markov_words 0.1.1 → 0.2.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/Gemfile +2 -2
- data/Gemfile.lock +1 -1
- data/Rakefile +6 -6
- data/bin/console +5 -5
- data/lib/markov_words/file_store.rb +39 -0
- data/lib/markov_words/version.rb +1 -1
- data/lib/markov_words/words.rb +213 -0
- data/lib/markov_words.rb +3 -223
- data/markov_words.gemspec +16 -17
- metadata +14 -12
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 5dda90c58c74410faf33ed06402150326be2afe8
|
4
|
+
data.tar.gz: 77cfaf41ef3314807e06b2ef12f5aa7bc42cbe4a
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 14d952d8db7fe1f162bccf3c6ab9f1fcc389ff5cf91000f5f0bb87761860592148f4cd2bdc1fdb82e83ba9037316181f8c6987e10f62dc52c248037f5a902b7c
|
7
|
+
data.tar.gz: 7ab05abcd9938125c56c76f82eec3eab0d46cc3588a4b120f9736830a19a334c988eac762a09360523128889654b76dad03002caf4c8be6ea4017d8ce6b9846f
|
data/Gemfile
CHANGED
@@ -1,6 +1,6 @@
|
|
1
|
-
source
|
1
|
+
source 'https://rubygems.org'
|
2
2
|
|
3
|
-
git_source(:github) {|repo_name| "https://github.com/#{repo_name}" }
|
3
|
+
git_source(:github) { |repo_name| "https://github.com/#{repo_name}" }
|
4
4
|
|
5
5
|
# Specify your gem's dependencies in markov_words.gemspec
|
6
6
|
gemspec
|
data/Gemfile.lock
CHANGED
data/Rakefile
CHANGED
@@ -1,10 +1,10 @@
|
|
1
|
-
require
|
2
|
-
require
|
1
|
+
require 'bundler/gem_tasks'
|
2
|
+
require 'rake/testtask'
|
3
3
|
|
4
4
|
Rake::TestTask.new(:test) do |t|
|
5
|
-
t.libs <<
|
6
|
-
t.libs <<
|
7
|
-
t.test_files = FileList[
|
5
|
+
t.libs << 'test'
|
6
|
+
t.libs << 'lib'
|
7
|
+
t.test_files = FileList['test/**/*_test.rb']
|
8
8
|
end
|
9
9
|
|
10
|
-
task :
|
10
|
+
task default: :test
|
data/bin/console
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
#!/usr/bin/env ruby
|
2
2
|
|
3
|
-
require
|
4
|
-
require
|
3
|
+
require 'bundler/setup'
|
4
|
+
require 'markov_words'
|
5
5
|
|
6
6
|
# You can add fixtures and/or initialization code here to make experimenting
|
7
7
|
# with your gem easier. You can also use a different console, if you like.
|
8
8
|
|
9
9
|
# (If you use this, don't forget to add pry to your Gemfile!)
|
10
|
-
require
|
10
|
+
require 'pry'
|
11
11
|
Pry.start
|
12
12
|
|
13
|
-
#require
|
14
|
-
#IRB.start(__FILE__)
|
13
|
+
# require 'irb'
|
14
|
+
# IRB.start(__FILE__)
|
@@ -0,0 +1,39 @@
|
|
1
|
+
require 'securerandom'
