Class: LoremIpsum::Generator
- Inherits:
-
Object
- Object
- LoremIpsum::Generator
- Defined in:
- lib/lorem-ipsum.rb
Instance Method Summary collapse
- #analyze(filename) ⇒ Object
- #generate(options) ⇒ Object
-
#initialize(data_files = [], opts = { :max_ngraph => 3 }) ⇒ Generator
constructor
A new instance of Generator.
- #next_char(prev) ⇒ Object
- #next_paragraph(word_count, variance = 0) ⇒ Object
- #next_sentence(word_count) ⇒ Object
- #next_word ⇒ Object
Constructor Details
#initialize(data_files = [], opts = { :max_ngraph => 3 }) ⇒ Generator
Returns a new instance of Generator.
5 6 7 8 9 10 |
# File 'lib/lorem-ipsum.rb', line 5 def initialize(data_files = [], opts = { :max_ngraph => 3 }) @letter_count = { :count => 0 } @max_ngraph = opts[:max_ngraph] data_files.each { |file| analyze(file) } end |
Instance Method Details
#analyze(filename) ⇒ Object
12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 |
# File 'lib/lorem-ipsum.rb', line 12 def analyze(filename) File.open(filename) do |file| while (line = file.gets) # todo - not handling punctuation line = line.strip.downcase.gsub(/[^a-z ]/, '') << ' ' word = "^" line.chars do |c| word << c n = [@max_ngraph, word.length].min ngraph = word[-n..-1] ngraph.chars.inject(@letter_count) do |hash, char| hash[char] ||= { :count => 0 } hash[char][:count] += 1 hash = hash[char] end if c == ' ' word = "^" end end end end end |
#generate(options) ⇒ Object
38 39 40 41 42 43 44 45 46 47 48 49 |
# File 'lib/lorem-ipsum.rb', line 38 def generate() [:paras] ||= 1 [:variance] ||= 0 [:words] ||= 100 str = "" 1.upto([:paras]) do |i| str << next_paragraph([:words], [:variance]) << "\n" end str end |
#next_char(prev) ⇒ Object
86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 |
# File 'lib/lorem-ipsum.rb', line 86 def next_char(prev) # Need to make sure our words don't get too long. Not everyone is Charles # Dickens, even in fake-Latin land. These parameters seem to look nice, # but salt to taste. return ' ' if prev.length > 4 && rand(9 + prev.length) < prev.length n = [@max_ngraph-1, prev.length].min prev_ngraph = prev[-n..-1] # If we don't have statistics for this n-graph, just use the stats # for the (n-1)-graph n_count = nil until n_count n_count = prev_ngraph.chars.inject(@letter_count) do |hash, char| break if ! hash[char] hash = hash[char] end prev_ngraph = prev_ngraph[1..-1] n_count = @letter_count if ! prev_ngraph end n_count = n_count.reject { |k,v| k == :count || prev.empty? && k == ' ' } num_letters ||= n_count.values.inject(0) { |s,c| s += c[:count] } index = rand(num_letters + 1) "abcdefghijklmnopqrstuvqxyz ".chars do |c| index -= n_count[c] && n_count[c][:count] || 0 return c if index <= 0 end end |
#next_paragraph(word_count, variance = 0) ⇒ Object
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 |
# File 'lib/lorem-ipsum.rb', line 51 def next_paragraph(word_count, variance = 0) str = "" if variance != 0 range = (2 * variance * word_count / 100.0).to_i word_count = word_count + rand(range) - range / 2 end while word_count > 0 sentence_count = [rand(10) + 5, word_count].min str << next_sentence(sentence_count) word_count -= sentence_count end str << "\n" end |
#next_sentence(word_count) ⇒ Object
68 69 70 71 72 73 74 75 76 77 78 |
# File 'lib/lorem-ipsum.rb', line 68 def next_sentence(word_count) str = "" 1.upto(word_count) do |i| word = next_word word.capitalize! if i == 1 word.gsub!(/ /,'. ') if i == word_count str << word end str end |
#next_word ⇒ Object
80 81 82 83 84 |
# File 'lib/lorem-ipsum.rb', line 80 def next_word word = "^" word << next_char(word) while word[-1..-1] != ' ' word[1..-1] end |