reorder a PDF, sorting pages by bookmark titles

notes.md

notes

This sorts the pages of a PDF, in order of the titles of the PDF bookmarks.

This was hacked together for a one-off task. Hardcoding the input name made sense, with a symlink. It was the first time I'd scripted anything re: PDFs.

pdf-sort-bookmark-pages.rb

#!/usr/bin/env ruby

# writing locally in ruby 3.0.2

require 'hexapdf'
require 'byebug'

doc = HexaPDF::Document.open('input.pdf')

pages_to_ref = Hash.new # page number to [oid, gen]
bookmarks_to_ref = Hash.new # bookmark title to [oid, gen]
counter = 0

print "iterating through pages, '.' per 100 pages "
doc.pages.each_with_index do |page, index|
  # byebug
  # puts "pausing"
  # page.methods
    # page.oid = 4
    # page.gen = 0
  pages_to_ref[index] = [page.oid, page.gen]
  counter += 1
  print "." if counter % 100 == 0
end
puts " done!"

counter = 0
print "parsing bookmarks / outlines, . per 100 "
doc.outline.each_item do |item, index|
  item.each_item do |inner_item, inner_index|
    # first one:
    # inner_item.value[:Title] = "040922 01"
    # which is the title of the first bookmark / outline
    # inner_item.value[:Dest][0] = #<HexaPDF::Reference [4, 0]>
    # inner_item.value[:Dest][0].oid = 4
    # inner_item.value[:Dest][0].gen = 0

    title = inner_item.value[:Title]
    next if title.empty?
    next if title == "DUPE"
    bookmarks_to_ref[title] = [inner_item.value[:Dest][0].oid, inner_item.value[:Dest][0].gen]

    counter += 1
    print "." if counter % 100 == 0
  end
end
puts " done!"

puts "total bookmark count: #{bookmarks_to_ref.keys.count}"

# copying pages, by iterating through sorted bookmark titles
counter = 0
target = HexaPDF::Document.new
print "copying pages, . per 100 "
bookmarks_to_ref.keys.sort.each do |title|
  # copy src page into dest pdf
  ref = bookmarks_to_ref[title]
  page_num = pages_to_ref.key(ref)
  target.pages << target.import(doc.pages[page_num])

  counter += 1
  print "." if counter % 100 == 0
end # /copying pages
puts " done!"


# copy bookmarks to dest pdf
counter = 0
print "copying bookmarks, . per 100 "
main = target.outline.add_item("Main") # do |main|
bookmarks_to_ref.keys.sort.each do |title|
  main.add_item(title, destination: counter)

  counter += 1
  print "." if counter % 100 == 0
end # /copying bookmarks
puts " done!"


puts "writing target pdf..."
target.write("sorted.pdf")
puts " done!"

# OCR dest pdf, outside of this