Module: RubyParserStuff

Included in:
RubyParser::Parser
Defined in:
lib/ruby_parser_extras.rb

Defined Under Namespace

Classes: Environment, Keyword, StackState

Constant Summary collapse

VERSION =
"3.14.1"
ARG_TYPES =
[:arglist, :call_args, :array, :args].map { |k|
  [k, true]
}.to_h
ENCODING_ORDER =

This is in sorted order of occurrence according to charlock_holmes against 500k files, with UTF_8 forced to the top.

Overwrite this contstant if you need something different.

[
  Encoding::UTF_8, # moved to top to reflect default in 2.0
  Encoding::ISO_8859_1,
  Encoding::ISO_8859_2,
  Encoding::ISO_8859_9,
  Encoding::SHIFT_JIS,
  Encoding::WINDOWS_1252,
  Encoding::EUC_JP
]
JUMP_TYPE =
[:return, :next, :break, :yield].map { |k| [k, true] }.to_h
TAB_WIDTH =
8

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Instance Attribute Details

#canonicalize_conditionsObject

Canonicalize conditionals. Eg:

not x ? a : b

becomes:

x ? b : a


46
47
48
# File 'lib/ruby_parser_extras.rb', line 46

def canonicalize_conditions
  @canonicalize_conditions
end

#commentsObject (readonly)

Returns the value of attribute comments.



35
36
37
# File 'lib/ruby_parser_extras.rb', line 35

def comments
  @comments
end

#envObject (readonly)

Returns the value of attribute env.



35
36
37
# File 'lib/ruby_parser_extras.rb', line 35

def env
  @env
end

#fileObject

Returns the value of attribute file.



33
34
35
# File 'lib/ruby_parser_extras.rb', line 33

def file
  @file
end

#in_defObject

Returns the value of attribute in_def.



33
34
35
# File 'lib/ruby_parser_extras.rb', line 33

def in_def
  @in_def
end

#in_kwargObject

Returns the value of attribute in_kwarg.



34
35
36
# File 'lib/ruby_parser_extras.rb', line 34

def in_kwarg
  @in_kwarg
end

#in_singleObject

Returns the value of attribute in_single.



33
34
35
# File 'lib/ruby_parser_extras.rb', line 33

def in_single
  @in_single
end

#lexerObject

Returns the value of attribute lexer.



33
34
35
# File 'lib/ruby_parser_extras.rb', line 33

def lexer
  @lexer
end

Class Method Details

.deprecate(old, new) ⇒ Object



59
60
61
62
63
64
# File 'lib/ruby_parser_extras.rb', line 59

def self.deprecate old, new
  define_method old do |*args|
    warn "DEPRECATED: #{old} -> #{new} from #{caller.first}"
    send new, *args
  end
end

Instance Method Details

#_racc_do_reduce(arg, act) ⇒ Object



74
75
76
77
78
79
80
81
# File 'lib/ruby_parser_extras.rb', line 74

def _racc_do_reduce arg, act
  x = super

  @racc_vstack.grep(Sexp).each do |sexp|
    sexp.check_line_numbers
  end
  x
end

#arg_blk_pass(node1, node2) ⇒ Object

TODO: nuke



127
128
129
130
131
# File 'lib/ruby_parser_extras.rb', line 127

def arg_blk_pass node1, node2 # TODO: nuke
  node1 = s(:arglist, node1) unless ARG_TYPES[node1.sexp_type]
  node1 << node2 if node2
  node1
end

#arg_concat(node1, node2) ⇒ Object

TODO: nuke



133
134
135
136
137
138
139
140
# File 'lib/ruby_parser_extras.rb', line 133

def arg_concat node1, node2 # TODO: nuke
  raise "huh" unless node2

  splat = s(:splat, node2)
  splat.line node2.line

  node1 << splat
end

#argl(x) ⇒ Object



142
143
144
145
# File 'lib/ruby_parser_extras.rb', line 142

def argl x
  x = s(:arglist, x) if x and x.sexp_type == :array
  x
end

#args(args) ⇒ Object



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
# File 'lib/ruby_parser_extras.rb', line 147

def args args
  result = s(:args)

  ss = args.grep Sexp
  if ss.empty? then
    result.line lexer.lineno
  else
    result.line ss.first.line
  end

  args.each do |arg|
    case arg
    when Sexp then
      case arg.sexp_type
      when :args, :block, :array, :call_args then # HACK call_args mismatch
        result.concat arg.sexp_body
      when :block_arg then
        result << :"&#{arg.last}"
      when :shadow then
        name = arg.last
        self.env[name] = :lvar
        if Sexp === result.last and result.last.sexp_type == :shadow then
          result.last << name
        else
          result << arg
        end
      when :masgn, :block_pass, :hash then # HACK: remove. prolly call_args
        result << arg
      else
        raise "unhandled: #{arg.sexp_type} in #{args.inspect}"
      end
    when Symbol then
      name = arg.to_s.delete("&*")
      self.env[name.to_sym] = :lvar unless name.empty?
      result << arg
    when ",", "|", ";", "(", ")", nil then
      # ignore
    else
      raise "unhandled: #{arg.inspect} in #{args.inspect}"
    end
  end

  result
end

#array_to_hash(array) ⇒ Object



192
193
194
195
196
197
198
199
# File 'lib/ruby_parser_extras.rb', line 192

def array_to_hash array
  case array.sexp_type
  when :kwsplat then
    array
  else
    s(:hash, *array.sexp_body).line array.line
  end
end

