srl_ruby 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/.rspec +4 -0
- data/.rubocop.yml +3 -0
- data/.yardopts +6 -0
- data/Gemfile +6 -0
- data/LICENSE.txt +21 -0
- data/README.md +66 -0
- data/Rakefile +16 -0
- data/bin/srl_ruby +58 -0
- data/lib/regex/abstract_method.rb +35 -0
- data/lib/regex/alternation.rb +27 -0
- data/lib/regex/anchor.rb +45 -0
- data/lib/regex/atomic_expression.rb +16 -0
- data/lib/regex/capturing_group.rb +51 -0
- data/lib/regex/char_class.rb +38 -0
- data/lib/regex/char_range.rb +51 -0
- data/lib/regex/char_shorthand.rb +50 -0
- data/lib/regex/character.rb +204 -0
- data/lib/regex/compound_expression.rb +57 -0
- data/lib/regex/concatenation.rb +29 -0
- data/lib/regex/expression.rb +60 -0
- data/lib/regex/lookaround.rb +50 -0
- data/lib/regex/match_option.rb +34 -0
- data/lib/regex/monadic_expression.rb +28 -0
- data/lib/regex/multiplicity.rb +91 -0
- data/lib/regex/non_capturing_group.rb +27 -0
- data/lib/regex/polyadic_expression.rb +60 -0
- data/lib/regex/quantifiable.rb +22 -0
- data/lib/regex/repetition.rb +29 -0
- data/lib/regex/wildcard.rb +23 -0
- data/lib/srl_ruby/ast_builder.rb +384 -0
- data/lib/srl_ruby/grammar.rb +106 -0
- data/lib/srl_ruby/regex_repr.rb +13 -0
- data/lib/srl_ruby/tokenizer.rb +147 -0
- data/lib/srl_ruby/version.rb +3 -0
- data/lib/srl_ruby.rb +4 -0
- data/spec/integration_spec.rb +451 -0
- data/spec/regex/character_spec.rb +166 -0
- data/spec/regex/multiplicity_spec.rb +79 -0
- data/spec/spec_helper.rb +16 -0
- data/spec/srl_ruby/srl_ruby_spec.rb +7 -0
- data/spec/srl_ruby/tokenizer_spec.rb +147 -0
- data/srl_ruby.gemspec +58 -0
- metadata +150 -0
@@ -0,0 +1,23 @@
|
|
1
|
+
# File: wildcard.rb
|
2
|
+
|
3
|
+
require_relative 'atomic_expression' # Access the superclass
|
4
|
+
|
5
|
+
module Regex # This module is used as a namespace
|
6
|
+
# A wildcard matches any character (except for the newline).
|
7
|
+
class Wildcard < AtomicExpression
|
8
|
+
# Constructor
|
9
|
+
def initialize()
|
10
|
+
super
|
11
|
+
end
|
12
|
+
|
13
|
+
protected
|
14
|
+
|
15
|
+
# Conversion method re-definition.
|
16
|
+
# Purpose: Return the String representation of the expression.
|
17
|
+
def text_repr()
|
18
|
+
return '.'
|
19
|
+
end
|
20
|
+
end # class
|
21
|
+
end # module
|
22
|
+
|
23
|
+
# End of file
|
@@ -0,0 +1,384 @@
|
|
1
|
+
require 'stringio'
|
2
|
+
require_relative 'regex_repr'
|
3
|
+
|
4
|
+
module SrlRuby
|
5
|
+
# The purpose of a ASTBuilder is to build piece by piece an AST
|
6
|
+
# (Abstract Syntax Tree) from a sequence of input tokens and
|
7
|
+
# visit events produced by walking over a GFGParsing object.
|
8
|
+
# Uses the Builder GoF pattern.
|
9
|
+
# The Builder pattern creates a complex object
|
10
|
+
# (say, a parse tree) from simpler objects (terminal and non-terminal
|
11
|
+
# nodes) and using a step by step approach.
|
12
|
+
class ASTBuilder < Rley::ParseRep::ASTBaseBuilder
|
13
|
+
Terminal2NodeClass = {}.freeze
|
14
|
+
|
15
|
+
attr_reader :options
|
16
|
+
|
17
|
+
protected
|
18
|
+
|
19
|
+
def terminal2node()