|
2
|
+
|
3
|
+
module MarkovWords
|
4
|
+
# Utility for persisting arbitrary data to disk as Marshal'ed Ruby objects
|
5
|
+
class FileStore
|
6
|
+
attr_reader :file_path
|
7
|
+
attr_reader :data
|
8
|
+
|
9
|
+
# @option opts [String] :file_path Path and name for where the file should
|
10
|
+
# be stored.
|
11
|
+
# @option opts [Boolean] :flush_data Do you want the file to be cleared on
|
12
|
+
# open?
|
13
|
+
def initialize(opts)
|
14
|
+
@file_path = opts.fetch :file_path, "/tmp/#{SecureRandom.base64}"
|
15
|
+
delete_if_exists(@file_path) if opts[:flush_data]
|
16
|
+
end
|
17
|
+
|
18
|
+
# Store arbitary data into file storage
|
19
|
+
# @param data [Object] Any Marshal-able object
|
20
|
+
def store_data(data)
|
21
|
+
File.open(@file_path, 'wb') { |f| Marshal.dump(data, f) }
|
22
|
+
end
|
23
|
+
|
24
|
+
# Retrieve whatever data is stored in the file + return it
|
25
|
+
def retrieve_data
|
26
|
+
result = nil
|
27
|
+
if File.exist?(@file_path)
|
28
|
+
File.open(@file_path, 'r') { |f| result = Marshal.load(f) }
|
29
|
+
end
|
30
|
+
result
|
31
|
+
end
|
32
|
+
|
33
|
+
private
|
34
|
+
|
35
|
+
def delete_if_exists(path)
|
36
|
+
File.delete path if File.exist? path
|
37
|
+
end
|
38
|
+
end
|
39
|
+
end
|
data/lib/markov_words/version.rb
CHANGED
@@ -0,0 +1,213 @@
|
|
1
|
+
module MarkovWords
|
2
|
+
# This class takes care of word generation, caching, and data storage.
|
3
|
+
class Words
|
4
|
+
# Perform caching? Defaults to true.
|
5
|
+
attr_reader :cache
|
6
|
+
# File location where you want to store the cache
|
7
|
+
attr_reader :cache_file
|
8
|
+
# How many words you want to store in the cache?
|
9
|
+
attr_reader :cache_size
|
10
|
+
# Object for storing + retrieving cache data from persistent storage
|
11
|
+
attr_reader :cache_store
|
12
|
+
# Your dictionary of words. Defaults to /usr/share/dict/words.
|
13
|
+
attr_reader :corpus_file
|
14
|
+
# Where should your database be stored on disk?
|
15
|
+
attr_reader :data_file
|
16
|
+
# Object for storing + retrieving n-gram data from persistent storage
|
17
|
+
attr_reader :data_store
|
18
|
+
# The database of "grams" (word/count combinations), stored on the disk and
|
19
|
+
# loaded into this variable in memory when generating words.
|
20
|
+
attr_reader :grams
|
21
|
+
# Number of n-grams to compute for your database. Defaults to 2
|
22
|
+
attr_reader :gram_size
|
23
|
+
# Max generated word length. Defaults to 16
|
24
|
+
attr_reader :max_length
|
25
|
+
# Minimum generated word length. Defaults to 3. NOTE: If your corpus size
|
26
|
+
# is very small (<1000 words or so), it's hard to guarantee a min_length
|
27
|
+
# because so many n-grams will have no association, which terminates word
|
28
|
+
# generation.
|
29
|
+
attr_reader :min_length
|
30
|
+
|
31
|
+
# Create a new "Words" object
|
32
|
+
# @param opts [Hash] options sent to the object. Any of the object
|
33
|
+
# attributes (eg `:cache_file` or `:gram_size`) are valid parameters to
|
34
|
+
# add to the `opts` hash.
|
35
|
+
# @return [Words] A `MarkovWords::Words` object.
|
36
|
+
def initialize(opts = {})
|
37
|
+
@grams = nil
|
38
|
+
@gram_size = opts.fetch :gram_size, 2
|
39
|
+
@max_length = opts.fetch :max_length, 16
|
40
|
+
@min_length = opts.fetch :min_length, 3
|
41
|
+
|
42
|
+
initialize_cache(opts)
|
43
|
+
initialize_data(opts)
|
44
|
+
end
|
45
|
+
|
46
|
+
# "Top off" the cache of stored words, and ensure that it's at
|
47
|
+
# `@cache_size`. If `@cache` is set to `false`, returns an empty array.
|
48
|
+
# @return [Array<String>] All words in the cache.
|
49
|
+
def refresh_cache
|
50
|
+
if @cache
|
51
|
+
words_array = @cache_store.retrieve_data
|
52
|
+
words_array << generate_word while words_array.length < @cache_size
|
53
|
+
@cache_store.store_data words_array
|
54
|
+
words_array
|
55
|
+
else
|
56
|
+
[]
