digest-ed2k 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- data/.gemtest +0 -0
- data/Rakefile +29 -0
- data/digest-ed2k.gemspec +17 -0
- data/lib/digest/ed2k.rb +193 -0
- data/t/zeroes.rb +41 -0
- metadata +69 -0
data/.gemtest
ADDED
File without changes
|
data/Rakefile
ADDED
@@ -0,0 +1,29 @@
|
|
1
|
+
require 'rake/testtask'
|
2
|
+
require 'rdoc/task'
|
3
|
+
|
4
|
+
spec = eval(File.read("digest-ed2k.gemspec"))
|
5
|
+
source = FileList["lib/**/*.rb"]
|
6
|
+
|
7
|
+
begin
|
8
|
+
# why is require documented to return true/false
|
9
|
+
# if it raises LoadError when it should return false
|
10
|
+
require 'yard'
|
11
|
+
|
12
|
+
YARD::Rake::YardocTask.new do |yd|
|
13
|
+
yd.files = source
|
14
|
+
end
|
15
|
+
|
16
|
+
task :doc => :yard
|
17
|
+
rescue LoadError
|
18
|
+
task :doc => :rdoc
|
19
|
+
end
|
20
|
+
|
21
|
+
RDoc::Task.new do |rd|
|
22
|
+
rd.rdoc_files = source
|
23
|
+
end
|
24
|
+
|
25
|
+
Rake::TestTask.new do |t|
|
26
|
+
t.libs << 't'
|
27
|
+
t.test_files = spec.test_files
|
28
|
+
t.verbose = false
|
29
|
+
end
|
data/digest-ed2k.gemspec
ADDED
@@ -0,0 +1,17 @@
|
|
1
|
+
Gem::Specification.new do |s|
|
2
|
+
s.name = 'digest-ed2k'
|
3
|
+
s.version = '1.0.0'
|
4
|
+
s.date = '2012-04-11'
|
5
|
+
s.license = 'ISC'
|
6
|
+
s.summary = 'Digest module for calculating ED2K hashes.'
|
7
|
+
s.description = s.summary + ' Uses openssl for MD4 hashing.'
|
8
|
+
s.authors = ['Kovensky']
|
9
|
+
s.email = ['diogomfranco@gmail.com']
|
10
|
+
s.test_files = Dir["t/*.rb"]
|
11
|
+
s.files = Dir["lib/**/*.rb"] + s.test_files +
|
12
|
+
["#{s.name}.gemspec","Rakefile", ".gemtest"]
|
13
|
+
s.require_path = 'lib'
|
14
|
+
s.homepage = 'https://github.com/Kovensky/digest-ed2k'
|
15
|
+
|
16
|
+
s.add_development_dependency 'yard'
|
17
|
+
end
|
data/lib/digest/ed2k.rb
ADDED
@@ -0,0 +1,193 @@
|
|
1
|
+
# Implementation of the ed2k hash algorithm.
|
2
|
+
|
3
|
+
# @author Kovensky (mailto:diogomfranco@gmail.com)
|
4
|
+
|
5
|
+
require 'openssl'
|
6
|
+
require 'digest'
|
7
|
+
|
8
|
+
# {Digest::Class Digest} subclass that calculates an ed2k hash.
|
9
|
+
#
|
10
|
+
# The implemented algorithm is as described on http://wiki.anidb.net/w/Ed2k,
|
11
|
+
# using the "red code" method (appending null to hashlist).
|
12
|
+
#
|
13
|
+
# Uses {OpenSSL::Digest OpenSSL::Digest::MD4} internally for the MD4 hashing.
|
14
|
+
#
|
15
|
+
# @note Due to ed2k's use of 9500KB chunks, this object can be up to around
|
16
|
+
# 10MB large in memory, not counting OpenSSL::Digest::MD4's state.
|
17
|
+
class Digest::ED2k < Digest::Class
|
18
|
+
# The version
|
19
|
+
VERSION = '1.0.0'