|
20
|
+
Terminal2NodeClass
|
21
|
+
end
|
22
|
+
|
23
|
+
# Overriding method.
|
24
|
+
# Factory method for creating a node object for the given
|
25
|
+
# input token.
|
26
|
+
# @param aTerminal [Terminal] Terminal symbol associated with the token
|
27
|
+
# @param aTokenPosition [Integer] Position of token in the input stream
|
28
|
+
# @param aToken [Token] The input token
|
29
|
+
def new_leaf_node(_production, _terminal, aTokenPosition, aToken)
|
30
|
+
node = Rley::PTree::TerminalNode.new(aToken, aTokenPosition)
|
31
|
+
|
32
|
+
return node
|
33
|
+
end
|
34
|
+
|
35
|
+
def multiplicity(lowerBound, upperBound)
|
36
|
+
return SRL::Regex::Multiplicity.new(lowerBound, upperBound, :greedy)
|
37
|
+
end
|
38
|
+
|
39
|
+
def string_literal(aString, to_escape = true)
|
40
|
+
if aString.size > 1
|
41
|
+
chars = []
|
42
|
+
aString.each_char do |ch|
|
43
|
+
if to_escape && Regex::Character::MetaChars.include?(ch)
|
44
|
+
chars << Regex::Character.new("\\")
|
45
|
+
end
|
46
|
+
chars << Regex::Character.new(ch)
|
47
|
+
end
|
48
|
+
result = Regex::Concatenation.new(*chars)
|
49
|
+
elsif to_escape && Regex::Character::MetaChars.include?(aString)
|
50
|
+
backslash = Regex::Character.new("\\")
|
51
|
+
a_string = Regex::Character.new(aString)
|
52
|
+
result = Regex::Concatenation.new(backslash, a_string)
|
53
|
+
else
|
54
|
+
result = Regex::Character.new(aString)
|
55
|
+
end
|
56
|
+
|
57
|
+
return result
|
58
|
+
end
|
59
|
+
|
60
|
+
def char_range(lowerBound, upperBound)
|
61
|
+
# TODO fix module nesting
|
62
|
+
lower = Regex::Character.new(lowerBound)
|
63
|
+
upper = Regex::Character.new(upperBound)
|
64
|
+
return Regex::CharRange.new(lower, upper)
|
65
|
+
end
|
66
|
+
|
67
|
+
def char_class(toNegate, *theChildren)
|
68
|
+
Regex::CharClass.new(toNegate, *theChildren)
|
69
|
+
end
|
70
|
+
|
71
|
+
def char_shorthand(shortName)
|
72
|
+
Regex::CharShorthand.new(shortName)
|
73
|
+
end
|
74
|
+
|
75
|
+
def wildcard()
|
76
|
+
Regex::Wildcard.new
|
77
|
+
end
|
78
|
+
|
79
|
+
def repetition(expressionToRepeat, aMultiplicity)
|
80
|
+
return Regex::Repetition.new(expressionToRepeat, aMultiplicity)
|
81
|
+
end
|
82
|
+
|
83
|
+
def begin_anchor
|
84
|
+
return Regex::Anchor.new('^')
|
85
|
+
end
|
86
|
+
|
87
|
+
# rule('expression' => %w[pattern separator flags]).as 'flagged_expr'
|
88
|
+
def reduce_flagged_expr(_production, aRange, theTokens, theChildren)
|
89
|
+
@options = theChildren[2] if theChildren[2]
|
90
|
+
return_first_child(aRange, theTokens, theChildren)
|
91
|
+
end
|
92
|
+
|
93
|
+
# rule('pattern' => %w[pattern separator quantifiable]).as 'pattern_sequence'
|
94
|
+
def reduce_pattern_sequence(_production, _range, _tokens, theChildren)
|
95
|
+
return Regex::Concatenation.new(theChildren[0], theChildren[2])
|
96
|
+
end
|
97
|
+
|
98
|
+
# rule('flags' => %[flags separator single_flag]).as 'flag_sequence'
|
99
|
+
def reduce_flag_sequence(_production, _range, _tokens, theChildren)
|
100
|
+
theChildren[0] << theChildren[2]
|
101
|
+
end
|
102
|
+
|
103
|
+
# rule('single_flag' => %w[CASE INSENSITIVE]).as 'case_insensitive'
|
104
|
+
def reduce_case_insensitive(_production, _range, _tokens, _children)
|
105
|
+
return [Regex::MatchOption.new(:IGNORECASE, true)]
|
106
|
+
end
|
107
|
+
|
108
|
+
# rule('single_flag' => %w[MULTI LINE]).as 'multi_line'
|
109
|
+
def reduce_multi_line(_production, _range, _tokens, _children)
|
110
|
+
return [Regex::MatchOption.new(:MULTILINE, true)]
|
111
|
+
end
|
112
|
+
|
113
|
+
# rule('single_flag' => %w[ALL LAZY]).as 'all_lazy'
|
114
|
+
def reduce_all_lazy(_production, _range, _tokens, _children)
|
115
|
+
return [Regex::MatchOption.new(:ALL_LAZY, true)]
|
116
|
+
end
|
117
|
+
|
118
|
+
# rule 'quantifiable' => %w[begin_anchor anchorable end_anchor]
|
119
|
+