#aryset(receiver, index) ⇒ Object



201
202
203
204
205
206
207
208
# File 'lib/ruby_parser_extras.rb', line 201

def aryset receiver, index
  index ||= s()
  l = receiver.line
  result = s(:attrasgn, receiver, :"[]=",
             *index.sexp_body).compact # [].sexp_body => nil
  result.line = l
  result
end

#assignable(lhs, value = nil) ⇒ Object

Raises:

  • (SyntaxError)


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
# File 'lib/ruby_parser_extras.rb', line 210

def assignable(lhs, value = nil)
  id = lhs.to_sym unless Sexp === lhs

  raise "WTF" if Sexp === id
  id = id.to_sym if Sexp === id

  raise "write a test 1" if id.to_s =~ /^(?:self|nil|true|false|__LINE__|__FILE__)$/

  raise SyntaxError, "Can't change the value of #{id}" if
    id.to_s =~ /^(?:self|nil|true|false|__LINE__|__FILE__)$/

  result = case id.to_s
           when /^@@/ then
             asgn = in_def || in_single > 0
             s((asgn ? :cvasgn : :cvdecl), id)
           when /^@/ then
             s(:iasgn, id)
           when /^\$/ then
             s(:gasgn, id)
           when /^[A-Z]/ then
             s(:cdecl, id)
           else
             case self.env[id]
             when :lvar, :dvar, nil then
               s(:lasgn, id)
             else
               raise "wtf? unknown type: #{self.env[id]}"
             end
           end

  self.env[id] ||= :lvar if result.sexp_type == :lasgn

  line = case lhs
         when Sexp then
           lhs.line
         else
           value && value.line || lexer.lineno
         end

  result << value if value
  result.line = line

  return result
end

#backref_assign_error(ref) ⇒ Object



255
256
257
258
259
260
261
262
263
264
265
266
267
# File 'lib/ruby_parser_extras.rb', line 255

def backref_assign_error ref
  # TODO: need a test for this... obviously
  case ref.sexp_type
  when :nth_ref then
    raise "write a test 2"
    raise SyntaxError, "Can't set variable %p" % ref.last
  when :back_ref then
    raise "write a test 3"
    raise SyntaxError, "Can't set back reference %p" % ref.last
  else
    raise "Unknown backref type: #{ref.inspect}"
  end
end

#block_append(head, tail) ⇒ Object



269
270
271
272
273
274
275
276
277
278
279
280
# File 'lib/ruby_parser_extras.rb', line 269

def block_append(head, tail)
  return head if tail.nil?
  return tail if head.nil?

  line = [head.line, tail.line].compact.min

  head = remove_begin(head)
  head = s(:block, head) unless head.node_type == :block

  head.line = line
  head << tail
end

#block_dup_check(call_or_args, block) ⇒ Object



282
283
284
285
# File 'lib/ruby_parser_extras.rb', line 282

def block_dup_check call_or_args, block
  syntax_error "Both block arg and actual block given." if
    block and call_or_args.block_pass?
end

#block_var(*args) ⇒ Object



287
288
289
290
291
# File 'lib/ruby_parser_extras.rb', line 287

def block_var *args
  result = self.args args
  result.sexp_type = :masgn
  result
end

#call_args(args) ⇒ Object



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
319
320
321
322
# File 'lib/ruby_parser_extras.rb', line 293

def call_args args
  result = s(:call_args)

  a = args.grep(Sexp).first
  if a then
    result.line a.line
  else
    result.line lexer.lineno
  end

  args.each do |arg|
    case arg
    when Sexp then
      case arg.sexp_type
      when :array, :args, :call_args then # HACK? remove array at some point
        result.concat arg.sexp_body
      else
        result << arg
      end
    when Symbol then
      result << arg
    when ",", nil then
      # ignore
    else
      raise "unhandled: #{arg.inspect} in #{args.inspect}"
    end
  end

  result
end

#clean_mlhs(sexp) ⇒ Object



324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
# File 'lib/ruby_parser_extras.rb', line 324

def clean_mlhs sexp
  case sexp.sexp_type
  when :masgn then
    if sexp.size == 2 and sexp[1].sexp_type == :array then
      s(:masgn, *sexp[1].sexp_body.map { |sub| clean_mlhs sub })
    else
      debug20 5
      sexp
    end
  when :gasgn, :iasgn, :lasgn, :cvasgn then
    if sexp.size == 2 then
      sexp.last
    else
      debug20 7
      sexp # optional value
    end
  else
    raise "unsupported type: #{sexp.inspect}"
  end
end

#cond(node) ⇒ Object



345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
# File 'lib/ruby_parser_extras.rb', line 345

def cond node
  return nil if node.nil?
  node = value_expr node

  case node.sexp_type
  when :lit then
    if Regexp === node.last then
      s(:match, node)
    else
      node
    end
  when :and then
    _, lhs, rhs = node
    s(:and,  cond(lhs), cond(rhs))
  when :or then
    _, lhs, rhs = node
    s(:or,  cond(lhs), cond(rhs))
  when :dot2 then
    label = "flip#{node.hash}"
    env[label] = :lvar
    _, lhs, rhs = node
    s(:flip2, lhs, rhs) # TODO: recurse?
  when :dot3 then
    label = "flip#{node.hash}"
    env[label] = :lvar
    _, lhs, rhs = node
    s(:flip3, lhs, rhs)
  else
    node
  end.line node.line
end

#debug20(n, v = nil, r = nil) ⇒ Object



55
56
57
# File 'lib/ruby_parser_extras.rb', line 55

def debug20 n, v = nil, r = nil
  raise "not yet #{n} #{v.inspect} => #{r.inspect}" unless $good20[n]
end

