Class: Rexical::Generator

Inherits:
Object
  • Object
show all
Defined in:
lib/rexical/generator.rb

Constant Summary collapse

REX_HEADER = 
<<-REX_EOT.gsub(/^ {6}/, '')
  #--
  # DO NOT MODIFY!!!!
  # This file is automatically generated by rex %s
  # from lexical definition file "%s".
  #++

REX_EOT
REX_UTIL = 
<<-REX_EOT
  require 'strscan'

  class ScanError < StandardError ; end

  attr_reader   :lineno
  attr_reader   :filename
  attr_accessor :state

  def scan_setup(str)
    @ss = StringScanner.new(str)
    @lineno =  1
    @state  = nil
  end

  def action
    yield
  end

  def scan_str(str)
    scan_setup(str)
    do_parse
  end
  alias :scan :scan_str

  def load_file( filename )
    @filename = filename
    File.open(filename, "r") do |f|
      scan_setup(f.read)
    end
  end

  def scan_file( filename )
    load_file(filename)
    do_parse
  end

REX_EOT
REX_STUB = 
<<-REX_EOT

  if __FILE__ == $0
    exit  if ARGV.size != 1
    filename = ARGV.shift
    rex = %s.new
    begin
      rex.load_file  filename
      while  token = rex.next_token
        p token
      end
    rescue
      $stderr.printf  %s, rex.filename, rex.lineno, $!.message
    end
  end
REX_EOT

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(opts) ⇒ Generator

Returns a new instance of Generator.



29
30
31
32
33
34
35
36
37
38
39
 
# File 'lib/rexical/generator.rb', line 29

def initialize(opts)
  @lineno  =  0
  @macro  =  {}
  @rules  =  []
  @exclusive_states = [nil]
  @grammar_lines  =  nil
  @scanner_header  =  ""
  @scanner_footer  =  ""
  @scanner_inner  =  ""
  @opt  =  opts
end

Instance Attribute Details

#class_name ⇒ Object

Returns the value of attribute class_name.



21
22
23
 
# File 'lib/rexical/generator.rb', line 21

def class_name
  @class_name
end

#debug ⇒ Object

Returns the value of attribute debug.



27
28
29
 
# File 'lib/rexical/generator.rb', line 27

def debug
  @debug
end

#exclusive_states ⇒ Object

Returns the value of attribute exclusive_states.



24
25
26
 
# File 'lib/rexical/generator.rb', line 24

def exclusive_states
  @exclusive_states
end

#grammar_file ⇒ Object

Returns the value of attribute grammar_file.



17
18
19
 
# File 'lib/rexical/generator.rb', line 17

def grammar_file
  @grammar_file
end

#grammar_lines ⇒ Object

Returns the value of attribute grammar_lines.



18
19
20
 
# File 'lib/rexical/generator.rb', line 18

def grammar_lines
  @grammar_lines
end

#ignorecase ⇒ Object

Returns the value of attribute ignorecase.



25
26
27
 
# File 'lib/rexical/generator.rb', line 25

def ignorecase
  @ignorecase
end

#independent ⇒ Object

Returns the value of attribute independent.



26
27
28
 
# File 'lib/rexical/generator.rb', line 26

def independent
  @independent
end

#lineno ⇒ Object

Returns the value of attribute lineno.



22
23
24
 
# File 'lib/rexical/generator.rb', line 22

def lineno
  @lineno
end

#module_name ⇒ Object

Returns the value of attribute module_name.



20
21
22
 
# File 'lib/rexical/generator.rb', line 20

def module_name
  @module_name
end

#rules ⇒ Object

Returns the value of attribute rules.



23
24
25
 
# File 'lib/rexical/generator.rb', line 23

def rules
  @rules
end

#scanner_file ⇒ Object

Returns the value of attribute scanner_file.



19
20
21
 
# File 'lib/rexical/generator.rb', line 19

def scanner_file
  @scanner_file
end

Instance Method Details



45
46
47
 
# File 'lib/rexical/generator.rb', line 45

def add_footer( st )
  @scanner_footer  +=  "#{st}\n"
end

#add_header(st) ⇒ Object 



41
42
43
 
# File 'lib/rexical/generator.rb', line 41

def add_header( st )
  @scanner_header  +=  "#{st}\n"
end

#add_inner(st) ⇒ Object 



49
50
51
 
# File 'lib/rexical/generator.rb', line 49

def add_inner( st )
  @scanner_inner  +=  "#{st}\n"
end

#add_macro(st) ⇒ Object 



69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
 
# File 'lib/rexical/generator.rb', line 69

def add_macro( st )
  ss  =  StringScanner.new(st)
  ss.scan(/\s+/)
  key = ss.scan(/\S+/)
  ss.scan(/\s+/)
  st = ss.post_match
  len  =  st.size
  ndx  =  0
  while ndx <= len
    c  =  st[ndx,1]
    ndx  +=  1
    case  c
    when '\\'
      ndx  +=  1
      next
    when '#', ' '
      ndx  -=  1
      break
    end
  end
  expr = st[0,ndx]
  expr.gsub!('\ ', ' ')
  key  =  '{' + key + '}'
  @macro.each_pair do |k, e|
    expr.gsub!(k) { |m| e }
  end
  @macro[key]  =  expr
rescue
  raise ParseError, "parse error in add_macro:'#{st}'"
end

