viirya commented on a change in pull request #3250: URL: https://github.com/apache/hadoop/pull/3250#discussion_r682860242
########## File path: hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/zlib/BuiltInGzipCompressor.java ########## @@ -0,0 +1,270 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.hadoop.io.compress.zlib; + +import java.io.IOException; +import java.util.zip.Checksum; +import java.util.zip.Deflater; +import java.util.zip.GZIPOutputStream; + +import org.apache.hadoop.conf.Configuration; +import org.apache.hadoop.io.compress.Compressor; +import org.apache.hadoop.io.compress.DoNotPool; +import org.apache.hadoop.util.DataChecksum; + +/** + * A {@link Compressor} based on the popular gzip compressed file format. + * http://www.gzip.org/ + */ +@DoNotPool +public class BuiltInGzipCompressor implements Compressor { + + /** + * Fixed ten-byte gzip header. See {@link GZIPOutputStream}'s source for + * details. + */ + private static final byte[] GZIP_HEADER = new byte[]{ + 0x1f, (byte) 0x8b, 0x08, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}; + + private byte[] GZIP_TRAILER = new byte[]{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}; + + private final int GZIP_HEADER_LEN = 10; + private final int GZIP_TRAILER_LEN = 8; + + private Deflater deflater; + + private int headerOff = 0; + private int trailerOff = 0; + + private byte[] userBuf = null; + private int userBufOff = 0; + private int userBufLen = 0; + + private int headerBytesWritten = 0; + private int trailerBytesWritten = 0; + + private int currentBufLen = 0; + + private final Checksum crc = DataChecksum.newCrc32(); + + private BuiltInGzipDecompressor.GzipStateLabel state; + + public BuiltInGzipCompressor(Configuration conf) { init(conf); } + + @Override + public boolean finished() { + // Only if the trailer is also written, it is thought as finished. + return deflater.finished() && state == BuiltInGzipDecompressor.GzipStateLabel.FINISHED; + } + + @Override + public boolean needsInput() { + if (state == BuiltInGzipDecompressor.GzipStateLabel.INFLATE_STREAM) { + return deflater.needsInput(); + } + + return (state != BuiltInGzipDecompressor.GzipStateLabel.FINISHED); + } + + @Override + public int compress(byte[] b, int off, int len) throws IOException { + int numAvailBytes = 0; + + // If we are not within uncompressed data yet, output the header. + if (state != BuiltInGzipDecompressor.GzipStateLabel.INFLATE_STREAM) { + if (userBufLen <= 0) { + return numAvailBytes; + } + + int outputHeaderSize = writeHeader(b, off, len); + headerBytesWritten += outputHeaderSize; + + // Completes header output. + if (headerOff == GZIP_HEADER_LEN) { + state = BuiltInGzipDecompressor.GzipStateLabel.INFLATE_STREAM; + } + + numAvailBytes += outputHeaderSize; + + if (outputHeaderSize == len) { + return numAvailBytes; + } + + off += outputHeaderSize; + len -= outputHeaderSize; + } + + if (state == BuiltInGzipDecompressor.GzipStateLabel.INFLATE_STREAM) { + // hand off user data (or what's left of it) to Deflater--but note that + // Deflater may not have consumed all of previous bufferload, in which case + // userBufLen will be zero + if (userBufLen > 0) { Review comment: We need `userBufLen` to record if we have new input for `deflater`. We set it to 0 once we call `setInput` at `deflater`. It is possibly that the deflater doesn't consume all input and we will call `compress` later with partial input in the user buffer. At the moment, `userBufLen` is 0 and we don't call `setInput` at `deflate` again. -- 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: common-issues-unsubscr...@hadoop.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: common-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-issues-h...@hadoop.apache.org