#dedent(sexp) ⇒ Object



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
# File 'lib/ruby_parser_extras.rb', line 377

def dedent sexp
  dedent_count = dedent_size sexp

  skip_one = false
  sexp.map { |obj|
    case obj
    when Symbol then
      obj
    when String then
      obj.lines.map { |l| remove_whitespace_width l, dedent_count }.join
    when Sexp then
      case obj.sexp_type
      when :evstr then
        skip_one = true
        obj
      when :str then
        _, str = obj
        str = if skip_one then
                skip_one = false
                s1, *rest = str.lines
                s1 + rest.map { |l| remove_whitespace_width l, dedent_count }.join
              else
                str.lines.map { |l| remove_whitespace_width l, dedent_count }.join
              end

        s(:str, str).line obj.line
      else
        warn "unprocessed sexp %p" % [obj]
      end
    else
      warn "unprocessed: %p" % [obj]
    end
  }
end

#dedent_size(sexp) ⇒ Object



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
# File 'lib/ruby_parser_extras.rb', line 412

def dedent_size sexp
  skip_one = false
  sexp.flat_map { |s|
    case s
    when Symbol then
      next
    when String then
      s.lines
    when Sexp then
      case s.sexp_type
      when :evstr then
        skip_one = true
        next
      when :str then
        _, str = s
        lines = str.lines
        if skip_one then
          skip_one = false
          lines.shift
        end
        lines
      else
        warn "unprocessed sexp %p" % [s]
      end
    else
      warn "unprocessed: %p" % [s]
    end.map { |l| whitespace_width l[/^[ \t]*/] }
  }.compact.min
end

#dedent_string(string, width) ⇒ Object



442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
# File 'lib/ruby_parser_extras.rb', line 442

def dedent_string string, width
  characters_skipped = 0
  indentation_skipped = 0

  string.chars.each do |char|
    break if indentation_skipped >= width
    if char == " "
      characters_skipped += 1
      indentation_skipped += 1
    elsif char == "\t"
      proposed = TAB_WIDTH * (indentation_skipped / TAB_WIDTH + 1)
      break if proposed > width
      characters_skipped += 1
      indentation_skipped = proposed
    end
  end
  string[characters_skipped..-1]
end

#do_parseObject

for pure ruby systems only



69
70
71
# File 'lib/ruby_parser_extras.rb', line 69

def do_parse
  _racc_do_parse_rb(_racc_setup, false)
end

#gettable(id) ⇒ Object



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
# File 'lib/ruby_parser_extras.rb', line 461

def gettable(id)
  lineno = id.lineno if id.respond_to? :lineno
  id = id.to_sym if String === id

  result = case id.to_s
           when /^@@/ then
             s(:cvar, id)
           when /^@/ then
             s(:ivar, id)
           when /^\$/ then
             s(:gvar, id)
           when /^[A-Z]/ then
             s(:const, id)
           else
             type = env[id]
             if type then
               s(type, id)
             else
               new_call(nil, id)
             end
           end

  result.line lineno if lineno

  raise "identifier #{id.inspect} is not valid" unless result

  result
end

#hack_encoding(str, extra = nil) ⇒ Object



490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
# File 'lib/ruby_parser_extras.rb', line 490

def hack_encoding str, extra = nil
  encodings = ENCODING_ORDER.dup
  encodings.unshift(extra) unless extra.nil?

  # terrible, horrible, no good, very bad, last ditch effort.
  encodings.each do |enc|
    begin
      str.force_encoding enc
      if str.valid_encoding? then
        str.encode! Encoding::UTF_8
        break
      end
    rescue Encoding::InvalidByteSequenceError
      # do nothing
    rescue Encoding::UndefinedConversionError
      # do nothing
    end
  end

  # no amount of pain is enough for you.
  raise "Bad encoding. Need a magic encoding comment." unless
    str.encoding.name == "UTF-8"
end

#handle_encoding(str) ⇒ Object

Returns a UTF-8 encoded string after processing BOMs and magic encoding comments.

Holy crap… ok. Here goes:

Ruby’s file handling and encoding support is insane. We need to be able to lex a file. The lexer file is explicitly UTF-8 to make things cleaner. This allows us to deal with extended chars in class and method names. In order to do this, we need to encode all input source files as UTF-8. First, we look for a UTF-8 BOM by looking at the first line while forcing its encoding to ASCII-8BIT. If we find a BOM, we strip it and set the expected encoding to UTF-8. Then, we search for a magic encoding comment. If found, it overrides the BOM. Finally, we force the encoding of the input string to whatever was found, and then encode that to UTF-8 for compatibility with the lexer.



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
558
559
560
561
# File 'lib/ruby_parser_extras.rb', line 532

def handle_encoding str
  str = str.dup
  has_enc = str.respond_to? :encoding
  encoding = nil

  header = str.each_line.first(2)
  header.map! { |s| s.force_encoding "ASCII-8BIT" } if has_enc

  first = header.first || ""
  encoding, str = "utf-8", str[3..-1] if first =~ /\A\xEF\xBB\xBF/

  encoding = $1.strip if header.find { |s|
    s[/^#.*?-\*-.*?coding:\s*([^ ;]+).*?-\*-/, 1] ||
    s[/^#.*(?:en)?coding(?:\s*[:=])\s*([\w-]+)/, 1]
  }

  if encoding then
    if has_enc then
      encoding.sub!(/utf-8-.+$/, "utf-8") # HACK for stupid emacs formats
      hack_encoding str, encoding
    else
      warn "Skipping magic encoding comment"
    end
  else
    # nothing specified... ugh. try to encode as utf-8
    hack_encoding str if has_enc
  end

  str
end

#initialize(options = {}) ⇒ Object



109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
# File 'lib/ruby_parser_extras.rb', line 109

def initialize(options = {})
  super()

  v = self.class.name[/2\d/]
  raise "Bad Class name #{self.class}" unless v

  self.lexer = RubyLexer.new v && v.to_i
  self.lexer.parser = self
  self.in_kwarg = false

