encapsulation of jpeg in tiff

jpg2tif.rb

#!/usr/bin/env ruby
# To the extent possible under law, the author has waived all copyright
# and related or neighboring rights to this software by associating the
# CC0 1.0 (http://creativecommons.org/publicdomain/zero/1.0/) with it.
#
# This script embeds (possibly multiple) JPEG files into a (possibly
# multi-page) TIFF file.  Each of the given files must be a baseline
# YCbCr JPEG file.  '-' means stdin.  Each JPEG file is embedded in the
# TIFF file as a huge single strip, which is allowed but not recommended
# according to TIFF 6.0 TechNote2.  The output is written to the stdout.

class JPG2TIF

  class Error < StandardError; end

  private

  def die(s)
    raise Error, s
  end

  class Buffer
    def initialize(n = 0); @pos, @a = n, []; end
    def write(s); @a.push(s); @pos += s.size; end
    def align(n); write([0].pack('c') * (n - 1 - (@pos + n - 1) % n)); end
    attr_reader :pos, :a
  end

  def parse_jpeg(src)
    pos = 0
    begin
      a = (src[pos,4] || '').unpack('nn')
      case a[0]
      when 0xffff then pos += 1; next
      when 0xffd8, 0xffd9 then a[1] = 0 # start of image, end of image
      when 0..0xff00, nil then die 'not in jpeg format'
      else die 'segment without length found' unless a[1] and a[1] >= 2
      end
      yield({:pos => pos, :mark => a[0], :len => a[1]})
      pos += 2 + a[1]
      case a[0] when 0xffda then # start of scan
        m = /(?=\xff[^\x00\xd0-\xd7])/mn.match(src, pos)
        pos = m ? m.begin(0) : die('unterminated entropy-coded segment')
      end
    end while pos < src.length
  end

  def process(output, jpeg, start, has_next)
    icc, sof = nil, nil
    parse_jpeg(jpeg) do |mark:0, pos:0, len:0|
      $stderr.printf "pos %08x mark %04x len %04x\n", pos, mark, len if $VERBOSE
      case mark
      when 0xffc0 then # SOF0
        die 'duplicate SOF0 found' if sof
        sof = jpeg[pos+4,len-2]
      when 0xffe2 then # ICCProfile
        if /\AICC_PROFILE\x00\x01\x01/n =~ jpeg[pos+4,len-2] then
          die 'duplicate ICC_PROFILE found' if icc
          icc = {:pos => pos + 18, :len => len - 16}
          icc[:copy] = jpeg[pos+18,len-16] if pos % 2 != 0
        end
      when 0xffd8..0xffdb, 0xffdd then next  # SOI, EOI, SOS, DQT, DRI
      when 0xffc4, 0xffcc then next          # DHT, DAC
      when 0xffe0..0xffef, 0xfffe then next  # APP, COM
      else die('not in baseline: %04x' % mark)
      end
    end

    case sof
    when /\A\x08....\x03\x01(.).\x02\x11.\x03\x11.\z/mn then
      sampling = (($1.getbyte(0) & 0xf) << 16) | (($1.getbyte(0) & 0xf0) >> 4)
    when nil then die 'no SOF0 found'
    else die 'not YCbCr 24 bit color with good subsampling'
    end
    height, width = sof[1,4].unpack('nn')

    bitsoff, rbwoff, jpegoff, iccoff, nextifd = 0, 0, 0, 0, 0
    ifd = proc do
      e = [
        [0x0100, 3, 1, width],        # ImageWidth
        [0x0101, 3, 1, height],       # ImageHeight
        [0x0102, 3, 3, bitsoff],      # BitsPerSample
        [0x0103, 3, 1, 7],            # Compression: JPEG
        [0x0106, 3, 1, 6],            # Photometric: YCbCr
        [0x0111, 4, 1, jpegoff],      # StripOffsets
        [0x0115, 3, 1, 3],            # SamplesPerPixel: 3
        [0x0116, 3, 1, height],       # RowsPerStrip
        [0x0117, 4, 1, jpeg.size],    # StripByteCounts
        [0x011c, 3, 1, 1],            # PlanarConfiguration: Chunky
        [0x0212, 3, 2, sampling],     # YCbCrSubSampling
        [0x0214, 5, 6, rbwoff],       # ReferenceBlackWhite
      ]
      iccoff = icc[:pos] + jpegoff if icc and not icc[:copy]
      e.push [0x8773, 7, icc[:len], iccoff] if icc  # ICCProfile
      [[e.size].pack('v'), e.map { |i| i.pack('vvVV') },
       [nextifd].pack('V')].join
    end