def reduce_pinned_quantifiable(_production, _range, _tokens, theChildren)
|
120
|
+
theChildren[1].begin_anchor = theChildren[0]
|
121
|
+
theChildren[1].end_anchor = theChildren[2]
|
122
|
+
return theChildren[1]
|
123
|
+
end
|
124
|
+
|
125
|
+
# rule 'quantifiable' => %w[begin_anchor anchorable]
|
126
|
+
def reduce_begin_anchor_quantifiable(_production, _range, _tokens, theChildren)
|
127
|
+
theChildren[1].begin_anchor = theChildren[0]
|
128
|
+
return theChildren[1]
|
129
|
+
end
|
130
|
+
|
131
|
+
# rule 'quantifiable' => %w[anchorable end_anchor]
|
132
|
+
def reduce_end_anchor_quantifiable(_production, _range, _tokens, theChildren)
|
133
|
+
theChildren[0].end_anchor = theChildren[1]
|
134
|
+
return theChildren[0]
|
135
|
+
end
|
136
|
+
|
137
|
+
# rule 'begin_anchor' => %w[STARTS WITH]
|
138
|
+
def reduce_starts_with(_production, _range, _tokens, _children)
|
139
|
+
begin_anchor
|
140
|
+
end
|
141
|
+
|
142
|
+
# rule 'begin_anchor' => %w[BEGIN WITH]
|
143
|
+
def reduce_begin_with(_production, _range, _tokens, _children)
|
144
|
+
begin_anchor
|
145
|
+
end
|
146
|
+
|
147
|
+
# rule 'end_anchor' => %w[MUST END].as 'end_anchor'
|
148
|
+
def reduce_end_anchor(_production, _range, _tokens, _children)
|
149
|
+
return Regex::Anchor.new('$')
|
150
|
+
end
|
151
|
+
|
152
|
+
# rule('anchorable' => %w[assertable assertion]).as 'asserted_anchorable'
|
153
|
+
def reduce_asserted_anchorable(_production, _range, _tokens, theChildren)
|
154
|
+
assertion = theChildren.last
|
155
|
+
assertion.children.unshift(theChildren[0])
|
156
|
+
return assertion
|
157
|
+
end
|
158
|
+
|
159
|
+
# rule('assertion' => %w[IF FOLLOWED BY assertable]).as 'if_followed'
|
160
|
+
def reduce_if_followed(_production, _range, _tokens, theChildren)
|
161
|
+
return Regex::Lookaround.new(theChildren.last, :ahead, :positive)
|
162
|
+
end
|
163
|
+
|
164
|
+
# rule('assertion' => %w[IF NOT FOLLOWED BY assertable]).as 'if_not_followed'
|
165
|
+
def reduce_if_not_followed(_production, _range, _tokens, theChildren)
|
166
|
+
return Regex::Lookaround.new(theChildren.last, :ahead, :negative)
|
167
|
+
end
|
168
|
+
|
169
|
+
# rule('assertion' => %w[IF ALREADY HAD assertable]).as 'if_had'
|
170
|
+
def reduce_if_had(_production, _range, _tokens, theChildren)
|
171
|
+
return Regex::Lookaround.new(theChildren.last, :behind, :positive)
|
172
|
+
end
|
173
|
+
|
174
|
+
# rule('assertion' => %w[IF NOT ALREADY HAD assertable]).as 'if_not_had'
|
175
|
+
def reduce_if_not_had(_production, _range, _tokens, theChildren)
|
176
|
+
return Regex::Lookaround.new(theChildren.last, :behind, :negative)
|
177
|
+
end
|
178
|
+
|
179
|
+
# rule('assertable' => %w[term quantifier]).as 'quantified_assertable'
|
180
|
+
def reduce_quantified_assertable(_production, _range, _tokens, theChildren)
|
181
|
+
quantifier = theChildren[1]
|
182
|
+
term = theChildren[0]
|
183
|
+
repetition(term, quantifier)
|
184
|
+
end
|
185
|
+
|
186
|
+
# rule('letter_range' => %w[LETTER FROM LETTER_LIT TO LETTER_LIT]).as 'lowercase_from_to'
|
187
|
+
def reduce_lowercase_from_to(_production, _range, _tokens, theChildren)
|
188
|
+
lower = theChildren[2].token.lexeme
|
189
|
+
upper = theChildren[4].token.lexeme
|
190
|
+
ch_range = char_range(lower, upper)
|
191
|
+
char_class(false, ch_range)
|
192
|
+
end
|
193
|
+
|
194
|
+
# rule('letter_range' => %w[UPPERCASE LETTER FROM LETTER_LIT TO LETTER_LIT]).as 'uppercase_from_to'
|
195
|
+
def reduce_uppercase_from_to(_production, _range, _tokens, theChildren)
|
196
|
+
lower = theChildren[3].token.lexeme
|
197
|
+
upper = theChildren[5].token.lexeme
|
198
|
+
ch_range = char_range(lower.upcase, upper.upcase)
|
199
|
+
char_class(false, ch_range)
|
200
|
+
end
|
201
|
+
|
202
|
+
# rule('letter_range' => 'LETTER').as 'any_lowercase'
|
203
|
+