  @env = RubyParserStuff::Environment.new
  @comments = []

  @canonicalize_conditions = true

  self.reset
end

#invert_block_call(val) ⇒ Object



563
564
565
566
567
568
569
570
571
572
# File 'lib/ruby_parser_extras.rb', line 563

def invert_block_call val
  ret, iter = val
  type, call = ret

  iter.insert 1, call

  ret = s(type).line ret.line

  [iter, ret]
end

#inverted?(val) ⇒ Boolean

Returns:

  • (Boolean)


574
575
576
# File 'lib/ruby_parser_extras.rb', line 574

def inverted? val
  JUMP_TYPE[val[0].sexp_type]
end

#list_append(list, item) ⇒ Object

TODO: nuke me sigh



578
579
580
581
582
# File 'lib/ruby_parser_extras.rb', line 578

def list_append list, item # TODO: nuke me *sigh*
  return s(:array, item) unless list
  list = s(:array, list) unless Sexp === list && list.sexp_type == :array
  list << item
end

#list_prepend(item, list) ⇒ Object

TODO: nuke me sigh



584
585
586
587
588
# File 'lib/ruby_parser_extras.rb', line 584

def list_prepend item, list # TODO: nuke me *sigh*
  list = s(:array, list) unless Sexp === list && list.sexp_type == :array
  list.insert 1, item
  list
end

#literal_concat(head, tail) ⇒ Object

TODO: ugh. rewrite



590
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/ruby_parser_extras.rb', line 590

def literal_concat head, tail # TODO: ugh. rewrite
  return tail unless head
  return head unless tail

  htype, ttype = head.sexp_type, tail.sexp_type

  head = s(:dstr, "", head).line head.line if htype == :evstr

  case ttype
  when :str then
    if htype == :str
      head.last << tail.last
    elsif htype == :dstr and head.size == 2 then
      head.last << tail.last
    else
      head << tail
    end
  when :dstr then
    if htype == :str then
      lineno = head.line
      tail[1] = head.last + tail[1]
      head = tail
      head.line = lineno
    else
      tail.sexp_type = :array
      tail[1] = s(:str, tail[1]).line tail.line
      tail.delete_at 1 if tail[1] == s(:str, "")

      head.push(*tail.sexp_body)
    end
  when :evstr then
    if htype == :str then
      f, l = head.file, head.line
      head = s(:dstr, *head.sexp_body).line head.line
      head.file = f
      head.line = l
    end

    if head.size == 2 and tail.size > 1 and tail[1].sexp_type == :str then
      head.last << tail[1].last
      head.sexp_type = :str if head.size == 2 # HACK ?
    else
      head.push(tail)
    end
  else
    x = [head, tail]
    raise "unknown type: #{x.inspect}"
  end

  return head
end

#logical_op(type, left, right) ⇒ Object



642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
# File 'lib/ruby_parser_extras.rb', line 642

def logical_op type, left, right
  left = value_expr left

  if left and left.sexp_type == type and not left.paren then
    node, rhs = left, nil

    loop do
      _, _lhs, rhs = node
      break unless rhs && rhs.sexp_type == type and not rhs.paren
      node = rhs
    end

    node.pop
    node << s(type, rhs, right).line(rhs.line)

    return left
  end

  result = s(type, left, right)
  result.line left.line if left.line
  result
end

#new_aref(val) ⇒ Object



665
666
667
668
669
# File 'lib/ruby_parser_extras.rb', line 665

def new_aref val
  val[2] ||= s(:arglist)
  val[2].sexp_type = :arglist if val[2].sexp_type == :array # REFACTOR
  new_call val[0], :"[]", val[2]
end

#new_assign(lhs, rhs) ⇒ Object



671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
# File 'lib/ruby_parser_extras.rb', line 671

def new_assign lhs, rhs
  return nil unless lhs

  rhs = value_expr rhs

  case lhs.sexp_type
  when :lasgn, :iasgn, :cdecl, :cvdecl, :gasgn, :cvasgn, :attrasgn, :safe_attrasgn then
    lhs << rhs
  when :const then
    lhs.sexp_type = :cdecl
    lhs << rhs
  else
    raise "unknown lhs #{lhs.inspect} w/ #{rhs.inspect}"
  end

  lhs
end

#new_attrasgn(recv, meth, call_op = :".") ⇒ Object



689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
# File 'lib/ruby_parser_extras.rb', line 689

def new_attrasgn recv, meth, call_op = :"."
  meth = :"#{meth}="

  result = case call_op.to_sym
           when :"."
             s(:attrasgn, recv, meth)
           when :"&."
             s(:safe_attrasgn, recv, meth)
           else
             raise "unknown call operator: `#{type.inspect}`"
           end

  result.line = recv.line
  result
end

#new_body(val) ⇒ Object



705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
# File 'lib/ruby_parser_extras.rb', line 705

def new_body val
  body, resbody, elsebody, ensurebody = val

  result = body

  if resbody then
    result = s(:rescue)
    result << body if body

    res = resbody

    while res do
      result << res
      res = res.resbody(true)
    end

    result << elsebody if elsebody

    result.line = (body || resbody).line
  end

  if elsebody and not resbody then
    warning("else without rescue is useless")
    result = s(:begin, result).line result.line if result
    result = block_append(result, elsebody)
  end

  result = s(:ensure, result, ensurebody).compact.line result.line if ensurebody

  result
end

#new_brace_body(args, body, lineno) ⇒ Object



737
738
739
# File 'lib/ruby_parser_extras.rb', line 737

def new_brace_body args, body, lineno
  new_iter(nil, args, body).line lineno
end

