Class: Puppet::Parser::Lexer

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

Defined Under Namespace

Modules: Contextual Classes: 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.



392
393
394
 
# File 'lib/puppet/parser/lexer.rb', line 392

def initialize
  initvars
end

Instance Attribute Details

#file ⇒ Object 



18
19
20
 
# File 'lib/puppet/parser/lexer.rb', line 18

def file
  @file
end

#indefine ⇒ Object Also known as: indefine? 



20
21
22
 
# File 'lib/puppet/parser/lexer.rb', line 20

def indefine
  @indefine
end

#last ⇒ Object (readonly) 



18
19
20
 
# File 'lib/puppet/parser/lexer.rb', line 18

def last
  @last
end

#lexing_context ⇒ Object (readonly) 



18
19
20
 
# File 'lib/puppet/parser/lexer.rb', line 18

def lexing_context
  @lexing_context
end

#line ⇒ Object 



20
21
22
 
# File 'lib/puppet/parser/lexer.rb', line 20

def line
  @line
end

#token_queue ⇒ Object (readonly) 



18
19
20
 
# File 'lib/puppet/parser/lexer.rb', line 18

def token_queue
  @token_queue
end

Instance Method Details

#clear ⇒ Object 



324
325
326
 
# File 'lib/puppet/parser/lexer.rb', line 324

def clear
  initvars
end

#commentpop ⇒ Object

returns the content of the currently accumulated content cache



585
586
587
 
# File 'lib/puppet/parser/lexer.rb', line 585

def commentpop
  @commentstack.pop[0]
end

#commentpush ⇒ Object 



599
600
601
 
# File 'lib/puppet/parser/lexer.rb', line 599

def commentpush
  @commentstack.push(['', @line])
end

#expected ⇒ Object 



328
329
330
331
332
 
# File 'lib/puppet/parser/lexer.rb', line 328

def expected
  return nil if @expected.empty?
  name = @expected[-1]
  TOKENS.lookup(name) or lex_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.



368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
 
# File 'lib/puppet/parser/lexer.rb', line 368

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 



356
357
358
359
360
361
362
363
364
365
 
# File 'lib/puppet/parser/lexer.rb', line 356

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.



388
389
390
 
# File 'lib/puppet/parser/lexer.rb', line 388

def find_token
  shift_token || find_regex_token || find_string_token
end

#fullscan ⇒ Object

scan the whole file basically just used for testing



336
337
338
339
340
341
342
343
344
345
 
# File 'lib/puppet/parser/lexer.rb', line 336

def fullscan
  array = []

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

#getcomment(line = nil) ⇒ Object 



589
590
591
592
593
594
595
596
597
 
# File 'lib/puppet/parser/lexer.rb', line 589

def getcomment(line = nil)
  comment = @commentstack.last
  if line.nil? or comment[1] <= line
    @commentstack.pop
    @commentstack.push(['', @line])
    return comment[0]
  end
  ''
end

#initvars ⇒ Object 



396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
 
# File 'lib/puppet/parser/lexer.rb', line 396

def initvars
  @line = 1
  @previous_token = nil
  @scanner = nil
  @file = nil
  # AAARRGGGG! okay, regexes in ruby are bloody annoying
  # no one else has "\n" =~ /\s/
  @skip = %r{[ \t\r]+}

  @namestack = []
  @token_queue = []
  @indefine = false
  @expected = []
  @commentstack = [ ['', @line] ]
  @lexing_context = {
    :after => nil,
    :start_of_line => true,
    :string_interpolation_depth => 0
    }
end

#lex_error(msg) ⇒ Object

Raises:



32
33
34
 
# File 'lib/puppet/parser/lexer.rb', line 32

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

#munge_token(token, value) ⇒ Object

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



418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
 
# File 'lib/puppet/parser/lexer.rb', line 418

def munge_token(token, value)
  @line += 1 if token.incr_line

  skip if token.skip_text

  return if token.skip and not token.accumulate?