|
57
|
+
end
|
58
|
+
end
|
59
|
+
|
60
|
+
# Generate a new word, or return one from the cache if available.
|
61
|
+
# @return [String] The word.
|
62
|
+
def word
|
63
|
+
if @cache
|
64
|
+
load_word_from_cache
|
65
|
+
else
|
66
|
+
generate_word
|
67
|
+
end
|
68
|
+
end
|
69
|
+
|
70
|
+
private
|
71
|
+
|
72
|
+
def initialize_cache(opts)
|
73
|
+
@cache = opts.fetch :cache, true
|
74
|
+
@cache_file = opts.fetch :cache_file,
|
75
|
+
"tmp/markov_words_#{@gram_size}.cache"
|
76
|
+
@cache_size = opts.fetch :cache_size, 70
|
77
|
+
@cache_store = FileStore.new(file_path: @cache_file)
|
78
|
+
end
|
79
|
+
|
80
|
+
def initialize_data(opts)
|
81
|
+
@corpus_file = opts.fetch :corpus_file,
|
82
|
+
'/usr/share/dict/words'
|
83
|
+
@data_file = opts.fetch :data_file,
|
84
|
+
"tmp/markov_words_#{@gram_size}.data"
|
85
|
+
@data_store = FileStore.new(file_path: @data_file)
|
86
|
+
end
|
87
|
+
|
88
|
+
def contains_vowel?(ary)
|
89
|
+
if ary.length < 2
|
90
|
+
true
|
91
|
+
else
|
92
|
+
ary.take(2).join.match(/[aeiou]/)
|
93
|
+
end
|
94
|
+
end
|
95
|
+
|
96
|
+
# Generates an English (by default) -sounding word.
|
97
|
+
def generate_word
|
98
|
+
set_grams if @grams.nil?
|
99
|
+
generate_gram_array(generate_word_length).join
|
100
|
+
end
|
101
|
+
|
102
|
+
def generate_gram_array(desired_length)
|
103
|
+
gram = ''
|
104
|
+
gram_array = generate_initial_gram_array
|
105
|
+
until gram_array.join.length == desired_length || gram.nil?
|
106
|
+
# grab last @gram_size (or possibly fewer if the array is too small)
|
107
|
+
# elements from the current gram_array, to use as the next key.
|
108
|
+
gal = gram_array.length
|
109
|
+
current_gram_size = gal >= @gram_size ? @gram_size : gal
|
110
|
+
key = gram_array[-current_gram_size..-1].join
|
111
|
+
|
112
|
+
gram = pick_random_char(@grams[key])
|
113
|
+
gram_array << gram
|
114
|
+
end
|
115
|
+
gram_array
|
116
|
+
end
|
117
|
+
|
118
|
+
# Set initial array of chars, which is taken from the @grams key list.
|
119
|
+
# must contain a vowel in the first 2 chars (unless @gram_size == 1 in
|
120
|
+
# which case any letter).
|
121
|
+
def generate_initial_gram_array
|
122
|
+
initial_gram_array = []
|
123
|
+
|
124
|
+
all_grams_array = @grams.to_a
|
125
|
+
gram_min_length = @gram_size < @min_length ? @gram_size : @min_length
|
126
|
+
until initial_gram_array.length >= gram_min_length &&
|
127
|
+
contains_vowel?(initial_gram_array)
|
128
|
+
initial_gram_array = all_grams_array.sample[0].chars
|
129
|
+
end
|
130
|
+
initial_gram_array
|
131
|
+
end
|
132
|
+
|
133
|
+
# The word must be a random length, between @min and @max
|
134
|
+
def generate_word_length
|
135
|
+
word_length = 0
|
136
|
+
until word_length >= @min_length
|
137
|
+
word_length = SecureRandom.rand(@max_length)
|
138
|
+
end
|
139
|
+
word_length
|
140
|
+
end
|
141
|
+
|
142
|
+
def load_word_from_cache
|
143
|
+
words_array = @cache_store.retrieve_data
|
144
|
+
if words_array.nil? || words_array.empty?
