org.eclipse.jgit.diff

Class RawText

java.lang.Object

org.eclipse.jgit.diff.Sequence

org.eclipse.jgit.diff.RawText

public class RawText
extends Sequence

A Sequence supporting UNIX formatted text in byte[] format.

Elements of the sequence are the lines of the file, as delimited by the UNIX newline character ('\n'). The file content is treated as 8 bit binary text, with no assumptions or requirements on character encoding.

Note that the first line of the file is element 0, as defined by the Sequence interface API. Traditionally in a text editor a patch file the first line is line number 1. Callers may need to subtract 1 prior to invoking methods if they are converting from "line number" to "element index".

Field Summary

Fields

Modifier and Type	Field and Description
protected byte[]	content The file content for this sequence.
static RawText	EMPTY_TEXT A RawText of length 0
protected IntList	lines Map of line number to starting position within content.

Constructor Summary

Constructors

Constructor and Description
RawText(byte[] input) Create a new sequence from an existing content byte array.
RawText(File file) Create a new sequence from a file.

Method Summary

Modifier and Type	Method and Description
protected String	decode(int start, int end) Decode a region of the text into a String.
String	getLineDelimiter() Get the line delimiter for the first line.
String	getString(int i) Get the text for a single line.
String	getString(int begin, int end, boolean dropLF) Get the text for a region of lines.
static boolean	isBinary(byte[] raw) Determine heuristically whether a byte array represents binary (as opposed to text) content.
static boolean	isBinary(byte[] raw, int length) Determine heuristically whether a byte array represents binary (as opposed to text) content.
static boolean	isBinary(InputStream raw) Determine heuristically whether the bytes contained in a stream represents binary (as opposed to text) content.
boolean	isMissingNewlineAtEnd() Determine if the file ends with a LF ('\n').
static RawText	load(ObjectLoader ldr, int threshold) Read a blob object into RawText, or throw BinaryBlobException if the blob is binary.
int	size() Get size
void	writeLine(OutputStream out, int i) Write a specific line to the output stream, without its trailing LF.

Methods inherited from class java.lang.Object

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

Field Detail

EMPTY_TEXT

public static final RawText EMPTY_TEXT

A RawText of length 0

content

protected final byte[] content

The file content for this sequence.

lines

protected final IntList lines

Map of line number to starting position within content.

Constructor Detail

RawText

public RawText(byte[] input)

Create a new sequence from an existing content byte array.

The entire array (indexes 0 through length-1) is used as the content.

Parameters:

input - the content array. The array is never modified, so passing through cached arrays is safe.

RawText

public RawText(File file)
        throws IOException

Create a new sequence from a file.

The entire file contents are used.

Parameters:

file - the text file.

Throws:

IOException - if Exceptions occur while reading the file

Method Detail

size

public int size()

Get size

Specified by:

size in class Sequence

Returns:

size

writeLine

public void writeLine(OutputStream out,
                      int i)
               throws IOException

Write a specific line to the output stream, without its trailing LF.

The specified line is copied as-is, with no character encoding translation performed.

If the specified line ends with an LF ('\n'), the LF is not copied. It is up to the caller to write the LF, if desired, between output lines.

Parameters:

out - stream to copy the line data onto.

i - index of the line to extract. Note this is 0-based, so line number 1 is actually index 0.

Throws:

IOException - the stream write operation failed.

isMissingNewlineAtEnd

public boolean isMissingNewlineAtEnd()

Determine if the file ends with a LF ('\n').

Returns:

true if the last line has an LF; false otherwise.

getString

public String getString(int i)

Get the text for a single line.

Parameters:

i - index of the line to extract. Note this is 0-based, so line number 1 is actually index 0.

Returns:

the text for the line, without a trailing LF.

getString

public String getString(int begin,
                        int end,
                        boolean dropLF)

Get the text for a region of lines.

Parameters:

begin - index of the first line to extract. Note this is 0-based, so line number 1 is actually index 0.

end - index of one past the last line to extract.

dropLF - if true the trailing LF ('\n') of the last returned line is dropped, if present.

Returns:

the text for lines [begin, end).

decode

protected String decode(int start,
                        int end)

Decode a region of the text into a String. The default implementation of this method tries to guess the character set by considering UTF-8, the platform default, and falling back on ISO-8859-1 if neither of those can correctly decode the region given.

Parameters:

start - first byte of the content to decode.

end - one past the last byte of the content to decode.

Returns:

the region [start, end) decoded as a String.

isBinary

public static boolean isBinary(byte[] raw)

Determine heuristically whether a byte array represents binary (as opposed to text) content.

Parameters:

raw - the raw file content.

Returns:

true if raw is likely to be a binary file, false otherwise

isBinary

public static boolean isBinary(InputStream raw)
                        throws IOException

Determine heuristically whether the bytes contained in a stream represents binary (as opposed to text) content. Note: Do not further use this stream after having called this method! The stream may not be fully read and will be left at an unknown position after consuming an unknown number of bytes. The caller is responsible for closing the stream.

Parameters:

raw - input stream containing the raw file content.

Returns:

true if raw is likely to be a binary file, false otherwise

Throws:

IOException - if input stream could not be read

isBinary

public static boolean isBinary(byte[] raw,
                               int length)

Determine heuristically whether a byte array represents binary (as opposed to text) content.

Parameters:

raw - the raw file content.

length - number of bytes in raw to evaluate. This should be raw.length unless raw was over-allocated by the caller.

Returns:

true if raw is likely to be a binary file, false otherwise

getLineDelimiter

public String getLineDelimiter()

Get the line delimiter for the first line.

Returns:

the line delimiter or null

Since:

2.0

load

public static RawText load(ObjectLoader ldr,
                           int threshold)
                    throws IOException,
                           BinaryBlobException

Read a blob object into RawText, or throw BinaryBlobException if the blob is binary.

Parameters:

ldr - the ObjectLoader for the blob

threshold - if the blob is larger than this size, it is always assumed to be binary.

Returns:

the RawText representing the blob.

Throws:

BinaryBlobException - if the blob contains binary data.

IOException - if the input could not be read.

Since:

4.10