Class: KPeg::StringEscape

Inherits:
Object
  • Object
show all
Defined in:
lib/kpeg/string_escape.rb

Defined Under Namespace

Classes: MemoEntry, ParseError, RuleInfo

Constant Summary collapse

Rules = 
{}

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(str, debug = false) ⇒ StringEscape

This is distinct from setup_parser so that a standalone parser can redefine #initialize and still have access to the proper parser setup code.



7
8
9
 
# File 'lib/kpeg/string_escape.rb', line 7

def initialize(str, debug=false)
  setup_parser(str, debug)
end

Instance Attribute Details

#failed_rule ⇒ Object (readonly)

Returns the value of attribute failed_rule.



153
154
155
 
# File 'lib/kpeg/string_escape.rb', line 153

def failed_rule
  @failed_rule
end

#failing_rule_offset ⇒ Object (readonly)

Returns the value of attribute failing_rule_offset.



26
27
28
 
# File 'lib/kpeg/string_escape.rb', line 26

def failing_rule_offset
  @failing_rule_offset
end

#pos ⇒ Object

Returns the value of attribute pos.



27
28
29
 
# File 'lib/kpeg/string_escape.rb', line 27

def pos
  @pos
end

#result ⇒ Object

Returns the value of attribute result.



27
28
29
 
# File 'lib/kpeg/string_escape.rb', line 27

def result
  @result
end

#string ⇒ Object (readonly)

Returns the value of attribute string.



25
26
27
 
# File 'lib/kpeg/string_escape.rb', line 25

def string
  @string
end

#text ⇒ Object (readonly)

Returns the value of attribute text.



359
360
361
 
# File 'lib/kpeg/string_escape.rb', line 359

def text
  @text
end

Class Method Details

.rule_info(name, rendered) ⇒ Object 



351
352
353
 
# File 'lib/kpeg/string_escape.rb', line 351

def self.rule_info(name, rendered)
  RuleInfo.new(name, rendered)
end

Instance Method Details

#_embed ⇒ Object

embed = embed_seg*:s { @text = s.join }



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
 
# File 'lib/kpeg/string_escape.rb', line 593

def _embed

  _save = self.pos
  while true # sequence
    _ary = []
    while true
      _tmp = apply(:_embed_seg)
      _ary << @result if _tmp
      break unless _tmp
    end
    _tmp = true
    @result = _ary
    s = @result
    unless _tmp
      self.pos = _save
      break
    end
    @result = begin;  @text = s.join ; end
    _tmp = true
    unless _tmp
      self.pos = _save
    end
    break
  end # end sequence

  set_failed_rule :_embed unless _tmp
  return _tmp
end

#_embed_seg ⇒ Object

embed_seg = (“#” { “\#” } | segment)



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
587
588
589
590
 
# File 'lib/kpeg/string_escape.rb', line 560

def _embed_seg

  _save = self.pos
  while true # choice

    _save1 = self.pos
    while true # sequence
      _tmp = match_string("#")
      unless _tmp
        self.pos = _save1
        break
      end
      @result = begin;  "\\#" ; end
      _tmp = true
      unless _tmp
        self.pos = _save1
      end
      break
    end # end sequence

    break if _tmp
    self.pos = _save
    _tmp = apply(:_segment)
    break if _tmp
    self.pos = _save
    break
  end # end choice

  set_failed_rule :_embed_seg unless _tmp
  return _tmp
end

#_root ⇒ Object

root = segment*:s { @text = s.join }



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
555
556
557
 
# File 'lib/kpeg/string_escape.rb', line 530

def _root

  _save = self.pos
  while true # sequence
    _ary = []
    while true
      _tmp = apply(:_segment)
      _ary << @result if _tmp
      break unless _tmp
    end
    _tmp = true
    @result = _ary
    s = @result
    unless _tmp
      self.pos = _save
      break
    end
    @result = begin;  @text = s.join ; end
    _tmp = true
    unless _tmp
      self.pos = _save
    end
    break
  end # end sequence

  set_failed_rule :_root unless _tmp
  return _tmp
end

#_segment ⇒ Object

segment = (< /[w ]+/ > { text } | “\” { “\\” } | “n” { “\n” } | “r” { “\r” } | “t” { “\t” } | “b” { “\b” } | “"” { “\"” } | < . > { text })