|
145
|
+
words_array = Array.new(@cache_size) { generate_word }
|
146
|
+
end
|
147
|
+
|
148
|
+
word = words_array.pop
|
149
|
+
cache_store.store_data words_array
|
150
|
+
|
151
|
+
word
|
152
|
+
end
|
153
|
+
|
154
|
+
# Generate a MarkovWords corpus from a datafile, with a given size of
|
155
|
+
# n-gram. Returns a hash of "grams", which are a map of a letter to the
|
156
|
+
# frequency of the letters that follow it, eg: {"c" => {"a" => 1, "b" =>
|
157
|
+
# 2}}
|
158
|
+
def markov_corpus(file, gram_size)
|
159
|
+
grams = {}
|
160
|
+
|
161
|
+
# Corpus contains a list of words, separated by newlines
|
162
|
+
File.foreach(file) do |word|
|
163
|
+
word = word.downcase.delete('-')
|
164
|
+
gram_size.downto(1) do |current_gram_size|
|
165
|
+
markov_update_count! grams, word, current_gram_size
|
166
|
+
end
|
167
|
+
end
|
168
|
+
|
169
|
+
grams
|
170
|
+
end
|
171
|
+
|
172
|
+
# Given a database of `grams` and a `word`, and the `gram_size` (the
|
173
|
+
# maximum n-gram size we want to compute), update the `grams` database with
|
174
|
+
# entries for each n-gram combination starting at `gram_size` and going
|
175
|
+
# down to 1.
|
176
|
+
def markov_update_count!(grams, word, gram_size)
|
177
|
+
word.chars.each_cons(gram_size + 1) do |gram|
|
178
|
+
l = gram[0..gram_size - 1].join
|
179
|
+
r = gram[gram_size]
|
180
|
+
|
181
|
+
unless l.empty? || r.empty? || line_ending?(r)
|
182
|
+
grams[l] = {} if grams[l].nil?
|
183
|
+
grams[l][r] = grams[l][r].nil? ? 1 : grams[l][r] += 1
|
184
|
+
end
|
185
|
+
end
|
186
|
+
end
|
187
|
+
|
188
|
+
# Given a hash in the format: {"c" => {"a" => 1, "b" => 2}}, grab a random
|
189
|
+
# element from the values hash, accurate to the distribution of counts.
|
190
|
+
# In the example hash above, "a" would have a 33% chance of being chosen,
|
191
|
+
# while "b" would have a 66% chance (1/2 ratio).
|
192
|
+
def pick_random_char(counts_hash = {})
|
193
|
+
total = counts_hash.values.sum
|
194
|
+
pick_num = SecureRandom.rand(total)
|
195
|
+
counter = 0
|
196
|
+
counts_hash.each do |char, count|
|
197
|
+
counter += count
|
198
|
+
return char if counter >= pick_num
|
199
|
+
end
|
200
|
+
end
|
201
|
+
|
202
|
+
def line_ending?(word)
|
203
|
+
word.include?("\n")
|
204
|
+
end
|
205
|
+
|
206
|
+
def set_grams
|
207
|
+
grams = @data_store.retrieve_data ||
|
208
|
+
markov_corpus(@corpus_file, @gram_size)
|
209
|
+
@data_store.store_data grams unless grams == @grams
|
210
|
+
@grams = grams
|
211
|
+
end
|
212
|
+
end
|
213
|
+
end
|
data/lib/markov_words.rb
CHANGED
@@ -1,228 +1,8 @@
|
|
1
|
-
require
|
1
|
+
require 'markov_words/version'
|
2
|
+
require 'markov_words/words'
|
3
|
+
require 'markov_words/file_store'
|
2
4
|
require 'securerandom'
