theosib-amazon commented on code in PR #960:
URL: https://github.com/apache/parquet-mr/pull/960#discussion_r929020852
##########
parquet-common/src/main/java/org/apache/parquet/bytes/MultiBufferInputStream.java:
##########
@@ -379,4 +427,120 @@ public void remove() {
second.remove();
}
}
+
+ @Override
+ public byte readByte() throws IOException {
+ return (byte)readUnsignedByte();
+ }
+
+ @Override
+ public int readUnsignedByte() throws IOException {
+ if (current == null) {
+ throw new EOFException();
+ }
+
+ this.position += 1;
+ while (true) {
+ try {
+ return current.get() & 0xFF;
+ } catch (BufferUnderflowException e) {
+ if (!nextBuffer()) {
+ // there are no more buffers
+ throw new EOFException();
+ }
+ }
+ }
+ }
+
+ /**
+ * When reading a short will cross a buffer boundary, read one byte at a
time.
+ * @return a short value
+ * @throws IOException
+ */
+ private int getShortSlow() throws IOException {
+ int c0 = readUnsignedByte();
+ int c1 = readUnsignedByte();
+ return ((c0 << 0) + (c1 << 8));
+ }
+
+ public short readShort() throws IOException {
+ if (current == null) {
+ throw new EOFException();
+ }
+
+ if (current.remaining() >= 2) {
+ // If the whole short can be read from the current buffer, use intrinsics
+ this.position += 2;
+ return current.getShort();
+ } else {
+ // Otherwise get the short one byte at a time
+ return (short)getShortSlow();
+ }
+ }
+
+ public int readUnsignedShort() throws IOException {
+ return readShort() & 0xffff;
+ }
+
+ /**
+ * When reading an int will cross a buffer boundary, read one byte at a time.
+ * @return an int value
+ * @throws IOException
+ */
+ private int getIntSlow() throws IOException {
+ int c0 = readUnsignedByte();
+ int c1 = readUnsignedByte();
+ int c2 = readUnsignedByte();
+ int c3 = readUnsignedByte();
+ return ((c0 << 0) + (c1 << 8)) + ((c2 << 16) + (c3 << 24));
+ }
+
+ @Override
+ public int readInt() throws IOException {
+ if (current == null) {
+ throw new EOFException();
+ }
+
+ if (current.remaining() >= 4) {
+ // If the whole int can be read from the current buffer, use intrinsics
+ this.position += 4;
+ return current.getInt();
+ } else {
+ // Otherwise get the int one byte at a time
+ return getIntSlow();
+ }
+ }
+
+ /**
+ * When reading a long will cross a buffer boundary, read one byte at a time.
+ * @return a long value
+ * @throws IOException
+ */
+ private long getLongSlow() throws IOException {
+ long ch0 = (long)readUnsignedByte() << 0;
+ long ch1 = (long)readUnsignedByte() << 8;
+ long ch2 = (long)readUnsignedByte() << 16;
+ long ch3 = (long)readUnsignedByte() << 24;
+ long ch4 = (long)readUnsignedByte() << 32;
+ long ch5 = (long)readUnsignedByte() << 40;
+ long ch6 = (long)readUnsignedByte() << 48;
+ long ch7 = (long)readUnsignedByte() << 56;
+ return ((ch0 + ch1) + (ch2 + ch3)) + ((ch4 + ch5) + (ch6 + ch7));
+ }
+
+ @Override
+ public long readLong() throws IOException {
+ if (current == null) {
+ throw new EOFException();
+ }
+
+ if (current.remaining() >= 8) {
+ // If the whole short can be read from the current buffer, use intrinsics
+ this.position += 8;
Review Comment:
Done.
--
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: dev-unsubscr...@parquet.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