def reduce_any_lowercase(_production, _range, _tokens, _children)
|
204
|
+
ch_range = char_range('a', 'z')
|
205
|
+
char_class(false, ch_range)
|
206
|
+
end
|
207
|
+
|
208
|
+
# rule('letter_range' => %w[UPPERCASE LETTER]).as 'any_uppercase'
|
209
|
+
def reduce_any_uppercase(_production, _range, _tokens, _children)
|
210
|
+
ch_range = char_range('A', 'Z')
|
211
|
+
char_class(false, ch_range)
|
212
|
+
end
|
213
|
+
|
214
|
+
# rule('digit_range' => %w[digit_or_number FROM DIGIT_LIT TO DIGIT_LIT]).as 'digits_from_to'
|
215
|
+
def reduce_digits_from_to(aProduction, aRange, theTokens, theChildren)
|
216
|
+
reduce_lowercase_from_to(aProduction, aRange, theTokens, theChildren)
|
217
|
+
end
|
218
|
+
|
219
|
+
# rule('digit_range' => 'digit_or_number').as 'simple_digit_range'
|
220
|
+
def reduce_simple_digit_range(_production, _range, _tokens, _children)
|
221
|
+
char_shorthand('d')
|
222
|
+
end
|
223
|
+
|
224
|
+
# rule('character_class' => %w[ANY CHARACTER]).as 'any_character'
|
225
|
+
def reduce_any_character(_production, _range, _tokens, _children)
|
226
|
+
char_shorthand('w')
|
227
|
+
end
|
228
|
+
|
229
|
+
# rule('character_class' => %w[NO CHARACTER]).as 'no_character'
|
230
|
+
def reduce_no_character(_production, _range, _tokens, _children)
|
231
|
+
char_shorthand('W')
|
232
|
+
end
|
233
|
+
|
234
|
+
# rule('character_class' => 'WHITESPACE').as 'whitespace'
|
235
|
+
def reduce_whitespace(_production, _range, _tokens, _children)
|
236
|
+
char_shorthand('s')
|
237
|
+
end
|
238
|
+
|
239
|
+
# rule('character_class' => %w[NO WHITESPACE]).as 'no_whitespace'
|
240
|
+
def reduce_no_whitespace(_production, _range, _tokens, _children)
|
241
|
+
char_shorthand('S')
|
242
|
+
end
|
243
|
+
|
244
|
+
# rule('character_class' => 'ANYTHING').as 'anything'
|
245
|
+
def reduce_anything(_production, _range, _tokens, _children)
|
246
|
+
wildcard
|
247
|
+
end
|
248
|
+
|
249
|
+
# rule('alternation' => %w[ANY OF LPAREN alternatives RPAREN]).as 'any_of'
|
250
|
+
def reduce_one_of(_production, _range, _tokens, theChildren)
|
251
|
+
raw_literal = theChildren[-1].token.lexeme.dup
|
252
|
+
alternatives = raw_literal.chars.map { |ch| Regex::Character.new(ch) }
|
253
|
+
# TODO check other implementations
|
254
|
+
return Regex::CharClass.new(false, *alternatives)
|
255
|
+
end
|
256
|
+
|
257
|
+
# rule('special_char' => 'TAB').as 'tab'
|
258
|
+
def reduce_tab(_production, _range, _tokens, _children)
|
259
|
+
Regex::Character.new('\t')
|
260
|
+
end
|
261
|
+
|
262
|
+
# rule('special_char' => 'BACKSLASH').as 'backslash'
|
263
|
+
def reduce_backslash(_production, _range, _tokens, _children)
|
264
|
+
Regex::Character.new('\\')
|
265
|
+
end
|
266
|
+
|
267
|
+
# rule('special_char' => %w[NEW LINE]).as 'new_line'
|
268
|
+
def reduce_new_line(_production, _range, _tokens, _children)
|
269
|
+
# TODO: control portability
|
270
|
+
Regex::Character.new('\n')
|
271
|
+
end
|
272
|
+
|
273
|
+
# rule('literal' => %w[LITERALLY STRING_LIT]).as 'literally'
|
274
|
+
def reduce_literally(_production, _range, _tokens, theChildren)
|
275
|
+
# What if literal is empty?...
|
276
|
+
|
277
|
+
raw_literal = theChildren[-1].token.lexeme.dup
|
278
|
+
return string_literal(raw_literal)
|
279
|
+
end
|
280
|
+
|
281
|
+
# rule('alternation' => %w[ANY OF LPAREN alternatives RPAREN]).as 'any_of'
|
282
|
+
def reduce_any_of(_production, _range, _tokens, theChildren)
|
283
|
+
return Regex::Alternation.new(*theChildren[3])
|
284
|
+
end
|
285
|
+
|
286
|
+
# rule('alternatives' => %w[alternatives separator quantifiable]).as 'alternative_list'
|
287
|
+
def reduce_alternative_list(_production, _range, _tokens, theChildren)
|
288
|
+
return theChildren[0] << theChildren[-1]
|
289
|
+
end
|
290
|
+
|
291
|
+
# rule('alternatives' => 'quantifiable').as 'simple_alternative'
|
