# encoding: utf-8

module Util
  class TextCleaner
    # Takes in text file, cleans data and writes to new text file.
    def clean_data(data)
      clean_controls(data)
    end

    # Cleans control information from passed in file
    def clean_controls(extracted_data)
      controls_data = isolate_controls_data(extracted_data)
      clean_section_header = remove_section_header(controls_data)
      clean_whitespace = remove_newline_in_controls(clean_section_header)
      clean_special = remove_special(clean_whitespace)
      clean_no_space = remove_extra_space(clean_special)
      clean_pagenum = remove_pagenum(clean_no_space)
      separate_controls(clean_pagenum)
    end

    # Removes everything before and after the controls
    def isolate_controls_data(extracted_data)
      extracted_data = extracted_data.gsub(/\| P a g e+/, "| P a g e\n")
      extracted_data = extracted_data.split("\n").map(&:strip).reject { |e| e.to_s.empty? }.join("\n")
      extracted_data = extracted_data.gsub('???', '')
      /^1\.1\s*[^)]*?(?=\)$)(.*\n)*?(?=\s*Appendix:)/.match(extracted_data).to_s
    end

    # Removes all pagenumbers between the controls
    def remove_pagenum(extracted_data)
      clean_pagenum = extracted_data.gsub(/(\d{1,3}\|Page|\d{1,3} \| P a g e)/, '').to_s
      clean_pagenum.gsub(/(\d{1,3} \| Page)/, '').to_s
    end

    # Removes section headers for each control
    def remove_section_header(extracted_data)
      extracted_data.gsub(/(?<!•)\s\n\d{1}\s.*(?:.*\n)*?(?=\d\.\d)/, "\n\n").to_s
    end

    # removes newlines between a control
    def remove_newline_in_controls(extracted_data)
      extracted_data.gsub(/\s\n.*?(?!d\.)/, "\n").to_s
    end

    # adds whitespace between different controls
    def separate_controls(extracted_data)
      extracted_data.gsub(/((?=^\s*?\d\.\d{1,}.*\n?.*?(?<=\)$)))/, "\n").to_s
    end

    def remove_special(extracted_data)
      extracted_data = extracted_data.gsub(//, '')
      extracted_data.gsub(/•/, '')
    end

    def remove_extra_space(extracted_data)
      clean_data = extracted_data.gsub(/\n\n\n/, "\n")
      clean_data = clean_data.gsub(/\t\n/, "\n")
      clean_data = clean_data.delete("\t")
      clean_data = clean_data.delete("\r")
      clean_data = clean_data.gsub(/\s\s/, ' ')
      clean_data.gsub(/(\n\n(?!^\d\.\d{1,}.*\n?.*?))/, '')
    end
  end
end