|
20
|
+
|
21
|
+
# Chunk size of the ed2k hash, 9500KB.
|
22
|
+
CHUNK_SIZE = 9728000
|
23
|
+
|
24
|
+
# Creates a reset object.
|
25
|
+
# @param initial_chunk optionally appends data to the new object
|
26
|
+
def initialize (initial_chunk = nil)
|
27
|
+
@md4 = OpenSSL::Digest::MD4.new
|
28
|
+
self.reset
|
29
|
+
self << initial_chunk if initial_chunk
|
30
|
+
end
|
31
|
+
|
32
|
+
# Reads the contents of `io`, 9500KB at a time,
|
33
|
+
# until EOF, and add to the digest.
|
34
|
+
#
|
35
|
+
# @param [IO] io the {IO} to be read
|
36
|
+
# @return self for convenient chaining.
|
37
|
+
def io (io)
|
38
|
+
# why do I have to use a while instead of an each_chunk or sth
|
39
|
+
buf = ""
|
40
|
+
while io.read CHUNK_SIZE, buf
|
41
|
+
self << buf
|
42
|
+
end
|
43
|
+
return self
|
44
|
+
end
|
45
|
+
|
46
|
+
# Opens the file pointed by path and calls {#io} with it.
|
47
|
+
#
|
48
|
+
# @param [String] path path to a file that will be read
|
49
|
+
# @return self for convenient chaining.
|
50
|
+
def file (path)
|
51
|
+
File.open(path) do |f|
|
52
|
+
return self.io f
|
53
|
+
end
|
54
|
+
end
|
55
|
+
|
56
|
+
# Resets the state; effectively the same as constructing a new object.
|
57
|
+
# @return self for convenient chaining.
|
58
|
+
def reset
|
59
|
+
@md4.reset
|
60
|
+
@buf = ""
|
61
|
+
@rounds = 0
|
62
|
+
@finalized = false
|
63
|
+
return self
|
64
|
+
end
|
65
|
+
|
66
|
+
# Append `data` to the digest. Will raise an {ArgumentError}
|
67
|
+
# if the object has been {#finalize}d before.
|
68
|
+
#
|
69
|
+
# Every 9500KB of accumulated data will be hashed as per the ed2k algorithm.
|
70
|
+
#
|
71
|
+
# @param [String] data the string to be appended to the digest
|
72
|
+
# @return self for convenient chaining.
|
73
|
+
def update (data)
|
74
|
+
if @finalized
|
75
|
+
raise ArgumentError.new("Can't add to an ed2k hash after finalizing. Call reset if you want to calculate a new hash.")
|
76
|
+
end
|
77
|
+
@buf += data
|
78
|
+
_sync
|
79
|
+
return self
|
80
|
+
end
|
81
|
+
alias << update
|
82
|
+
|
83
|
+
# {#finalize}s the digest and returns it.
|
84
|
+
#
|
85
|
+
# @note Due to how ed2k hashes work, once the digest has been obtained,
|
86
|
+
# the object must be {#reset} before being reused. The {#digest!} family
|
87
|
+
# of methods automatically resets the object after being called.
|
88
|
+
#
|
89
|
+
# @param str if present and non-nil, will cause digest to reset self,
|
90
|
+
# update it with `str` and return the new digest.
|
91
|
+
# @return [String] the raw digest
|
92
|
+
def digest(str = nil)
|
93
|
+
unless str
|
94
|
+
self.finalize
|
95
|
+
@md4.digest
|
96
|
+
else
|
97
|
+
reset
|
98
|
+
self << str
|
99
|
+
digest
|
100
|
+
end
|
101
|
+
end
|
102
|
+
|
103
|
+
# (see #digest)
|
104
|
+
# @note calls {#reset} on self after obtaining digest
|
105
|
+
def digest!
