henrik · February 6, 2009 21:49 · kugland · Apr 6, 2012 · henrik · Apr 9, 2012
diff --git a/README.markdown b/README.markdown
diff --git a/dezoomify.rb b/dezoomify.rb
 #!/usr/bin/env ruby
 # Dezoomify. See README.markdown.
 # By Henrik Nyh <http://henrik.nyh.se> 2009-02-06 under the MIT License.

 require 'cgi'
 require 'open-uri'
 require 'rubygems'
 require 'nokogiri'

 module Kernel
  def windows?
    RbConfig::CONFIG['host_os'].match(/mswin|windows|mingw/i)
  end
 end

 if windows?
  # Case-sensitive. Use forward slashes, or double-escape backslashes.
  TEMP_DIRECTORY = "C:/WINDOWS/Temp"
 else
  TEMP_DIRECTORY = "/tmp"
 end

 ARGV.each_with_index do |page_url, page_url_index|
  if page_url.include?("zoomifyImagePath")
    puts "#{page_url_index}. Extracting path from URL"
    source = page_url
  else
    puts "#{page_url_index}. Visiting #{page_url}"
    source = open(page_url).read
  end

  paths = source.scan(/zoomifyImagePath=([^"'&]+)/).flatten.map {|path| path.gsub(' ', '%20') }.uniq

  paths.each_with_index do |path, path_index|
    path = CGI.unescape(path)
    full_path = URI.join(page_url, path+'/')
    puts "  #{page_url_index}.#{path_index} Found image path #{full_path}"


    # <IMAGE_PROPERTIES WIDTH="1737" HEIGHT="2404" NUMTILES="99" NUMIMAGES="1" VERSION="1.8" TILESIZE="256"/>
    xml_url = URI.join(full_path.to_s, 'ImageProperties.xml')
    doc = Nokogiri::XML(open(xml_url))
    props = doc.at('IMAGE_PROPERTIES')

    width = props[:WIDTH].to_i
    height = props[:HEIGHT].to_i
    tilesize = props[:TILESIZE].to_f

    tiles_wide = (width/tilesize).ceil
    tiles_high = (height/tilesize).ceil

    # Determine max zoom level.
    # Also determine tile_counts per zoom level, used to determine tile group.
    # With thanks to http://trac.openlayers.org/attachment/ticket/1285/zoomify.patch.
    zoom = 0
    w = width
    h = height
    tile_counts = []
    while w > tilesize || h > tilesize
      zoom += 1

      t_wide = (w / tilesize).ceil
      t_high = (h / tilesize).ceil
      tile_counts.unshift t_wide*t_high

      w = (w / 2.0).floor
      h = (h / 2.0).floor
    end
    tile_counts.unshift 1  # Zoom level 0 has a single tile.
    tile_count_before_level = tile_counts[0..-2].inject(0) {|sum, num| sum + num }

    files_by_row = []
    tiles_high.times do |y|
      row = []
      tiles_wide.times do |x|
        filename = '%s-%s-%s.jpg' % [zoom, x, y]
        local_filepath = "#{TEMP_DIRECTORY}/zoomify-#{filename}"
        row << local_filepath

        tile_group = ((x + y * tiles_wide + tile_count_before_level) / tilesize).floor

        tile_url = URI.join(full_path.to_s, "TileGroup#{tile_group}/#{filename}")
        url = URI.join(tile_url.to_s, filename)
        puts "    Getting #{url}..."
        File.open(local_filepath, 'wb') {|f| f.print url.read }
      end
      files_by_row << row
    end


    # `montage` is ImageMagick.
    # We first stitch together the tiles of each row, then stitch all rows.
    # Stitching the full image all at once can get extremely inefficient for large images.

    puts "    Stitching #{tiles_wide} x #{tiles_high} = #{tiles_wide*tiles_high} tiles..."

    row_files = []
    files_by_row.each_with_index do |row, index|
      filename = "#{TEMP_DIRECTORY}/zoomify-row-#{index}.jpg"
      `montage #{row.join(' ')} -geometry +0+0 -tile #{tiles_wide}x1 #{filename}`
      row_files << filename
    end

    filename = "#{TEMP_DIRECTORY}/zoomified-#{page_url_index}.#{path_index}.jpg"

    `montage #{row_files.join(' ')} -geometry +0+0 -tile 1x#{tiles_high} #{filename}`

    unless windows? || `which xattr`.empty?
      # Set "Downloaded from" Finder metadata, like Safari does.
      system('xattr', '-w', 'com.apple.metadata:kMDItemWhereFroms', page_url, filename)
    end

    puts "    Done: #{filename}"

