src.org.tukaani.xz.XZInputStream Maven / Gradle / Ivy
/*
* XZInputStream
*
* Author: Lasse Collin
*
* This file has been put into the public domain.
* You can do whatever you want with this file.
*/
package org.tukaani.xz;
import java.io.InputStream;
import java.io.DataInputStream;
import java.io.IOException;
import java.io.EOFException;
import org.tukaani.xz.common.DecoderUtil;
/**
* Decompresses a .xz file in streamed mode (no seeking).
*
* Use this to decompress regular standalone .xz files. This reads from
* its input stream until the end of the input or until an error occurs.
* This supports decompressing concatenated .xz files.
*
*
Typical use cases
*
* Getting an input stream to decompress a .xz file:
*
* InputStream infile = new FileInputStream("foo.xz");
* XZInputStream inxz = new XZInputStream(infile);
*
*
* It's important to keep in mind that decompressor memory usage depends
* on the settings used to compress the file. The worst-case memory usage
* of XZInputStream is currently 1.5 GiB. Still, very few files will
* require more than about 65 MiB because that's how much decompressing
* a file created with the highest preset level will need, and only a few
* people use settings other than the predefined presets.
*
* It is possible to specify a memory usage limit for
* XZInputStream. If decompression requires more memory than
* the specified limit, MemoryLimitException will be thrown when reading
* from the stream. For example, the following sets the memory usage limit
* to 100 MiB:
*
* InputStream infile = new FileInputStream("foo.xz");
* XZInputStream inxz = new XZInputStream(infile, 100 * 1024);
*
*
* When uncompressed size is known beforehand
*
* If you are decompressing complete files and your application knows
* exactly how much uncompressed data there should be, it is good to try
* reading one more byte by calling read() and checking
* that it returns -1. This way the decompressor will parse the
* file footers and verify the integrity checks, giving the caller more
* confidence that the uncompressed data is valid. (This advice seems to
* apply to java.util.zip.GZIPInputStream too.)
*
* @see SingleXZInputStream
*/
public class XZInputStream extends InputStream {
private final int memoryLimit;
private InputStream in;
private SingleXZInputStream xzIn;
private boolean endReached = false;
private IOException exception = null;
/**
* Creates a new XZ decompressor without a memory usage limit.
*
* This constructor reads and parses the XZ Stream Header (12 bytes)
* from in. The header of the first Block is not read
* until read is called.
*
* @param in input stream from which XZ-compressed
* data is read
*
* @throws XZFormatException
* input is not in the XZ format
*
* @throws CorruptedInputException
* XZ header CRC32 doesn't match
*
* @throws UnsupportedOptionsException
* XZ header is valid but specifies options
* not supported by this implementation
*
* @throws EOFException
* less than 12 bytes of input was available
* from in
*
* @throws IOException may be thrown by in
*/
public XZInputStream(InputStream in) throws IOException {
this(in, -1);
}
/**
* Creates a new XZ decompressor with an optional memory usage limit.
*
* This is identical to XZInputStream(InputStream) except
* that this takes also the memoryLimit argument.
*
* @param in input stream from which XZ-compressed
* data is read
*
* @param memoryLimit memory usage limit in kibibytes (KiB)
* or -1 to impose no
* memory usage limit
*
* @throws XZFormatException
* input is not in the XZ format
*
* @throws CorruptedInputException
* XZ header CRC32 doesn't match
*
* @throws UnsupportedOptionsException
* XZ header is valid but specifies options
* not supported by this implementation
*
* @throws EOFException
* less than 12 bytes of input was available
* from in
*
* @throws IOException may be thrown by in
*/
public XZInputStream(InputStream in, int memoryLimit) throws IOException {
this.in = in;
this.memoryLimit = memoryLimit;
this.xzIn = new SingleXZInputStream(in, memoryLimit);
}
/**
* Decompresses the next byte from this input stream.
*
* Reading lots of data with read() from this input stream
* may be inefficient. Wrap it in {@link java.io.BufferedInputStream}
* if you need to read lots of data one byte at a time.
