Project

crc

0.01
No release in over 3 years
Low commit activity in last 3 years
Pure ruby implemented generic CRC (Cyclic Redundancy Check) calculator. Customization is posible for 1 to 64 bit width, any polynomials, and with/without bit reflection input/output. If you need more speed, please use crc-turbo.
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
 Dependencies

Development

>= 12
 Project Readme

crc - CRC calculator for ruby

This is a generic CRC (Cyclic Redundancy Check) calculator for ruby.

It is written by pure ruby with based on slice-by-eight algorithm (slice-by-16 algorithm with byte-order free).

Included built-in CRC models are CRC-32, CRC-32C, CRC-64-XZ, CRC-16, CRC-8-MAXIM, CRC-5-USB and many more.

Customization is posible for 1 to 64 bit width, any polynomials, and with/without bit reflection input/output.

This library is slower than ×85+ of zlib/crc32, and slower than ×120+ of extlzma/crc32 on FreeBSD 10.3R amd64.

If you need more speed, please use with crc-turbo.

Summary

  • package name: crc
  • author: dearblue (mailto:dearblue@users.noreply.github.com)
  • report issue to: https://github.com/dearblue/ruby-crc/issues
  • how to install: gem install crc
  • version: 0.4.2
  • production quality: TECHNICAL PREVIEW
  • licensing:
    • BSD-2-Clause : MAIN LICENSE
    • zlib-style License : lib/crc/_combine.rb
    • Creative Commons License Zero (CC0 / Public Domain) : lib/crc/_byruby.rb, lib/crc/_models.rb
  • dependency gems: none
  • dependency external C libraries: none
  • bundled external C libraries: none
  • installed executable file:
    • rbcrc: CRC calculation source code generator for c, ruby and javascript

API Guide

This examples are used CRC-32 model. Please see CRC for more details.

Calculate by direct

  • CRC.crc32(seq, init = CRC::CRC32.initial_crc) => crc-32 integer (likely as Zlib.crc32)
  • CRC.crc32.crc(seq, init = CRC::CRC32.initial_crc) => crc-32 integer (likely as Zlib.crc32)
  • CRC.crc32.digest(seq, init = CRC::CRC32.initial_crc) => crc-32 digest (likely as Digest::XXXX.digest)
  • CRC.crc32.hexdigest(seq, init = CRC::CRC32.initial_crc) -> crc-32 hex-digest (likely as Digest::XXXX.hexdigest)
  • CRC.crc32[seq, init = CRC::CRC32.initial_crc, current_length = 0] -> crc-32 calculator

Calculate by streaming

  • CRC.crc32.new(init = 0, current_length = 0) => crc-32 calculator
  • CRC::CRC32#update(seq) => self (likely as Digest::XXXX.update)
  • CRC::CRC32#finish => crc-32 integer (likely as Digest::XXXX.finish)
  • CRC::CRC32#crc => crc-32 integer (same as CRC::CRC32#finish)
  • CRC::CRC32#digest => crc-32 digest (likely as Digest::XXXX.digest)
  • CRC::CRC32#hexdigest => crc-32 hex-digest (likely as Digest::XXXX.hexdigest)

Example ::

x = CRC.crc32.new     # => #<CRC::CRC32:00000000>
x.update "123"        # => #<CRC::CRC32:884863D2>
x.update "456789"     # => #<CRC::CRC32:CBF43926>
x.crc                 # => 3421780262
x.digest              # => "\xCB\xF49&"
x.hexdigest           # => "CBF43926"

Combine

  • CRC.combine(crc1, crc2, len2) => combined crc integer (likely as Zlib.crc32_comibne)
  • CRC#+(right_crc) => combined crc calculator

Example-1 ::

CRC.crc32.combine(CRC.crc32("123"), CRC.crc32("456789"), 6) # => 3421780262

Example-2 ::

CRC.crc32["123"] + CRC.crc32["456"] + CRC.crc32["789"] # => #<CRC::CRC32:CBF43926>

Create customized crc model

  • CRC.new(bitsize, poly, initial_crc = 0, refin = true, refout = true, xor_output = ~0) => new crc model class

Example ::

MyCRC32 = CRC.new(32, 0x04C11DB7)
MyCRC32.class                     # => Class
MyCRC32.hexdigest("123456789")    # => "CBF43926"
MyCRC32["123456789"]              # => #<MyCRC32:CBF43926>

Calculate arc-crc

  • CRC::XXX.acrc(pre, post = nil, want_crc = 0) => arc-crc byte string

Example ::

a = "12"
c = "789"
wantcrc = CRC.crc32("123456789")
b = CRC.crc32.acrc(a, c, wantcrc)   # => "3456"
CRC.crc32[a + b + c]                # => #<CRC::CRC32:CBF43926>

See CRC::Calculate.acrc or below for more detail.

