Class: Crass::Tokenizer

Inherits:
Object
  • Object
show all
Defined in:
lib/crass/tokenizer.rb

Overview

Tokenizes a CSS string.

  1. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#tokenization

Constant Summary collapse

RE_COMMENT_CLOSE = 
/\*\//
RE_DIGIT = 
/[0-9]+/
RE_ESCAPE = 
/\\[^\n]/
RE_HEX = 
/[0-9A-Fa-f]{1,6}/
RE_NAME = 
/[0-9A-Za-z_\u0080-\u{10ffff}-]+/
RE_NAME_START = 
/[A-Za-z_\u0080-\u{10ffff}]+/
RE_NON_PRINTABLE = 
/[\u0000-\u0008\u000b\u000e-\u001f\u007f]+/
RE_NUMBER_DECIMAL = 
/\.[0-9]+/
RE_NUMBER_EXPONENT = 
/[Ee][+-]?[0-9]+/
RE_NUMBER_SIGN = 
/[+-]/
RE_NUMBER_STR = 
/\A
  (?<sign> [+-]?)
  (?<integer> [0-9]*)
  (?:\.
    (?<fractional> [0-9]*)
  )?
  (?:[Ee]
    (?<exponent_sign> [+-]?)
    (?<exponent> [0-9]*)
  )?
\z/x
RE_QUOTED_URL_START = 
/\A[\n\u0009\u0020]?["']/
RE_UNICODE_RANGE_START = 
/\+(?:[0-9A-Fa-f]|\?)/
RE_UNICODE_RANGE_END = 
/-[0-9A-Fa-f]/
RE_WHITESPACE = 
/[\n\u0009\u0020]+/
RE_WHITESPACE_ANCHORED = 
/\A[\n\u0009\u0020]+\z/

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(input, options = {}) ⇒ Tokenizer

Initializes a new Tokenizer.

Options:

  • :preserve_comments - If true, comments will be preserved as :comment tokens.

  • :preserve_hacks - If true, certain non-standard browser hacks such as the IE "*" hack will be preserved even though they violate CSS 3 syntax rules.



62
63
64
65
 
# File 'lib/crass/tokenizer.rb', line 62

def initialize(input, options = {})
  @s       = Scanner.new(preprocess(input))
  @options = options
end

Class Method Details

.tokenize(input, options = {}) ⇒ Object

Tokenizes the given input as a CSS string and returns an array of tokens.

See #initialize for options.



45
46
47
 
# File 'lib/crass/tokenizer.rb', line 45

def self.tokenize(input, options = {})
  Tokenizer.new(input, options).tokenize
end

Instance Method Details

#consume ⇒ Object

Consumes a token and returns the token that was consumed.

4.3.1. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-token



70
71
72
73
74
75
76
77
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
104
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
131
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
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
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
 
# File 'lib/crass/tokenizer.rb', line 70

def consume
  # Skip over any comments. When comments aren't being preserved this is
  # done iteratively rather than recursively so that a long run of adjacent
  # comments can't exhaust the Ruby stack and raise a `SystemStackError`.
  loop do
    return nil if @s.eos?

    @s.mark

    comment_token = consume_comments
    break unless comment_token

    return comment_token if @options[:preserve_comments]
  end

  # Consume whitespace.
  return create_token(:whitespace) if @s.scan(RE_WHITESPACE)

  char = @s.consume

  case char.to_sym
  when :'"'
    consume_string

  when :'#'
    if @s.peek =~ RE_NAME || valid_escape?(@s.peek(2))
      create_token(:hash,
        :type  => start_identifier?(@s.peek(3)) ? :id : :unrestricted,
        :value => consume_name)
    else
      create_token(:delim, :value => char)
    end

  when :'$'
    if @s.peek == '='
      @s.consume
      create_token(:suffix_match)
    else
      create_token(:delim, :value => char)
    end

  when :"'"
    consume_string

  when :'('
    create_token(:'(')

  when :')'
    create_token(:')')

  when :*
    if @s.peek == '='
      @s.consume
      create_token(:substring_match)

    # Non-standard: Preserve the IE * hack.
    elsif @options[:preserve_hacks] && @s.peek =~ RE_NAME_START
      @s.reconsume
      consume_ident

    else
      create_token(:delim, :value => char)
    end

  when :+
    if start_number?
      @s.reconsume
      consume_numeric
    else
      create_token(:delim, :value => char)
    end

  when :','
    create_token(:comma)

  when :-
    nextTwoChars   = @s.peek(2)
    nextThreeChars = char + nextTwoChars

    if start_number?(nextThreeChars)
      @s.reconsume
      consume_numeric
    elsif nextTwoChars == '->'
      @s.consume
      @s.consume
      create_token(:cdc)
    elsif start_identifier?(nextThreeChars)
      @s.reconsume
      consume_ident
    else
      create_token(:delim, :value => char)
    end