    # Reveal in Finder if on OS X.
    unless windows?
      `which osascript && osascript -e 'tell app "Finder"' -e 'reveal POSIX file "#{filename}"' -e 'activate' -e 'end'`
    end

  end

 end
	#!/usr/bin/env ruby
	# Dezoomify. See README.markdown.
	# By Henrik Nyh <http://henrik.nyh.se> 2009-02-06 under the MIT License.

	require 'cgi'
	require 'open-uri'
	require 'rubygems'
	require 'nokogiri'

	module Kernel
	def windows?
	RbConfig::CONFIG['host_os'].match(/mswin\|windows\|mingw/i)
	end
	end

	if windows?
	# Case-sensitive. Use forward slashes, or double-escape backslashes.
	TEMP_DIRECTORY = "C:/WINDOWS/Temp"
	else
	TEMP_DIRECTORY = "/tmp"
	end

	ARGV.each_with_index do \|page_url, page_url_index\|
	if page_url.include?("zoomifyImagePath")
	puts "#{page_url_index}. Extracting path from URL"
	source = page_url
	else
	puts "#{page_url_index}. Visiting #{page_url}"
	source = open(page_url).read
	end

	paths = source.scan(/zoomifyImagePath=([^"'&]+)/).flatten.map {\|path\| path.gsub(' ', '%20') }.uniq

	paths.each_with_index do \|path, path_index\|
	path = CGI.unescape(path)
	full_path = URI.join(page_url, path+'/')
	puts " #{page_url_index}.#{path_index} Found image path #{full_path}"


	# <IMAGE_PROPERTIES WIDTH="1737" HEIGHT="2404" NUMTILES="99" NUMIMAGES="1" VERSION="1.8" TILESIZE="256"/>
	xml_url = URI.join(full_path.to_s, 'ImageProperties.xml')
	doc = Nokogiri::XML(open(xml_url))
	props = doc.at('IMAGE_PROPERTIES')

	width = props[:WIDTH].to_i
	height = props[:HEIGHT].to_i
	tilesize = props[:TILESIZE].to_f

	tiles_wide = (width/tilesize).ceil
	tiles_high = (height/tilesize).ceil

	# Determine max zoom level.
	# Also determine tile_counts per zoom level, used to determine tile group.
	# With thanks to http://trac.openlayers.org/attachment/ticket/1285/zoomify.patch.
	zoom = 0
	w = width
	h = height
	tile_counts = []
	while w > tilesize \|\| h > tilesize
	zoom += 1

	t_wide = (w / tilesize).ceil
	t_high = (h / tilesize).ceil
	tile_counts.unshift t_wide*t_high

	w = (w / 2.0).floor
	h = (h / 2.0).floor
	end
	tile_counts.unshift 1 # Zoom level 0 has a single tile.
	tile_count_before_level = tile_counts[0..-2].inject(0) {\|sum, num\| sum + num }

	files_by_row = []
	tiles_high.times do \|y\|
	row = []
	tiles_wide.times do \|x\|
	filename = '%s-%s-%s.jpg' % [zoom, x, y]
	local_filepath = "#{TEMP_DIRECTORY}/zoomify-#{filename}"
	row << local_filepath

	tile_group = ((x + y * tiles_wide + tile_count_before_level) / tilesize).floor

	tile_url = URI.join(full_path.to_s, "TileGroup#{tile_group}/#{filename}")
	url = URI.join(tile_url.to_s, filename)
	puts " Getting #{url}..."
	File.open(local_filepath, 'wb') {\|f\| f.print url.read }
	end
	files_by_row << row
	end


	# `montage` is ImageMagick.
	# We first stitch together the tiles of each row, then stitch all rows.
	# Stitching the full image all at once can get extremely inefficient for large images.

	puts " Stitching #{tiles_wide} x #{tiles_high} = #{tiles_wide*tiles_high} tiles..."

	row_files = []
	files_by_row.each_with_index do \|row, index\|
	filename = "#{TEMP_DIRECTORY}/zoomify-row-#{index}.jpg"
	`montage #{row.join(' ')} -geometry +0+0 -tile #{tiles_wide}x1 #{filename}`
	row_files << filename
	end

	filename = "#{TEMP_DIRECTORY}/zoomified-#{page_url_index}.#{path_index}.jpg"

	`montage #{row_files.join(' ')} -geometry +0+0 -tile 1x#{tiles_high} #{filename}`

	unless windows? \|\| `which xattr`.empty?
	# Set "Downloaded from" Finder metadata, like Safari does.
	system('xattr', '-w', 'com.apple.metadata:kMDItemWhereFroms', page_url, filename)
	end

	puts " Done: #{filename}"

	# Reveal in Finder if on OS X.
	unless windows?
	`which osascript && osascript -e 'tell app "Finder"' -e 'reveal POSIX file "#{filename}"' -e 'activate' -e 'end'`
	end

	end

	end