Module: Bidi2pdf::TestHelpers::Images::TIFFHelper

Included in:
Extractor
Defined in:
lib/bidi2pdf/test_helpers/images/tiff_helper.rb

Overview

rubocop: disable Metrics/ModuleLength, Metrics/AbcSize

Constant Summary collapse

IMAGE_WIDTH =

TIFF Tag IDs

256
IMAGE_LENGTH = 
257
BITS_PER_SAMPLE = 
258
COMPRESSION = 
259
PHOTOMETRIC_INTERPRETATION = 
262
STRIP_OFFSETS = 
273
SAMPLES_PER_PIXEL = 
277
ROWS_PER_STRIP = 
278
STRIP_BYTE_COUNTS = 
279
PLANAR_CONFIGURATION = 
284
INK_SET = 
332
TYPE_SHORT =

TIFF Data Types

3
TYPE_LONG = 
4
COMPRESSION_NONE =

TIFF Compression Types

1
COMPRESSION_CCITT_G3 = 
3
COMPRESSION_CCITT_G4 = 
4
PHOTO_WHITE_IS_ZERO =

TIFF Photometric Interpretations

0
PHOTO_BLACK_IS_ZERO = 
1
PHOTO_RGB = 
2
PHOTO_SEPARATION = 
5
PLANAR_CHUNKY =

Planar Configuration

1

Instance Method Summary collapse

Instance Method Details

#pack_tiff(entries) ⇒ Object

See:

* https://gist.github.com/gstorer/f6a9f1dfe41e8e64dcf58d07afa9ab2a
* https://github.com/yob/pdf-reader/blob/main/examples/extract_images.rb



62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
 
# File 'lib/bidi2pdf/test_helpers/images/tiff_helper.rb', line 62

def pack_tiff(entries)
  fields = entries.size
  fmt = [
    "a2", # Byte order ("II")
    "S<", # TIFF magic (42)
    "L<", # Offset to first IFD (8)
    "S<", # Number of directory entries
    ("S< S< L< L<" * fields), # each tag: id, type, count, value
    "L<" # Next IFD offset (0 = end)
  ].join(" ")

  # Build the flat array: ['II', 42, 8, fields, tag1, type1, count1, value1, …, 0]
  values = ["II", 42, 8, fields] + entries.flatten + [0]
  values.pack(fmt)
end

#tiff_header(hash, data) ⇒ Object 



39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
 
# File 'lib/bidi2pdf/test_helpers/images/tiff_helper.rb', line 39

def tiff_header(hash, data)
  cs_entry = hash[:ColorSpace]

  cs = if cs_entry.is_a?(Array) && cs_entry.first == :ICCBased
         icc_stream = cs_entry[1]
         icc_stream.hash[:Alternate]
       else
         cs_entry
       end

  case cs
  when :DeviceCMYK then tiff_header_for_CMYK(hash, data)
  when :DeviceGray then tiff_header_for_gray(hash, data)
  when :DeviceRGB then tiff_header_for_rgb(hash, data)
  else
    logger.warn("Unsupported color space '#{cs}' for compressed image with filter '#{hash[:Filter]}'. Skipping image.")
    nil # Skip processing this image
  end
end

#tiff_header_for_ccitt(hash, data) ⇒ Object 



105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
 
# File 'lib/bidi2pdf/test_helpers/images/tiff_helper.rb', line 105

def tiff_header_for_ccitt(hash, data)
  dp = hash[:DecodeParms] || {}
  width = dp[:Columns] || hash[:Width]
  height = hash[:Height]
  k = dp[:K] || 0
  group = (k.positive? ? COMPRESSION_CCITT_G3 : COMPRESSION_CCITT_G4)
  img_size = data.bytesize

  # We’ll emit exactly 8 tags:
  fields = 8
  # Calculate where the image data will start:
  header_size = 2 + 2 + 4 + 2 + (fields * 12) + 4

  entries = [
    [IMAGE_WIDTH, TYPE_LONG, 1, width], # ImageWidth
    [IMAGE_LENGTH, TYPE_LONG, 1, height], # ImageLength
    [BITS_PER_SAMPLE, TYPE_SHORT, 1, 1], # BitsPerSample
    [COMPRESSION, TYPE_SHORT, 1, group], # Compression (3=G3, 4=G4)
    [PHOTOMETRIC_INTERPRETATION, TYPE_SHORT, 1, PHOTO_WHITE_IS_ZERO], # PhotometricInterpretation (0 = WhiteIsZero)
    [STRIP_OFFSETS, TYPE_LONG, 1, header_size], # StripOffsets
    [ROWS_PER_STRIP, TYPE_LONG, 1, height], # RowsPerStrip
    [STRIP_BYTE_COUNTS, TYPE_LONG, 1, img_size] # StripByteCounts
  ]

  pack_tiff(entries)
end

#tiff_header_for_cmyk(hash, data) ⇒ Object 



132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
 
# File 'lib/bidi2pdf/test_helpers/images/tiff_helper.rb', line 132

def tiff_header_for_cmyk(hash, data)
  width = hash[:Width]
  height = hash[:Height]
  bpc = hash[:BitsPerComponent] || 8
  img_size = data.bytesize

