Class: Fontisan::Tables::Cmap

Inherits:
Binary::BaseRecord show all
Extended by:
Registered
Defined in:
lib/fontisan/tables/cmap.rb

Overview

Parser for the 'cmap' (Character to Glyph Index Mapping) table

The cmap table maps character codes to glyph indices. It supports multiple encoding formats to accommodate different character sets and Unicode planes.

This implementation focuses on:

  • Format 4: Segment mapping for BMP (Basic Multilingual Plane, U+0000-U+FFFF)
  • Format 12: Segmented coverage for full Unicode support

Reference: OpenType specification, cmap table

Constant Summary collapse

PLATFORM_UNICODE =

Platform IDs

0
PLATFORM_MACINTOSH = 
1
PLATFORM_MICROSOFT = 
3
ENC_MS_UNICODE_BMP =

Microsoft Encoding IDs

1
ENC_MS_UNICODE_UCS4 =

Unicode BMP (UCS-2)

10

Instance Attribute Summary

Attributes inherited from Binary::BaseRecord

#raw_data

Instance Method Summary collapse

Methods included from Registered

register_tag

Methods inherited from Binary::BaseRecord

read, #valid?

Instance Method Details

#extract_subtable_data(record, data) ⇒ Object

Extract subtable data if record exists and offset is valid



126
127
128
129
130
131
 
# File 'lib/fontisan/tables/cmap.rb', line 126

def extract_subtable_data(record, data)
  return nil unless record
  return nil unless record[:offset] < data.length

  data[record[:offset]..]
end

#find_best_unicode_subtable(records, data) ⇒ Object

Find the best Unicode subtable from encoding records



94
95
96
97
98
99
 
# File 'lib/fontisan/tables/cmap.rb', line 94

def find_best_unicode_subtable(records, data)
  # Try in priority order: UCS-4, BMP, Unicode
  find_ucs4_subtable(records, data) ||
    find_bmp_subtable(records, data) ||
    find_unicode_subtable(records, data)
end

#find_bmp_subtable(records, data) ⇒ Object

Find Microsoft Unicode BMP subtable



111
112
113
114
115
116
117
 
# File 'lib/fontisan/tables/cmap.rb', line 111

def find_bmp_subtable(records, data)
  record = records.find do |r|
    r[:platform_id] == PLATFORM_MICROSOFT &&
      r[:encoding_id] == ENC_MS_UNICODE_BMP
  end
  extract_subtable_data(record, data)
end

#find_ucs4_subtable(records, data) ⇒ Object

Find Microsoft Unicode UCS-4 subtable (full Unicode)



102
103
104
105
106
107
108
 
# File 'lib/fontisan/tables/cmap.rb', line 102

def find_ucs4_subtable(records, data)
  record = records.find do |r|
    r[:platform_id] == PLATFORM_MICROSOFT &&
      r[:encoding_id] == ENC_MS_UNICODE_UCS4
  end
  extract_subtable_data(record, data)
end

#find_unicode_subtable(records, data) ⇒ Object

Find Unicode platform subtable (any encoding)



120
121
122
123
 
# File 'lib/fontisan/tables/cmap.rb', line 120

def find_unicode_subtable(records, data)
  record = records.find { |r| r[:platform_id] == PLATFORM_UNICODE }
  extract_subtable_data(record, data)
end

#has_bmp_coverage? ⇒ Boolean

Validation helper: Check if BMP coverage exists

Checks if the Basic Multilingual Plane (U+0000-U+FFFF) has mappings

Returns:

  • (Boolean) —

    True if BMP characters are mapped



310
311
312
313
314
315
316
 
# File 'lib/fontisan/tables/cmap.rb', line 310

def has_bmp_coverage?
  mappings = unicode_mappings
  return false if mappings.nil? || mappings.empty?

  # Check if any BMP characters (0x0000-0xFFFF) are mapped
  mappings.keys.any? { |code| code.between?(0x0000, 0xFFFF) }
end

#has_format_4_subtable? ⇒ Boolean

Validation helper: Check if format 4 subtable exists

Format 4 is the minimum requirement for Unicode BMP support

Returns:

  • (Boolean) —

    True if format 4 subtable is found



336
337
338
339
340
341
342
343
344
345
346
347
348
349
 
# File 'lib/fontisan/tables/cmap.rb', line 336

def has_format_4_subtable?
  data = to_binary_s
  records = read_encoding_records(data)

  records.any? do |record|
    subtable_data = extract_subtable_data(record, data)
    next false unless subtable_data && subtable_data.length >= 2

    format = subtable_data[0, 2].unpack1("n")
    format == 4
  end
rescue StandardError
  false
end