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

  return unless token

  if token.accumulate?
    comment = @commentstack.pop
    comment[0] << value + "\n"
    @commentstack.push(comment)
  end

  return if token.skip

  return token, { :value => value, :line => @line }
end

#namespace ⇒ Object

Collect the current namespace.



448
449
450
 
# File 'lib/puppet/parser/lexer.rb', line 448

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

#pos ⇒ Object

Returns the position on the line. This implementation always returns nil. It is here for API reasons in Puppet::Error which needs to support both –parser current, and –parser future.



27
28
29
30
 
# File 'lib/puppet/parser/lexer.rb', line 27

def pos
  # Make the lexer comply with newer API. It does not produce a pos...
  nil
end

#scan {|[false,false]| ... } ⇒ Object

this is the heart of the lexer

Yields:

  • ([false,false]) 


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
 
# File 'lib/puppet/parser/lexer.rb', line 455

def scan
  #Puppet.debug("entering scan")
  lex_error "Invalid or empty string" unless @scanner

  # Skip any initial whitespace.
  skip

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

    # 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

    # this matches a blank line; eat the previously accumulated comments
    getcomment if lexing_context[:start_of_line] and newline
    lexing_context[:start_of_line] = newline

    final_token, token_value = munge_token(matched_token, value)

    unless final_token
      skip
      next
    end

    final_token_name = final_token.name
    lexing_context[:after]         = final_token_name unless newline
    lexing_context[:string_interpolation_depth] += 1 if final_token_name == :DQPRE
    lexing_context[:string_interpolation_depth] -= 1 if final_token_name == :DQPOST

    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

    if final_token_name == :LBRACE or final_token_name == :LPAREN
      commentpush
    end
    if final_token_name == :RPAREN
      commentpop
    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
  @scanner = nil

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

#skip ⇒ Object

Skip any skipchars in our remaining string.



526
527
528
 
# File 'lib/puppet/parser/lexer.rb', line 526

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



536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
 
# File 'lib/puppet/parser/lexer.rb', line 536

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
  str = @scanner.scan_until(/([^\\]|^|[^\\])([\\]{2})*[#{terminators}]/) or lex_error "Unclosed quote after '#{last}' in '#{rest}'"
  @line += str.count("\n") # literal carriage returns add to the line count.
  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 "Unrecognised escape sequence '\\#{ch}'#{file && " in file #{file}"}#{line && " at line #{line}"}" unless ignore_invalid_escapes
      "\\#{ch}"
    end
  }
  [ str[0..-2],str[-1,1] ]
end

#string=(string) ⇒ Object

just parse a string, not a whole file



580
581
582
 
# File 'lib/puppet/parser/lexer.rb', line 580

def string=(string)
  @scanner = StringScanner.new(string)
end

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



560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
 
# File 'lib/puppet/parser/lexer.rb', line 560

def tokenize_interpolated_string(token_type,preamble='')
  value,terminator = slurpstring('"$')
  token_queue << [TOKENS[token_type[terminator]],preamble+value]
  variable_regex = if Puppet[:allow_variables_with_dashes]
                     TOKENS[:VARIABLE_WITH_DASH].regex
                   else
                     TOKENS[:VARIABLE].regex
                   end
  if terminator != '$' or @scanner.scan(/\{/)
    token_queue.shift
  elsif var_name = @scanner.scan(variable_regex)
    warn_if_variable_has_hyphen(var_name)
    token_queue << [TOKENS[:VARIABLE],var_name]
    tokenize_interpolated_string(DQ_continuation_token_types)
  else
    tokenize_interpolated_string(token_type,token_queue.pop.last + terminator)
  end
end

#warn_if_variable_has_hyphen(var_name) ⇒ Object 



603
604
605
606
607
 
# File 'lib/puppet/parser/lexer.rb', line 603

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