  # CMYK needs 10 tags + a 4×SHORT BitsPerSample array
  fields = 10
  bits_array_size = 4 * 2 # 4 channels × 2 bytes each

  # Size of header + IFD (before the bits array)
  header_ifd_size = 2 + 2 + 4 + 2 + (fields * 12) + 4
  # Where the pixel data will really start:
  data_offset = header_ifd_size + bits_array_size

  entries = [
    [IMAGE_WIDTH, TYPE_LONG, 1, width], # ImageWidth
    [IMAGE_LENGTH, TYPE_LONG, 1, height], # ImageLength
    [BITS_PER_SAMPLE, TYPE_SHORT, 4, header_ifd_size], # BitsPerSample (pointer to array)
    [COMPRESSION, TYPE_SHORT, 1, COMPRESSION_NONE], # Compression (1 = none)
    [PHOTOMETRIC_INTERPRETATION, TYPE_SHORT, 1, PHOTO_SEPARATION], # PhotometricInterpretation (5 = Separation)
    [STRIP_OFFSETS, TYPE_LONG, 1, data_offset], # StripOffsets
    [SAMPLES_PER_PIXEL, TYPE_SHORT, 1, 4], # SamplesPerPixel
    [STRIP_BYTE_COUNTS, TYPE_LONG, 1, img_size], # StripByteCounts
    [PLANAR_CONFIGURATION, TYPE_SHORT, 1, PLANAR_CHUNKY], # PlanarConfiguration (1 = chunky)
    [INK_SET, TYPE_SHORT, 1, 1] # InkSet (1 = CMYK)
  ]

  header = pack_tiff(entries)
  # Append the 4-channel BitsPerSample array as little‐endian SHORTs:
  header << [bpc, bpc, bpc, bpc].pack("S<S<S<S<")
  header
end

#tiff_header_for_gray(hash, data) ⇒ Object 



78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
 
# File 'lib/bidi2pdf/test_helpers/images/tiff_helper.rb', line 78

def tiff_header_for_gray(hash, data)
  width = hash[:Width]
  height = hash[:Height]
  bpc = hash[:BitsPerComponent] || 8
  img_size = data.bytesize

  # 9 tags, no extra arrays needed
  fields = 9
  # size of header+IFD before the image data starts
  header_ifd_size = 2 + 2 + 4 + 2 + (fields * 12) + 4
  data_offset = header_ifd_size

  entries = [
    [IMAGE_WIDTH, TYPE_LONG, 1, width], # ImageWidth
    [IMAGE_LENGTH, TYPE_LONG, 1, height], # ImageLength
    [BITS_PER_SAMPLE, TYPE_SHORT, 1, bpc], # BitsPerSample
    [COMPRESSION, TYPE_SHORT, 1, COMPRESSION_NONE], # Compression (1 = none)
    [PHOTOMETRIC_INTERPRETATION, TYPE_SHORT, 1, PHOTO_BLACK_IS_ZERO], # PhotometricInterpretation (1 = BlackIsZero)
    [STRIP_OFFSETS, TYPE_LONG, 1, data_offset], # StripOffsets
    [SAMPLES_PER_PIXEL, TYPE_SHORT, 1, 1], # SamplesPerPixel
    [STRIP_BYTE_COUNTS, TYPE_LONG, 1, img_size], # StripByteCounts
    [PLANAR_CONFIGURATION, TYPE_SHORT, 1, PLANAR_CHUNKY] # PlanarConfiguration (1 = chunky)
  ]

  pack_tiff(entries)
end

#tiff_header_for_rgb(hash, data) ⇒ Object 



166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
 
# File 'lib/bidi2pdf/test_helpers/images/tiff_helper.rb', line 166

def tiff_header_for_rgb(hash, data)
  width = hash[:Width]
  height = hash[:Height]
  bpc = hash[:BitsPerComponent] || 8
  img_size = data.bytesize

  # 8 tags + a 3×SHORT BitsPerSample array
  fields = 8
  bits_array_size = 3 * 2 # 3 channels × 2 bytes each

  # size of header + IFD before the bits array
  header_ifd_size = 2 + 2 + 4 + 2 + (fields * 12) + 4
  # where the pixel data really starts:
  data_offset = header_ifd_size + bits_array_size

  entries = [
    [IMAGE_WIDTH, TYPE_LONG, 1, width], # ImageWidth
    [IMAGE_LENGTH, TYPE_LONG, 1, height], # ImageLength
    [BITS_PER_SAMPLE, TYPE_SHORT, 3, header_ifd_size], # BitsPerSample → pointer to our array
    [COMPRESSION, TYPE_SHORT, 1, COMPRESSION_NONE], # Compression (1 = none)
    [PHOTOMETRIC_INTERPRETATION, TYPE_SHORT, 1, PHOTO_RGB], # PhotometricInterpretation (2 = RGB)
    [STRIP_OFFSETS, TYPE_LONG, 1, data_offset], # StripOffsets
    [SAMPLES_PER_PIXEL, TYPE_SHORT, 1, 3], # SamplesPerPixel
    [STRIP_BYTE_COUNTS, TYPE_LONG, 1, img_size] # StripByteCounts
  ]

  # pack the IFD
  header = pack_tiff(entries)

  # append the 3-channel BitsPerSample as little-endian SHORTs
  header << [bpc, bpc, bpc].pack("S<S<S<")

  header
end