  when :'.'
    if start_number?
      @s.reconsume
      consume_numeric
    else
      create_token(:delim, :value => char)
    end

  when :':'
    create_token(:colon)

  when :';'
    create_token(:semicolon)

  when :<
    if @s.peek(3) == '!--'
      @s.consume
      @s.consume
      @s.consume

      create_token(:cdo)
    else
      create_token(:delim, :value => char)
    end

  when :'@'
    if start_identifier?(@s.peek(3))
      create_token(:at_keyword, :value => consume_name)
    else
      create_token(:delim, :value => char)
    end

  when :'['
    create_token(:'[')

  when :'\\'
    if valid_escape?
      @s.reconsume
      consume_ident
    else
      # Parse error.
      create_token(:delim,
        :error => true,
        :value => char)
    end

  when :']'
    create_token(:']')

  when :'^'
    if @s.peek == '='
      @s.consume
      create_token(:prefix_match)
    else
      create_token(:delim, :value => char)
    end

  when :'{'
    create_token(:'{')

  when :'}'
    create_token(:'}')

  when :U, :u
    if @s.peek(2) =~ RE_UNICODE_RANGE_START
      @s.consume
      consume_unicode_range
    else
      @s.reconsume
      consume_ident
    end

  when :|
    case @s.peek
    when '='
      @s.consume
      create_token(:dash_match)

    when '|'
      @s.consume
      create_token(:column)

    else
      create_token(:delim, :value => char)
    end

  when :~
    if @s.peek == '='
      @s.consume
      create_token(:include_match)
    else
      create_token(:delim, :value => char)
    end

  else
    case char
    when RE_DIGIT
      @s.reconsume
      consume_numeric

    when RE_NAME_START
      @s.reconsume
      consume_ident

    else
      create_token(:delim, :value => char)
    end
  end
end

#consume_bad_url ⇒ Object

Consumes the remnants of a bad URL and returns the consumed text.

4.3.15. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-the-remnants-of-a-bad-url



276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
 
# File 'lib/crass/tokenizer.rb', line 276

def consume_bad_url
  text = String.new

  until @s.eos?
    if valid_escape?
      text << consume_escaped
    elsif valid_escape?(@s.peek(2))
      @s.consume
      text << consume_escaped
    else
      char = @s.consume

      if char == ')'
        break
      else
        text << char
      end
    end
  end

  text
end

#consume_comments ⇒ Object

Consumes comments and returns them, or nil if no comments were consumed.

4.3.2. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-comments



302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
 
# File 'lib/crass/tokenizer.rb', line 302

def consume_comments
  if @s.peek(2) == '/*'
    @s.consume
    @s.consume

    if text = @s.scan_until(RE_COMMENT_CLOSE)
      text.slice!(-2, 2)
    else
      # Parse error.
      text = @s.consume_rest
    end

    return create_token(:comment, :value => text)
  end

  nil
end

#consume_escaped ⇒ Object

Consumes an escaped code point and returns its unescaped value.

This method assumes that the \ has already been consumed, and that the next character in the input has already been verified not to be a newline or EOF.

4.3.8. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-an-escaped-code-point



327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
 
# File 'lib/crass/tokenizer.rb', line 327

def consume_escaped
  return "\ufffd" if @s.eos?

  if hex_str = @s.scan(RE_HEX)
    @s.consume if @s.peek =~ RE_WHITESPACE

    codepoint = hex_str.hex

    if codepoint == 0 ||
        codepoint.between?(0xD800, 0xDFFF) ||
        codepoint > 0x10FFFF

      return "\ufffd"
    else
      return codepoint.chr(Encoding::UTF_8)
    end
  end

  @s.consume
end

#consume_ident ⇒ Object

Consumes an ident-like token and returns it.

4.3.4. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-an-ident-like-token



351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
 
# File 'lib/crass/tokenizer.rb', line 351

def consume_ident
  value = consume_name

  if @s.peek == '('
    @s.consume

    if value.downcase == 'url'
      @s.consume while @s.peek(2) =~ RE_WHITESPACE_ANCHORED

      if @s.peek(2) =~ RE_QUOTED_URL_START
        create_token(:function, :value => value)
      else
        consume_url
      end
    else
      create_token(:function, :value => value)
    end
  else
    create_token(:ident, :value => value)
  end
end

#consume_name ⇒ Object

Consumes a name and returns it.