292
|
+
def reduce_simple_alternative(_production, _range, _tokens, theChildren)
|
293
|
+
return [theChildren.last]
|
294
|
+
end
|
295
|
+
|
296
|
+
# rule('grouping' => %w[LPAREN pattern RPAREN]).as 'grouping_parenthenses'
|
297
|
+
def reduce_grouping_parenthenses(_production, _range, _tokens, theChildren)
|
298
|
+
return Regex::NonCapturingGroup.new(theChildren[1])
|
299
|
+
end
|
300
|
+
|
301
|
+
# rule('capturing_group' => %w[CAPTURE assertable]).as 'capture'
|
302
|
+
def reduce_capture(_production, _range, _tokens, theChildren)
|
303
|
+
return Regex::CapturingGroup.new(theChildren[1])
|
304
|
+
end
|
305
|
+
|
306
|
+
# rule('capturing_group' => %w[CAPTURE assertable UNTIL assertable]).as
|
307
|
+
# 'capture_until'
|
308
|
+
def reduce_capture_until(_production, _range, _tokens, theChildren)
|
309
|
+
group = Regex::CapturingGroup.new(theChildren[1])
|
310
|
+
return Regex::Concatenation.new(group, theChildren[3])
|
311
|
+
end
|
312
|
+
|
313
|
+
# rule('capturing_group' => %w[CAPTURE assertable AS var_name]).as
|
314
|
+
# 'named_capture'
|
315
|
+
def reduce_named_capture(_production, _range, _tokens, theChildren)
|
316
|
+
name = theChildren[3].token.lexeme.dup
|
317
|
+
return Regex::CapturingGroup.new(theChildren[1], name)
|
318
|
+
end
|
319
|
+
|
320
|
+
# rule('capturing_group' => %w[CAPTURE assertable AS var_name
|
321
|
+
# UNTIL assertable]).as 'named_capture_until'
|
322
|
+
def reduce_named_capture_until(_production, _range, _tokens, theChildren)
|
323
|
+
name = theChildren[3].token.lexeme.dup
|
324
|
+
group = Regex::CapturingGroup.new(theChildren[1], name)
|
325
|
+
return Regex::Concatenation.new(group, theChildren[5])
|
326
|
+
end
|
327
|
+
|
328
|
+
# rule('quantifier' => 'ONCE').as 'once'
|
329
|
+
def reduce_once(_production, _range, _tokens, _children)
|
330
|
+
multiplicity(1, 1)
|
331
|
+
end
|
332
|
+
|
333
|
+
# rule('quantifier' => 'TWICE').as 'twice'
|
334
|
+
def reduce_twice(_production, _range, _tokens, _children)
|
335
|
+
multiplicity(2, 2)
|
336
|
+
end
|
337
|
+
|
338
|
+
# rule('quantifier' => %w[EXACTLY count TIMES]).as 'exactly'
|
339
|
+
def reduce_exactly(_production, _range, _tokens, theChildren)
|
340
|
+
count = theChildren[1].token.lexeme.to_i
|
341
|
+
multiplicity(count, count)
|
342
|
+
end
|
343
|
+
|
344
|
+
# rule('quantifier' => %w[BETWEEN count AND count times_suffix]).as
|
345
|
+
# 'between_and'
|
346
|
+
def reduce_between_and(_production, _range, _tokens, theChildren)
|
347
|
+
lower = theChildren[1].token.lexeme.to_i
|
348
|
+
upper = theChildren[3].token.lexeme.to_i
|
349
|
+
multiplicity(lower, upper)
|
350
|
+
end
|
351
|
+
|
352
|
+
# rule('quantifier' => 'OPTIONAL').as 'optional'
|
353
|
+
def reduce_optional(_production, _range, _tokens, _children)
|
354
|
+
multiplicity(0, 1)
|
355
|
+
end
|
356
|
+
|
357
|
+
# rule('quantifier' => %w[ONCE OR MORE]).as 'once_or_more'
|
358
|
+
def reduce_once_or_more(_production, _range, _tokens, _children)
|
359
|
+
multiplicity(1, :more)
|
360
|
+
end
|
361
|
+
|
362
|
+
# rule('quantifier' => %w[NEVER OR MORE]).as 'never_or_more'
|
363
|
+
def reduce_never_or_more(_production, _range, _tokens, _children)
|
364
|
+
multiplicity(0, :more)
|
365
|
+
end
|
366
|
+
|
367
|
+
# rule('quantifier' => %w[AT LEAST count TIMES]).as 'at_least'
|
368
|
+
def reduce_at_least(_production, _range, _tokens, theChildren)
|
369
|
+
count = theChildren[2].token.lexeme.to_i
|
370
|
+
multiplicity(count, :more)
|
371
|
+
end
|
372
|
+
|
373
|
+
# rule('times_suffix' => 'TIMES').as 'times_keyword'
|
374
|
+
def reduce_times_keyword(_production, _range, _tokens, _children)
|
375
|
+
return nil
|
376
|
+
end
|
377
|
+
|
378
|
+
# rule('times_suffix' => []).as 'times_dropped'
|
379
|
+
def reduce_times_dropped(_production, _range, _tokens, _children)
