Class: Puppet::Pops::Parser::Lexer

Inherits:
Object
  • Object
show all
Extended by:
Forwardable
Defined in:
lib/puppet/pops/parser/lexer.rb

Defined Under Namespace

Modules: Contextual Classes: Locator, Token, TokenList

Constant Summary collapse

TOKENS = 
TokenList.new
DQ_initial_token_types = 
{'$' => :DQPRE,'"' => :STRING}
DQ_continuation_token_types = 
{'$' => :DQMID,'"' => :DQPOST}
KEYWORDS = 
TokenList.new
@@pairs = 
{
  "{"   => "}",
  "("   => ")",
  "["   => "]",
  "<|"  => "|>",
  "<<|" => "|>>",
  "|"   => "|"
}

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize ⇒ Lexer

Returns a new instance of Lexer.



464
465
466
467
 
# File 'lib/puppet/pops/parser/lexer.rb', line 464

def initialize
  @multibyte = init_multibyte
  initvars
end

Instance Attribute Details

#file ⇒ Object 



15
16
17
 
# File 'lib/puppet/pops/parser/lexer.rb', line 15

def file
  @file
end

#indefine ⇒ Object Also known as: indefine? 



19
20
21
 
# File 'lib/puppet/pops/parser/lexer.rb', line 19

def indefine
  @indefine
end

#lexing_context ⇒ Object (readonly) 



15
16
17
 
# File 'lib/puppet/pops/parser/lexer.rb', line 15

def lexing_context
  @lexing_context
end

#locator ⇒ Object (readonly) 



17
18
19
 
# File 'lib/puppet/pops/parser/lexer.rb', line 17

def locator
  @locator
end

#token_queue ⇒ Object (readonly) 



15
16
17
 
# File 'lib/puppet/pops/parser/lexer.rb', line 15

def token_queue
  @token_queue
end

Instance Method Details

#assert_numeric(value) ⇒ Object 



469
470
471
472
473
474
475
476
477
 
# File 'lib/puppet/pops/parser/lexer.rb', line 469

def assert_numeric(value)
  if value =~ /^0[xX].*$/
    lex_error (positioned_message("Not a valid hex number #{value}")) unless value =~ /^0[xX][0-9A-Fa-f]+$/
  elsif value =~ /^0[^.].*$/
    lex_error(positioned_message("Not a valid octal number #{value}")) unless value =~ /^0[0-7]+$/
  else
    lex_error(positioned_message("Not a valid decimal number #{value}")) unless value =~ /0?\d+(?:\.\d+)?(?:[eE]-?\d+)?/
  end
end

#clear ⇒ Object 



396
397
398
 
# File 'lib/puppet/pops/parser/lexer.rb', line 396

def clear
  initvars
end

#expected ⇒ Object 



400
401
402
403
404
 
# File 'lib/puppet/pops/parser/lexer.rb', line 400

def expected
  return nil if @expected.empty?
  name = @expected[-1]
  TOKENS.lookup(name) or lex_error "Internal Lexer Error: Could not find expected token #{name}"
end

#find_regex_token ⇒ Object

Find the next token that matches a regex. We look for these first.



440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
 
# File 'lib/puppet/pops/parser/lexer.rb', line 440

def find_regex_token
  best_token = nil
  best_length = 0

  # I tried optimizing based on the first char, but it had
  # a slightly negative affect and was a good bit more complicated.
  TOKENS.regex_tokens.each do |token|
    if length = @scanner.match?(token.regex) and token.acceptable?(lexing_context)
      # We've found a longer match
      if length > best_length
        best_length = length
        best_token = token
      end
    end
  end

  return best_token, @scanner.scan(best_token.regex) if best_token
end

#find_string_token ⇒ Object 



428
429
430
431
432
433
434
435
436
437
 
# File 'lib/puppet/pops/parser/lexer.rb', line 428

def find_string_token
  # We know our longest string token is three chars, so try each size in turn
  # until we either match or run out of chars.  This way our worst-case is three
  # tries, where it is otherwise the number of string token we have.  Also,
  # the lookups are optimized hash lookups, instead of regex scans.
  #
  s = @scanner.peek(3)
  token = TOKENS.lookup(s[0,3]) || TOKENS.lookup(s[0,2]) || TOKENS.lookup(s[0,1])
  [ token, token && @scanner.scan(token.regex) ]
end

#find_token ⇒ Object

Find the next token, returning the string and the token.