366
367
368
369
370
371
372
373
374
375
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
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
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
456
457
458
459
460
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
488
489
490
491
492
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
 
# File 'lib/kpeg/string_escape.rb', line 366

def _segment

  _save = self.pos
  while true # choice

    _save1 = self.pos
    while true # sequence
      _text_start = self.pos
      _tmp = scan(/\A(?-mix:[\w ]+)/)
      if _tmp
        text = get_text(_text_start)
      end
      unless _tmp
        self.pos = _save1
        break
      end
      @result = begin;  text ; end
      _tmp = true
      unless _tmp
        self.pos = _save1
      end
      break
    end # end sequence

    break if _tmp
    self.pos = _save

    _save2 = self.pos
    while true # sequence
      _tmp = match_string("\\")
      unless _tmp
        self.pos = _save2
        break
      end
      @result = begin;  "\\\\" ; end
      _tmp = true
      unless _tmp
        self.pos = _save2
      end
      break
    end # end sequence

    break if _tmp
    self.pos = _save

    _save3 = self.pos
    while true # sequence
      _tmp = match_string("\n")
      unless _tmp
        self.pos = _save3
        break
      end
      @result = begin;  "\\n" ; end
      _tmp = true
      unless _tmp
        self.pos = _save3
      end
      break
    end # end sequence

    break if _tmp
    self.pos = _save

    _save4 = self.pos
    while true # sequence
      _tmp = match_string("\r")
      unless _tmp
        self.pos = _save4
        break
      end
      @result = begin;  "\\r" ; end
      _tmp = true
      unless _tmp
        self.pos = _save4
      end
      break
    end # end sequence

    break if _tmp
    self.pos = _save

    _save5 = self.pos
    while true # sequence
      _tmp = match_string("\t")
      unless _tmp
        self.pos = _save5
        break
      end
      @result = begin;  "\\t" ; end
      _tmp = true
      unless _tmp
        self.pos = _save5
      end
      break
    end # end sequence

    break if _tmp
    self.pos = _save

    _save6 = self.pos
    while true # sequence
      _tmp = match_string("\b")
      unless _tmp
        self.pos = _save6
        break
      end
      @result = begin;  "\\b" ; end
      _tmp = true
      unless _tmp
        self.pos = _save6
      end
      break
    end # end sequence

    break if _tmp
    self.pos = _save

    _save7 = self.pos
    while true # sequence
      _tmp = match_string("\"")
      unless _tmp
        self.pos = _save7
        break
      end
      @result = begin;  "\\\"" ; end
      _tmp = true
      unless _tmp
        self.pos = _save7
      end
      break
    end # end sequence

    break if _tmp
    self.pos = _save

    _save8 = self.pos
    while true # sequence
      _text_start = self.pos
      _tmp = get_byte
      if _tmp
        text = get_text(_text_start)
      end
      unless _tmp
        self.pos = _save8
        break
      end
      @result = begin;  text ; end
      _tmp = true
      unless _tmp
        self.pos = _save8
      end
      break
    end # end sequence

    break if _tmp
    self.pos = _save
    break
  end # end choice

  set_failed_rule :_segment unless _tmp
  return _tmp
end

#apply(rule) ⇒ Object 



286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
 
# File 'lib/kpeg/string_escape.rb', line 286

def apply(rule)
  if m = @memoizations[rule][@pos]
    @pos = m.pos
    if !m.set
      m.left_rec = true
      return nil
    end

    @result = m.result

    return m.ans
  else
    m = MemoEntry.new(nil, @pos)
    @memoizations[rule][@pos] = m
    start_pos = @pos

    ans = __send__ rule

    lr = m.left_rec

    m.move! ans, @pos, @result

    # Don't bother trying to grow the left recursion
    # if it's failing straight away (thus there is no seed)
    if ans and lr
      return grow_lr(rule, nil, start_pos, m)
    else
      return ans
    end

    return ans
  end
end

#apply_with_args(rule, *args) ⇒ Object 



251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
 
# File 'lib/kpeg/string_escape.rb', line 251

def apply_with_args(rule, *args)
  memo_key = [rule, args]
  if m = @memoizations[memo_key][@pos]
    @pos = m.pos
    if !m.set
      m.left_rec = true
      return nil
    end