|
3
5
|
|
4
6
|
# @author Donald L. Merand
|
5
|
-
# A nice library for generating random words (not sentences) using Markov
|
6
|
-
# chains.
|
7
7
|
module MarkovWords
|
8
|
-
|
9
|
-
# This class takes care of word generation, caching, and data storage.
|
10
|
-
class Words
|
11
|
-
# Perform caching? Defaults to true.
|
12
|
-
attr :cache
|
13
|
-
# File location where you want to store the cache
|
14
|
-
attr :cache_file
|
15
|
-
# How many words you want to store in the cache?
|
16
|
-
attr :cache_size
|
17
|
-
# Your dictionary of words. Defaults to /usr/share/dict/words.
|
18
|
-
attr :corpus_file
|
19
|
-
# Where should your database be stored on disk?
|
20
|
-
attr :data_file
|
21
|
-
# The database of "grams" (word/count combinations), stored on the disk and
|
22
|
-
# loaded into this variable in memory when generating words.
|
23
|
-
attr :grams
|
24
|
-
# Number of n-grams to compute for your database. Defaults to 2
|
25
|
-
attr :gram_size
|
26
|
-
# Max generated word length. Defaults to 16
|
27
|
-
attr :max_length
|
28
|
-
# Minimum generated word length. Defaults to 3. NOTE: If your corpus size
|
29
|
-
# is very small (<1000 words or so), it's hard to guarantee a min_length
|
30
|
-
# because so many n-grams will have no association, which terminates word
|
31
|
-
# generation.
|
32
|
-
attr :min_length
|
33
|
-
|
34
|
-
# Create a new "Words" object
|
35
|
-
# @param opts [Hash] options sent to the object. Any of the object
|
36
|
-
# attributes (eg `:cache_file` or `:gram_size`) are valid parameters to
|
37
|
-
# add to the `opts` hash.
|
38
|
-
# @return [Words] A `MarkovWords::Words` object.
|
39
|
-
def initialize(opts = {})
|
40
|
-
@gram_size = opts.fetch :gram_size, 2
|
41
|
-
@max_length = opts.fetch :max_length, 16
|
42
|
-
@min_length = opts.fetch :min_length, 3
|
43
|
-
|
44
|
-
@cache = opts.fetch :cache, true
|
45
|
-
@cache_file = opts.fetch :cache_file,
|
46
|
-
"tmp/markov_words_#{@gram_size}.cache"
|
47
|
-
@cache_size = opts.fetch :cache_size, 70
|
48
|
-
@corpus_file = opts.fetch :corpus_file,
|
49
|
-
'/usr/share/dict/words'
|
50
|
-
@data_file = opts.fetch :data_file,
|
51
|
-
"tmp/markov_words_#{@gram_size}.data"
|
52
|
-
@grams = nil
|
53
|
-
end
|
54
|
-
|
55
|
-
# "Top off" the cache of stored words, and ensure that it's at
|
56
|
-
# `@cache_size`. If `@cache` is set to `false`, returns an empty array.
|
57
|
-
# @return [Array<String>] All words in the cache.
|
58
|
-
def refresh_cache
|
59
|
-
if @cache
|
60
|
-
words_array = load_from_file(@cache_file) || []
|
61
|
-
|
62
|
-
while words_array.length < @cache_size
|
63
|
-
words_array << generate_word
|
64
|
-
end
|
65
|
-
|
66
|
-
save_to_file(@cache_file, words_array)
|
67
|
-
words_array
|
68
|
-
else
|
69
|
-
[]
|
70
|
-
end
|
71
|
-
end
|
72
|
-
|
73
|
-
# Generate a new word, or return one from the cache if available.
|
74
|
-
# @return [String] The word.
|
75
|
-
def word
|
76
|
-
if @cache
|
77
|
-
load_word_from_cache
|
78
|
-
else
|
79
|
-
generate_word
|
80
|
-
end
|
81
|
-
end
|
82
|
-
|
83
|
-
private
|
84
|
-
|
85
|
-
def contains_vowel?(ary)
|
86
|
-
if ary.length < 2
|
87
|
-
true
|
88
|
-
else
|
89
|
-
ary.take(2).join.match(/[aeiou]/)
|
90
|
-
end
|
91
|
-
end
|
92
|
-
|
93
|
-
# Generates an English (by default)- sounding word.
|
94
|
-
def generate_word
|
95
|
-
set_grams if @grams.nil?
