Simple grep for word in a directory of textfiles. Work in progress...

grepword.rb

#!/usr/bin/env ruby

# Change SEARCH_ROOT to point to proper directory.
# USAGE: ruby grepword.rb <word>

SEARCH_ROOT = "/path/to/books/directory"

word = Regexp.new(ARGV[0].encode('utf-8', 'external'))

def clean line
  ['sjis', 'utf-8', 'utf-16'].each do |encoding|
    begin
      cleaned = line.encode('utf-8', encoding)
    rescue Exception => e
      next
    end
    return cleaned if cleaned.valid_encoding?
  end
  ""
end

def match_words word
  Dir.glob("#{SEARCH_ROOT}/**/*.txt").each do |file|
    next unless File.file?(file)
    File.open(file) do |f|
      f.each_line do |line|

        line = clean(line)
        
        if match = line.match(word)
          yield file, line, match
        end
      end
    end
  end
end

match_words(word) do |file, line, match|
  puts File.basename file
  puts line
  puts
end