#has_required_characters?(*required_chars) ⇒ Boolean

Validation helper: Check if required characters are mapped

Checks for essential characters like space (U+0020)

Parameters:

  • required_chars (Array<Integer>) —

    Character codes that must be present

Returns:

  • (Boolean) —

    True if all required characters are mapped



324
325
326
327
328
329
 
# File 'lib/fontisan/tables/cmap.rb', line 324

def has_required_characters?(*required_chars)
  mappings = unicode_mappings
  return false if mappings.nil?

  required_chars.all? { |code| mappings.key?(code) }
end

#has_subtables? ⇒ Boolean

Validation helper: Check if at least one subtable exists

Returns:

  • (Boolean) —

    True if num_tables > 0



294
295
296
 
# File 'lib/fontisan/tables/cmap.rb', line 294

def has_subtables?
  num_tables&.positive?
end

#has_unicode_mapping? ⇒ Boolean

Validation helper: Check if Unicode mapping exists

Returns:

  • (Boolean) —

    True if Unicode mappings are present



301
302
303
 
# File 'lib/fontisan/tables/cmap.rb', line 301

def has_unicode_mapping?
  !unicode_mappings.nil? && !unicode_mappings.empty?
end

#map_character_range(start_char, end_char, start_glyph, mappings) ⇒ Object

Map a range of characters to glyphs



275
276
277
278
279
280
 
# File 'lib/fontisan/tables/cmap.rb', line 275

def map_character_range(start_char, end_char, start_glyph, mappings)
  (start_char..end_char).each do |code|
    glyph_index = start_glyph + (code - start_char)
    mappings[code] = glyph_index if glyph_index != 0
  end
end

#parse_format_12(data, mappings) ⇒ Object

Parse Format 12 subtable (segmented coverage) Format 12 supports full Unicode range



234
235
236
237
238
239
240
 
# File 'lib/fontisan/tables/cmap.rb', line 234

def parse_format_12(data, mappings)
  header = parse_format_12_header(data)
  return unless header

  parse_format_12_groups(data, header[:num_groups], header[:length],
                         mappings)
end

#parse_format_12_groups(data, num_groups, length, mappings) ⇒ Object

Parse Format 12 sequential map groups



258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
 
# File 'lib/fontisan/tables/cmap.rb', line 258

def parse_format_12_groups(data, num_groups, length, mappings)
  offset = 16
  num_groups.times do
    break if offset + 12 > length

    start_char_code = data[offset, 4].unpack1("N")
    end_char_code = data[offset + 4, 4].unpack1("N")
    start_glyph_id = data[offset + 8, 4].unpack1("N")

    map_character_range(start_char_code, end_char_code, start_glyph_id,
                        mappings)

    offset += 12
  end
end

#parse_format_12_header(data) ⇒ Object

Parse Format 12 header



243
244
245
246
247
248
249
250
251
252
253
254
255
 
# File 'lib/fontisan/tables/cmap.rb', line 243

def parse_format_12_header(data)
  return nil if data.length < 16

  format = data[0, 2].unpack1("n")
  return nil unless format == 12

  length = data[4, 4].unpack1("N")
  return nil if length > data.length

  num_groups = data[12, 4].unpack1("N")

  { length: length, num_groups: num_groups }
end

#parse_format_4(data, mappings) ⇒ Object

Parse Format 4 subtable (segment mapping to delta values) Format 4 is the most common format for BMP Unicode fonts rubocop:disable Metrics/MethodLength rubocop:disable Metrics/CyclomaticComplexity rubocop:disable Metrics/PerceivedComplexity



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
165
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
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
 
# File 'lib/fontisan/tables/cmap.rb', line 138

def parse_format_4(data, mappings)
  return if data.length < 14

  # Format 4 header
  format = data[0, 2].unpack1("n")
  return unless format == 4

  length = data[2, 2].unpack1("n")
  return if length > data.length

  seg_count_x2 = data[6, 2].unpack1("n")
  seg_count = seg_count_x2 / 2

  # Arrays start at offset 14
  offset = 14

  # Read endCode array
  end_codes = []
  seg_count.times do
    break if offset + 2 > length

    end_codes << data[offset, 2].unpack1("n")
    offset += 2
  end

  # Skip reservedPad (2 bytes)
  offset += 2

  # Read startCode array
  start_codes = []
  seg_count.times do
    break if offset + 2 > length

    start_codes << data[offset, 2].unpack1("n")
    offset += 2
  end

  # Read idDelta array
  id_deltas = []
  seg_count.times do
    break if offset + 2 > length

    id_deltas << data[offset, 2].unpack1("n")
    offset += 2
  end

  # Read idRangeOffset array
  id_range_offsets = []
  id_range_offset_pos = offset
  seg_count.times do
    break if offset + 2 > length

    id_range_offsets << data[offset, 2].unpack1("n")
    offset += 2
  end

  # Process each segment
  seg_count.times do |i|
    start_code = start_codes[i]
    end_code = end_codes[i]
    id_delta = id_deltas[i]
    id_range_offset = id_range_offsets[i]

