Class: Puppet::Parser::Lexer

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

Defined Under Namespace

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

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize ⇒ Lexer

Returns a new instance of Lexer.



369
370
371
372
373
 
# File 'lib/puppet/parser/lexer.rb', line 369

def initialize
  @find = 0
  @regex = 0
  initvars
end

Instance Attribute Details

#file ⇒ Object

Returns the value of attribute file.



14
15
16
 
# File 'lib/puppet/parser/lexer.rb', line 14

def file
  @file
end

#indefine ⇒ Object

Returns the value of attribute indefine.



16
17
18
 
# File 'lib/puppet/parser/lexer.rb', line 16

def indefine
  @indefine
end

#last ⇒ Object (readonly)

Returns the value of attribute last.



14
15
16
 
# File 'lib/puppet/parser/lexer.rb', line 14

def last
  @last
end

#lexing_context ⇒ Object (readonly)

Returns the value of attribute lexing_context.



14
15
16
 
# File 'lib/puppet/parser/lexer.rb', line 14

def lexing_context
  @lexing_context
end

#line ⇒ Object

Returns the value of attribute line.



16
17
18
 
# File 'lib/puppet/parser/lexer.rb', line 16

def line
  @line
end

#token_queue ⇒ Object (readonly)

Returns the value of attribute token_queue.



14
15
16
 
# File 'lib/puppet/parser/lexer.rb', line 14

def token_queue
  @token_queue
end

Class Method Details

.acceptable?(context = {}) ⇒ Boolean

:stopdoc: # Issue #4161

Returns:



168
169
170
 
# File 'lib/puppet/parser/lexer.rb', line 168

def (TOKENS[:NUMBER]).acceptable?(context={})
  ![:DQPRE,:DQMID].include? context[:after]
end

Instance Method Details

#clear ⇒ Object 



289
290
291
 
# File 'lib/puppet/parser/lexer.rb', line 289

def clear
  initvars
end

#commentpop ⇒ Object

returns the content of the currently accumulated content cache



566
567
568
 
# File 'lib/puppet/parser/lexer.rb', line 566

def commentpop
  @commentstack.pop[0]
end

#commentpush ⇒ Object 



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

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

#expected ⇒ Object 



293
294
295
296
297
 
# File 'lib/puppet/parser/lexer.rb', line 293

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.



335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
 
# File 'lib/puppet/parser/lexer.rb', line 335

def find_regex_token
  @regex += 1
  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 



323
324
325
326
327
328
329
330
331
332
 
# File 'lib/puppet/parser/lexer.rb', line 323

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.



356
357
358
359
 
# File 'lib/puppet/parser/lexer.rb', line 356

def find_token
  @find += 1
  shift_token || find_regex_token || find_string_token
end

#fullscan ⇒ Object

scan the whole file basically just used for testing



301
302
303
304
305
306
307
308
309
310
 
# File 'lib/puppet/parser/lexer.rb', line 301

def fullscan
  array = []

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

#getcomment(line = nil) ⇒ Object 



570
571
572
573
574
575
576
577
578
 
# File 'lib/puppet/parser/lexer.rb', line 570

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

#indefine? ⇒ Boolean

Returns:



361
362
363
364
365
366
367
 
# File 'lib/puppet/parser/lexer.rb', line 361

def indefine?
  if defined?(@indefine)
    @indefine
  else
    false
  end
end

#initvars ⇒ Object 



375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
 
# File 'lib/puppet/parser/lexer.rb', line 375

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:



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

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.



397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
 
# File 'lib/puppet/parser/lexer.rb', line 397

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

#namepop ⇒ Object

Go up one in the namespace.



420
421
422
 
# File 'lib/puppet/parser/lexer.rb', line 420

def namepop
  @namestack.pop
end

#namespace ⇒ Object

Collect the current namespace.



425
426
427
 
# File 'lib/puppet/parser/lexer.rb', line 425

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

#namestack(value) ⇒ Object

This value might have

in it, but we don’t care – it’ll be

handled normally when joining, and when popping we want to pop this full value, however long the namespace is.



432
433
434
 
# File 'lib/puppet/parser/lexer.rb', line 432

def namestack(value)
  @namestack << value
end

#rest ⇒ Object 



436
437
438
 
# File 'lib/puppet/parser/lexer.rb', line 436

def rest
  @scanner.rest
end

#scan {|[false,false]| ... } ⇒ Object

this is the heart of the lexer

Yields:

  • ([false,false]) 


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
 
# File 'lib/puppet/parser/lexer.rb', line 441

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
    yielded = false
    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

    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

#scan_until(regex) ⇒ Object

Provide some limited access to the scanner, for those tokens that need it.



518
519
520
 
# File 'lib/puppet/parser/lexer.rb', line 518

def scan_until(regex)
  @scanner.scan_until(regex)
end

#shift_token ⇒ Object 



319
320
321
 
# File 'lib/puppet/parser/lexer.rb', line 319

def shift_token
  @token_queue.shift
end

#skip ⇒ Object

Skip any skipchars in our remaining string.



512
513
514
 
# File 'lib/puppet/parser/lexer.rb', line 512

def skip
  @scanner.skip(@skip)
end

#slurpstring(terminators, escapes = %w{ \\ $ ' " 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



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

def slurpstring(terminators,escapes=%w{ \\  $ ' " 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 '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



561
562
563
 
# File 'lib/puppet/parser/lexer.rb', line 561

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

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



547
548
549
550
551
552
553
554
555
556
557
558
 
# File 'lib/puppet/parser/lexer.rb', line 547

def tokenize_interpolated_string(token_type,preamble='')
  value,terminator = slurpstring('"$')
  token_queue << [TOKENS[token_type[terminator]],preamble+value]
  if terminator != '$' or @scanner.scan(/\{/)
    token_queue.shift
  elsif var_name = @scanner.scan(TOKENS[:VARIABLE].regex)
    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