460
461
462
 
# File 'lib/puppet/pops/parser/lexer.rb', line 460

def find_token
  shift_token || find_regex_token || find_string_token
end

#followed_by ⇒ Object

Returns “<eof>” if at end of input, else the following 5 characters with n r t escaped



699
700
701
702
703
704
705
706
 
# File 'lib/puppet/pops/parser/lexer.rb', line 699

def followed_by
  return "<eof>" if @scanner.eos?
  result = @scanner.rest[0,5] + "..."
  result.gsub!("\t", '\t')
  result.gsub!("\n", '\n')
  result.gsub!("\r", '\r')
  result
end

#format_quote(q) ⇒ Object 



708
709
710
711
712
713
714
 
# File 'lib/puppet/pops/parser/lexer.rb', line 708

def format_quote q
  if q == "'"
    '"\'"'
  else
    "'#{q}'"
  end
end

#fullscan ⇒ Object

scan the whole file basically just used for testing



408
409
410
411
412
413
414
415
416
417
 
# File 'lib/puppet/pops/parser/lexer.rb', line 408

def fullscan
  array = []

  self.scan { |token, str|
    # Ignore any definition nesting problems
    @indefine = false
    array.push([token,str])
  }
  array
end

#init_multibyte ⇒ Object

Returns true if ruby version >= 1.9.3 since regexp supports multi-byte matches and expanded character categories like [[:blank:]].

This implementation will fail if there are more than 255 minor or micro versions of ruby



484
485
486
487
488
 
# File 'lib/puppet/pops/parser/lexer.rb', line 484

def init_multibyte
  numver = RUBY_VERSION.split(".").collect {|s| s.to_i }
  return true if (numver[0] << 16 | numver[1] << 8 | numver[2]) >= (1 << 16 | 9 << 8 | 3)
  false
end

#initvars ⇒ Object 



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
 
# File 'lib/puppet/pops/parser/lexer.rb', line 494

def initvars
  @previous_token = nil
  @scanner = nil
  @file = nil

  # AAARRGGGG! okay, regexes in ruby are bloody annoying
  # no one else has "\n" =~ /\s/

  if multibyte?
    # Skip all kinds of space, and CR, but not newlines
    @skip = %r{[[:blank:]\r]+}
  else
    @skip = %r{[ \t\r]+}
  end

  @namestack = []
  @token_queue = []
  @indefine = false
  @expected = []
  @lexing_context = {
    :after => nil,
    :start_of_line => true,
    :offset => 0,      # byte offset before where token starts
    :end_offset => 0,  # byte offset after scanned token
    :brace_count => 0,  # nested depth of braces
    :interpolation_stack => []   # matching interpolation brace level
  }
end

#lex_error(msg) ⇒ Object

Raises:



22
23
24
 
# File 'lib/puppet/pops/parser/lexer.rb', line 22

def lex_error msg
  raise Puppet::LexError.new(msg)
end

#line ⇒ Object

Returns the line number (starting from 1) for the current position in the scanned text (at the end of the last produced, but not necessarily consumed.



782
783
784
785
 
# File 'lib/puppet/pops/parser/lexer.rb', line 782

def line
  return 1 unless lexing_context && locator
  locator.line_for_offset(lexing_context[:end_offset])
end

#match?(r) ⇒ Boolean

Returns:

  • (Boolean) 


656
657
658
 
# File 'lib/puppet/pops/parser/lexer.rb', line 656

def match? r
  @scanner.match?(r)
end

#multibyte? ⇒ Boolean

Returns:

  • (Boolean) 


490
491
492
 
# File 'lib/puppet/pops/parser/lexer.rb', line 490

def multibyte?
  @multibyte
end

#munge_token(token, value) ⇒ Object

Make any necessary changes to the token and/or value.



