Variable-length integer used in Apple `'dcmp' (0)` and `'dcmp' (1)` compressed resource formats: format specification

A variable-length integer, in the format used by the 0xfe chunks in the 'dcmp' (0) and 'dcmp' (1) resource compression formats. See the dcmp_0 and dcmp_1 specs for more information about these compression formats.

This variable-length integer format can store an integer x in any of the following ways:

  • In a single byte, if 0 <= x <= 0x7f (7-bit unsigned integer)
  • In 2 bytes, if -0x4000 <= x <= 0x3eff (15-bit signed integer with the highest 0x100 values unavailable)
  • In 5 bytes, if -0x80000000 <= x <= 0x7fffffff (32-bit signed integer)

In practice, values are always stored in the smallest possible format, but technically any of the larger formats could be used as well.

Application

Mac OS

KS implementation details

License: MIT
Minimal Kaitai Struct required: 0.8

This page hosts a formal specification of Variable-length integer used in Apple `'dcmp' (0)` and `'dcmp' (1)` compressed resource formats using Kaitai Struct. This specification can be automatically translated into a variety of programming languages to get a parsing library.

Block diagram

Format specification in Kaitai Struct YAML

meta:
  id: dcmp_variable_length_integer
  title: Variable-length integer used in Apple `'dcmp' (0)` and `'dcmp' (1)` compressed resource formats
  application: Mac OS
  license: MIT
  ks-version: "0.8"
  endian: be
doc: |
  A variable-length integer,
  in the format used by the 0xfe chunks in the `'dcmp' (0)` and `'dcmp' (1)` resource compression formats.
  See the dcmp_0 and dcmp_1 specs for more information about these compression formats.

  This variable-length integer format can store an integer `x` in any of the following ways:

  * In a single byte,
    if `0 <= x <= 0x7f`
    (7-bit unsigned integer)
  * In 2 bytes,
    if `-0x4000 <= x <= 0x3eff`
    (15-bit signed integer with the highest `0x100` values unavailable)
  * In 5 bytes, if `-0x80000000 <= x <= 0x7fffffff`
    (32-bit signed integer)

  In practice,
  values are always stored in the smallest possible format,
  but technically any of the larger formats could be used as well.
doc-ref: 'https://github.com/dgelessus/python-rsrcfork/blob/f891a6e/src/rsrcfork/compress/common.py'
seq:
  - id: first
    type: u1
    doc: |
      The first byte of the variable-length integer.
      This determines which storage format is used.

      * For the 1-byte format,
        this encodes the entire value of the value.
      * For the 2-byte format,
        this encodes the high 7 bits of the value,
        minus `0xc0`.
        The highest bit of the value,
        i. e. the second-highest bit of this field,
        is the sign bit.
      * For the 5-byte format,
        this is always `0xff`.
  - id: more
    type:
      switch-on: first
      cases:
        0xff: s4
        _: u1
    if: first >= 0x80
    doc: |
      The remaining bytes of the variable-length integer.

      * For the 1-byte format,
        this is not present.
      * For the 2-byte format,
        this encodes the low 8 bits of the value.
      * For the 5-byte format,
        this encodes the entire value.
instances:
  value:
    value: |
      first == 0xff ? more
      : first >= 0x80 ? (first << 8 | more) - 0xc000
      : first
    doc: |
      The decoded value of the variable-length integer.