Class: PuppetLint::Lexer::StringSlurper

Inherits:
Object
  • Object
show all
Defined in:
lib/puppet-lint/lexer/string_slurper.rb

Overview

Document this TODO

Defined Under Namespace

Classes: UnterminatedStringError

Constant Summary collapse

START_INTERP_PATTERN = 
%r{\$\{}
END_INTERP_PATTERN = 
%r{\}}
END_STRING_PATTERN = 
%r{(\A|[^\\])(\\\\)*"}
UNENC_VAR_PATTERN = 
%r{(\A|[^\\])\$(::)?(\w+(-\w+)*::)*\w+(-\w+)*}
ESC_DQUOTE_PATTERN = 
%r{\\+"}
LBRACE_PATTERN = 
%r{\{}

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(string) ⇒ StringSlurper

Returns a new instance of StringSlurper.



21
22
23
24
25
26
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 21

def initialize(string)
  @scanner = StringScanner.new(string)
  @results = []
  @interp_stack = []
  @segment = []
end

Instance Attribute Details

#interp_stack ⇒ Object

Returns the value of attribute interp_stack.



12
13
14
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 12

def interp_stack
  @interp_stack
end

#results ⇒ Object

Returns the value of attribute results.



11
12
13
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 11

def results
  @results
end

#scanner ⇒ Object

Returns the value of attribute scanner.



10
11
12
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 10

def scanner
  @scanner
end

Instance Method Details

#consumed_chars ⇒ Object

Get the number of characters consumed by the StringSlurper.

StringScanner from Ruby 2.0 onwards supports #charpos which returns the number of characters and is multibyte character aware.

Prior to this, Ruby’s multibyte character support in Strings was a bit unusual and neither String#length nor String#split behave as expected, so we use String#scan to split all the consumed text using a UTF-8 aware regex and use the length of the result



110
111
112
113
114
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 110

def consumed_chars
  return scanner.charpos if scanner.respond_to?(:charpos)

  (scanner.pre_match + scanner.matched).scan(%r{.}mu).length
end

#end_heredoc(pattern) ⇒ Object 



158
159
160
161
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 158

def end_heredoc(pattern)
  results << [:HEREDOC, @segment.join]
  results << [:HEREDOC_TERM, scanner.scan(pattern)]
end

#end_interp ⇒ Object 



133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 133

def end_interp
  if interp_stack.empty?
    @segment << scanner.scan(END_INTERP_PATTERN)
    return
  else
    interp_stack.pop
  end

  if interp_stack.empty?
    results << [:INTERP, @segment.join]
    @segment = []
    scanner.skip(END_INTERP_PATTERN)
  else
    @segment << scanner.scan(END_INTERP_PATTERN)
  end
end

#end_string ⇒ Object 



163
164
165
166
167
168
169
170
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 163

def end_string
  if interp_stack.empty?
    @segment << scanner.scan(END_STRING_PATTERN).gsub!(%r{"\Z}, '')
    results << [@segment_type, @segment.join]
  else
    @segment << scanner.scan(END_STRING_PATTERN)
  end
end

#parse ⇒ Object

Raises:



28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 28

def parse
  @segment_type = :STRING

  until scanner.eos?
    if scanner.match?(START_INTERP_PATTERN)
      start_interp
    elsif !interp_stack.empty? && scanner.match?(LBRACE_PATTERN)
      read_char
    elsif scanner.match?(END_INTERP_PATTERN)
      end_interp
    elsif unenclosed_variable?
      unenclosed_variable
    elsif scanner.match?(END_STRING_PATTERN)
      end_string
      break if interp_stack.empty?
    elsif scanner.match?(ESC_DQUOTE_PATTERN)
      @segment << scanner.scan(ESC_DQUOTE_PATTERN)
    else
      read_char
    end
  end

  raise UnterminatedStringError if results.empty? && scanner.matched?

  results
end

#parse_heredoc(heredoc_tag) ⇒ Object 



61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 61

def parse_heredoc(heredoc_tag)
  heredoc_name = heredoc_tag[%r{\A"?(.+?)"?(:.+?)?#{PuppetLint::Lexer::WHITESPACE_RE}*(/.*)?\Z}, 1]
  end_heredoc_pattern = %r{^\|?\s*-?\s*#{Regexp.escape(heredoc_name)}$}
  interpolation = heredoc_tag.start_with?('"')

  @segment_type = :HEREDOC

  until scanner.eos?
    if scanner.match?(end_heredoc_pattern)
      end_heredoc(end_heredoc_pattern)
      break if interp_stack.empty?
    elsif interpolation && scanner.match?(START_INTERP_PATTERN)
      start_interp
    elsif interpolation && !interp_stack.empty? && scanner.match?(LBRACE_PATTERN)
      read_char
    elsif interpolation && unenclosed_variable?
      unenclosed_variable
    elsif interpolation && scanner.match?(END_INTERP_PATTERN)
      end_interp
    else
      read_char
    end
  end

  results
end

#read_char ⇒ Object 



88
89
90
91
92
93
94
95
96
97
98
99
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 88

def read_char
  @segment << scanner.getch

  return if interp_stack.empty?

  case @segment.last
  when '{'
    interp_stack.push(true)
  when '}'
    interp_stack.pop
  end
end

#start_interp ⇒ Object 



116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 116

def start_interp
  if @segment.last && @segment.last == '\\'
    read_char
    return
  end

  if interp_stack.empty?
    scanner.skip(START_INTERP_PATTERN)
    results << [@segment_type, @segment.join]
    @segment = []
  else
    @segment << scanner.scan(START_INTERP_PATTERN)
  end

  interp_stack.push(true)
end

#unenclosed_variable ⇒ Object 



150
151
152
153
154
155
156
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 150

def unenclosed_variable
  read_char if scanner.match?(%r{.\$})

  results << [@segment_type, @segment.join]
  results << [:UNENC_VAR, scanner.scan(UNENC_VAR_PATTERN)]
  @segment = []
end

#unenclosed_variable? ⇒ Boolean

Returns:

  • (Boolean) 


55
56
57
58
59
 
# File 'lib/puppet-lint/lexer/string_slurper.rb', line 55

def unenclosed_variable?
  interp_stack.empty? &&
    scanner.match?(UNENC_VAR_PATTERN) &&
    (@segment.last.nil? ? true : !@segment.last.end_with?('\\'))
end