|
96
|
-
|
97
|
-
gram = ''
|
98
|
-
gram_array = []
|
99
|
-
|
100
|
-
# The word must be a random length, between @min and @max
|
101
|
-
desired_length = 0
|
102
|
-
until desired_length >= @min_length
|
103
|
-
desired_length = SecureRandom.rand(@max_length)
|
104
|
-
end
|
105
|
-
|
106
|
-
# Set initial array of chars, which is taken from the @grams key list. must
|
107
|
-
# contain a vowel in the first 2 chars (unless @gram_size == 1 in which
|
108
|
-
# case any letter).
|
109
|
-
all_grams_array = @grams.to_a
|
110
|
-
gram_min_length = @gram_size < @min_length ? @gram_size : @min_length
|
111
|
-
until gram_array.length >= gram_min_length && contains_vowel?(gram_array)
|
112
|
-
gram_array = all_grams_array.sample[0].chars
|
113
|
-
end
|
114
|
-
|
115
|
-
until gram_array.join.length == desired_length || gram.nil?
|
116
|
-
# grab last @gram_size (or possibly fewer if the array is too small)
|
117
|
-
# elements from the current gram_array, to use as the next key.
|
118
|
-
gal = gram_array.length
|
119
|
-
current_gram_size = gal >= @gram_size ? @gram_size : gal
|
120
|
-
key = gram_array[-current_gram_size..-1].join
|
121
|
-
|
122
|
-
gram = pick_random_char(@grams[key])
|
123
|
-
gram_array << gram
|
124
|
-
end
|
125
|
-
|
126
|
-
gram_array.join
|
127
|
-
end
|
128
|
-
|
129
|
-
def generate_words_array
|
130
|
-
@cache_size.times.map { generate_word }
|
131
|
-
end
|
132
|
-
|
133
|
-
def load_from_file(file)
|
134
|
-
result = nil
|
135
|
-
if File.exist?(file)
|
136
|
-
File.open(file, 'r') {|f| result = Marshal.load(f)}
|
137
|
-
end
|
138
|
-
result
|
139
|
-
end
|
140
|
-
|
141
|
-
def load_word_from_cache
|
142
|
-
words_array = load_from_file(@cache_file)
|
143
|
-
if words_array.nil? || words_array.empty?
|
144
|
-
words_array = generate_words_array
|
145
|
-
end
|
146
|
-
|
147
|
-
word = words_array.pop
|
148
|
-
save_to_file(@cache_file, words_array)
|
149
|
-
|
150
|
-
word
|
151
|
-
end
|
152
|
-
|
153
|
-
# Generate a MarkovWords corpus from a datafile, with a given size of n-gram.
|
154
|
-
# Returns a hash of "grams", which are a map of a letter to the frequency of
|
155
|
-
# the letters that follow it, eg: {"c" => {"a" => 1, "b" => 2}}
|
156
|
-
def markov_corpus(file, gram_size)
|
157
|
-
grams = {}
|
158
|
-
|
159
|
-
# Corpus contains a list of words, separated by newlines
|
160
|
-
File.foreach(file) do |word|
|
161
|
-
word = word.downcase.gsub(/-/, '')
|
162
|
-
gram_size.downto(1) do |current_size|
|
163
|
-
word.chars.each_cons(current_size + 1) do |gram|
|
164
|
-
first = gram[0..current_size - 1].join
|
165
|
-
second = gram[current_size]
|
166
|
-
|
167
|
-
unless first.empty? || second.empty? || is_line_ending?(second)
|
168
|
-
update_count(grams, first, second)
|
169
|
-
end
|
170
|
-
end
|
171
|
-
end
|
172
|
-
end
|
173
|
-
|
174
|
-
grams
|
175
|
-
end
|
176
|
-
|
177
|
-
# Given a hash in the format: {"c" => {"a" => 1, "b" => 2}}, grab a random
|
178
|
-
# element from the values hash, accurate to the distribution of counts.
|
179
|
-
# In the example hash above, "a" would have a 33% chance of being chosen,
|
180
|
-
# while "b" would have a 66% chance (1/2 ratio).
|
181
|
-
def pick_random_char(counts_hash = {})
|
182
|
-
if counts_hash.nil?
