Class: PuppetLint::Lexer

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

Overview

Internal: The puppet-lint lexer. Converts your manifest into its tokenised form.

Defined Under Namespace

Classes: Token

Constant Summary collapse

KEYWORDS =

Internal: A Hash whose keys are Strings representing reserved keywords in the Puppet DSL. From github.com/puppetlabs/puppet/blob/master/lib/puppet/pops/parser/lexer2.rb#L116-L137 or thereabouts

{
  'case'     => true,
  'class'    => true,
  'default'  => true,
  'define'   => true,
  'import'   => true,
  'if'       => true,
  'elsif'    => true,
  'else'     => true,
  'inherits' => true,
  'node'     => true,
  'and'      => true,
  'or'       => true,
  'undef'    => true,
  'false'    => true,
  'true'     => true,
  'in'       => true,
  'unless'   => true,
  'function' => true,
  'type'     => true,
  'attr'     => true,
  'private'  => true,
}
APP_MANAGEMENT_TOKENS =

Internal: A Hash whose keys are Strings representing reserved keywords in the Puppet DSL when Application Management is enabled From github.com/puppetlabs/puppet/blob/master/lib/puppet/pops/parser/lexer2.rb#L142-L159 or therabouts Currently unused

{
  'application' => true,
  'consumes'    => true,
  'produces'    => true,
  'site'        => true,
}
REGEX_PREV_TOKENS =

Internal: A Hash whose keys are Symbols representing token types which a regular expression can follow.

{
  :NODE    => true,
  :LBRACE  => true,
  :RBRACE  => true,
  :MATCH   => true,
  :NOMATCH => true,
  :COMMA   => true,
  :LBRACK  => true,
  :IF      => true,
  :ELSIF   => true,
}
KNOWN_TOKENS =

Internal: An Array of Arrays containing tokens that can be described by a single regular expression. Each sub-Array contains 2 elements, the name of the token as a Symbol and a regular expression describing the value of the token.

[
  [:TYPE, /\A(Integer|Float|Boolean|Regexp|String|Array|Hash|Resource|Class|Collection|Scalar|Numeric|CatalogEntry|Data|Tuple|Struct|Optional|NotUndef|Variant|Enum|Pattern|Any|Callable|Type|Runtime|Undef|Default)\b/],
  [:CLASSREF, /\A(((::){0,1}[A-Z][-\w]*)+)/],
  [:NUMBER, /\A\b((?:0[xX][0-9A-Fa-f]+|0?\d+(?:\.\d+)?(?:[eE]-?\d+)?))\b/],
  [:NAME, /\A(((::)?[_a-z0-9][-\w]*)(::[a-z0-9][-\w]*)*)/],
  [:LBRACK, /\A(\[)/],
  [:RBRACK, /\A(\])/],
  [:LBRACE, /\A(\{)/],
  [:RBRACE, /\A(\})/],
  [:LPAREN, /\A(\()/],
  [:RPAREN, /\A(\))/],
  [:ISEQUAL, /\A(==)/],
  [:MATCH, /\A(=~)/],
  [:FARROW, /\A(=>)/],
  [:EQUALS, /\A(=)/],
  [:APPENDS, /\A(\+=)/],
  [:PARROW, /\A(\+>)/],
  [:PLUS, /\A(\+)/],
  [:GREATEREQUAL, /\A(>=)/],
  [:RSHIFT, /\A(>>)/],
  [:GREATERTHAN, /\A(>)/],
  [:LESSEQUAL, /\A(<=)/],
  [:LLCOLLECT, /\A(<<\|)/],
  [:OUT_EDGE, /\A(<-)/],
  [:OUT_EDGE_SUB, /\A(<~)/],
  [:LCOLLECT, /\A(<\|)/],
  [:LSHIFT, /\A(<<)/],
  [:LESSTHAN, /\A(<)/],
  [:NOMATCH, /\A(!~)/],
  [:NOTEQUAL, /\A(!=)/],
  [:NOT, /\A(!)/],
  [:RRCOLLECT, /\A(\|>>)/],
  [:RCOLLECT, /\A(\|>)/],
  [:IN_EDGE, /\A(->)/],
  [:IN_EDGE_SUB, /\A(~>)/],
  [:MINUS, /\A(-)/],
  [:COMMA, /\A(,)/],
  [:DOT, /\A(\.)/],
  [:COLON, /\A(:)/],
  [:AT, /\A(@)/],
  [:SEMIC, /\A(;)/],
  [:QMARK, /\A(\?)/],
  [:BACKSLASH, /\A(\\)/],
  [:TIMES, /\A(\*)/],
  [:MODULO, /\A(%)/],
  [:PIPE, /\A(\|)/],
]
FORMATTING_TOKENS =

Internal: A Hash whose keys are Symbols representing token types which are considered to be formatting tokens (i.e. tokens that don’t contain code).

