org.apache.commons.compress.archivers.tar

Class TarUtils

java.lang.Object

org.apache.commons.compress.archivers.tar.TarUtils

public class TarUtils
extends java.lang.Object

This class provides static utility methods to work with byte streams.

Method Summary

Modifier and Type	Method and Description
static long	computeCheckSum(byte[] buf) Computes the checksum of a tar entry header.
static int	formatCheckSumOctalBytes(long value, byte[] buf, int offset, int length) Writes an octal value into a buffer.
static int	formatLongOctalBytes(long value, byte[] buf, int offset, int length) Writes an octal long integer into a buffer.
static int	formatLongOctalOrBinaryBytes(long value, byte[] buf, int offset, int length) Writes a long integer into a buffer as an octal string if this will fit, or as a binary number otherwise.
static int	formatNameBytes(java.lang.String name, byte[] buf, int offset, int length) Copies a name into a buffer.
static int	formatNameBytes(java.lang.String name, byte[] buf, int offset, int length, ZipEncoding encoding) Copies a name into a buffer.
static int	formatOctalBytes(long value, byte[] buf, int offset, int length) Writes an octal integer into a buffer.
static void	formatUnsignedOctalString(long value, byte[] buffer, int offset, int length) Fills a buffer with unsigned octal number, padded with leading zeroes.
static boolean	parseBoolean(byte[] buffer, int offset) Parses a boolean byte from a buffer.
static java.lang.String	parseName(byte[] buffer, int offset, int length) Parses an entry name from a buffer.
static java.lang.String	parseName(byte[] buffer, int offset, int length, ZipEncoding encoding) Parses an entry name from a buffer.
static long	parseOctal(byte[] buffer, int offset, int length) Parses an octal string from a buffer.
static long	parseOctalOrBinary(byte[] buffer, int offset, int length) Computes the value contained in a byte buffer.
static TarArchiveStructSparse	parseSparse(byte[] buffer, int offset) Parses the content of a PAX 1.0 sparse block.
static boolean	verifyCheckSum(byte[] header) Wikipedia says: The checksum is calculated by taking the sum of the unsigned byte values of the header block with the eight checksum bytes taken to be ASCII spaces (decimal value 32).

Methods inherited from class java.lang.Object

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

Method Detail

computeCheckSum

public static long computeCheckSum(byte[] buf)

Computes the checksum of a tar entry header.

Parameters:

buf - The tar entry's header buffer.

Returns:

The computed checksum.

formatCheckSumOctalBytes

public static int formatCheckSumOctalBytes(long value,
                                           byte[] buf,
                                           int offset,
                                           int length)

Writes an octal value into a buffer. Uses formatUnsignedOctalString(long, byte[], int, int) to format the value as an octal string with leading zeros. The converted number is followed by NUL and then space.

Parameters:

value - The value to convert

buf - The destination buffer

offset - The starting offset into the buffer.

length - The size of the buffer.

Returns:

The updated value of offset, i.e. offset+length

Throws:

java.lang.IllegalArgumentException - if the value (and trailer) will not fit in the buffer

formatLongOctalBytes

public static int formatLongOctalBytes(long value,
                                       byte[] buf,
                                       int offset,
                                       int length)

Writes an octal long integer into a buffer. Uses formatUnsignedOctalString(long, byte[], int, int) to format the value as an octal string with leading zeros. The converted number is followed by a space.

Parameters:

value - The value to write as octal

buf - The destinationbuffer.

offset - The starting offset into the buffer.

length - The length of the buffer

Returns:

The updated offset

Throws:

java.lang.IllegalArgumentException - if the value (and trailer) will not fit in the buffer

formatLongOctalOrBinaryBytes

public static int formatLongOctalOrBinaryBytes(long value,
                                               byte[] buf,
                                               int offset,
                                               int length)

Writes a long integer into a buffer as an octal string if this will fit, or as a binary number otherwise. Uses formatUnsignedOctalString(long, byte[], int, int) to format the value as an octal string with leading zeros. The converted number is followed by a space.

Parameters:

value - The value to write into the buffer.

buf - The destination buffer.

offset - The starting offset into the buffer.

length - The length of the buffer.

Returns:

The updated offset.

Throws:

java.lang.IllegalArgumentException - if the value (and trailer) will not fit in the buffer.

Since:

1.4

formatNameBytes

public static int formatNameBytes(java.lang.String name,
                                  byte[] buf,
                                  int offset,
                                  int length)

Copies a name into a buffer. Copies characters from the name into the buffer starting at the specified offset. If the buffer is longer than the name, the buffer is filled with trailing NULs. If the name is longer than the buffer, the output is truncated.

Parameters:

name - The header name from which to copy the characters.

buf - The buffer where the name is to be stored.

offset - The starting offset into the buffer

length - The maximum number of header bytes to copy.

Returns:

The updated offset, i.e. offset + length

formatNameBytes

public static int formatNameBytes(java.lang.String name,
                                  byte[] buf,
                                  int offset,
                                  int length,
                                  ZipEncoding encoding)
                           throws java.io.IOException

Copies a name into a buffer. Copies characters from the name into the buffer starting at the specified offset. If the buffer is longer than the name, the buffer is filled with trailing NULs. If the name is longer than the buffer, the output is truncated.

