Class: Crass::Tokenizer

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

Overview

Tokenizes a CSS string.

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_UNICODE_RANGE_START = 
/\+(?:[0-9A-Fa-f]|\?)/
RE_UNICODE_RANGE_END = 
/-[0-9A-Fa-f]/
RE_URL_QUOTE = 
/["']/
RE_WHITESPACE = 
/[\n\u0009\u0020]+/

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.



61
62
63
64
 
# File 'lib/crass/tokenizer.rb', line 61

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.



44
45
46
 
# File 'lib/crass/tokenizer.rb', line 44

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.

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-token0



69
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
 
# File 'lib/crass/tokenizer.rb', line 69

def consume
  return token(:eof) if @s.eos?

  @s.mark
  return token(:whitespace) if @s.scan(RE_WHITESPACE)

  case char = @s.consume
  when '"'
    consume_string('"')

  when '#'
    if @s.peek =~ RE_NAME || valid_escape?
      value = consume_name

      token(:hash,
        :type  => start_identifier? ? :id : :unrestricted,
        :value => value)
    else
      token(:delim, :value => char)
    end

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

  when "'"
    consume_string("'")

  when '('
    token(:'(')

  when ')'
    token(:')')

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

    elsif @options[:preserve_hacks] && @s.peek =~ RE_NAME_START
      # NON-STANDARD: IE * hack
      @s.reconsume
      consume_ident

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

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

  when ','
    token(:comma)

  when '-'
    if start_number?
      @s.reconsume
      consume_numeric
    elsif start_identifier?
      @s.reconsume
      consume_ident
    elsif @s.peek(2) == '->'
      @s.consume
      @s.consume
      token(:cdc)
    else
      token(:delim, :value => char)
    end

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

  when '/'
    if @s.peek == '*'
      @s.consume

      if text = @s.scan_until(RE_COMMENT_CLOSE)
        text.slice!(-2, 2)
      else
        text = @s.rest
      end

      if @options[:preserve_comments]
        token(:comment, :value => text)
      else
        consume
      end
    else
      token(:delim, :value => char)
    end

  when ':'
    token(:colon)

  when ';'
    token(:semicolon)

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

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

  when '@'
    if start_identifier?
      token(:at_keyword, :value => consume_name)
    else
      token(:delim, :value => char)
    end

  when '['
    token(:'[')

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

  when ']'
    token(:']')

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

  when '{'
    token(:'{')

  when '}'
    token(:'}')

  when RE_DIGIT
    @s.reconsume
    consume_numeric

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

  when RE_NAME_START
    @s.reconsume
    consume_ident

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

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

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

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

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

#consume_bad_url ⇒ Object

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

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-the-remnants-of-a-bad-url0



275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
 
# File 'lib/crass/tokenizer.rb', line 275

def consume_bad_url
  text = ''

  while true
    return text if @s.eos?

    if valid_escape?
      text << consume_escaped
    else
      char = @s.consume

      if char == ')'
        return text
      else
        text << char
      end
    end
  end
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.

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-an-escaped-code-point0



302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
 
# File 'lib/crass/tokenizer.rb', line 302

def consume_escaped
  case
  when @s.eos?
    "\ufffd"

  when 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

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

  else
    @s.consume
  end
end

#consume_ident ⇒ Object

Consumes an ident-like token and returns it.

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-an-ident-like-token0



329
330
331
332
333
334
335
336
337
338
339
340
341
 
# File 'lib/crass/tokenizer.rb', line 329

def consume_ident
  value = consume_name

  if value.downcase == 'url' && @s.peek == '('
    @s.consume
    consume_url
  elsif @s.peek == '('
    @s.consume
    token(:function, :value => value)
  else
    token(:ident, :value => value)
  end
end

#consume_name ⇒ Object

Consumes a name and returns it.

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-name0



346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
 
# File 'lib/crass/tokenizer.rb', line 346

def consume_name
  result = ''

  while char = @s.peek
    if char =~ RE_NAME
      result << @s.consume

    elsif char == '\\' && valid_escape?
      result << @s.consume
      result << consume_escaped

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

    else
      return result
    end
  end
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`).

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-number0



372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
 
# File 'lib/crass/tokenizer.rb', line 372

def consume_number
  repr = ''
  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.