#new_call(recv, meth, args = nil, call_op = :".") ⇒ Object



741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
# File 'lib/ruby_parser_extras.rb', line 741

def new_call recv, meth, args = nil, call_op = :"."
  result = case call_op.to_sym
           when :"."
             s(:call, recv, meth)
           when :"&."
             s(:safe_call, recv, meth)
           else
             raise "unknown call operator: `#{type.inspect}`"
           end

  # TODO: need a test with f(&b) to produce block_pass
  # TODO: need a test with f(&b) { } to produce warning

  if args
    if ARG_TYPES[args.sexp_type] then
      result.concat args.sexp_body
    else
      result << args
    end
  end

  # line = result.grep(Sexp).map(&:line).compact.min
  result.line = recv.line if recv
  result.line ||= lexer.lineno

  result
end

#new_case(expr, body, line) ⇒ Object



769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
# File 'lib/ruby_parser_extras.rb', line 769

def new_case expr, body, line
  result = s(:case, expr)

  while body and body.node_type == :when
    result << body
    body = body.delete_at 3
  end

  result[2..-1].each do |node|
    block = node.block(:delete)
    node.concat block.sexp_body if block
  end

  # else
  body = nil if body == s(:block)
  result << body

  result.line = line
  result
end

#new_class(val) ⇒ Object



790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
# File 'lib/ruby_parser_extras.rb', line 790

def new_class val
  line, path, superclass, body = val[1], val[2], val[3], val[5]

  result = s(:class, path, superclass)

  if body then
    if body.sexp_type == :block then
      result.push(*body.sexp_body)
    else
      result.push body
    end
  end

  result.line = line
  result.comments = self.comments.pop
  result
end

#new_compstmt(val) ⇒ Object



808
809
810
811
812
# File 'lib/ruby_parser_extras.rb', line 808

def new_compstmt val
  result = void_stmts(val.grep(Sexp)[0])
  result = remove_begin(result) if result
  result
end

#new_const_op_asgn(val) ⇒ Object



814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
# File 'lib/ruby_parser_extras.rb', line 814

def new_const_op_asgn val
  lhs, asgn_op, rhs = val[0], val[1].to_sym, val[2]

  result = case asgn_op
           when :"||" then
             s(:op_asgn_or, lhs, rhs)
           when :"&&" then
             s(:op_asgn_and, lhs, rhs)
           else
             s(:op_asgn, lhs, asgn_op, rhs)
           end

  result.line = lhs.line
  result
end

#new_defn(val) ⇒ Object



830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
# File 'lib/ruby_parser_extras.rb', line 830

def new_defn val
  (_, line), name, _, args, body, nil_body_line, * = val
  body ||= s(:nil).line nil_body_line

  args.line line

  result = s(:defn, name.to_sym, args).line line

  if body then
    if body.sexp_type == :block then
      result.push(*body.sexp_body)
    else
      result.push body
    end
  end

  result.comments = self.comments.pop

  result
end

#new_defs(val) ⇒ Object



851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
# File 'lib/ruby_parser_extras.rb', line 851

def new_defs val
  recv, (name, _line), args, body = val[1], val[4], val[6], val[7]
  line, _ = val[5]
  body ||= s(:nil).line line

  args.line line

  result = s(:defs, recv, name.to_sym, args)

  if body then
    if body.sexp_type == :block then
      result.push(*body.sexp_body)
    else
      result.push body
    end
  end

  result.line = recv.line
  result.comments = self.comments.pop
  result
end

#new_do_body(args, body, lineno) ⇒ Object



873
874
875
# File 'lib/ruby_parser_extras.rb', line 873

def new_do_body args, body, lineno
  new_iter(nil, args, body).line(lineno)
end

#new_for(expr, var, body) ⇒ Object



877
878
879
880
881
# File 'lib/ruby_parser_extras.rb', line 877

def new_for expr, var, body
  result = s(:for, expr, var).line(var.line)
  result << body if body
  result
end

#new_hash(val) ⇒ Object



883
884
885
# File 'lib/ruby_parser_extras.rb', line 883

def new_hash val
  s(:hash, *val[2].values).line(val[1])
end

#new_if(c, t, f) ⇒ Object



887
888
889
890
891
892
# File 'lib/ruby_parser_extras.rb', line 887

def new_if c, t, f
  l = [c.line, t && t.line, f && f.line].compact.min
  c = cond c
  c, t, f = c.last, f, t if c.sexp_type == :not and canonicalize_conditions
  s(:if, c, t, f).line(l)
end

#new_iter(call, args, body) ⇒ Object



894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
# File 'lib/ruby_parser_extras.rb', line 894

def new_iter call, args, body
  body ||= nil

  args ||= s(:args)
  args = s(:args, args) if Symbol === args

  result = s(:iter)
  result << call if call
  result << args
  result << body if body

  result.line call.line if call

  unless args == 0 then
    args.line call.line if call
    args.sexp_type = :args
  end

  result
end

#new_masgn(lhs, rhs, wrap = false) ⇒ Object



915
916
917
918
919
920
921
922
923
924
925
926
927
# File 'lib/ruby_parser_extras.rb', line 915

def new_masgn lhs, rhs, wrap = false
  _, ary = lhs

  line = rhs.line
  rhs = value_expr(rhs)
  rhs = ary ? s(:to_ary, rhs) : s(:array, rhs) if wrap
  rhs.line line if wrap

  lhs.delete_at 1 if ary.nil?
  lhs << rhs

  lhs
end

#new_masgn_arg(rhs, wrap = false) ⇒ Object



929
930
931
932
933
934
# File 'lib/ruby_parser_extras.rb', line 929

def new_masgn_arg rhs, wrap = false
  rhs = value_expr(rhs)
  # HACK: could be array if lhs isn't right
  rhs = s(:to_ary, rhs).line rhs.line if wrap
  rhs
end