Parameters:

name - The header name from which to copy the characters.

buf - The buffer where the name is to be stored.

offset - The starting offset into the buffer

length - The maximum number of header bytes to copy.

encoding - name of the encoding to use for file names

Returns:

The updated offset, i.e. offset + length

Throws:

java.io.IOException - on error

Since:

1.4

formatOctalBytes

public static int formatOctalBytes(long value,
                                   byte[] buf,
                                   int offset,
                                   int length)

Writes an octal integer into a buffer. Uses formatUnsignedOctalString(long, byte[], int, int) to format the value as an octal string with leading zeros. The converted number is followed by space and NUL

Parameters:

value - The value to write

buf - The buffer to receive the output

offset - The starting offset into the buffer

length - The size of the output buffer

Returns:

The updated offset, i.e. offset+length

Throws:

java.lang.IllegalArgumentException - if the value (and trailer) will not fit in the buffer

formatUnsignedOctalString

public static void formatUnsignedOctalString(long value,
                                             byte[] buffer,
                                             int offset,
                                             int length)

Fills a buffer with unsigned octal number, padded with leading zeroes.

Parameters:

value - number to convert to octal - treated as unsigned

buffer - destination buffer

offset - starting offset in buffer

length - length of buffer to fill

Throws:

java.lang.IllegalArgumentException - if the value will not fit in the buffer

parseBoolean

public static boolean parseBoolean(byte[] buffer,
                                   int offset)

Parses a boolean byte from a buffer. Leading spaces and NUL are ignored. The buffer may contain trailing spaces or NULs.

Parameters:

buffer - The buffer from which to parse.

offset - The offset into the buffer from which to parse.

Returns:

The boolean value of the bytes.

Throws:

java.lang.IllegalArgumentException - if an invalid byte is detected.

parseName

public static java.lang.String parseName(byte[] buffer,
                                         int offset,
                                         int length)

Parses an entry name from a buffer. Parsing stops when a NUL is found or the buffer length is reached.

Parameters:

buffer - The buffer from which to parse.

offset - The offset into the buffer from which to parse.

length - The maximum number of bytes to parse.

Returns:

The entry name.

parseName

public static java.lang.String parseName(byte[] buffer,
                                         int offset,
                                         int length,
                                         ZipEncoding encoding)
                                  throws java.io.IOException

Parses an entry name from a buffer. Parsing stops when a NUL is found or the buffer length is reached.

Parameters:

buffer - The buffer from which to parse.

offset - The offset into the buffer from which to parse.

length - The maximum number of bytes to parse.

encoding - name of the encoding to use for file names

Returns:

The entry name.

Throws:

java.io.IOException - on error

Since:

1.4

parseOctal

public static long parseOctal(byte[] buffer,
                              int offset,
                              int length)

Parses an octal string from a buffer.

Leading spaces are ignored. The buffer must contain a trailing space or NUL, and may contain an additional trailing space or NUL.

The input buffer is allowed to contain all NULs, in which case the method returns 0L (this allows for missing fields).

To work-around some tar implementations that insert a leading NUL this method returns 0 if it detects a leading NUL since Commons Compress 1.4.

Parameters:

buffer - The buffer from which to parse.

offset - The offset into the buffer from which to parse.

length - The maximum number of bytes to parse - must be at least 2 bytes.

Returns:

The long value of the octal string.

Throws:

java.lang.IllegalArgumentException - if the trailing space/NUL is missing or if an invalid byte is detected.

parseOctalOrBinary

public static long parseOctalOrBinary(byte[] buffer,
                                      int offset,
                                      int length)

Computes the value contained in a byte buffer. If the most significant bit of the first byte in the buffer is set, this bit is ignored and the rest of the buffer is interpreted as a binary number. Otherwise, the buffer is interpreted as an octal number as per the parseOctal function above.

Parameters:

buffer - The buffer from which to parse.

offset - The offset into the buffer from which to parse.

length - The maximum number of bytes to parse.

Returns:

The long value of the octal or binary string.

Throws:

java.lang.IllegalArgumentException - if the trailing space/NUL is missing or an invalid byte is detected in an octal number, or if a binary number would exceed the size of a signed long 64-bit integer.

Since:

1.4

parseSparse

public static TarArchiveStructSparse parseSparse(byte[] buffer,
                                                 int offset)

Parses the content of a PAX 1.0 sparse block.

Parameters:

buffer - The buffer from which to parse.

offset - The offset into the buffer from which to parse.

Returns:

a parsed sparse struct

Since:

1.20

verifyCheckSum

public static boolean verifyCheckSum(byte[] header)

Wikipedia says:

The checksum is calculated by taking the sum of the unsigned byte values of the header block with the eight checksum bytes taken to be ASCII spaces (decimal value 32). It is stored as a six digit octal number with leading zeroes followed by a NUL and then a space. Various implementations do not adhere to this format. For better compatibility, ignore leading and trailing whitespace, and get the first six digits. In addition, some historic tar implementations treated bytes as signed. Implementations typically calculate the checksum both ways, and treat it as good if either the signed or unsigned sum matches the included checksum.

The return value of this method should be treated as a best-effort heuristic rather than an absolute and final truth. The checksum verification logic may well evolve over time as more special cases are encountered.

Parameters:

header - tar header

Returns:

whether the checksum is reasonably good

Since:

1.5

See Also:

COMPRESS-191