|
183
|
-
return nil
|
184
|
-
else
|
185
|
-
total = counts_hash.values.sum
|
186
|
-
pick_num = SecureRandom.rand(total)
|
187
|
-
counter = 0
|
188
|
-
counts_hash.each do |char, count|
|
189
|
-
counter += count
|
190
|
-
return char if counter >= pick_num
|
191
|
-
end
|
192
|
-
end
|
193
|
-
end
|
194
|
-
|
195
|
-
def is_line_ending?(word)
|
196
|
-
word.include?("\n")
|
197
|
-
end
|
198
|
-
|
199
|
-
# Marshal a Ruby object to file storage
|
200
|
-
def save_to_file(file, data)
|
201
|
-
File.open(file, 'wb') {|f| Marshal.dump(data, f)}
|
202
|
-
end
|
203
|
-
|
204
|
-
def set_grams
|
205
|
-
if File.exist? @data_file
|
206
|
-
@grams = load_from_file(@data_file)
|
207
|
-
else
|
208
|
-
@grams = markov_corpus(@corpus_file, @gram_size)
|
209
|
-
save_to_file(@data_file, @grams)
|
210
|
-
end
|
211
|
-
end
|
212
|
-
|
213
|
-
# Given a @grams entry, update the count of "second" in "first"
|
214
|
-
#
|
215
|
-
# Example:
|
216
|
-
# update_count({"a" => {"b" => 1}}, "a", "b")
|
217
|
-
# => {"a" => {"b" => 2}}
|
218
|
-
def update_count(grams, first, second)
|
219
|
-
grams[first] = {} if grams[first].nil?
|
220
|
-
if grams[first][second].nil?
|
221
|
-
grams[first][second] = 1
|
222
|
-
else
|
223
|
-
grams[first][second] += 1
|
224
|
-
end
|
225
|
-
end
|
226
|
-
|
227
|
-
end
|
228
8
|
end
|
data/markov_words.gemspec
CHANGED
@@ -1,29 +1,28 @@
|
|
1
|
-
|
2
|
-
lib = File.expand_path("../lib", __FILE__)
|
1
|
+
lib = File.expand_path('../lib', __FILE__)
|
3
2
|
$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
|
4
|
-
require
|
3
|
+
require 'markov_words/version'
|
5
4
|
|
6
5
|
Gem::Specification.new do |spec|
|
7
|
-
spec.name =
|
6
|
+
spec.name = 'markov_words'
|
8
7
|
spec.version = MarkovWords::VERSION
|
9
|
-
spec.authors = [
|
10
|
-
spec.email = [
|
8
|
+
spec.authors = ['Donald Merand']
|
9
|
+
spec.email = ['dmerand@explo.org']
|
11
10
|
|
12
|
-
spec.summary = %
|
13
|
-
spec.description = %
|
14
|
-
spec.homepage =
|
15
|
-
spec.license =
|
11
|
+
spec.summary = %{Generate words (not sentences) using Markov-chain techniques.}
|
12
|
+
spec.description = %{It's often nice to have random English-sounding words, eg. for password generators. This library uses Markov-chain techniques on words, as opposed to many others which focus on sentences.}
|
13
|
+
spec.homepage = 'https://github.com/exploration/markov_words'
|
14
|
+
spec.license = 'MIT'
|
16
15
|
|
17
16
|
spec.files = `git ls-files -z`.split("\x0").reject do |f|
|
18
17
|
f.match(%r{^(test|spec|features)/})
|
19
18
|
end
|
20
|
-
spec.bindir =
|
19
|
+
spec.bindir = 'exe'
|
21
20
|
spec.executables = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
|
22
|
-
spec.require_paths = [
|
21
|
+
spec.require_paths = ['lib']
|
23
22
|
|
24
|
-
spec.add_development_dependency
|
25
|
-
spec.add_development_dependency
|
26
|
-
spec.add_development_dependency
|
27
|
-
spec.add_development_dependency
|
28
|
-
spec.add_development_dependency
|
23
|
+
spec.add_development_dependency 'bundler', '~> 1.16'
|
24
|
+
spec.add_development_dependency 'minitest', '~> 5.0'
|
25
|
+
spec.add_development_dependency 'pry', '~> 0.11'
|
26
|
+
spec.add_development_dependency 'rake', '~> 10.0'
|
27
|
+
spec.add_development_dependency 'yard', '~> 0.6'
|
29
28
|
end
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: markov_words
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.