#new_match(lhs, rhs) ⇒ Object



936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
# File 'lib/ruby_parser_extras.rb', line 936

def new_match lhs, rhs
  if lhs then
    case lhs.sexp_type
    when :dregx, :dregx_once then
      # TODO: no test coverage
      return s(:match2, lhs, rhs).line(lhs.line)
    when :lit then
      return s(:match2, lhs, rhs).line(lhs.line) if Regexp === lhs.last
    end
  end

  if rhs then
    case rhs.sexp_type
    when :dregx, :dregx_once then
      # TODO: no test coverage
      return s(:match3, rhs, lhs).line(lhs.line)
    when :lit then
      return s(:match3, rhs, lhs).line(lhs.line) if Regexp === rhs.last
    end
  end

  new_call(lhs, :"=~", argl(rhs)).line lhs.line
end

#new_module(val) ⇒ Object



960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
# File 'lib/ruby_parser_extras.rb', line 960

def new_module val
  line, path, body = val[1], val[2], val[4]

  result = s(:module, path)

  if body then # REFACTOR?
    if body.sexp_type == :block then
      result.push(*body.sexp_body)
    else
      result.push body
    end
  end

  result.line = line
  result.comments = self.comments.pop
  result
end

#new_op_asgn(val) ⇒ Object



978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
# File 'lib/ruby_parser_extras.rb', line 978

def new_op_asgn val
  lhs, asgn_op, arg = val[0], val[1].to_sym, val[2]
  name = gettable(lhs.value).line lhs.line
  arg = remove_begin(arg)
  result = case asgn_op # REFACTOR
           when :"||" then
             lhs << arg
             s(:op_asgn_or, name, lhs)
           when :"&&" then
             lhs << arg
             s(:op_asgn_and, name, lhs)
           else
             lhs << new_call(name, asgn_op, argl(arg))
             lhs
           end
  result.line = lhs.line
  result
end

#new_op_asgn1(val) ⇒ Object



997
998
999
1000
1001
1002
1003
1004
1005
# File 'lib/ruby_parser_extras.rb', line 997

def new_op_asgn1 val
  lhs, _, args, _, op, rhs = val

  args.sexp_type = :arglist if args

  result = s(:op_asgn1, lhs, args, op.to_sym, rhs)
  result.line lhs.line
  result
end

#new_op_asgn2(val) ⇒ Object



1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
# File 'lib/ruby_parser_extras.rb', line 1007

def new_op_asgn2 val
  recv, call_op, meth, op, arg = val
  meth = :"#{meth}="

  result = case call_op.to_sym
           when :"."
             s(:op_asgn2, recv, meth, op.to_sym, arg)
           when :"&."
             s(:safe_op_asgn2, recv, meth, op.to_sym, arg)
           else
             raise "unknown call operator: `#{type.inspect}`"
           end

  result.line = recv.line
  result
end

#new_qsym_listObject



1024
1025
1026
1027
1028
# File 'lib/ruby_parser_extras.rb', line 1024

def new_qsym_list
  result = s(:array).line lexer.lineno
  self.lexer.fixup_lineno
  result
end

#new_qsym_list_entry(val) ⇒ Object



1030
1031
1032
1033
1034
1035
# File 'lib/ruby_parser_extras.rb', line 1030

def new_qsym_list_entry val
  _, str, _ = val
  result = s(:lit, str.to_sym).line lexer.lineno
  self.lexer.fixup_lineno
  result
end

#new_qword_listObject



1037
1038
1039
1040
1041
# File 'lib/ruby_parser_extras.rb', line 1037

def new_qword_list
  result = s(:array).line lexer.lineno
  self.lexer.fixup_lineno
  result
end

#new_qword_list_entry(val) ⇒ Object



1043
1044
1045
1046
1047
1048
1049
# File 'lib/ruby_parser_extras.rb', line 1043

def new_qword_list_entry val
  _, str, _ = val
  str.force_encoding("ASCII-8BIT") unless str.valid_encoding?
  result = s(:str, str).line lexer.lineno # TODO: problematic? grab from parser
  self.lexer.fixup_lineno
  result
end

#new_regexp(val) ⇒ Object



1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
# File 'lib/ruby_parser_extras.rb', line 1051

def new_regexp val
  _, node, options = val

  node ||= s(:str, "").line lexer.lineno

  o, k = 0, nil
  options.split(//).uniq.each do |c| # FIX: this has a better home
    v = {
      "x" => Regexp::EXTENDED,
      "i" => Regexp::IGNORECASE,
      "m" => Regexp::MULTILINE,
      "o" => Regexp::ONCE,
      "n" => Regexp::ENC_NONE,
      "e" => Regexp::ENC_EUC,
      "s" => Regexp::ENC_SJIS,
      "u" => Regexp::ENC_UTF8,
    }[c]
    raise "unknown regexp option: #{c}" unless v
    o += v
  end

