Quick/dirty script to organize files by year/month.

mv_files.rb

#!/usr/bin/env ruby

# Quick-and-dirty script that moves HTML files produced by Thunderbird
# into year/month directories;
#TODO: Update script to be more memory-efficient

require 'fileutils'
# COMMENTING THIS OUT. WE NEED TO UPDATE IMPLEMENTATION 
# TO ONLY CREATE FOLDERS FOR MONTHS WHERE EMAIL EXISTS#years = (2010..2014).to_a.collect { |year| year.to_s }
#months = (1..12).to_a.collect { |num| num.to_s.rjust(2,'0') }#years.each do |year|
#   months.each do |month|
#    `mkdir -p Inbox/#{year}/#{month}`
#   end
#end

# glob all *.html files outside of year/month dirs
def files_to_move
  all_files = Dir.glob("Inbox/**/*.html").reject do |pth| 
    encoded_path = pth.encode('UTF-8', 'binary', invalid: :replace, undef: :replace, replace: '')
    encoded_path.match %r{\d{4}/\d{2}/\d{4}}
  end
  by_yearmonth =  {}
  all_files.each do |pth|
    base_fname = File.basename(pth)
    year, month = base_fname[0..3], base_fname[4..5]
    key = [year, month]
    begin
      by_yearmonth[key] << pth
    rescue NoMethodError
      by_yearmonth[key] = [pth]
    end
  end
  by_yearmonth
end

# Move html files to year/month dirs
files_to_move.each_pair do |key,files|
  year, month = key
  target = "Inbox/#{year}/#{month}/"
  files.each do |file|
    FileUtils.mv file, target
  end
end