#add_option(st) ⇒ Object 



53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
 
# File 'lib/rexical/generator.rb', line 53

def add_option( st )
  opts = st.split
  opts.each do |opt|
    case opt
    when /ignorecase/i
      @opt['--ignorecase'] = true
    when /stub/i
      @opt['--stub'] = true
    when /independent/i
      @opt['--independent'] = true
    when /matcheos/i
      @opt['--matcheos'] = true
    end
  end
end

#add_rule(rule_state, rule_expr, rule_action = nil) ⇒ Object 



100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
 
# File 'lib/rexical/generator.rb', line 100

def add_rule( rule_state, rule_expr, rule_action=nil )
  st = rule_expr.dup
  @macro.each_pair do |k, e|
    rule_expr.gsub!(k) { |m| e }
  end
  if rule_state.to_s[1,1] =~ /[A-Z]/
    @exclusive_states << rule_state  unless @exclusive_states.include?(rule_state)
    exclusive_state = rule_state
    start_state = nil
  else
    exclusive_state = nil
    start_state = rule_state
  end
  rule = [exclusive_state, start_state, rule_expr, rule_action]
  @rules << rule
rescue
  raise ParseError, "parse error in add_rule:'#{st}'"
end

#next_line ⇒ Object 



123
124
125
126
127
128
 
# File 'lib/rexical/generator.rb', line 123

def next_line
  @lineno += 1
  @grammar_lines.scan_until(/\n/).chomp
rescue
  nil
end

#parse ⇒ Object 



130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
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
 
# File 'lib/rexical/generator.rb', line 130

def parse
  state1  =  :HEAD
  state2  =  nil
  state3  =  nil
  lastmodes  =  []
  while st = next_line
    case state1
    when :FOOT
      add_footer  st

    when :HEAD
      ss  =  StringScanner.new(st)
      if ss.scan(/class/)
        state1  =  :CLASS
        st  =  ss.post_match.strip
        @class_name  =  st
      else
        add_header  st
      end

    when :CLASS
      s = st.strip
      next  if s.size == 0 or s[0,1] == '#'

      ss  =  StringScanner.new(st)
      if ss.scan(/option.*$/)
        state2 = :OPTION
        next
      end
      if ss.scan(/inner.*$/)
        state2 = :INNER
        next
      end
      if ss.scan(/macro.*$/)
        state2 = :MACRO
        next
      end
      if ss.scan(/rule.*$/)
        state2 = :RULE
        next
      end
      if ss.scan(/end.*$/)
        state1 = :FOOT
        next
      end

      case state2
      when :OPTION
        add_option  st

      when :INNER
        add_inner  st

      when :MACRO
        add_macro  st

      when :RULE
        case state3
        when nil
          rule_state, rule_expr, rule_action  =  parse_rule(st)
          if rule_action =~ /\s*\{/
            lastmodes = parse_action(rule_action, lastmodes)
            if lastmodes.empty?
              add_rule  rule_state, rule_expr, rule_action
            else
              state3  =  :CONT
              rule_action  +=  "\n"
            end
          else
            add_rule  rule_state, rule_expr
          end

        when :CONT
          rule_action  +=  "#{st}\n"
          lastmodes = parse_action(st, lastmodes)
          if lastmodes.empty?
            state3  =  nil
            add_rule  rule_state, rule_expr, rule_action
          else
          end

        end # case state3
      end # case state2
    end # case state1
  end # while
end

#parse_action(st, lastmodes = []) ⇒ Object 



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
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
 
# File 'lib/rexical/generator.rb', line 229

def parse_action(st, lastmodes=[])
  modes  =  lastmodes
  mode  =  lastmodes[-1]
  ss  =  StringScanner.new(st)
  until ss.eos?
    c  =  ss.scan(/./)
    case  c
    when '#'
      if (mode == :brace) or (mode == nil)
        #p [c, mode, modes]
        return  modes
      end
    when '{'
      if (mode == :brace) or (mode == nil)
        mode = :brace
        modes.push  mode
      end
    when '}'
      if (mode == :brace)
        modes.pop
        mode = modes[0]
      end
    when "'"
      if (mode == :brace)
        mode = :quote
        modes.push  mode
      elsif (mode == :quote)
        modes.pop
        mode = modes[0]
      end
    when '"'
      if (mode == :brace)
        mode = :doublequote
        modes.push  mode
      elsif (mode == :doublequote)
        modes.pop
        mode = modes[0]
      end
    when '`'
      if (mode == :brace)
        mode = :backquote
        modes.push  mode
      elsif (mode == :backquote)
        modes.pop
        mode = modes[0]
      end
    end
  end
  #p [c, mode, modes]
  return  modes
end