    # Skip the final segment (0xFFFF)
    next if start_code == 0xFFFF

    if id_range_offset.zero?
      # Use idDelta directly
      (start_code..end_code).each do |code|
        glyph_index = (code + id_delta) & 0xFFFF
        mappings[code] = glyph_index if glyph_index != 0
      end
    else
      # Use glyphIdArray
      (start_code..end_code).each do |code|
        # Calculate position in glyphIdArray
        array_offset = id_range_offset_pos + (i * 2) + id_range_offset
        array_offset += (code - start_code) * 2

        next if array_offset + 2 > length

        glyph_index = data[array_offset, 2].unpack1("n")
        next if glyph_index.zero?

        glyph_index = (glyph_index + id_delta) & 0xFFFF
        mappings[code] = glyph_index if glyph_index != 0
      end
    end
  end
end

#parse_mappings ⇒ Object

Parse all encoding records and extract Unicode mappings



41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
 
# File 'lib/fontisan/tables/cmap.rb', line 41

def parse_mappings
  mappings = {}

  # Get the full binary data
  data = to_binary_s

  # Read encoding records
  records = read_encoding_records(data)

  # Try to find the best Unicode subtable
  # Prefer Microsoft Unicode UCS-4 (format 12), then Unicode BMP (format 4)
  subtable_data = find_best_unicode_subtable(records, data)

  return mappings unless subtable_data

  # Parse the subtable based on its format
  format = subtable_data[0, 2].unpack1("n")

  case format
  when 4
    parse_format_4(subtable_data, mappings)
  when 12
    parse_format_12(subtable_data, mappings)
  end

  mappings
end

#read_encoding_records(data) ⇒ Object

Read encoding records from the beginning of the table



70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
 
# File 'lib/fontisan/tables/cmap.rb', line 70

def read_encoding_records(data)
  records = []
  offset = 4 # Skip version and numTables

  num_tables.times do
    break if offset + 8 > data.length

    platform_id = data[offset, 2].unpack1("n")
    encoding_id = data[offset + 2, 2].unpack1("n")
    subtable_offset = data[offset + 4, 4].unpack1("N")

    records << {
      platform_id: platform_id,
      encoding_id: encoding_id,
      offset: subtable_offset,
    }

    offset += 8
  end

  records
end

#subtable_formats ⇒ Array<Integer>

Distinct subtable formats present in this cmap, sorted ascending.

Each encoding record's subtable begins with a uint16 format tag. Returns the unique set of those formats across all records. Useful for audit reports that need to record which cmap subtables contributed to the Unicode mappings.

Returns:

  • (Array<Integer>) —

    sorted unique format numbers



359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
 
# File 'lib/fontisan/tables/cmap.rb', line 359

def subtable_formats
  data = to_binary_s
  records = read_encoding_records(data)

  formats = records.filter_map do |record|
    subtable_data = extract_subtable_data(record, data)
    next nil unless subtable_data && subtable_data.length >= 2

    subtable_data[0, 2].unpack1("n")
  end

  formats.uniq.sort
rescue StandardError
  []
end

#unicode_mappings ⇒ Object

Parse encoding records and subtables



36
37
38
 
# File 'lib/fontisan/tables/cmap.rb', line 36

def unicode_mappings
  @unicode_mappings ||= parse_mappings
end

#valid_glyph_indices?(max_glyph_id) ⇒ Boolean

Validation helper: Check if glyph indices are within bounds

Parameters:

  • max_glyph_id (Integer) —

    Maximum valid glyph ID from maxp table

Returns:

  • (Boolean) —

    True if all mapped glyph IDs are valid



379
380
381
382
383
384
385
386
 
# File 'lib/fontisan/tables/cmap.rb', line 379

def valid_glyph_indices?(max_glyph_id)
  mappings = unicode_mappings
  return true if mappings.nil? || mappings.empty?

  mappings.values.all? do |glyph_id|
    glyph_id >= 0 && glyph_id < max_glyph_id
  end
end

#valid_version? ⇒ Boolean

Validation helper: Check if version is valid

cmap version should be 0

Returns:

  • (Boolean) —

    True if version is 0



287
288
289
 
# File 'lib/fontisan/tables/cmap.rb', line 287

def valid_version?
  version.zero?
end