  case node.sexp_type
  when :str then
    node.sexp_type = :lit
    node[1] = if k then
                Regexp.new(node[1], o, k)
              else
                begin
                  Regexp.new(node[1], o)
                rescue RegexpError => e
                  warn "WA\RNING: #{e.message} for #{node[1].inspect} #{options.inspect}"
                  begin
                    warn "WA\RNING: trying to recover with ENC_UTF8"
                    Regexp.new(node[1], Regexp::ENC_UTF8)
                  rescue RegexpError => e
                    warn "WA\RNING: trying to recover with ENC_NONE"
                    Regexp.new(node[1], Regexp::ENC_NONE)
                  end
                end
              end
  when :dstr then
    if options =~ /o/ then
      node.sexp_type = :dregx_once
    else
      node.sexp_type = :dregx
    end
    node << o if o and o != 0
  else
    node = s(:dregx, "", node).line node.line
    node.sexp_type = :dregx_once if options =~ /o/
    node << o if o and o != 0
  end

  node
end

#new_resbody(cond, body) ⇒ Object



1107
1108
1109
1110
1111
1112
1113
1114
1115
# File 'lib/ruby_parser_extras.rb', line 1107

def new_resbody cond, body
  if body && body.sexp_type == :block then
    body.shift # remove block and splat it in directly
  else
    body = [body]
  end

  s(:resbody, cond, *body).line cond.line
end

#new_rescue(body, resbody) ⇒ Object



1117
1118
1119
# File 'lib/ruby_parser_extras.rb', line 1117

def new_rescue body, resbody
  s(:rescue, body, resbody).line body.line
end

#new_sclass(val) ⇒ Object



1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
# File 'lib/ruby_parser_extras.rb', line 1121

def new_sclass val
  recv, in_def, in_single, body = val[3], val[4], val[6], val[7]

  result = s(:sclass, recv)

  if body then
    if body.sexp_type == :block then
      result.push(*body.sexp_body)
    else
      result.push body
    end
  end

  result.line = val[2]
  self.in_def = in_def
  self.in_single = in_single
  result
end

#new_string(val) ⇒ Object



1140
1141
1142
1143
1144
1145
1146
1147
# File 'lib/ruby_parser_extras.rb', line 1140

def new_string val
  str, = val
  str.force_encoding("UTF-8")
  str.force_encoding("ASCII-8BIT") unless str.valid_encoding?
  result = s(:str, str).line lexer.lineno
  self.lexer.fixup_lineno str.count("\n")
  result
end

#new_super(args) ⇒ Object



1149
1150
1151
1152
1153
1154
1155
1156
# File 'lib/ruby_parser_extras.rb', line 1149

def new_super args
  if args && args.node_type == :block_pass then
    s(:super, args).line args.line
  else
    args ||= s(:arglist).line lexer.lineno
    s(:super, *args.sexp_body).line args.line
  end
end

#new_symbol_listObject



1158
1159
1160
1161
1162
# File 'lib/ruby_parser_extras.rb', line 1158

def new_symbol_list
  result = s(:array).line lexer.lineno
  self.lexer.fixup_lineno
  result
end

#new_symbol_list_entry(val) ⇒ Object



1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
# File 'lib/ruby_parser_extras.rb', line 1164

def new_symbol_list_entry val
  _, sym, _ = val

  sym ||= s(:str, "")

  line = lexer.lineno

  case sym.sexp_type
  when :dstr then
    sym.sexp_type = :dsym
  when :str then
    sym = s(:lit, sym.last.to_sym)
  else
    sym = s(:dsym, "", sym || s(:str, "").line(line))
  end

  sym.line line

  self.lexer.fixup_lineno

  sym
end

#new_undef(n, m = nil) ⇒ Object



1187
1188
1189
1190
1191
1192
1193
# File 'lib/ruby_parser_extras.rb', line 1187

def new_undef n, m = nil
  if m then
    block_append(n, s(:undef, m).line(m.line))
  else
    s(:undef, n).line n.line
  end
end

#new_until(block, expr, pre) ⇒ Object



1195
1196
1197
# File 'lib/ruby_parser_extras.rb', line 1195

def new_until block, expr, pre
  new_until_or_while :until, block, expr, pre
end

#new_until_or_while(type, block, expr, pre) ⇒ Object



1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
# File 'lib/ruby_parser_extras.rb', line 1199

def new_until_or_while type, block, expr, pre
  other = type == :until ? :while : :until
  line = [block && block.line, expr.line].compact.min
  block, pre = block.last, false if block && block.sexp_type == :begin

  expr = cond expr

  result = unless expr.sexp_type == :not and canonicalize_conditions then
             s(type,  expr,      block, pre)
           else
             s(other, expr.last, block, pre)
           end

  result.line = line
  result
end

#new_when(cond, body) ⇒ Object



1216
1217
1218
# File 'lib/ruby_parser_extras.rb', line 1216

def new_when cond, body
  s(:when, cond, body)
end

#new_while(block, expr, pre) ⇒ Object



1220
1221
1222
# File 'lib/ruby_parser_extras.rb', line 1220

def new_while block, expr, pre
  new_until_or_while :while, block, expr, pre
end

#new_word_listObject



1224
1225
1226
1227
1228
# File 'lib/ruby_parser_extras.rb', line 1224

def new_word_list
  result = s(:array).line lexer.lineno
  self.lexer.fixup_lineno
  result
end

#new_word_list_entry(val) ⇒ Object



1230
1231
1232
1233
1234
1235
# File 'lib/ruby_parser_extras.rb', line 1230

def new_word_list_entry val
  _, word, _ = val
  result = word.sexp_type == :evstr ? s(:dstr, "", word).line(word.line) : word
  self.lexer.fixup_lineno
  result
end