|
4
|
+
version: 0.2.1
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Donald Merand
|
@@ -25,47 +25,47 @@ dependencies:
|
|
25
25
|
- !ruby/object:Gem::Version
|
26
26
|
version: '1.16'
|
27
27
|
- !ruby/object:Gem::Dependency
|
28
|
-
name:
|
28
|
+
name: minitest
|
29
29
|
requirement: !ruby/object:Gem::Requirement
|
30
30
|
requirements:
|
31
31
|
- - "~>"
|
32
32
|
- !ruby/object:Gem::Version
|
33
|
-
version: '
|
33
|
+
version: '5.0'
|
34
34
|
type: :development
|
35
35
|
prerelease: false
|
36
36
|
version_requirements: !ruby/object:Gem::Requirement
|
37
37
|
requirements:
|
38
38
|
- - "~>"
|
39
39
|
- !ruby/object:Gem::Version
|
40
|
-
version: '
|
40
|
+
version: '5.0'
|
41
41
|
- !ruby/object:Gem::Dependency
|
42
|
-
name:
|
42
|
+
name: pry
|
43
43
|
requirement: !ruby/object:Gem::Requirement
|
44
44
|
requirements:
|
45
45
|
- - "~>"
|
46
46
|
- !ruby/object:Gem::Version
|
47
|
-
version: '
|
47
|
+
version: '0.11'
|
48
48
|
type: :development
|
49
49
|
prerelease: false
|
50
50
|
version_requirements: !ruby/object:Gem::Requirement
|
51
51
|
requirements:
|
52
52
|
- - "~>"
|
53
53
|
- !ruby/object:Gem::Version
|
54
|
-
version: '
|
54
|
+
version: '0.11'
|
55
55
|
- !ruby/object:Gem::Dependency
|
56
|
-
name:
|
56
|
+
name: rake
|
57
57
|
requirement: !ruby/object:Gem::Requirement
|
58
58
|
requirements:
|
59
59
|
- - "~>"
|
60
60
|
- !ruby/object:Gem::Version
|
61
|
-
version: '0
|
61
|
+
version: '10.0'
|
62
62
|
type: :development
|
63
63
|
prerelease: false
|
64
64
|
version_requirements: !ruby/object:Gem::Requirement
|
65
65
|
requirements:
|
66
66
|
- - "~>"
|
67
67
|
- !ruby/object:Gem::Version
|
68
|
-
version: '0
|
68
|
+
version: '10.0'
|
69
69
|
- !ruby/object:Gem::Dependency
|
70
70
|
name: yard
|
71
71
|
requirement: !ruby/object:Gem::Requirement
|
@@ -80,9 +80,9 @@ dependencies:
|
|
80
80
|
- - "~>"
|
81
81
|
- !ruby/object:Gem::Version
|
82
82
|
version: '0.6'
|
83
|
-
description:
|
83
|
+
description: It's often nice to have random English-sounding words, eg. for password
|
84
84
|
generators. This library uses Markov-chain techniques on words, as opposed to many
|
85
|
-
others which focus on sentences.
|
85
|
+
others which focus on sentences.
|
86
86
|
email:
|
87
87
|
- dmerand@explo.org
|
88
88
|
executables: []
|
@@ -101,7 +101,9 @@ files:
|
|
101
101
|
- bin/console
|
102
102
|
- bin/setup
|
103
103
|
- lib/markov_words.rb
|
104
|
+
- lib/markov_words/file_store.rb
|
104
105
|
- lib/markov_words/version.rb
|
106
|
+
- lib/markov_words/words.rb
|
105
107
|
- markov_words.gemspec
|
106
108
|
homepage: https://github.com/exploration/markov_words
|
107
109
|
licenses:
|