Built-in CRC models

$ rbcrc -lq

CRC-1, CRC-3-ROHC, CRC-4-INTERLAKEN, CRC-4-ITU, CRC-5-EPC, CRC-5-ITU, CRC-5-USB, CRC-6-CDMA2000-A, CRC-6-CDMA2000-B, CRC-6-DARC, CRC-6-ITU, CRC-7, CRC-7-ROHC, CRC-7-UMTS, CRC-8-CCITT, CRC-8-MAXIM, CRC-8-DARC, CRC-8-SAE, CRC-8-WCDMA, CRC-8-CDMA2000, CRC-8-DVB-S2, CRC-8-EBU, CRC-8-I-CODE, CRC-8-ITU, CRC-8-LTE, CRC-8-ROHC, CRC-10, CRC-10-CDMA2000, CRC-11, CRC-11-UMTS, CRC-12-CDMA2000, CRC-12-DECT, CRC-12-UMTS, CRC-13-BBC, CRC-14-DARC, CRC-15, CRC-15-MPT1327, CRC-16, CRC-16-AUG-CCITT, CRC-16-CDMA2000, CRC-16-DECT-R, CRC-16-DECT-X, CRC-16-T10-DIF, CRC-16-DNP, CRC-16-BUYPASS, CRC-16-CCITT-FALSE, CRC-16-DDS-110, CRC-16-EN-13757, CRC-16-GENIBUS, CRC-16-LJ1200, CRC-16-MAXIM, CRC-16-MCRF4XX, CRC-16-RIELLO, CRC-16-TELEDISK, CRC-16-TMS37157, CRC-16-USB, CRC-16-A, CRC-16-KERMIT, CRC-16-MODBUS, CRC-16-X-25, CRC-16-XMODEM, CRC-24-Radix-64, CRC-24-OPENPGP, CRC-24-BLE, CRC-24-FLEXRAY-A, CRC-24-FLEXRAY-B, CRC-24-INTERLAKEN, CRC-24-LTE-A, CRC-24-LTE-B, CRC-30-CDMA, CRC-31-PHILIPS, CRC-32, CRC-32-BZIP2, CRC-32C, CRC-32D, CRC-32-MPEG-2, CRC-32-POSIX, CRC-32Q, CRC-32-JAMCRC, CRC-32-XFER, CRC-40-GSM, CRC-64-XZ, CRC-64-JONES, CRC-64-ECMA, CRC-64-WE, CRC-64-ISO

Environment variables for behavior

  • RUBY_CRC_NOFAST=0: Use "crc-turbo" if posible. When failure required, same as RUBY_CRC_NOFAST=1.
  • RUBY_CRC_NOFAST=1: Force use ruby implementation with slice-by-16 algorithm. Not used "crc-turbo".
  • RUBY_CRC_NOFAST=2: Switch to lookup table algorithm from slice-by-16 algorithm. Slower than about 52% (when CRC-32).
  • RUBY_CRC_NOFAST=3: Switch to reference algorithm from slice-by-16 algorithm. Slower than about 7% (when CRC-32).

About CRC.combine

CRC.combine is ported from Mark Adler's crccomb.c in https://stackoverflow.com/questions/29915764/generic-crc-8-16-32-64-combine-implementation#29928573.

Source code generator for the specific CRC calculator (from crc-0.3.1)

Add source code generator for the specific CRC calculator to c, ruby, and javascript.

Algorithm is slicing-by-16 only for ruby and javascript.

For C, be able choose into bit-by-bit, bit-by-bit-fast, halfbyte-table, standard-table, slicing-by-4, slicing-by-8, and slicing-by-16.