#new_xstring(str) ⇒ Object



1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
# File 'lib/ruby_parser_extras.rb', line 1237

def new_xstring str
  if str then
    case str.sexp_type
    when :str
      str.sexp_type = :xstr
    when :dstr
      str.sexp_type = :dxstr
    else
      str = s(:dxstr, "", str).line str.line
    end
    str
  else
    s(:xstr, "")
  end
end

#new_yield(args = nil) ⇒ Object

Raises:

  • (SyntaxError)


1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
# File 'lib/ruby_parser_extras.rb', line 1253

def new_yield args = nil
  # TODO: raise args.inspect unless [:arglist].include? args.first # HACK
  raise "write a test 4" if args && args.node_type == :block_pass
  raise SyntaxError, "Block argument should not be given." if
    args && args.node_type == :block_pass

  args ||= s(:arglist).line lexer.lineno

  args.sexp_type = :arglist if [:call_args, :array].include? args.sexp_type
  args = s(:arglist, args).line args.line unless args.sexp_type == :arglist

  s(:yield, *args.sexp_body).line args.line
end

#next_tokenObject



1267
1268
1269
1270
1271
1272
1273
1274
1275
# File 'lib/ruby_parser_extras.rb', line 1267

def next_token
  token = self.lexer.next_token

  if token and token.first != RubyLexer::EOF then
    return token
  else
    return [false, false]
  end
end

#on_error(et, ev, values) ⇒ Object



1277
1278
1279
1280
1281
1282
1283
1284
# File 'lib/ruby_parser_extras.rb', line 1277

def on_error(et, ev, values)
  super
rescue Racc::ParseError => e
  # I don't like how the exception obscures the error message
  e.message.replace "%s:%p :: %s" % [self.file, lexer.lineno, e.message.strip]
  warn e.message if $DEBUG
  raise
end

#process(str, file = "(string)", time = 10) ⇒ Object Also known as: parse

Parse str at path file and return a sexp. Raises Timeout::Error if it runs for more than time seconds.



1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
# File 'lib/ruby_parser_extras.rb', line 1290

def process(str, file = "(string)", time = 10)
  Timeout.timeout time do
    raise "bad val: #{str.inspect}" unless String === str

    str = handle_encoding str

    self.file = file.dup

    @yydebug = ENV.has_key? "DEBUG"

    # HACK -- need to get tests passing more than have graceful code
    self.lexer.ss = RPStringScanner.new str

    do_parse
  end
end

#remove_begin(node) ⇒ Object Also known as: value_expr



1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
# File 'lib/ruby_parser_extras.rb', line 1309

def remove_begin node
  line = node.line

  node = node.last while node and node.sexp_type == :begin and node.size == 2

  node = s(:nil) if node == s(:begin)

  node.line ||= line

  node
end

#resetObject



1323
1324
1325
1326
1327
1328
1329
# File 'lib/ruby_parser_extras.rb', line 1323

def reset
  lexer.reset
  self.in_def = false
  self.in_single = 0
  self.env.reset
  self.comments.clear
end

#ret_args(node) ⇒ Object



1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
# File 'lib/ruby_parser_extras.rb', line 1331

def ret_args node
  if node then
    raise "write a test 5" if node.sexp_type == :block_pass

    raise SyntaxError, "block argument should not be given" if
      node.sexp_type == :block_pass

    node.sexp_type = :array if node.sexp_type == :call_args
    node = node.last if node.sexp_type == :array && node.size == 2

    # HACK matz wraps ONE of the FOUR splats in a newline to
    # distinguish. I use paren for now. ugh
    node = s(:svalue, node).line node.line if node.sexp_type == :splat and not node.paren
    node.sexp_type = :svalue if node.sexp_type == :arglist && node[1].sexp_type == :splat
  end

  node
end

#s(*args) ⇒ Object



1350
1351
1352
1353
1354
1355
# File 'lib/ruby_parser_extras.rb', line 1350

def s(*args)
  result = Sexp.new(*args)
  # result.line ||= lexer.lineno if lexer.ss unless ENV["CHECK_LINE_NUMS"] # otherwise...
  result.file = self.file
  result
end

#syntax_error(msg) ⇒ Object Also known as: yyerror



1357
1358
1359
# File 'lib/ruby_parser_extras.rb', line 1357

def syntax_error msg
  raise RubyParser::SyntaxError, msg
end

#void_stmts(node) ⇒ Object



1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
# File 'lib/ruby_parser_extras.rb', line 1363

def void_stmts node
  return nil unless node
  return node unless node.sexp_type == :block

  if node.respond_to? :sexp_body= then
    node.sexp_body = node.sexp_body.map { |n| remove_begin n }
  else
    node[1..-1] = node[1..-1].map { |n| remove_begin(n) }
  end

  node
end

#warning(s) ⇒ Object



1376
1377
1378
# File 'lib/ruby_parser_extras.rb', line 1376

def warning s
  # do nothing for now
end

#whitespace_width(line, remove_width = nil) ⇒ Object Also known as: remove_whitespace_width



1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
# File 'lib/ruby_parser_extras.rb', line 1380

def whitespace_width line, remove_width = nil
  col = 0
  idx = 0

  line.chars.each do |c|
    break if remove_width && col >= remove_width
    case c
    when " " then
      col += 1
    when "\t" then
      n = TAB_WIDTH * (col / TAB_WIDTH + 1)
      break if remove_width && n > remove_width
      col = n
    else
      break
    end
    idx += 1
  end

  if remove_width then
    line[idx..-1]
  else
    col
  end
end