Crcs

This gem provides C extension for calculating CRC32 digests. It's based on table-driven slice-by-8 CRC32 algorithm.

Nothing special. Simply run

gem install crcs

or add folowing line to your Gemfile

gem 'crcs'

This gem integrates to Ruby Digest module. This way you can easily use all methods that this module provides. Just use Digest::CRC32 class after requiring gem into your code. For example:

# Generate hexdigest for sample string
Digest::CRC32.hexdigest('Your data')
# or
crc32 = Digest::CRC32.new
crc32 << 'Data'
crc32 << 'More data'
crc32.hexdigest

# Generate file digest
Digest::CRC32.file('filename.txt').hexdigest

Let's compare this gem with few others:

• crc32 which provides C extension with the same algorithm but without all that fancy Digest module staff. It has only one method calculate. So you should care about previous crc32 digests and data length by yourself
• Standard zlib ruby implementation
• digest-crc which is pure Ruby implementation of few CRC-algorithms

First of all we will create a string with random data provided by SecureRandom.random_bytes. I choosed variable with 5 megabytes of random bytes Next let's create digest for that string 1000 times by each of gems.

All benchmarks performed on MacBook Pro 15 mid 2015 with 2,2 GHz Intel Core i7

                           user      system     total       real
crcs                    12.660000   0.030000  12.690000 ( 12.709954)
tdobrovolskij/crc32      3.550000   0.010000   3.560000 (  3.574127)
zlib                     0.470000   0.000000   0.470000 (  0.471863)
postmodern/digest-crc  947.670000   1.330000 949.000000 (950.414528)

For some reasons my gem is terribly slower than crc32 and zlib libraries. I think it's caused by overhead of creating digest objects and other things that happening inside of that module. Maybe I should extract my code to separate class and try again? Fortunately, as you can see, my gem isn't so bad comparing with pure ruby implementation.