    @result = m.result

    return m.ans
  else
    m = MemoEntry.new(nil, @pos)
    @memoizations[memo_key][@pos] = m
    start_pos = @pos

    ans = __send__ rule, *args

    lr = m.left_rec

    m.move! ans, @pos, @result

    # Don't bother trying to grow the left recursion
    # if it's failing straight away (thus there is no seed)
    if ans and lr
      return grow_lr(rule, args, start_pos, m)
    else
      return ans
    end

    return ans
  end
end

#current_column(target = pos) ⇒ Object 



29
30
31
32
33
34
35
 
# File 'lib/kpeg/string_escape.rb', line 29

def current_column(target=pos)
  if c = string.rindex("\n", target-1)
    return target - c - 1
  end

  target + 1
end

#current_line(target = pos) ⇒ Object 



37
38
39
40
41
42
43
44
45
46
47
48
 
# File 'lib/kpeg/string_escape.rb', line 37

def current_line(target=pos)
  cur_offset = 0
  cur_line = 0

  string.each_line do |line|
    cur_line += 1
    cur_offset += line.size
    return cur_line if cur_offset >= target
  end

  -1
end

#external_invoke(other, rule, *args) ⇒ Object 



232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
 
# File 'lib/kpeg/string_escape.rb', line 232

def external_invoke(other, rule, *args)
  old_pos = @pos
  old_string = @string

  set_string other.string, other.pos

  begin
    if val = __send__(rule, *args)
      other.pos = @pos
      other.result = @result
    else
      other.set_failed_rule "#{self.class}##{rule}"
    end
    val
  ensure
    set_string old_string, old_pos
  end
end

#failure_caret ⇒ Object 



90
91
92
93
94
95
96
 
# File 'lib/kpeg/string_escape.rb', line 90

def failure_caret
  l = current_line @failing_rule_offset
  c = current_column @failing_rule_offset

  line = lines[l-1]
  "#{line}\n#{' ' * (c - 1)}^"
end

#failure_character ⇒ Object 



98
99
100
101
102
 
# File 'lib/kpeg/string_escape.rb', line 98

def failure_character
  l = current_line @failing_rule_offset
  c = current_column @failing_rule_offset
  lines[l-1][c-1, 1]
end

#failure_info ⇒ Object 



78
79
80
81
82
83
84
85
86
87
88
 
# File 'lib/kpeg/string_escape.rb', line 78

def failure_info
  l = current_line @failing_rule_offset
  c = current_column @failing_rule_offset

  if @failed_rule.kind_of? Symbol
    info = self.class::Rules[@failed_rule]
    "line #{l}, column #{c}: failed rule '#{info.name}' = '#{info.rendered}'"
  else
    "line #{l}, column #{c}: failed rule '#{@failed_rule}'"
  end
end

#failure_oneline ⇒ Object 



104
105
106
107
108
109
110
111
112
113
114
115
116
 
# File 'lib/kpeg/string_escape.rb', line 104

def failure_oneline
  l = current_line @failing_rule_offset
  c = current_column @failing_rule_offset

  char = lines[l-1][c-1, 1]

  if @failed_rule.kind_of? Symbol
    info = self.class::Rules[@failed_rule]
    "@#{l}:#{c} failed rule '#{info.name}', got '#{char}'"
  else
    "@#{l}:#{c} failed rule '#{@failed_rule}', got '#{char}'"
  end
end

#get_byte ⇒ Object 



176
177
178
179
180
181
182
183
184
 
# File 'lib/kpeg/string_escape.rb', line 176

def get_byte
  if @pos >= @string_size
    return nil
  end

  s = @string[@pos].ord
  @pos += 1
  s
end

#get_text(start) ⇒ Object 



58
59
60
 
# File 'lib/kpeg/string_escape.rb', line 58

def get_text(start)
  @string[start..@pos-1]
end

#grow_lr(rule, args, start_pos, m) ⇒ Object 



320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
 
# File 'lib/kpeg/string_escape.rb', line 320

def grow_lr(rule, args, start_pos, m)
  while true
    @pos = start_pos
    @result = m.result

