ವರ್ಡಲ್ಲಾ ಸಹಾಯಕ

kannada_wordalla_helper.rb

# encoding: utf-8
#
# The Kannada dictionary file needed for this script is available here : https://github.com/alar-dict/data
# Download the YAML file and place it in the same directory as this script.
#
# Using this script :
# This script is useful in an IRB console to be used interactively.
#
# 1. Lanuch irb
# 2. Load this script with require 'wordalla.rb'
# 3. It will take a few seconds to parse the big Kannada dictionary file.
# 4. After that it will filter out all the 5 letter words and keep it ready for further filtering.
# 5. Based on what the wordalla website is showing you, you can filter out the words by calling the method filter_with_includes_and_excludes.
# 6. Inputs for this method are :
#   a. words : Input the list of words to be filtered. You can pass `fives` here. That is where script stores the words of length five
#   b. must_include : Array of letters that must be included
#   c. must_exclude : Array of letters that must be excluded
#

require 'yaml'

ANUSVARA = "\u0C82".freeze
VISARGA = "\u0C83".freeze
VOTTU = "್".freeze
VOWEL_SIGNS = %w( ಾ ಿ ೀ ು ೂ ೆ ೇ ೈ ೊ ೋ ೌ ೃ).freeze

def kannada_word_length(word)
  length = 0
  is_vattakshara = false
  word.each_char do |letter|
    next if (letter == ANUSVARA || letter == VISARGA)
    next if VOWEL_SIGNS.include?(letter)

    if (letter == VOTTU)
      is_vattakshara = true
      next
    end

    if (is_vattakshara)
      is_vattakshara = false
    else
      length+= 1
    end
  end

  return length
end

def filter_with_includes_and_excludes(words, must_include = [], must_exclude = [])
  filtered_words = words.select do |w|
    has_all_include = true
    must_include.each do |letter|
      unless w['entry'].include?(letter)
        has_all_include = false
        break
      end
    end
    next unless has_all_include

    has_any_excluded = false
    must_exclude.each do |letter|
      if w['entry'].include?(letter)
        has_any_excluded = true
        break
      end
    end
    next if has_any_excluded

    true
  end

  return filtered_words
end

dict = YAML.load_file('alar.yml')
fives = dict.select { |w| kannada_word_length(w['entry']) == 5 }