524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
 
# File 'lib/puppet/pops/parser/lexer.rb', line 524

def munge_token(token, value)
  # A token may already have been munged (converted and positioned)
  #
  return token, value if value.is_a? Hash

  skip if token.skip_text

  return if token.skip

  token, value = token.convert(self, value) if token.respond_to?(:convert)

  return unless token

  return if token.skip

  # If the conversion performed the munging/positioning
  return token, value if value.is_a? Hash

  pos_hash = position_in_source
  pos_hash[:value] = value

  # Add one to pos, first char on line is 1
  return token, pos_hash
end

#namespace ⇒ Object

Collect the current namespace.



573
574
575
 
# File 'lib/puppet/pops/parser/lexer.rb', line 573

def namespace
  @namestack.join("::")
end

#pos ⇒ Object 



560
561
562
 
# File 'lib/puppet/pops/parser/lexer.rb', line 560

def pos
  @locator.pos_on_line(lexing_context[:offset])
end

#position_in_source ⇒ Object

Returns a hash with the current position in source based on the current lexing context



551
552
553
554
555
556
557
558
 
# File 'lib/puppet/pops/parser/lexer.rb', line 551

def position_in_source
  pos        = @locator.pos_on_line(lexing_context[:offset])
  offset     = @locator.char_offset(lexing_context[:offset])
  length     = @locator.char_length(lexing_context[:offset], lexing_context[:end_offset])
  start_line = @locator.line_for_offset(lexing_context[:offset])

  return { :line => start_line, :pos => pos, :offset => offset, :length => length}
end

#positioned_message(msg) ⇒ Object

Formats given message by appending file, line and position if available.



691
692
693
694
695
696
 
# File 'lib/puppet/pops/parser/lexer.rb', line 691

def positioned_message msg
  result = [msg]
  result << "in file #{file}" if file
  result << "at line #{line}:#{pos}" if line
  result.join(" ")
end

#replace_false_start_with_text(appendix) ⇒ Object 



756
757
758
759
760
761
762
763
764
 
# File 'lib/puppet/pops/parser/lexer.rb', line 756

def replace_false_start_with_text(appendix)
  last_token = token_queue.pop
  value = last_token.last
  if value.is_a? Hash
    value[:value] + appendix
  else
    value + appendix
  end
end

#scan {|[false,false]| ... } ⇒ Object

this is the heart of the lexer

Yields:

  • ([false,false]) 


579
580
581
582
583
584
585
586
587
588
589
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
641
642
643
644
645
646
647
648
649
 
# File 'lib/puppet/pops/parser/lexer.rb', line 579

def scan
  #Puppet.debug("entering scan")
  lex_error "Internal Error: No string or file given to lexer to process." unless @scanner

  # Skip any initial whitespace.
  skip

  until token_queue.empty? and @scanner.eos? do
    offset = @scanner.pos
    matched_token, value = find_token
    end_offset = @scanner.pos

    # error out if we didn't match anything at all
    lex_error "Could not match #{@scanner.rest[/^(\S+|\s+|.*)/]}" unless matched_token

    newline = matched_token.name == :RETURN

    lexing_context[:start_of_line] = newline
    lexing_context[:offset] = offset
    lexing_context[:end_offset] = end_offset

    final_token, token_value = munge_token(matched_token, value)
    # update end position since munging may have moved the end offset
    lexing_context[:end_offset] = @scanner.pos

    unless final_token
      skip
      next
    end

    lexing_context[:after] = final_token.name unless newline
    if final_token.name == :DQPRE
      lexing_context[:interpolation_stack] << lexing_context[:brace_count]
    elsif final_token.name == :DQPOST
      lexing_context[:interpolation_stack].pop
    end

    value = token_value[:value]

    if match = @@pairs[value] and final_token.name != :DQUOTE and final_token.name != :SQUOTE
      @expected << match
    elsif exp = @expected[-1] and exp == value and final_token.name != :DQUOTE and final_token.name != :SQUOTE
      @expected.pop
    end

    yield [final_token.name, token_value]

    if @previous_token
      namestack(value) if @previous_token.name == :CLASS and value != '{'

      if @previous_token.name == :DEFINE
        if indefine?
          msg = "Cannot nest definition #{value} inside #{@indefine}"
          self.indefine = false
          raise Puppet::ParseError, msg
        end

        @indefine = value
      end
    end
    @previous_token = final_token
    skip
  end
  # Cannot reset @scanner to nil here - it is needed to answer questions about context after
  # completed parsing.
  # Seems meaningless to do this. Everything will be gc anyway.
  #@scanner = nil

  # This indicates that we're done parsing.
  yield [false,false]
end

#skip ⇒ Object

Skip any skipchars in our remaining string.



