wiithon.binary.writer module

Write numbers, strings and raw bytes into a binary stream

This module exposes BinaryWriter, the counterpart of BinaryReader Everything is written in big-endian since it is what the Wii uses

class wiithon.binary.writer.BinaryWriter(stream, encoding='utf-8')

Bases: object

A writer that appends structured values to a stream

The writer never owns the stream, it only writes in it, so closing it is up to the caller Contrary to the reader there is no from_bytes helper, wrap a io.BytesIO yourself and read it back with getvalue()

Example

>>> buf = BytesIO()
>>> writer = BinaryWriter(buf)
>>> writer.u32(0x5D1C9EA3)
>>> writer.string("GAME", 0x10)
>>> data = buf.getvalue()

Parameters:

• stream (BinaryIO)

• encoding (str, default: 'utf-8')

stream

Stream to write for operations

encoding

Encoding used by default for writing

seek(offset)

Change the stream cursor position to a given offset

Parameters:

offset (int) – The new stream cursor

Return type:

None

Note

Seeking past the end of the stream does not grow it, the gap is filled only when you write there

tell()

Get the current stream cursor position

Return type:

int

Returns:

The current stream cursor position within the stream

pad(count, byte=b'\\x00')

Write a filler of count bytes to the stream of the same byte, used to align structure on a boundary

Parameters:

• count (int) – How many times byte is written

• byte (bytes, default: b'\\x00') – The filler byte. Default is null byte

Return type:

None

Note

count is the number of repetitions, not a number of bytes.

size()

Get the total size of the stream

The cursor is moved to the end to measure the stream then put back where it was.

Return type:

int

Returns:

The size of the whole stream in bytes, not the number of bytes written

u8(data)

Write a single unsigned byte to the stream in big-endian at the current stream position The cursor is placed after the written byte

Parameters:

data (int) – The number to write, from 0 to 255

Return type:

None

u16(data)

Write two unsigned bytes to the stream in big-endian at the current stream position The cursor is placed after the written bytes

Parameters:

data (int) – The number to write, from 0 to 65535

Return type:

None

u32(data)

Write four unsigned bytes to the stream in big-endian at the current stream position The cursor is placed after the written bytes

Parameters:

data (int) – The number to write, from 0 to 4294967295

Return type:

None

u64(data)

Write eight unsigned bytes to the stream in big-endian at the current stream position The cursor is placed after the written bytes

Parameters:

data (int) – The number to write, from 0 to 18446744073709551615

Return type:

None

s8(data)

Write a single signed byte to the stream in big-endian at the current stream position The cursor is placed after the written byte

Parameters:

data (int) – The number to write, from -128 to 127

Return type:

None

s16(data)

Write two signed bytes to the stream in big-endian at the current stream position The cursor is placed after the written bytes

Parameters:

data (int) – The number to write, from -32768 to 32767

Return type:

None

s32(data)

Write four signed bytes to the stream in big-endian at the current stream position The cursor is placed after the written bytes

Parameters:

data (int) – The number to write, from -2147483648 to 2147483647

Return type:

None

s64(data)

Write eight signed bytes to the stream in big-endian at the current stream position The cursor is placed after the written bytes

Parameters:

data (int) – The number to write, from -9223372036854775808 to 9223372036854775807

Return type:

None

float(data)

Write a 32 bits float to the stream in big-endian at the current stream position The cursor is placed after the written bytes

Parameters:

data (float) – The number to write

Return type:

None

list_u32(numbers)

Write a list of unsigned 32 bits numbers one after the other

Parameters:

numbers (list[int]) – The numbers to write, in the order they are stored

Return type:

None

Note

No count is written before the list. If the format needs one, write it yourself

u32_shifted(data)

Right shift a number 2 times then write it as an unsigned 32 bits number

This is the inverse of u32_shifted(), for the formats that store offsets divided by 4

Parameters:

data (int) – The real value, before the shift

Return type:

None

Warning

The two low bits are lost.

u32_le(data)

Write four unsigned bytes to the stream in little-endian at the current stream position

Parameters:

data (int) – The number to write

Return type:

None

Note

Almost everything on a Wii disc is big-endian. This is only for the few structures that are not

raw(data)

Write bytes to the stream as they are, without any conversion

Parameters:

data (bytes) – The bytes to write

Return type:

None

string(value, size=None, padding=b'\\x00', encoding=None, *, add_null_byte=False)

Write a string in a field of a fixed size, padded if it is too short

Parameters:

• value (str) – The string to write

• size (int | None, default: None) – The size of the field in bytes. Defaults to the encoded length of value, which writes it without any padding

• padding (bytes, default: b'\\x00') – The byte used to fill the end of the field

• encoding (str | None, default: None) – The encoding to use. Defaults to the encoding of the writer

• add_null_byte (bool, default: False) – Write an extra null byte after the field, for the formats that store strings one after the other in a string table

Raises:

BinaryError – If the encoded string is longer than size

Return type:

None

Note

size counts encoded bytes, not characters. A japanese string in shift_jis takes more room than its length suggests

Warning

With add_null_byte the field takes size + 1 bytes, the null byte is not counted inside