    output.write [0x4949,0x2a,8].pack('vvV') if start == 0
    start = 8 if start == 0
    buf = Buffer.new(start + ifd.call.size)
    bitsoff = buf.pos
    rbwoff = buf.write([8,8,8].pack('v*'))
    jpegoff = buf.write([0,1,255,1,128,1,255,1,128,1,255,1].pack('V*'))
    buf.write jpeg
    iccoff = buf.align(2) if icc and icc[:copy]
    buf.write icc[:copy] if icc and icc[:copy]
    nextifd = buf.align(2) if has_next
    output.write ifd.call
    buf.a.each { |i| output.write i }
    nextifd
  end

  public

  def create(output, args)
    off = 0
    args.each_with_index do |bin, i|
      $stderr.printf "image %d offset %d\n", i + 1, off if $VERBOSE
      jpeg = block_given? ? yield(bin) : bin
      off = process(output, jpeg, off, i + 1 < args.size)
    end
  end

end

if __FILE__ == $0 then
  $stdout.binmode
  $stdin.set_encoding Encoding::ASCII_8BIT
  files = ARGV.empty? ? ['-'] : ARGV
  JPG2TIF.new.create($stdout, files) do |input|
    input == '-' ? $stdin.read : File.binread(input)
  end
end

tif2jpg.rb

#!/usr/bin/env ruby
# To the extent possible under law, the author has waived all copyright
# and related or neighboring rights to this software by associating the
# CC0 1.0 (http://creativecommons.org/publicdomain/zero/1.0/) with it.
#
# This script extracts JPEG files embedded in a TIFF file by jpg2tif.rb.
# it reads a TIFF file from stdin and writes each jPEG image to each file
# specified in the arguments.  '-' means stdout.  To extract more than one
# images from a multi-page TIFF file, multiple output filenames must be
# speficied in the arguments.  The N-th page in the TIFF file is written
# to the N-th file.

class TIF2JPG

  class Error < StandardError; end

  private

  def die(s)
    raise Error, s
  end

  public

  def read(tiff)
    sig, ver, off = tiff.unpack('vvV')
    die 'not a tiff file' unless sig == 0x4949 and ver == 0x2a and off > 0

    while off > 0
      num_entries = tiff[off,2].unpack('v')[0]
      off += 2
      die 'IFD not found' unless num_entries
      strip_offset, strip_length = nil
      num_entries.times do |i|
        tag, ty, count, offset = (tiff[off,12] || '').unpack('vvVV')
        off += 12
        $stderr.printf "tag %04x type %04x count %08x offset %08x\n",
                       tag, ty, count, offset if $VERBOSE
        case tag
        when 0x0111  # StripOffsets
          die 'unexpected StripOffsets' unless ty == 4 and count == 1
          strip_offset = offset
        when 0x0117  # StripByteCounts
          die 'unexpected StripByteCounts' unless ty == 4 and count == 1
          strip_length = offset
        end
      end
      die 'no strip found' unless strip_offset and strip_length
      off = (tiff[off,4] || '').unpack('V')[0] || 0
      yield tiff[strip_offset, strip_length]
    end
  end

end

if __FILE__ == $0 then
  $stdout.binmode
  $stdin.set_encoding Encoding::ASCII_8BIT
  files = ARGV.empty? ? ['-'] : ARGV
  TIF2JPG.new.read($stdin.read) do |jpeg|
    break if files.empty?
    output = files.shift
    if output == '-' then
      $stdout.write jpeg
    else
      File.binwrite output, jpeg
    end
  end
end