{
  :WHITESPACE    => true,
  :NEWLINE       => true,
  :COMMENT       => true,
  :MLCOMMENT     => true,
  :SLASH_COMMENT => true,
  :INDENT        => true,
}

Instance Method Summary collapse

Constructor Details

#initialize ⇒ Lexer

Returns a new instance of Lexer.



29
30
31
32
 
# File 'lib/puppet-lint/lexer.rb', line 29

def initialize
  @line_no = 1
  @column = 1
end

Instance Method Details

#get_string_segment(string, terminators) ⇒ Object

Internal: Split a string on multiple terminators, excluding escaped terminators.

string - The String to be split. terminators - The String of terminators that the String should be split

on.

Returns an Array consisting of two Strings, the String up to the first terminator and the terminator that was found.



334
335
336
337
338
339
340
341
 
# File 'lib/puppet-lint/lexer.rb', line 334

def get_string_segment(string, terminators)
  str = string.scan_until(/([^\\]|^|[^\\])([\\]{2})*[#{terminators}]+/)
  begin
    [str[0..-2], str[-1,1]]
  rescue
    [nil, nil]
  end
end

#interpolate_string(string, line, column) ⇒ Object

Internal: Tokenise the contents of a double quoted string.

string - The String to be tokenised. line - The Integer line number of the start of the passed string. column - The Integer column number of the start of the passed string.

Returns nothing.



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
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
 
# File 'lib/puppet-lint/lexer.rb', line 350

def interpolate_string(string, line, column)
  ss = StringScanner.new(string)
  first = true
  value, terminator = get_string_segment(ss, '"$')
  until value.nil?
    if terminator == "\""
      if first
        tokens << new_token(:STRING, value, value.size + 2, :line => line, :column => column)
        first = false
      else
        line += value.scan(/(\r\n|\r|\n)/).size
        token_column = column + (ss.pos - value.size)
        tokens << new_token(:DQPOST, value, value.size + 1, :line => line, :column => token_column)
      end
    else
      if first
        tokens << new_token(:DQPRE, value, value.size + 1, :line => line, :column => column)
        first = false
      else
        line += value.scan(/(\r\n|\r|\n)/).size
        token_column = column + (ss.pos - value.size)
        tokens << new_token(:DQMID, value, value.size, :line => line, :column => token_column)
      end
      if ss.scan(/\{/).nil?
        var_name = ss.scan(/(::)?(\w+(-\w+)*::)*\w+(-\w+)*/)
        if var_name.nil?
          token_column = column + ss.pos - 1
          tokens << new_token(:DQMID, "$", 1, :line => line, :column => token_column)
        else
          token_column = column + (ss.pos - var_name.size)
          tokens << new_token(:UNENC_VARIABLE, var_name, var_name.size, :line => line, :column => token_column)
        end
      else
        contents = ss.scan_until(/\}/)[0..-2]
        if contents.match(/\A(::)?([\w-]+::)*[\w-]+(\[.+?\])*/)
          contents = "$#{contents}"
        end
        lexer = PuppetLint::Lexer.new
        lexer.tokenise(contents)
        lexer.tokens.each do |token|
          tok_col = column + token.column + (ss.pos - contents.size - 1)
          tok_line = token.line + line - 1
          tokens << new_token(token.type, token.value, token.value.size, :line => tok_line, :column => tok_col)
        end
      end
    end
    value, terminator = get_string_segment(ss, '"$')
  end
end

#new_token(type, value, length, opts = {}) ⇒ Object

Internal: Create a new PuppetLint::Lexer::Token object, calculate its line number and column and then add it to the Linked List of tokens.

type - The Symbol token type. value - The token value. length - The Integer length of the token’s value. opts - A Hash of additional values required to determine line number and

      column:
:line   - The Integer line number if calculated externally.
:column - The Integer column number if calculated externally.

Returns the instantiated PuppetLint::Lexer::Token object.



292
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
323
 
# File 'lib/puppet-lint/lexer.rb', line 292

def new_token(type, value, length, opts = {})
  column = opts[:column] || @column
  line_no = opts[:line] || @line_no

  token = Token.new(type, value, line_no, column)
  unless tokens.last.nil?
    token.prev_token = tokens.last
    tokens.last.next_token = token

    unless FORMATTING_TOKENS.include?(token.type)
      prev_nf_idx = tokens.rindex { |r| ! FORMATTING_TOKENS.include? r.type }
      unless prev_nf_idx.nil?
        prev_nf_token = tokens[prev_nf_idx]
        prev_nf_token.next_code_token = token
        token.prev_code_token = prev_nf_token
      end
    end
  end

  @column += length
  if type == :NEWLINE
    @line_no += 1
    @column = 1
  end
  if [:MLCOMMENT, :SSTRING, :STRING].include? type and /(?:\r\n|\r|\n)/.match(value)
    lines = value.split(/(?:\r\n|\r|\n)/, -1)
    @line_no += lines.length-1
    @column = lines.last.length
  end

  token
end

#possible_regex? ⇒ Boolean

Internal: Given the tokens already processed, determine if the next token could be a regular expression.