|
380
|
+
return nil
|
381
|
+
end
|
382
|
+
end # class
|
383
|
+
end # module
|
384
|
+
# End of file
|
@@ -0,0 +1,106 @@
|
|
1
|
+
# Grammar for SRL (Simple Regex Language)
|
2
|
+
require 'rley' # Load the gem
|
3
|
+
module SrlRuby
|
4
|
+
########################################
|
5
|
+
# Work in progress.
|
6
|
+
# This is a very partial grammar of SRL.
|
7
|
+
# It will be expanded with the coming versions of Rley
|
8
|
+
builder = Rley::Syntax::GrammarBuilder.new do
|
9
|
+
add_terminals('LPAREN', 'RPAREN', 'COMMA')
|
10
|
+
add_terminals('DIGIT_LIT', 'INTEGER', 'LETTER_LIT')
|
11
|
+
add_terminals('LITERALLY', 'STRING_LIT')
|
12
|
+
add_terminals('BEGIN', 'STARTS', 'WITH')
|
13
|
+
add_terminals('MUST', 'END')
|
14
|
+
add_terminals('UPPERCASE', 'LETTER', 'FROM', 'TO')
|
15
|
+
add_terminals('DIGIT', 'NUMBER', 'ANY', 'NO')
|
16
|
+
add_terminals('CHARACTER', 'WHITESPACE', 'ANYTHING')
|
17
|
+
add_terminals('TAB', 'BACKSLASH', 'NEW', 'LINE')
|
18
|
+
add_terminals('OF', 'ONE')
|
19
|
+
add_terminals('EXACTLY', 'TIMES', 'ONCE', 'TWICE')
|
20
|
+
add_terminals('BETWEEN', 'AND', 'OPTIONAL', 'OR')
|
21
|
+
add_terminals('MORE', 'NEVER', 'AT', 'LEAST')
|
22
|
+
add_terminals('IF', 'FOLLOWED', 'BY', 'NOT')
|
23
|
+
add_terminals('ALREADY', 'HAD')
|
24
|
+
add_terminals('CAPTURE', 'AS', 'UNTIL')
|
25
|
+
add_terminals('CASE', 'INSENSITIVE', 'MULTI', 'ALL')
|
26
|
+
add_terminals('LAZY')
|
27
|
+
|
28
|
+
rule('srl' => 'expression').as 'start_rule'
|
29
|
+
rule('expression' => %w[pattern separator flags]).as 'flagged_expr'
|
30
|
+
rule('expression' => 'pattern').as 'simple_expr'
|
31
|
+
rule('pattern' => %w[pattern separator quantifiable]).as 'pattern_sequence'
|
32
|
+
rule('pattern' => 'quantifiable').as 'basic_pattern'
|
33
|
+
rule('separator' => 'COMMA').as 'comma_separator'
|
34
|
+
rule('separator' => []).as 'void_separator'
|
35
|
+
rule('flags' => %w[flags separator single_flag]).as 'flag_sequence'
|
36
|
+
rule('single_flag' => %w[CASE INSENSITIVE]).as 'case_insensitive'
|
37
|
+
rule('single_flag' => %w[MULTI LINE]).as 'multi_line'
|
38
|
+
rule('single_flag' => %w[ALL LAZY]).as 'all_lazy'
|
39
|
+
rule('quantifiable' => %w[begin_anchor anchorable end_anchor]).as 'pinned_quantifiable'
|
40
|
+
rule('quantifiable' => %w[begin_anchor anchorable]).as 'begin_anchor_quantifiable'
|
41
|
+
rule('quantifiable' => %w[anchorable end_anchor]).as 'end_anchor_quantifiable'
|
42
|
+
rule('quantifiable' => 'anchorable').as 'simple_quantifiable'
|
43
|
+
rule('begin_anchor' => %w[STARTS WITH]).as 'starts_with'
|
44
|
+
rule('begin_anchor' => %w[BEGIN WITH]).as 'begin_with'
|
45
|
+
rule('end_anchor' => %w[MUST END]).as 'end_anchor'
|
46
|
+
rule('anchorable' => 'assertable').as 'simple_anchorable'
|
47
|
+
rule('anchorable' => %w[assertable assertion]).as 'asserted_anchorable'
|
48
|
+
rule('assertion' => %w[IF FOLLOWED BY assertable]).as 'if_followed'
|
49
|
+
rule('assertion' => %w[IF NOT FOLLOWED BY assertable]).as 'if_not_followed'
|
50
|
+
rule('assertion' => %w[IF ALREADY HAD assertable]).as 'if_had'
|
51
|
+
rule('assertion' => %w[IF NOT ALREADY HAD assertable]).as 'if_not_had'
|
52
|
+
rule('assertable' => 'term').as 'simple_assertable'
|
53
|
+
rule('assertable' => %w[term quantifier]).as 'quantified_assertable'
|
54
|
+
rule('term' => 'atom').as 'atom_term'
|
55
|
+
rule('term' => 'alternation').as 'alternation_term'
|
56
|
+
rule('term' => 'grouping').as 'grouping_term'
|
57
|
+
rule('term' => 'capturing_group').as 'capturing_group_atom'
|
58
|
+
rule('atom' => 'letter_range').as 'letter_range_atom'
|
59
|
+
rule('atom' => 'digit_range').as 'digit_range_atom'
|
60
|
+
rule('atom' => 'character_class').as 'character_class_atom'
|
61
|
+
rule('atom' => 'special_char').as 'special_char_atom'
|
62
|
+
rule('atom' => 'literal').as 'literal_atom'
|
63
|
+
