Class: Bismas::CLI::Chardiff

Inherits:
Bismas::CLI show all
Defined in:
lib/bismas/cli/chardiff.rb

Class Method Summary collapse

Instance Method Summary collapse

Methods inherited from Bismas::CLI

#execute_options, #input_options, #require_gem, #type_option, #unsupported_type

Class Method Details

.defaults ⇒ Object 



35
36
37
38
39
40
41
 
# File 'lib/bismas/cli/chardiff.rb', line 35

def self.defaults
  super.merge(
    file1_encoding:  DEFAULT_ENCODING,
    file2_encoding:  DEFAULT_ENCODING,
    output_encoding: Encoding.default_external
  )
end

Instance Method Details

#run(arguments) ⇒ Object 



43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
 
# File 'lib/bismas/cli/chardiff.rb', line 43

def run(arguments)
  quit unless arguments.empty?

  output_encoding, frequencies = options[:output_encoding],
    Hash.new { |h, k| h[k] = Hash.new { |i, j| i[j] = [0, 0] } }

  2.times { |index|
    file = "file#{index + 1}"

    reader_options = {
      encoding: "#{options[:"#{file}_encoding"]}:#{output_encoding}",
      key: key = options[:"#{file}_key"]
    }

    Reader.parse_file(options[file.to_sym], reader_options) { |id, record|
      frequency = Hash.new(0); record.delete(key)

      record.each_value { |array| array.each { |value|
        value.each_char { |char| frequency[char] += 1 } } }

      frequencies.values_at(0, key ? id.to_i : $.).each { |hash|
        frequency.each { |char, count| hash[char][index] += count } }
    }
  }

  File.open_file(options[:output], {}, 'w') { |io|
    begin
      csv = CSV.new(io) << %w[id char count1 count2 diff]

      frequencies.sort_by { |id,| id }.each { |id, hash|
        hash.sort_by { |char,| char }.each { |char, (count1, count2)|
          unless count1 == count2
            csv << [id, char, count1, count2, count2 - count1]
          end
        }
      }
    ensure
      csv.close if csv
    end
  }
end