    if args
      ans = __send__ rule, *args
    else
      ans = __send__ rule
    end
    return nil unless ans

    break if @pos <= m.pos

    m.move! ans, @pos, @result
  end

  @result = m.result
  @pos = m.pos
  return m.ans
end

#lines ⇒ Object 



50
51
52
53
54
 
# File 'lib/kpeg/string_escape.rb', line 50

def lines
  lines = []
  string.each_line { |l| lines << l }
  lines
end

#match_string(str) ⇒ Object 



155
156
157
158
159
160
161
162
163
 
# File 'lib/kpeg/string_escape.rb', line 155

def match_string(str)
  len = str.size
  if @string[pos,len] == str
    @pos += len
    return str
  end

  return nil
end

#parse(rule = nil) ⇒ Object 



197
198
199
200
201
202
203
204
205
206
207
208
209
 
# File 'lib/kpeg/string_escape.rb', line 197

def parse(rule=nil)
  # We invoke the rules indirectly via apply
  # instead of by just calling them as methods because
  # if the rules use left recursion, apply needs to
  # manage that.

  if !rule
    apply(:_root)
  else
    method = rule.gsub("-","_hyphen_")
    apply :"_#{method}"
  end
end

#raise_error ⇒ Object

Raises:



121
122
123
 
# File 'lib/kpeg/string_escape.rb', line 121

def raise_error
  raise ParseError, failure_oneline
end

#scan(reg) ⇒ Object 



165
166
167
168
169
170
171
172
173
 
# File 'lib/kpeg/string_escape.rb', line 165

def scan(reg)
  if m = reg.match(@string[@pos..-1])
    width = m.end(0)
    @pos += width
    return true
  end

  return nil
end

#set_failed_rule(name) ⇒ Object 



146
147
148
149
150
151
 
# File 'lib/kpeg/string_escape.rb', line 146

def set_failed_rule(name)
  if @pos > @failing_rule_offset
    @failed_rule = name
    @failing_rule_offset = @pos
  end
end

#set_string(string, pos) ⇒ Object

Sets the string and current parsing position for the parser.



63
64
65
66
67
 
# File 'lib/kpeg/string_escape.rb', line 63

def set_string string, pos
  @string = string
  @string_size = string ? string.size : 0
  @pos = pos
end

#setup_foreign_grammar ⇒ Object

:stopdoc:



363
 
# File 'lib/kpeg/string_escape.rb', line 363

def setup_foreign_grammar; end

#setup_parser(str, debug = false) ⇒ Object

Prepares for parsing str. If you define a custom initialize you must call this method before #parse



15
16
17
18
19
20
21
22
23
 
# File 'lib/kpeg/string_escape.rb', line 15

def setup_parser(str, debug=false)
  set_string str, 0
  @memoizations = Hash.new { |h,k| h[k] = {} }
  @result = nil
  @failed_rule = nil
  @failing_rule_offset = -1

  setup_foreign_grammar
end

#show_error(io = STDOUT) ⇒ Object 



125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
 
# File 'lib/kpeg/string_escape.rb', line 125

def show_error(io=STDOUT)
  error_pos = @failing_rule_offset
  line_no = current_line(error_pos)
  col_no = current_column(error_pos)

  io.puts "On line #{line_no}, column #{col_no}:"

  if @failed_rule.kind_of? Symbol
    info = self.class::Rules[@failed_rule]
    io.puts "Failed to match '#{info.rendered}' (rule '#{info.name}')"
  else
    io.puts "Failed to match rule '#{@failed_rule}'"
  end

  io.puts "Got: #{string[error_pos,1].inspect}"
  line = lines[line_no-1]
  io.puts "=> #{line}"
  io.print(" " * (col_no + 3))
  io.puts "^"
end

#show_pos ⇒ Object 



69
70
71
72
73
74
75
76
 
# File 'lib/kpeg/string_escape.rb', line 69

def show_pos
  width = 10
  if @pos < width
    "#{@pos} (\"#{@string[0,@pos]}\" @ \"#{@string[@pos,width]}\")"
  else
    "#{@pos} (\"... #{@string[@pos - width, width]}\" @ \"#{@string[@pos,width]}\")"
  end
end