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-numeric-token0



395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
 
# File 'lib/crass/tokenizer.rb', line 395

def consume_numeric
  number = consume_number

  if start_identifier?
    token(:dimension,
      :repr  => number[0],
      :type  => number[2],
      :unit  => consume_name,
      :value => number[1])

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

    token(:percentage,
      :repr  => number[0],
      :value => number[1])

  else
    token(:number,
      :repr  => number[0],
      :type  => number[2],
      :value => number[1])
  end
end

#consume_string(ending) ⇒ Object

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

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-string-token0



424
425
426
427
428
429
430
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
 
# File 'lib/crass/tokenizer.rb', line 424

def consume_string(ending)
  value = ''

  while char = @s.consume
    case char
    when ending then break

    when "\n"
      return 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

  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.

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-unicode-range-token0



461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
 
# File 'lib/crass/tokenizer.rb', line 461

def consume_unicode_range
  value = @s.scan(RE_HEX)

  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 token(:unicode_range, range)
  end

  range[:start] = value.hex

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

  token(:unicode_range, range)
end

#consume_url ⇒ Object

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

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#consume-a-url-token0



493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
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
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
 
# File 'lib/crass/tokenizer.rb', line 493

def consume_url
  value = ''

  @s.scan(RE_WHITESPACE)
  return token(:url, :value => value) if @s.eos?

  # Quoted URL.
  if @s.peek =~ RE_URL_QUOTE
    string = consume_string(@s.consume)

    if string[:node] == :bad_string
      return token(:bad_url, :value => string[:value] + consume_bad_url)
    end

    value = string[:value]
    @s.scan(RE_WHITESPACE)

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

  # Unquoted URL.
  while !@s.eos?
    case char = @s.consume
    when ')' then break

    when RE_WHITESPACE
      @s.scan(RE_WHITESPACE)

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

    when '"', "'", '(', RE_NON_PRINTABLE
      return token(:bad_url,
        :error => true,
        :value => value + consume_bad_url)

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

    else
      value << char
    end
  end

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

#convert_string_to_number(str) ⇒ Object

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

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#convert-a-string-to-a-number0



559
560
561
562
563
564
565
566
567
568
569
570
571
572
 
# File 'lib/crass/tokenizer.rb', line 559

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

  # I know this looks nutty, but it's exactly what's defined in the spec,
  # and it works.
  s * (i + f * 10**-d) * 10**(t * e)
end

#preprocess(input) ⇒ Object

Preprocesses input to prepare it for the tokenizer.

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



577
578
579
580
581
582
583
584
585
 
# File 'lib/crass/tokenizer.rb', line 577

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 next three characters in the input stream will be checked, but will not be consumed.

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#check-if-three-code-points-would-start-an-identifier

Returns:

  • (Boolean) 


592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
 
# File 'lib/crass/tokenizer.rb', line 592

def start_identifier?(text = nil)
  text = @s.peek(3) if text.nil?

  case text[0]
  when '-'
    !!(text[1] =~ 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 next three characters in the input stream will be checked, but will not be consumed.

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#check-if-three-code-points-would-start-a-number

Returns:

  • (Boolean) 


615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
 
# File 'lib/crass/tokenizer.rb', line 615

def start_number?(text = nil)
  text = @s.peek(3) 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

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

Creates and returns a new token with the given properties.



634
635
636
637
638
639
640
 
# File 'lib/crass/tokenizer.rb', line 634

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

#tokenize ⇒ Object

Tokenizes the input stream and returns an array of tokens.



643
644
645
646
647
648
649
650
651
652
653
654
655
 
# File 'lib/crass/tokenizer.rb', line 643

def tokenize
  @s.reset

  tokens = []
  token  = consume

  while token && token[:node] != :eof
    tokens << token
    token = consume
  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 next two characters in the input stream will be checked, but will not be consumed.

www.w3.org/TR/2013/WD-css-syntax-3-20130919/#check-if-two-code-points-are-a-valid-escape

Returns:

  • (Boolean) 


662
663
664
665
 
# File 'lib/crass/tokenizer.rb', line 662

def valid_escape?(text = nil)
  text = @s.peek(2) if text.nil?
  !!(text[0] == '\\' && text[1] != "\n")
end