
{
- // TODO: properly vectorize this
- for (int i = 0; i < total; i++) {
- c.putBoolean(rowId + i, readBoolean());
+ int i = 0;
+ if (bitOffset > 0) {
+ i = Math.min(8 - bitOffset, total);
+ c.putBooleans(rowId, i, currentByte, bitOffset);
+ bitOffset = (bitOffset + i) & 7;
+ }
+ for (; i + 7 < total; i += 8) {
+ updateCurrentByte();
+ c.putBooleans(rowId + i, currentByte);
+ }
+ if (i < total) {
+ updateCurrentByte();
+ bitOffset = total - i;
+ c.putBooleans(rowId + i, bitOffset, currentByte, 0);
}
}
@Override
public final void skipBooleans(int total) {
- // TODO: properly vectorize this
- for (int i = 0; i < total; i++) {
- readBoolean();
+ int skipBytes = (bitOffset + total) / 8 - 1;
Review comment:
nit: maybe name to `numBytesToSkip`
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