Returns true if the next token could be a regex, otherwise return false.

Returns:

  • (Boolean) 


266
267
268
269
270
271
272
273
274
275
276
277
278
 
# File 'lib/puppet-lint/lexer.rb', line 266

def possible_regex?
  prev_token = tokens.reject { |r|
    FORMATTING_TOKENS.include? r.type
  }.last

  return true if prev_token.nil?

  if REGEX_PREV_TOKENS.include? prev_token.type
    true
  else
    false
  end
end

#tokenise(code) ⇒ Object

Internal: Convert a Puppet manifest into tokens.

code - The Puppet manifest to be tokenised as a String.

Returns an Array of PuppetLint::Lexer::Token objects. Raises PuppetLint::LexerError if it encounters unexpected characters (usually the result of syntax errors).



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
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
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
254
255
256
257
258
259
260
 
# File 'lib/puppet-lint/lexer.rb', line 166

def tokenise(code)
  i = 0

  while i < code.size
    chunk = code[i..-1]

    found = false

    KNOWN_TOKENS.each do |type, regex|
      if value = chunk[regex, 1]
        length = value.size
        if type == :NAME
          if KEYWORDS.include? value
            tokens << new_token(value.upcase.to_sym, value, length)
          else
            tokens << new_token(type, value, length)
          end
        else
          tokens << new_token(type, value, length)
        end
        i += length
        found = true
        break
      end
    end

    unless found
      if var_name = chunk[/\A\$((::)?(\w+(-\w+)*::)*\w+(-\w+)*(\[.+?\])*)/, 1]
        length = var_name.size + 1
        tokens << new_token(:VARIABLE, var_name, length)

      elsif chunk.match(/\A'(.*?)'/m)
        str_content = StringScanner.new(code[i+1..-1]).scan_until(/(\A|[^\\])(\\\\)*'/m)
        length = str_content.size + 1
        tokens << new_token(:SSTRING, str_content[0..-2], length)

      elsif chunk.match(/\A"/)
        str_contents = StringScanner.new(code[i+1..-1]).scan_until(/(\A|[^\\])(\\\\)*"/m)
        _ = code[0..i].split("\n")
        interpolate_string(str_contents, _.count, _.last.length)
        length = str_contents.size + 1

      elsif comment = chunk[/\A(#.*)/, 1]
        length = comment.size
        comment.sub!(/#/, '')
        tokens << new_token(:COMMENT, comment, length)

      elsif slash_comment = chunk[/\A(\/\/.*)/, 1]
        length = slash_comment.size
        slash_comment.sub!(/\/\//, '')
        tokens << new_token(:SLASH_COMMENT, slash_comment, length)

      elsif mlcomment = chunk[/\A(\/\*.*?\*\/)/m, 1]
        length = mlcomment.size
        mlcomment_raw = mlcomment.dup
        mlcomment.sub!(/\A\/\* ?/, '')
        mlcomment.sub!(/ ?\*\/\Z/, '')
        mlcomment.gsub!(/^ *\*/, '')
        tokens << new_token(:MLCOMMENT, mlcomment, length)
        tokens.last.raw = mlcomment_raw

      elsif chunk.match(/\A\/.*?\//) && possible_regex?
        str_content = StringScanner.new(code[i+1..-1]).scan_until(/(\A|[^\\])(\\\\)*\//m)
        length = str_content.size + 1
        tokens << new_token(:REGEX, str_content[0..-2], length)

      elsif eolindent = chunk[/\A((\r\n|\r|\n)[ \t]+)/m, 1]
        eol = eolindent[/\A([\r\n]+)/m, 1]
        indent = eolindent[/\A[\r\n]+([ \t]+)/m, 1]
        tokens << new_token(:NEWLINE, eol, eol.size)
        tokens << new_token(:INDENT, indent, indent.size)
        length = indent.size + eol.size

      elsif whitespace = chunk[/\A([ \t]+)/, 1]
        length = whitespace.size
        tokens << new_token(:WHITESPACE, whitespace, length)

      elsif eol = chunk[/\A(\r\n|\r|\n)/, 1]
        length = eol.size
        tokens << new_token(:NEWLINE, eol, length)

      elsif chunk.match(/\A\//)
        length = 1
        tokens << new_token(:DIV, '/', length)

      else
        raise PuppetLint::LexerError.new(@line_no, @column)
      end

      i += length
    end
  end

  tokens
end

#tokens ⇒ Object

Internal: Access the internal token storage.

Returns an Array of PuppetLint::Lexer::Toxen objects.



155
156
157
 
# File 'lib/puppet-lint/lexer.rb', line 155

def tokens
  @tokens ||= []
end