rule('letter_range' => %w[LETTER FROM LETTER_LIT TO LETTER_LIT]).as 'lowercase_from_to'
|
64
|
+
rule('letter_range' => %w[UPPERCASE LETTER FROM LETTER_LIT TO LETTER_LIT]).as 'uppercase_from_to'
|
65
|
+
rule('letter_range' => 'LETTER').as 'any_lowercase'
|
66
|
+
rule('letter_range' => %w[UPPERCASE LETTER]).as 'any_uppercase'
|
67
|
+
rule('digit_range' => %w[digit_or_number FROM DIGIT_LIT TO DIGIT_LIT]).as 'digits_from_to'
|
68
|
+
rule('digit_range' => 'digit_or_number').as 'simple_digit_range'
|
69
|
+
rule('character_class' => %w[ANY CHARACTER]).as 'any_character'
|
70
|
+
rule('character_class' => %w[NO CHARACTER]).as 'no_character'
|
71
|
+
rule('character_class' => 'WHITESPACE').as 'whitespace'
|
72
|
+
rule('character_class' => %w[NO WHITESPACE]).as 'no_whitespace'
|
73
|
+
rule('character_class' => 'ANYTHING').as 'anything'
|
74
|
+
rule('character_class' => %w[ONE OF STRING_LIT]).as 'one_of'
|
75
|
+
rule('special_char' => 'TAB').as 'tab'
|
76
|
+
rule('special_char' => 'BACKSLASH').as 'backslash'
|
77
|
+
rule('special_char' => %w[NEW LINE]).as 'new_line'
|
78
|
+
rule('literal' => %w[LITERALLY STRING_LIT]).as 'literally'
|
79
|
+
rule('alternation' => %w[ANY OF LPAREN alternatives RPAREN]).as 'any_of'
|
80
|
+
rule('alternatives' => %w[alternatives separator quantifiable]).as 'alternative_list'
|
81
|
+
rule('alternatives' => 'quantifiable').as 'simple_alternative'
|
82
|
+
rule('grouping' => %w[LPAREN pattern RPAREN]).as 'grouping_parenthenses'
|
83
|
+
rule('capturing_group' => %w[CAPTURE assertable]).as 'capture'
|
84
|
+
rule('capturing_group' => %w[CAPTURE assertable UNTIL assertable]).as 'capture_until'
|
85
|
+
rule('capturing_group' => %w[CAPTURE assertable AS var_name]).as 'named_capture'
|
86
|
+
rule('capturing_group' => %w[CAPTURE assertable AS var_name UNTIL assertable]).as 'named_capture_until'
|
87
|
+
rule('var_name' => 'STRING_LIT').as 'var_name'
|
88
|
+
rule('quantifier' => 'ONCE').as 'once'
|
89
|
+
rule('quantifier' => 'TWICE').as 'twice'
|
90
|
+
rule('quantifier' => %w[EXACTLY count TIMES]).as 'exactly'
|
91
|
+
rule('quantifier' => %w[BETWEEN count AND count times_suffix]).as 'between_and'
|
92
|
+
rule('quantifier' => 'OPTIONAL').as 'optional'
|
93
|
+
rule('quantifier' => %w[ONCE OR MORE]).as 'once_or_more'
|
94
|
+
rule('quantifier' => %w[NEVER OR MORE]).as 'never_or_more'
|
95
|
+
rule('quantifier' => %w[AT LEAST count TIMES]).as 'at_least'
|
96
|
+
rule('digit_or_number' => 'DIGIT').as 'digit_keyword'
|
97
|
+
rule('digit_or_number' => 'NUMBER').as 'number_keyword'
|
98
|
+
rule('count' => 'DIGIT_LIT').as 'single_digit'
|
99
|
+
rule('count' => 'INTEGER').as 'integer_count'
|
100
|
+
rule('times_suffix' => 'TIMES').as 'times_keyword'
|
101
|
+
rule('times_suffix' => []).as 'times_dropped'
|
102
|
+
end
|
103
|
+
|
104
|
+
# And now build the grammar and make it accessible via a global constant
|
105
|
+
Grammar = builder.grammar
|
106
|
+
end # module
|
@@ -0,0 +1,13 @@
|
|
1
|
+
require_relative '../regex/character'
|
2
|
+
require_relative '../regex/char_range'
|
3
|
+
require_relative '../regex/concatenation'
|
4
|
+
require_relative '../regex/multiplicity'
|
5
|
+
require_relative '../regex/repetition'
|
6
|
+
require_relative '../regex/char_class'
|
7
|
+
require_relative '../regex/char_shorthand'
|
8
|
+
require_relative '../regex/wildcard'
|
9
|
+
require_relative '../regex/alternation'
|
10
|
+
require_relative '../regex/non_capturing_group'
|
11
|
+
require_relative '../regex/anchor'
|
12
|
+
require_relative '../regex/lookaround'
|
13
|
+
require_relative '../regex/capturing_group'
|
@@ -0,0 +1,147 @@
|
|
1
|
+
# File: srl_tokenizer.rb
|
2
|
+
# Tokenizer for SRL (Simple Regex Language)
|
3
|
+
require 'strscan'
|
4
|
+
require 'rley' # Load the Rley gem
|
5
|
+
|
6
|
+
module SrlRuby
|
7
|
+
# The tokenizer should recognize:
|