4.3.12. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-name



376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
 
# File 'lib/crass/tokenizer.rb', line 376

def consume_name
  result = String.new

  until @s.eos?
    if match = @s.scan(RE_NAME)
      result << match
      next
    end

    char = @s.consume

    if valid_escape?
      result << consume_escaped

    # Non-standard: IE * hack
    elsif char == '*' && @options[:preserve_hacks]
      result << @s.consume

    else
      @s.reconsume
      return result
    end
  end

  result
end

#consume_number ⇒ Object

Consumes a number and returns a 3-element array containing the number's original representation, its numeric value, and its type (either :integer or :number).

4.3.13. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-number



408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
 
# File 'lib/crass/tokenizer.rb', line 408

def consume_number
  repr = String.new
  type = :integer

  repr << @s.consume if @s.peek =~ RE_NUMBER_SIGN
  repr << (@s.scan(RE_DIGIT) || '')

  if match = @s.scan(RE_NUMBER_DECIMAL)
    repr << match
    type = :number
  end

  if match = @s.scan(RE_NUMBER_EXPONENT)
    repr << match
    type = :number
  end

  [repr, convert_string_to_number(repr), type]
end

#consume_numeric ⇒ Object

Consumes a numeric token and returns it.

4.3.3. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-numeric-token



431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
 
# File 'lib/crass/tokenizer.rb', line 431

def consume_numeric
  number = consume_number
  repr = number[0]
  value = number[1]
  type = number[2]

  if type == :integer
    value = value.to_i
  else
    value = value.to_f
  end

  if start_identifier?(@s.peek(3))
    create_token(:dimension,
      :repr => repr,
      :type => type,
      :unit => consume_name,
      :value => value)

  elsif @s.peek == '%'
    @s.consume

    create_token(:percentage,
      :repr => repr,
      :type => type,
      :value => value)

  else
    create_token(:number,
      :repr => repr,
      :type => type,
      :value => value)
  end
end

#consume_string(ending = nil) ⇒ Object

Consumes a string token that ends at the given character, and returns the token.

4.3.5. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-string-token



470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
 
# File 'lib/crass/tokenizer.rb', line 470

def consume_string(ending = nil)
  ending = @s.current if ending.nil?
  value  = String.new

  until @s.eos?
    case char = @s.consume
    when ending
      break

    when "\n"
      # Parse error.
      @s.reconsume
      return create_token(:bad_string,
        :error => true,
        :value => value)

    when '\\'
      case @s.peek
      when ''
        # End of the input, so do nothing.
        next

      when "\n"
        @s.consume

      else
        value << consume_escaped
      end

    else
      value << char
    end
  end

  create_token(:string, :value => value)
end

#consume_unicode_range ⇒ Object

Consumes a Unicode range token and returns it. Assumes the initial "u+" or "U+" has already been consumed.

4.3.7. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-unicode-range-token



511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
 
# File 'lib/crass/tokenizer.rb', line 511

def consume_unicode_range
  value = @s.scan(RE_HEX) || String.new

  while value.length < 6
    break unless @s.peek == '?'
    value << @s.consume
  end

  range = {}

  if value.include?('?')
    range[:start] = value.gsub('?', '0').hex
    range[:end]   = value.gsub('?', 'F').hex
    return create_token(:unicode_range, range)
  end

  range[:start] = value.hex

  if @s.peek(2) =~ RE_UNICODE_RANGE_END
    @s.consume
    range[:end] = (@s.scan(RE_HEX) || '').hex
  else
    range[:end] = range[:start]
  end

  create_token(:unicode_range, range)
end

#consume_url ⇒ Object

Consumes a URL token and returns it. Assumes the original "url(" has already been consumed.

4.3.6. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-url-token



543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
 
# File 'lib/crass/tokenizer.rb', line 543

def consume_url
  value = String.new

  @s.scan(RE_WHITESPACE)

  until @s.eos?
    case char = @s.consume
    when ')'
      break

    when RE_WHITESPACE
      @s.scan(RE_WHITESPACE)

      if @s.eos? || @s.peek == ')'
        @s.consume
        break
      else
        return create_token(:bad_url, :value => value + consume_bad_url)
      end

    when '"', "'", '(', RE_NON_PRINTABLE
      # Parse error.
      return create_token(:bad_url,
        :error => true,
        :value => value + consume_bad_url)

    when '\\'
      if valid_escape?
        value << consume_escaped
      else
        # Parse error.
        return create_token(:bad_url,
          :error => true,
          :value => value + consume_bad_url
        )
      end

    else
      value << char
    end
  end

  create_token(:url, :value => value)
end

#convert_string_to_number(str) ⇒ Object

Converts a valid CSS number string into a number and returns the number.