$ rbcrc --help
usage: rbcrc [options] output-filename...
 -m crcname   choose included crc name in library (``-l'' to print list)
 -n crcname   declare function name or class name [DEFAULT is filename]
 -s bitsize   declare crc bit size [REQUIRED for customized crc]
 -p polynom   declare crc polynomial [REQUIRED for customized crc]
 -c initcrc   declare initial crc (not internal state) [DEFAULT: 0]
 -S initstate  declare initial state (internal state) [DEFAULT: unset]
 -x xormask   declare xor bit mask for when output [DEFAULT: ~0]
 -i           reflect input [DEFAULT]
 -I           normal input (not reflect)
 -o           reflect output [DEFAULT]
 -O           normal output (not reflect)
 -a algorithm  switch algorithm (see below) (C file type only)

 -l           print crc names
 -f           force overwrite
 -v           increment verbosery level
 -q           quiet mode (reset verbosery level to zero)

About LICENSE for generated source code:
  Generated code is under Creative Commons License Zero (CC0 / Public Domain).
  See https://creativecommons.org/publicdomain/zero/1.0/

Algorithms (C file type only):
  bit-by-bit, bit-by-bit-fast, halfbyte-table, standard-table,
  slicing-by-4, slicing-by-8, slicing-by-16, slicing-by-{2..999}

Support export file types:
  * .c for C (support C89, but required ``stdint.h'')
  * .js for javascript (required ECMAScript 6th edition)
  * .rb for ruby (for ruby-2.1+, jruby, and rubinius)
                 (executable for ruby-1.8, ruby-1.9 and ruby-2.0)
                 (executable for mruby and limitation bitsize by fixnum)

examples:
  * create crc-32 calculator to c source (and header file)
    $ rbcrc crc32.c

  * create crc-32c calculator to ruby source
    $ rbcrc crc32c.rb

  * create crc-30-cdma calculator to javascript source
    $ rbcrc crc30cdma.js

  * create crc-32 calculator to ``crc.c'', ``crc.rb'' and ``crc.js''
    $ rbcrc -mcrc32 crc.c crc.rb crc.js

  * create customized crc calculator (as mycrc function) to ``mycrc.c''
    $ rbcrc -s15 -p0x6789 -io -x~0 mycrc.c

  * create customized crc calculator (as MyCRC class) to ``mycrc_1.rb''
    $ rbcrc -s39 -p0x987654321 -IO -x1 -nMyCRC mycrc_1.rb

And, this command has feature too that is print for each CRC specifications as YAML format.

$ rbcrc -lvv
...snip...
"CRC-32":
  bitsize:              32
  polynomial:           0x04C11DB7  # 0xEDB88320 (bit reflected)
  reversed reciprocal:  0x82608EDB  # 0xDB710641 (bit reflected)
  reflect input:        true
  reflect output:       true
  initial crc:          0x00000000  # 0xFFFFFFFF (initial state)
  xor output:           0xFFFFFFFF
  magic number:         0x2144DF1C  # 0xDEBB20E3 (internal state)
  another names:
  - "CRC-32-ADCCP"
  - "CRC-32-PKZIP"
  - "PKZIP"
...snip...

arc-crc

(Written in japanese from here)

crc-0.4 にて、任意の CRC となるバイト列を逆算する機能が正式に追加されました。

require "crc/acrc" にて、その機能が利用可能となります。

名前の由来は、arc-sin などの C 関数である asin と同様に、arc-crc => acrc となっています。

以下は使用例です。

  • 文字列 "123456789????" を CRC32 した場合に 0 となるような、???? の部分を逆算する

    require "crc/acrc"
    
    seq = "123456789"
    seq << CRC.crc32.acrc(seq)
    p CRC.crc32[seq] # => #<CRC::CRC32:00000000>
    
  • 文字列 "123456789????ABCDEFG" の、???? の部分を逆算する

    require "crc/acrc"
    
    seq1 = "123456789"
    seq2 = "ABCDEFG"
    seq = seq1 + CRC.crc32.acrc(seq1, seq2) + seq2
    p CRC.crc32[seq] # => #<CRC::CRC32:00000000>
    
  • 文字列 "123456789????ABCDEFG" を CRC32 した場合に 0x12345678 となるような、???? の部分を逆算する

    require "crc/acrc"
    
    seq1 = "123456789"
    seq2 = "ABCDEFG"
    target_crc = 0x12345678
    seq = seq1 + CRC.crc32.acrc(seq1, seq2, target_crc) + seq2
    p CRC.crc32[seq] # => #<CRC::CRC32:12345678>
    
  • 独自仕様の CRC モジュールにも対応

    require "crc/acrc"
    
    seq1 = "123456789"
    seq2 = "ABCDEFG"
    target_crc = 0x12345678
    MyCRC = CRC.new(29, rand(1 << 29) | 1)
    seq = seq1 + MyCRC.acrc(seq1, seq2, target_crc) + seq2
    p MyCRC[seq] # => #<MyCRC:12345678>
    

[EOF]