8
|
+
# Keywords: as, capture, letter
|
9
|
+
# Integer literals including single digit
|
10
|
+
# String literals (quote delimited)
|
11
|
+
# Single character literal
|
12
|
+
# Delimiters: parentheses '(' and ')'
|
13
|
+
# Separators: comma (optional)
|
14
|
+
class Tokenizer
|
15
|
+
attr_reader(:scanner)
|
16
|
+
attr_reader(:lineno)
|
17
|
+
attr_reader(:line_start)
|
18
|
+
|
19
|
+
@@lexeme2name = {
|
20
|
+
'(' => 'LPAREN',
|
21
|
+
')' => 'RPAREN',
|
22
|
+
',' => 'COMMA'
|
23
|
+
}.freeze
|
24
|
+
|
25
|
+
# Here are all the SRL keywords (in uppercase)
|
26
|
+
@@keywords = %w[
|
27
|
+
ALL
|
28
|
+
ALREADY
|
29
|
+
AND
|
30
|
+
ANY
|
31
|
+
ANYTHING
|
32
|
+
AS
|
33
|
+
AT
|
34
|
+
BACKSLASH
|
35
|
+
BEGIN
|
36
|
+
BETWEEN
|
37
|
+
BY
|
38
|
+
CAPTURE
|
39
|
+
CASE
|
40
|
+
CHARACTER
|
41
|
+
DIGIT
|
42
|
+
END
|
43
|
+
EXACTLY
|
44
|
+
FOLLOWED
|
45
|
+
FROM
|
46
|
+
HAD
|
47
|
+
IF
|
48
|
+
INSENSITIVE
|
49
|
+
LAZY
|
50
|
+
LEAST
|
51
|
+
LETTER
|
52
|
+
LINE
|
53
|
+
LITERALLY
|
54
|
+
MORE
|
55
|
+
MULTI
|
56
|
+
MUST
|
57
|
+
NEVER
|
58
|
+
NEW
|
59
|
+
NO
|
60
|
+
NOT
|
61
|
+
NUMBER
|
62
|
+
OF
|
63
|
+
ONCE
|
64
|
+
ONE
|
65
|
+
OPTIONAL
|
66
|
+
OR
|
67
|
+
STARTS
|
68
|
+
TAB
|
69
|
+
TIMES
|
70
|
+
TO
|
71
|
+
TWICE
|
72
|
+
UNTIL
|
73
|
+
UPPERCASE
|
74
|
+
WHITESPACE
|
75
|
+
WITH
|
76
|
+
].map { |x| [x, x] } .to_h
|
77
|
+
|
78
|
+
class ScanError < StandardError; end
|
79
|
+
|
80
|
+
def initialize(source)
|
81
|
+
@scanner = StringScanner.new(source)
|
82
|
+
@lineno = 1
|
83
|
+
end
|
84
|
+
|
85
|
+
def tokens()
|
86
|
+
tok_sequence = []
|
87
|
+
until @scanner.eos?
|
88
|
+
token = _next_token
|
89
|
+
tok_sequence << token unless token.nil?
|
90
|
+
end
|
91
|
+
|
92
|
+
return tok_sequence
|
93
|
+
end
|
94
|
+
|
95
|
+
private
|
96
|
+
|
97
|
+
def _next_token()
|
98
|
+
skip_whitespaces
|
99
|
+
curr_ch = scanner.peek(1)
|
100
|
+
return nil if curr_ch.nil? || curr_ch.empty?
|
101
|
+
|
102
|
+
token = nil
|
103
|
+
|
104
|
+
if '(),'.include? curr_ch
|
105
|
+
# Delimiters, separators => single character token
|
106
|
+
token = build_token(@@lexeme2name[curr_ch], scanner.getch)
|
107
|
+
elsif (lexeme = scanner.scan(/[0-9]{2,}/))
|
108
|
+
token = build_token('INTEGER', lexeme) # An integer has 2..* digits
|
109
|
+
elsif (lexeme = scanner.scan(/[0-9]/))
|
110
|
+
token = build_token('DIGIT_LIT', lexeme)
|
111
|
+
elsif (lexeme = scanner.scan(/[a-zA-Z]{2,}/))
|
112
|
+
token = build_token(@@keywords[lexeme.upcase], lexeme)
|
113
|
+
# TODO: handle case unknown identifier
|
114
|
+
elsif (lexeme = scanner.scan(/[a-zA-Z]((?=\s)|$)/))
|
115
|
+
token = build_token('LETTER_LIT', lexeme)
|
116
|
+
elsif (lexeme = scanner.scan(/"([^"]|\\")*"/)) # Double quotes literal?
|
117
|
+
unquoted = lexeme.gsub(/(^")|("$)/, '')
|
118
|
+
token = build_token('STRING_LIT', unquoted)
|
119
|
+
elsif (lexeme = scanner.scan(/'([^']|\\')*'/)) # Single quotes literal?
|
120
|
+
unquoted = lexeme.gsub(/(^')|('$)/, '')
|
121
|
+
token = build_token('STRING_LIT', unquoted)
|
122
|
+
else # Unknown token
|
123
|
+
erroneous = curr_ch.nil? ? '' : curr_ch
|
124
|
+
sequel = scanner.scan(/.{1,20}/)
|
125
|
+
erroneous += sequel unless sequel.nil?
|
126
|
+
raise ScanError.new("Unknown token #{erroneous}")
|
127
|
+
end
|
128
|
+
|
129
|
+
return token
|
130
|
+
end
|
131
|
+
|
132
|
+
def build_token(aSymbolName, aLexeme)
|
133
|
+
begin
|
134
|
+
token = Rley::Lexical::Token.new(aLexeme, aSymbolName)
|
135
|
+
rescue StandardError
|
136
|
+
puts "Failing with '#{aSymbolName}' and '#{aLexeme}'"
|
137
|
+
raise ex
|
138
|
+
end
|
139
|
+
|
140
|
+
return token
|
141
|
+
end
|
142
|
+
|
143
|
+
def skip_whitespaces()
|
144
|
+
scanner.scan(/[ \t\f\n\r]+/)
|
145
|
+
end
|
146
|
+
end # class
|
147
|
+
end # module
|
data/lib/srl_ruby.rb
ADDED