#parse_rule(st) ⇒ Object 



217
218
219
220
221
222
223
224
225
226
227
 
# File 'lib/rexical/generator.rb', line 217

def parse_rule(st)
  st.strip!
  return  if st.size == 0 or st[0,1] == '#'
  ss  =  StringScanner.new(st)
  ss.scan(/\s+/)
  rule_state  =  ss.scan(/\:\S+/)
  ss.scan(/\s+/)
  rule_expr  =  ss.scan(/\S+/)
  ss.scan(/\s+/)
  [rule_state, rule_expr, ss.post_match]
end

#read_grammar ⇒ Object 



119
120
121
 
# File 'lib/rexical/generator.rb', line 119

def read_grammar
  @grammar_lines = StringScanner.new File.read(grammar_file)
end

#write_scanner(f = scanner_io) ⇒ Object 



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
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
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
 
# File 'lib/rexical/generator.rb', line 354

def write_scanner f = scanner_io
  flag = ""
  flag += "i"  if @opt['--ignorecase']

  f.printf REX_HEADER, Rexical::VERSION, grammar_file

  unless @opt['--independent']
    f.printf "require 'racc/parser'\n"
  end

  @scanner_header.each_line do |s|
    f.print s
  end
  if @opt['--independent']
    f.puts "class #{@class_name}"
  else
    f.puts "class #{@class_name} < Racc::Parser"
  end

  f.print REX_UTIL

  eos_check = @opt["--matcheos"] ? "" : "return if @ss.eos?"

  ## scanner method
  f.print <<-REX_EOT

    def next_token
      #{eos_check}

      # skips empty actions
      until token = _next_token or @ss.eos?; end
      token
    end

    def _next_token
      text = @ss.peek(1)
      @lineno  +=  1  if text == "\\n"
      token = case @state
  REX_EOT

  exclusive_states.each do |es|
    if es.nil?
      f.printf <<-REX_EOT
        when #{(["nil"] + rules.collect{ |rule| rule[1].nil? ? "nil" : rule[1] }).uniq.join(', ')}
      REX_EOT
    else
      f.printf <<-REX_EOT
        when #{es}
      REX_EOT
    end
    f.printf <<-REX_EOT
      case
    REX_EOT

    rules.each do |rule|
      exclusive_state, start_state, rule_expr, rule_action = *rule
      if es == exclusive_state

        if rule_action
          if start_state
            f.print <<-REX_EOT
              when((state == #{start_state}) and (text = @ss.scan(/#{rule_expr}/#{flag})))
                 action #{rule_action}

            REX_EOT
          else
            f.print <<-REX_EOT
              when (text = @ss.scan(/#{rule_expr}/#{flag}))
                 action #{rule_action}

            REX_EOT
          end
        else
          if start_state
            f.print <<-REX_EOT
              when (@state == #{start_state}) && (text = @ss.scan(/#{rule_expr}/#{flag}))
                ;

            REX_EOT
          else
            f.print <<-REX_EOT
              when (text = @ss.scan(/#{rule_expr}/#{flag}))
                ;

            REX_EOT
          end
        end

      end
    end # rules.each

    if @opt["--matcheos"]
      eos_check = <<-REX_EOT
        when @@ss.scan(/$/)
           ;
      REX_EOT
    else
      eos_check = ""
    end

    f.print <<-REX_EOT
      #{eos_check}
      else
        text = @ss.string[@ss.pos .. -1]
        raise  ScanError, "can not match: '" + text + "'"
      end  # if

    REX_EOT
  end # exclusive_states.each

  f.print <<-REX_EOT
    else
      raise  ScanError, "undefined state: '" + state.to_s + "'"
    end  # case state
  REX_EOT

  if @opt['--debug']
    f.print <<-REX_EOT
      p token
    REX_EOT
  end

  f.print <<-REX_EOT
      token
    end  # def _next_token

  REX_EOT

  @scanner_inner.each_line do |s|
    f.print s
  end
  f.puts "end # class"

  @scanner_footer.each_line do |s|
    f.print s
  end

  f.printf REX_STUB, @class_name, '"%s:%d:%s\n"'  if @opt['--stub']
  f.close

end