652
653
654
 
# File 'lib/puppet/pops/parser/lexer.rb', line 652

def skip
  @scanner.skip(@skip)
end

#slurpstring(terminators, escapes = %w{ \\ $ ' " r n t s }+["\n"], ignore_invalid_escapes = false) ⇒ Object

we’ve encountered the start of a string… slurp in the rest of the string and return it



666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
 
# File 'lib/puppet/pops/parser/lexer.rb', line 666

def slurpstring(terminators,escapes=%w{ \\  $ ' " r n t s }+["\n"],ignore_invalid_escapes=false)
  # we search for the next quote that isn't preceded by a
  # backslash; the caret is there to match empty strings
  last = @scanner.matched
  str = @scanner.scan_until(/([^\\]|^|[^\\])([\\]{2})*[#{terminators}]/) || lex_error(positioned_message("Unclosed quote after #{format_quote(last)} followed by '#{followed_by}'"))
  str.gsub!(/\\(.)/m) {
    ch = $1
    if escapes.include? ch
      case ch
      when 'r'; "\r"
      when 'n'; "\n"
      when 't'; "\t"
      when 's'; " "
      when "\n"; ''
      else      ch
      end
    else
      Puppet.warning(positioned_message("Unrecognized escape sequence '\\#{ch}'")) unless ignore_invalid_escapes
      "\\#{ch}"
    end
  }
  [ str[0..-2],str[-1,1] ]
end

#string=(string) ⇒ Object

just parse a string, not a whole file



767
768
769
770
 
# File 'lib/puppet/pops/parser/lexer.rb', line 767

def string=(string)
  @scanner = StringScanner.new(string)
  @locator = Locator.new(string, multibyte?)
end

#tokenize_interpolated_string(token_type, preamble = '') ⇒ Object 



716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
 
# File 'lib/puppet/pops/parser/lexer.rb', line 716

def tokenize_interpolated_string(token_type,preamble='')
  # Expecting a (possibly empty) stretch of text terminated by end of string ", a variable $, or expression ${
  # The length of this part includes the start and terminating characters.
  value,terminator = slurpstring('"$')

  # Advanced after '{' if this is in expression ${} interpolation
  braced = terminator == '$' && @scanner.scan(/\{/)
  # make offset to end_ofset be the length of the pre expression string including its start and terminating chars
  lexing_context[:end_offset] = @scanner.pos

  token_queue << [TOKENS[token_type[terminator]],position_in_source().merge!({:value => preamble+value})]
  variable_regex = if Puppet[:allow_variables_with_dashes]
    TOKENS[:VARIABLE_WITH_DASH].regex
  else
    TOKENS[:VARIABLE].regex
  end
  if terminator != '$' or braced
    return token_queue.shift
  end

  tmp_offset = @scanner.pos
  if var_name = @scanner.scan(variable_regex)
    lexing_context[:offset] = tmp_offset
    lexing_context[:end_offset] = @scanner.pos
    warn_if_variable_has_hyphen(var_name)
    # If the varname after ${ is followed by (, it is a function call, and not a variable
    # reference.
    #
    if braced && @scanner.match?(%r{[ \t\r]*\(})
      token_queue << [TOKENS[:NAME], position_in_source().merge!({:value=>var_name})]
    else
      token_queue << [TOKENS[:VARIABLE],position_in_source().merge!({:value=>var_name})]
    end
    lexing_context[:offset] = @scanner.pos
    tokenize_interpolated_string(DQ_continuation_token_types)
  else
    tokenize_interpolated_string(token_type, replace_false_start_with_text(terminator))
  end
end

#warn_if_variable_has_hyphen(var_name) ⇒ Object 



772
773
774
775
776
 
# File 'lib/puppet/pops/parser/lexer.rb', line 772

def warn_if_variable_has_hyphen(var_name)
  if var_name.include?('-')
    Puppet.deprecation_warning("Using `-` in variable names is deprecated at #{file || '<string>'}:#{line}. See http://links.puppetlabs.com/puppet-hyphenated-variable-deprecation")
  end
end