rename_halved_double_page_scans.rb

require 'fileutils'

# For use in cleaning scanned book/magazine/etc pages and preparing the scans 
# for distribution/OCR. 
# 
# Assume you've been scanning a book as double-page scans. You can  
# automatically split them into two files via ImageMagick `mogrify`: 
# 
#     mogrify -path "./Auto-Halved Pages" -format png -crop 50%x100% +repage "./Double-Page Scans/*.png"
# 
# ...which vertically divides an file named (for example), 
# 
#     "./Double-Page Scans/22.png" (pages 22 and 23)
# 
# into two files named 
# 
#     "./Auto-Halved Pages/22-0.png" (page 22, left)
#     "./Auto-Halved Pages/22-1.png" (page 23, right)
# 
# The Ruby script below will then correctly rename `22-0.png` to `22.png` and 
# `22-1.png` to `23.png`.
# 
# WARNING: This will not work on documents that read right-to-left! The 
# '-1' and '-0' suffixes will be swapped! Lazy way to fix this: rename all the 
# `*-0.png`s to `*-2.png` or something.)

def rename_halved(src_dir='Auto-Halved Pages', 
                  dest_dir: 'Correctly-Numbered Pages', # nil okay
                  ext: '.png', # nil okay
                  copy: nil, 
                  digits: nil, 
                  start_page_num: nil, 
                  noop: false)
  if ext && !ext.empty? && ext[0]!='.'
    ext='.'+ext
  end

  fnames=Dir.glob(File.join(src_dir,"*#{ext || '.*'}"),File::FNM_CASEFOLD).sort

  if (!start_page_num || !digits)
    m = /(?<left_page_num>\d+)-(?<diff>0|1)/.match(File.basename fnames.first)
    digits ||= m[:left_page_num].length
    start_page_num ||= m[:left_page_num].to_i
  end

  FileUtils.mkdir_p(dest_dir) if (dest_dir && !noop)

  page = start_page_num
  fnames.map do |old_fname|
    new_fname = File.join(
      (dest_dir || src_dir), 
      "#{page}".rjust(digits,"0")+(ext || File.extname(old_fname))
    )

    if noop
      # do nothing
    elsif dest_dir || copy
      FileUtils.cp(old_fname, new_fname)
    else
      FileUtils.mv(old_fname, new_fname)
    end
    page+=1
    new_fname
  end
end