4.3.14. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#convert-a-string-to-a-number



591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
 
# File 'lib/crass/tokenizer.rb', line 591

def convert_string_to_number(str)
  matches = RE_NUMBER_STR.match(str)

  s = matches[:sign] == '-' ? -1 : 1
  i = matches[:integer].to_i
  f = matches[:fractional].to_i
  d = matches[:fractional] ? matches[:fractional].length : 0
  t = matches[:exponent_sign] == '-' ? -1 : 1
  e = matches[:exponent].to_i

  exponent = t * e

  # Guard against denial of service via attacker-controlled exponentiation.
  # A tiny input like "1e5000000" would otherwise force Ruby to compute
  # `10**exponent` (an enormous integer) before the value is clamped below,
  # consuming disproportionate CPU and memory. When the exponent's magnitude
  # is far outside the range a finite double can represent, we saturate or
  # underflow without performing the exponentiation.
  if i == 0 && f == 0
    # The mantissa is zero, so the value is zero regardless of the exponent.
    value = s * 0.0

  elsif exponent - d > Float::MAX_10_EXP + 1
    # The value is larger than `Float::MAX` and would be clamped anyway.
    value = s * Float::MAX

  elsif exponent + matches[:integer].length < -325
    # The value is smaller than the smallest representable double and would
    # round to zero. (-325 is one less than the smallest subnormal base-10
    # exponent of roughly -324; `matches[:integer].length` is an upper bound
    # on the mantissa's magnitude, ensuring we never zero a representable
    # value.)
    value = s * 0.0

  else
    # I know this formula looks nutty, but it's exactly what's defined in
    # the spec, and it works.
    value = s * (i + f * 10**-d) * 10**exponent

    # Maximum and minimum values aren't defined in the spec, but are
    # enforced here for sanity.
    if value > Float::MAX
      value = Float::MAX
    elsif value < -Float::MAX
      value = -Float::MAX
    end
  end

  value
end

#create_token(type, properties = {}) ⇒ Object

Creates and returns a new token with the given properties.



643
644
645
646
647
648
649
 
# File 'lib/crass/tokenizer.rb', line 643

def create_token(type, properties = {})
  {
    :node => type,
    :pos  => @s.marker,
    :raw  => @s.marked
  }.merge!(properties)
end

#preprocess(input) ⇒ Object

Preprocesses input to prepare it for the tokenizer.

3.3. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#input-preprocessing



654
655
656
657
658
659
660
661
662
 
# File 'lib/crass/tokenizer.rb', line 654

def preprocess(input)
  input = input.to_s.encode('UTF-8',
    :invalid => :replace,
    :undef   => :replace)

  input.gsub!(/(?:\r\n|[\r\f])/, "\n")
  input.gsub!("\u0000", "\ufffd")
  input
end

#start_identifier?(text = nil) ⇒ Boolean

Returns true if the given three-character text would start an identifier. If text is nil, the current and next two characters in the input stream will be checked, but will not be consumed.

4.3.10. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#would-start-an-identifier

Returns:

  • (Boolean) 


669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
 
# File 'lib/crass/tokenizer.rb', line 669

def start_identifier?(text = nil)
  text = @s.current + @s.peek(2) if text.nil?

  case text[0]
  when '-'
    nextChar = text[1]
    !!(nextChar == '-' || nextChar =~ RE_NAME_START || valid_escape?(text[1, 2]))

  when RE_NAME_START
    true

  when '\\'
    valid_escape?(text[0, 2])

  else
    false
  end
end

#start_number?(text = nil) ⇒ Boolean

Returns true if the given three-character text would start a number. If text is nil, the current and next two characters in the input stream will be checked, but will not be consumed.

4.3.11. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#starts-with-a-number

Returns:

  • (Boolean) 


693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
 
# File 'lib/crass/tokenizer.rb', line 693

def start_number?(text = nil)
  text = @s.current + @s.peek(2) if text.nil?

  case text[0]
  when '+', '-'
    !!(text[1] =~ RE_DIGIT || (text[1] == '.' && text[2] =~ RE_DIGIT))

  when '.'
    !!(text[1] =~ RE_DIGIT)

  when RE_DIGIT
    true

  else
    false
  end
end

#tokenize ⇒ Object

Tokenizes the input stream and returns an array of tokens.



712
713
714
715
716
717
718
719
720
721
722
 
# File 'lib/crass/tokenizer.rb', line 712

def tokenize
  @s.reset

  tokens = []

  while token = consume
    tokens << token
  end

  tokens
end

#valid_escape?(text = nil) ⇒ Boolean

Returns true if the given two-character text is the beginning of a valid escape sequence. If text is nil, the current and next character in the input stream will be checked, but will not be consumed.

4.3.9. https://www.w3.org/TR/2013/WD-css-syntax-3-20130919/#starts-with-a-valid-escape

Returns:

  • (Boolean) 


729
730
731
732
 
# File 'lib/crass/tokenizer.rb', line 729

def valid_escape?(text = nil)
  text = @s.current + @s.peek if text.nil?
  !!(text[0] == '\\' && text[1] != "\n")
end