Package com.activeviam.io.file.impl

Class BufferedGZipOutputStream

java.lang.Object

java.io.OutputStream

java.io.FilterOutputStream

java.io.BufferedOutputStream

com.activeviam.io.file.impl.ABufferedArchiveOutputStream

com.activeviam.io.file.impl.BufferedGZipOutputStream

All Implemented Interfaces:

IArchiveOutputStream, Closeable, Flushable, AutoCloseable

public class BufferedGZipOutputStream
extends ABufferedArchiveOutputStream

Managed stream implementation of GZIPOutputStream for ensuring an optimal generation of a single-entry archive file.

NOTICE - How to use this?... Once an instance of this stream has been created, simply use write(byte[], int, int) (or equivalent write methods) to append more content into the stream. Lastly, use ABufferedArchiveOutputStream.close() to properly finalize the archive file generation.

We use buffered intermediary streams after the GZip output stream logic itself. All writing operations happen into a temporary file. We rename this temporary file into the final archive file (specified when creating this instance) upon closure of this stream.

We use an intermediary buffered stream AFTER the ZIP logic, because GZIPOutputStream is delegating to an underlying output stream for the final writing of byte[] content into system resources: this is either RandomAccessFile (if default constructor of GZIPOutputStream is used and instantiation of RandomAccessFile possible) or FileOutputStream (in all other cases).

Now, when the Deflater output (e.g. compressed output) is ready, GZIPOutputStream is using an intermediary buffer to read the byte[] of "compressed content" and pass that to the underlying output stream's write(byte[], int, int) (or equivalent). This buffer is of size BUFFER_SIZE = 512, which is quite small.

This means that (since RandomAccessFile and FileOutputStream are not buffered) we would be going to write to the system resource by blocks of BUFFER_SIZE bytes, which might be sub-optimal. So, instead, we use a BufferedOutputStream (piping into a FileOutputStream) as underlying output stream for the GZIPOutputStream that this stream wraps. The size of this output buffer is by default equal to ABufferedArchiveOutputStream.DEFAULT_SYSTEM_OUTPUT_BUFFER_SIZE (which is the default size suggested in BufferedOutputStream implementation) but this can be specified differently at instantiation time.

Author:

ActiveViam

See Also:

GZIPOutputStream, ABufferedArchiveOutputStream

Field Summary

Fields inherited from class com.activeviam.io.file.impl.ABufferedArchiveOutputStream

DEFAULT_SYSTEM_OUTPUT_BUFFER_SIZE, TMP_FILE_SUFFIX, ZIP_DEFLATER_INPUT_BUFFER_SIZE

Fields inherited from class java.io.BufferedOutputStream

buf, count

Fields inherited from class java.io.FilterOutputStream

out

Constructor Summary

Constructors

Constructor	Description
BufferedGZipOutputStream(File archiveFile, boolean keepOutputFileIfFailure)	Default constructor.
BufferedGZipOutputStream(File outputFile, int systemOutputBufferSize, boolean keepOutputFileIfFailure)	Alternative constructor.

Method Summary

Modifier and Type	Method	Description
long	getBytesWritten()
void	write(byte[] b)
void	write(byte[] b, int off, int len)
void	write(int b)

Methods inherited from class com.activeviam.io.file.impl.ABufferedArchiveOutputStream

beforeClose, close, getCurrentArchiveFile, getFinalArchiveFile, setOutputSuccess

Methods inherited from class java.io.BufferedOutputStream

flush

Methods inherited from class java.io.OutputStream

nullOutputStream

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

BufferedGZipOutputStream

public BufferedGZipOutputStream(File archiveFile,
                                boolean keepOutputFileIfFailure)
                         throws IOException

Default constructor.

Archive will be created in ZIP32 mode and the system output buffer (buffer used for writing to filesystem resource) will be sized with ABufferedArchiveOutputStream.DEFAULT_SYSTEM_OUTPUT_BUFFER_SIZE.

Parameters:

archiveFile - abstract representation of the target archive file pathname

keepOutputFileIfFailure - TRUE if the generated file needs to be kept even in case of failure of the extraction

Throws:

IOException - in case of stream creation failure

BufferedGZipOutputStream

public BufferedGZipOutputStream(File outputFile,
                                int systemOutputBufferSize,
                                boolean keepOutputFileIfFailure)
                         throws IOException

Alternative constructor.

Parameters:

outputFile - abstract representation of the final archive file pathname (see ABufferedArchiveOutputStream.getCurrentArchiveFile())

systemOutputBufferSize - size of the intermediary buffer used to write compressed results into the system resources (e.g. final archive file(s) on the filesystem)

keepOutputFileIfFailure - TRUE if the generated file needs to be kept even in case of failure of the extraction

Throws:

IOException - in case of stream creation failure

Method Detail

getBytesWritten

public long getBytesWritten()

Returns:

the number of bytes written into this stream (a.k.a. the volume of uncompressed data that has been processed). If the stream is still open when this is called this will return the bytes written so far (depends on what has been passed for writing and on the buffer fill/flush cycles), otherwise (called after stream closure) it will return the final number of bytes written.

write

public void write(byte[] b)
           throws IOException

Overrides:

write in class FilterOutputStream

Throws:

IOException

write

public void write(byte[] b,
                  int off,
                  int len)
           throws IOException

Overrides:

write in class BufferedOutputStream

Throws:

IOException

write

public void write(int b)
           throws IOException

Overrides:

write in class BufferedOutputStream

Throws:

IOException