|
106
|
+
ret = digest
|
107
|
+
reset
|
108
|
+
return ret
|
109
|
+
end
|
110
|
+
|
111
|
+
# {include:#digest}
|
112
|
+
# @note (see #digest)
|
113
|
+
# @param (see #digest)
|
114
|
+
# @return [String] the digest as a hexadecimal string
|
115
|
+
def hexdigest(str = nil)
|
116
|
+
unless str
|
117
|
+
self.finalize
|
118
|
+
@md4.hexdigest
|
119
|
+
else
|
120
|
+
reset
|
121
|
+
self << str
|
122
|
+
hexdigest
|
123
|
+
end
|
124
|
+
end
|
125
|
+
|
126
|
+
# {include:#hexdigest}
|
127
|
+
# @param (see #hexdigest)
|
128
|
+
# @return (see #hexdigest)
|
129
|
+
# @note (see #digest!)
|
130
|
+
def hexdigest!
|
131
|
+
ret = hexdigest
|
132
|
+
reset
|
133
|
+
return ret
|
134
|
+
end
|
135
|
+
|
136
|
+
# Finalizes the digest and prevents any new data from being added
|
137
|
+
# to it. This is automatically called by the {#digest} family of methods.
|
138
|
+
# The only way to unlock the object is to {#reset} it.
|
139
|
+
#
|
140
|
+
# @return self for convenient chaining.
|
141
|
+
def finalize
|
142
|
+
unless @finalized
|
143
|
+
if @rounds > 0
|
144
|
+
@md4 << OpenSSL::Digest::MD4.digest(@buf)
|
145
|
+
else
|
146
|
+
@md4.reset
|
147
|
+
@md4 << @buf
|
148
|
+
end
|
149
|
+
@buf = nil
|
150
|
+
@finalized = true
|
151
|
+
end
|
152
|
+
return self
|
153
|
+
end
|
154
|
+
|
155
|
+
# Override for {Object#inspect Object's inspect}
|
156
|
+
def inspect
|
157
|
+
return "#<ed2k unfinalized>" unless @finalized
|
158
|
+
return "#<ed2k hash=\"#{digest}\">"
|
159
|
+
end
|
160
|
+
|
161
|
+
private
|
162
|
+
# When the internal buffer exceeds {CHUNK_SIZE}, feed CHUNK_SIZE chunks
|
163
|
+
# to the MD4 hashing function until the buffer is below CHUNK_SIZE
|
164
|
+
# of length.
|
165
|
+
def _sync
|
166
|
+
while @buf.length >= CHUNK_SIZE
|
167
|
+
@md4 << OpenSSL::Digest::MD4.digest(@buf[0...CHUNK_SIZE])
|
168
|
+
@buf = @buf[CHUNK_SIZE...@buf.length] || ""
|
169
|
+
@rounds += 1
|
170
|
+
end
|
171
|
+
end
|
172
|
+
|
173
|
+
public
|
174
|
+
# Calculates and returns the {#digest} on data.
|
175
|
+
def self.digest (data)
|
176
|
+
return new(data).digest
|
177
|
+
end
|
178
|
+
|
179
|
+
# Calculates and returns the {#hexdigest} on data.
|
180
|
+
def self.hexdigest (data)
|
181
|
+
return new(data).hexdigest
|
182
|
+
end
|
183
|
+
|
184
|
+
# The same as calling {#io}(io) on a new object.
|
185
|
+
def self.io(io)
