com.davidehrmann.vcdiff.engine

Class VCDiffCodeTableWriterImpl

java.lang.Object

com.davidehrmann.vcdiff.engine.VCDiffCodeTableWriterImpl

All Implemented Interfaces:

VCDiffCodeTableWriter<OutputStream>

public class VCDiffCodeTableWriterImpl
extends Object
implements VCDiffCodeTableWriter<OutputStream>

Author:

David Ehrmann The method calls after construction *must* conform to the following pattern: {{add|copy|run}* [addChecksum] output}* When output has been called in this sequence, a complete target window (as defined in RFC 3284 section 4.3) will have been appended to out (unless no calls to add, run, or copy were made, in which case output will do nothing.) The output will not be available for use until after each call to output(). NOT threadsafe.

Constructor Summary

Constructors

Constructor and Description
VCDiffCodeTableWriterImpl(boolean interleaved) This constructor uses the default code table.

Method Summary

Modifier and Type	Method and Description
void	add(byte[] data, int offset, int length) encode an ADD opcode with the "size" bytes starting at data
void	addChecksum(int checksum) add a checksum to the code table
void	copy(int offset, int size) encode a COPY opcode with args "offset" (into dictionary) and "size" bytes.
void	finishEncoding(OutputStream out) There should not be any need to output more data since encodeChunk() encodes a complete target window and there is no end-of-delta-file marker.
int	getDeltaWindowSize()
void	init(int dictionarySize) Initializes the constructed object for use.
void	output(OutputStream out) Appends the encoded delta window to the output string.
void	run(int size, byte b) encode a RUN opcode for "size" copies of the value "byte".
int	targetLength()
void	writeHeader(OutputStream out, EnumSet<VCDiffFormatExtension> formatExtensions) Write the header (as defined in section 4.1 of the RFC) to *out.

Methods inherited from class java.lang.Object

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

Constructor Detail

VCDiffCodeTableWriterImpl

public VCDiffCodeTableWriterImpl(boolean interleaved)

This constructor uses the default code table. If interleaved is true, the encoder writes each delta file window by interleaving instructions and sizes with their corresponding addresses and data, rather than placing these elements into three separate sections. This facilitates providing partially decoded results when only a portion of a delta file window is received (e.g. when HTTP over TCP is used as the transmission protocol.) The interleaved format is not consistent with the VCDIFF draft standard.

Parameters:

interleaved - Whether or not to interleave the output data

Method Detail

init

public void init(int dictionarySize)

Initializes the constructed object for use. This method must be called after a VCDiffCodeTableWriter is constructed and before any of its other methods can be called. It will return false if there was an error initializing the object, or true if it was successful. After the object has been initialized and used, init() can be called again to restore the initial state of the object.

Specified by:

init in interface VCDiffCodeTableWriter<OutputStream>

Parameters:

dictionarySize - size of the dictionary being used

add

public void add(byte[] data,
                int offset,
                int length)

encode an ADD opcode with the "size" bytes starting at data

Specified by:

add in interface VCDiffCodeTableWriter<OutputStream>

Parameters:

data - data to add

offset - offset in data to start from

length - total bytes to add

addChecksum

public void addChecksum(int checksum)

Description copied from interface: VCDiffCodeTableWriter

add a checksum to the code table

Specified by:

addChecksum in interface VCDiffCodeTableWriter<OutputStream>

Parameters:

checksum - checksum to write to the writer

copy

public void copy(int offset,
                 int size)

encode a COPY opcode with args "offset" (into dictionary) and "size" bytes.

Specified by:

copy in interface VCDiffCodeTableWriter<OutputStream>

Parameters:

offset - offset into the dictionary to copy data from

size - number of bytes to copy from dictionary

finishEncoding

public void finishEncoding(OutputStream out)
                    throws IOException

There should not be any need to output more data since encodeChunk() encodes a complete target window and there is no end-of-delta-file marker.

Specified by:

finishEncoding in interface VCDiffCodeTableWriter<OutputStream>

Parameters:

out -

Throws:

IOException - if there's an exception writing to out

getDeltaWindowSize

public int getDeltaWindowSize()

output

public void output(OutputStream out)
            throws IOException

Appends the encoded delta window to the output string. The output string is not null-terminated and may contain embedded '\0' characters.

Specified by:

output in interface VCDiffCodeTableWriter<OutputStream>

Parameters:

out - writer mechanism to write to

Throws:

IOException - if there's an exception writing to out

run

public void run(int size,
                byte b)

encode a RUN opcode for "size" copies of the value "byte".

Specified by:

run in interface VCDiffCodeTableWriter<OutputStream>

Parameters:

size - number of copies of the value to write

b - byte value to write

writeHeader

public void writeHeader(OutputStream out,
                        EnumSet<VCDiffFormatExtension> formatExtensions)
                 throws IOException

Write the header (as defined in section 4.1 of the RFC) to *out. This includes information that can be gathered before the first chunk of input is available.

Specified by:

writeHeader in interface VCDiffCodeTableWriter<OutputStream>

Parameters:

out - writer mechanism to write to

formatExtensions - Flags for enabling features that are extensions to the format

Throws:

IOException - if there's an exception writing to out

targetLength

public int targetLength()