*
* @return the next decompressed byte, or -1
* to indicate the end of the compressed stream
*
* @throws CorruptedInputException
* @throws UnsupportedOptionsException
* @throws MemoryLimitException
*
* @throws XZIOException if the stream has been closed
*
* @throws EOFException
* compressed input is truncated or corrupt
*
* @throws IOException may be thrown by in
*/
public int read() throws IOException {
byte[] buf = new byte[1];
return read(buf, 0, 1) == -1 ? -1 : (buf[0] & 0xFF);
}
/**
* Decompresses into an array of bytes.
*
* If len is zero, no bytes are read and 0
* is returned. Otherwise this will try to decompress len
* bytes of uncompressed data. Less than len bytes may
* be read only in the following situations:
*
* - The end of the compressed data was reached successfully.
* - An error is detected after at least one but less
len
* bytes have already been successfully decompressed.
* The next call with non-zero len will immediately
* throw the pending exception.
* - An exception is thrown.
*
*
* @param buf target buffer for uncompressed data
* @param off start offset in buf
* @param len maximum number of uncompressed bytes to read
*
* @return number of bytes read, or -1 to indicate
* the end of the compressed stream
*
* @throws CorruptedInputException
* @throws UnsupportedOptionsException
* @throws MemoryLimitException
*
* @throws XZIOException if the stream has been closed
*
* @throws EOFException
* compressed input is truncated or corrupt
*
* @throws IOException may be thrown by in
*/
public int read(byte[] buf, int off, int len) throws IOException {
if (off < 0 || len < 0 || off + len < 0 || off + len > buf.length)
throw new IndexOutOfBoundsException();
if (len == 0)
return 0;
if (in == null)
throw new XZIOException("Stream closed");
if (exception != null)
throw exception;
if (endReached)
return -1;
int size = 0;
try {
while (len > 0) {
if (xzIn == null) {
prepareNextStream();
if (endReached)
return size == 0 ? -1 : size;
}
int ret = xzIn.read(buf, off, len);
if (ret > 0) {
size += ret;
off += ret;
len -= ret;
} else if (ret == -1) {
xzIn = null;
}
}
} catch (IOException e) {
exception = e;
if (size == 0)
throw e;
}
return size;
}
private void prepareNextStream() throws IOException {
DataInputStream inData = new DataInputStream(in);
byte[] buf = new byte[DecoderUtil.STREAM_HEADER_SIZE];
// The size of Stream Padding must be a multiple of four bytes,
// all bytes zero.
do {
// First try to read one byte to see if we have reached the end
// of the file.
int ret = inData.read(buf, 0, 1);
if (ret == -1) {
endReached = true;
return;
}
// Since we got one byte of input, there must be at least
// three more available in a valid file.
inData.readFully(buf, 1, 3);
} while (buf[0] == 0 && buf[1] == 0 && buf[2] == 0 && buf[3] == 0);
// Not all bytes are zero. In a valid Stream it indicates the
// beginning of the next Stream. Read the rest of the Stream Header
// and initialize the XZ decoder.
inData.readFully(buf, 4, DecoderUtil.STREAM_HEADER_SIZE - 4);
try {
xzIn = new SingleXZInputStream(in, memoryLimit, buf);
} catch (XZFormatException e) {
// Since this isn't the first .xz Stream, it is more
// logical to tell that the data is corrupt.
throw new CorruptedInputException(
"Garbage after a valid XZ Stream");
}
}
/**
* Returns the number of uncompressed bytes that can be read
* without blocking. The value is returned with an assumption
* that the compressed input data will be valid. If the compressed
* data is corrupt, CorruptedInputException may get
* thrown before the number of bytes claimed to be available have
* been read from this input stream.
*
* @return the number of uncompressed bytes that can be read
* without blocking
*/
public int available() throws IOException {
if (in == null)
throw new XZIOException("Stream closed");
if (exception != null)
throw exception;
return xzIn == null ? 0 : xzIn.available();
}
/**
* Closes the stream and calls in.close().
* If the stream was already closed, this does nothing.
*
* @throws IOException if thrown by in.close()
*/
public void close() throws IOException {
if (in != null) {
try {
in.close();
} finally {
in = null;
}
}
}
}