|
186
|
+
return new.io io
|
187
|
+
end
|
188
|
+
|
189
|
+
# The same as calling {#file}(path) on a new object.
|
190
|
+
def self.file(path)
|
191
|
+
return new.file path
|
192
|
+
end
|
193
|
+
end
|
data/t/zeroes.rb
ADDED
@@ -0,0 +1,41 @@
|
|
1
|
+
#require 'minitest/autorun'
|
2
|
+
require 'test/unit'
|
3
|
+
require 'digest/ed2k'
|
4
|
+
|
5
|
+
class Zeroes < Test::Unit::TestCase
|
6
|
+
def test_zero_length_file
|
7
|
+
assert_equal '31d6cfe0d16ae931b73c59d7e0c089c0',
|
8
|
+
Digest::ED2k.hexdigest('')
|
9
|
+
end
|
10
|
+
|
11
|
+
def add_by_random_chunks (str, ed2k = Digest::ED2k.new)
|
12
|
+
loops = 2000 # don't let bad random numbers freeze the test
|
13
|
+
while str.length > 1 && loops > 0
|
14
|
+
r = rand(str.length)
|
15
|
+
ed2k << str[0...r]
|
16
|
+
str = str[r...str.length]
|
17
|
+
loops -= 1
|
18
|
+
end
|
19
|
+
ed2k << str
|
20
|
+
end
|
21
|
+
|
22
|
+
def test_single_chunk_zero_file
|
23
|
+
assert_equal 'fc21d9af828f92a8df64beac3357425d',
|
24
|
+
add_by_random_chunks("\0" * Digest::ED2k::CHUNK_SIZE).hexdigest
|
25
|
+
end
|
26
|
+
|
27
|
+
def test_two_chunk_zero_file
|
28
|
+
assert_equal '114b21c63a74b6ca922291a11177dd5c',
|
29
|
+
add_by_random_chunks("\0" * (2 * Digest::ED2k::CHUNK_SIZE)).hexdigest
|
30
|
+
end
|
31
|
+
|
32
|
+
def test_two_and_half_chunk_zero_file
|
33
|
+
assert_equal 'a4b0b40d5b13bb045b0ce2a8fad8b393',
|
34
|
+
add_by_random_chunks("\0" * (5 * Digest::ED2k::CHUNK_SIZE / 2)).hexdigest
|
35
|
+
end
|
36
|
+
|
37
|
+
def test_three_and_half_chunk_by_object_random
|
38
|
+
assert_equal '57d514e8e464257aca18e50e47194e42',
|
39
|
+
add_by_random_chunks("\0" * (7 * Digest::ED2k::CHUNK_SIZE / 2)).hexdigest
|
40
|
+
end
|
41
|
+
end
|
metadata
ADDED
@@ -0,0 +1,69 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: digest-ed2k
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 1.0.0
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- Kovensky
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2012-04-11 00:00:00.000000000 Z
|
13
|
+
dependencies:
|
14
|
+
- !ruby/object:Gem::Dependency
|
15
|
+
name: yard
|
16
|
+
requirement: !ruby/object:Gem::Requirement
|
17
|
+
none: false
|
18
|
+
requirements:
|
19
|
+
- - ! '>='
|
20
|
+
- !ruby/object:Gem::Version
|
21
|
+
version: '0'
|
22
|
+
type: :development
|
23
|
+
prerelease: false
|
24
|
+
version_requirements: !ruby/object:Gem::Requirement
|
25
|
+
none: false
|
26
|
+
requirements:
|
27
|
+
- - ! '>='
|
28
|
+
- !ruby/object:Gem::Version
|
29
|
+
version: '0'
|
30
|
+
description: Digest module for calculating ED2K hashes. Uses openssl for MD4 hashing.
|
31
|
+
email:
|
32
|
+
- diogomfranco@gmail.com
|
33
|
+
executables: []
|
34
|
+
extensions: []
|
35
|
+
extra_rdoc_files: []
|
36
|
+
files:
|
37
|
+
- lib/digest/ed2k.rb
|
38
|
+
- t/zeroes.rb
|
39
|
+
- digest-ed2k.gemspec
|
40
|
+
- Rakefile
|
41
|
+
- .gemtest
|
42
|
+
homepage: https://github.com/Kovensky/digest-ed2k
|
43
|
+
licenses:
|
44
|
+
- ISC
|
45
|
+
post_install_message:
|
46
|
+
rdoc_options: []
|
47
|
+
require_paths:
|
48
|
+
- lib
|
49
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
50
|
+
none: false
|
51
|
+
requirements:
|
52
|
+
- - ! '>='
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: '0'
|
55
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
56
|
+
none: false
|
57
|
+
requirements:
|
58
|
+
- - ! '>='
|
59
|
+
- !ruby/object:Gem::Version
|
60
|
+
version: '0'
|
61
|
+
requirements: []
|
62
|
+
rubyforge_project:
|
63
|
+
rubygems_version: 1.8.21
|
64
|
+
signing_key:
|
65
|
+
specification_version: 3
|
66
|
+
summary: Digest module for calculating ED2K hashes.
|
67
|
+
test_files:
|
68
|
+
- t/zeroes.rb
|
69
|
+
has_rdoc:
|