ruby_parser 3.14.2 → 3.17.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- checksums.yaml.gz.sig +0 -0
- data.tar.gz.sig +0 -0
- data/History.rdoc +60 -0
- data/Manifest.txt +4 -0
- data/Rakefile +83 -16
- data/bin/ruby_parse_extract_error +8 -3
- data/compare/normalize.rb +45 -5
- data/debugging.md +172 -0
- data/lib/ruby20_parser.rb +2953 -2924
- data/lib/ruby20_parser.y +99 -59
- data/lib/ruby21_parser.rb +3008 -2977
- data/lib/ruby21_parser.y +99 -59
- data/lib/ruby22_parser.rb +3011 -2976
- data/lib/ruby22_parser.y +99 -59
- data/lib/ruby23_parser.rb +2955 -2923
- data/lib/ruby23_parser.y +99 -59
- data/lib/ruby24_parser.rb +3024 -2984
- data/lib/ruby24_parser.y +99 -59
- data/lib/ruby25_parser.rb +3023 -2984
- data/lib/ruby25_parser.y +99 -59
- data/lib/ruby26_parser.rb +2954 -2913
- data/lib/ruby26_parser.y +100 -59
- data/lib/ruby27_parser.rb +7393 -0
- data/lib/ruby27_parser.y +2715 -0
- data/lib/ruby30_parser.rb +7393 -0
- data/lib/ruby30_parser.y +2715 -0
- data/lib/ruby_lexer.rb +90 -39
- data/lib/ruby_lexer.rex +6 -7
- data/lib/ruby_lexer.rex.rb +7 -9
- data/lib/ruby_parser.rb +4 -0
- data/lib/ruby_parser.yy +164 -59
- data/lib/ruby_parser_extras.rb +57 -18
- data/test/test_ruby_lexer.rb +64 -16
- data/test/test_ruby_parser.rb +277 -3
- data/tools/munge.rb +9 -4
- metadata +55 -36
- metadata.gz.sig +0 -0
data/lib/ruby_lexer.rb
CHANGED
@@ -25,11 +25,10 @@ class RubyLexer
|
|
25
25
|
|
26
26
|
HAS_ENC = "".respond_to? :encoding
|
27
27
|
|
28
|
-
|
29
|
-
|
30
|
-
|
31
|
-
|
32
|
-
end
|
28
|
+
BTOKENS = {
|
29
|
+
".." => :tBDOT2,
|
30
|
+
"..." => :tBDOT3,
|
31
|
+
}
|
33
32
|
|
34
33
|
TOKENS = {
|
35
34
|
"!" => :tBANG,
|
@@ -137,6 +136,10 @@ class RubyLexer
|
|
137
136
|
ss.eos?
|
138
137
|
end
|
139
138
|
|
139
|
+
def expr_beg?
|
140
|
+
lex_state =~ EXPR_BEG
|
141
|
+
end
|
142
|
+
|
140
143
|
def expr_dot?
|
141
144
|
lex_state =~ EXPR_DOT
|
142
145
|
end
|
@@ -162,7 +165,7 @@ class RubyLexer
|
|
162
165
|
indent = func =~ STR_FUNC_INDENT ? "[ \t]*" : nil
|
163
166
|
expand = func =~ STR_FUNC_EXPAND
|
164
167
|
eol = last_line && last_line.end_with?("\r\n") ? "\r\n" : "\n"
|
165
|
-
eos_re = /#{indent}#{Regexp.escape eos}(
|
168
|
+
eos_re = /#{indent}#{Regexp.escape eos}(\r*\n|\z)/
|
166
169
|
err_msg = "can't match #{eos_re.inspect} anywhere in "
|
167
170
|
|
168
171
|
rb_compile_error err_msg if end_of_stream?
|
@@ -177,10 +180,15 @@ class RubyLexer
|
|
177
180
|
|
178
181
|
if expand then
|
179
182
|
case
|
180
|
-
when scan(/#[
|
181
|
-
|
183
|
+
when scan(/#(?=\$(-.|[a-zA-Z_0-9~\*\$\?!@\/\\;,\.=:<>\"\&\`\'+]))/) then
|
184
|
+
# TODO: !ISASCII
|
185
|
+
# ?! see parser_peek_variable_name
|
186
|
+
return :tSTRING_DVAR, matched
|
187
|
+
when scan(/#(?=\@\@?[a-zA-Z_])/) then
|
188
|
+
# TODO: !ISASCII
|
182
189
|
return :tSTRING_DVAR, matched
|
183
190
|
when scan(/#[{]/) then
|
191
|
+
self.command_start = true
|
184
192
|
return :tSTRING_DBEG, matched
|
185
193
|
when scan(/#/) then
|
186
194
|
string_buffer << "#"
|
@@ -320,6 +328,11 @@ class RubyLexer
|
|
320
328
|
lpar_beg && lpar_beg == paren_nest
|
321
329
|
end
|
322
330
|
|
331
|
+
def is_local_id id
|
332
|
+
# maybe just make this false for now
|
333
|
+
self.parser.env[id.to_sym] == :lvar # HACK: this isn't remotely right
|
334
|
+
end
|
335
|
+
|
323
336
|
def lvar_defined? id
|
324
337
|
# TODO: (dyna_in_block? && dvar_defined?(id)) || local_id?(id)
|
325
338
|
self.parser.env[id.to_sym] == :lvar
|
@@ -338,9 +351,9 @@ class RubyLexer
|
|
338
351
|
|
339
352
|
if scan(/[a-z0-9]{1,2}/i) then # Long-hand (e.g. %Q{}).
|
340
353
|
rb_compile_error "unknown type of %string" if ss.matched_size == 2
|
341
|
-
c, beg, short_hand = matched,
|
354
|
+
c, beg, short_hand = matched, getch, false
|
342
355
|
else # Short-hand (e.g. %{, %., %!, etc)
|
343
|
-
c, beg, short_hand = "Q",
|
356
|
+
c, beg, short_hand = "Q", getch, true
|
344
357
|
end
|
345
358
|
|
346
359
|
if end_of_stream? or c == RubyLexer::EOF or beg == RubyLexer::EOF then
|
@@ -457,7 +470,7 @@ class RubyLexer
|
|
457
470
|
if text =~ check then
|
458
471
|
content.gsub(ESC) { unescape $1 }
|
459
472
|
else
|
460
|
-
content.gsub(/\\\\/, "\\").gsub(
|
473
|
+
content.gsub(/\\\\/, "\\").gsub(/\\\'/, "'")
|
461
474
|
end
|
462
475
|
end
|
463
476
|
|
@@ -495,16 +508,19 @@ class RubyLexer
|
|
495
508
|
end
|
496
509
|
|
497
510
|
def process_brace_close text
|
498
|
-
# matching compare/parse23.y:8561
|
499
|
-
cond.lexpop
|
500
|
-
cmdarg.lexpop
|
501
|
-
|
502
511
|
case matched
|
503
512
|
when "}" then
|
504
513
|
self.brace_nest -= 1
|
505
|
-
self.lex_state = ruby24minus? ? EXPR_ENDARG : EXPR_END
|
506
|
-
|
507
514
|
return :tSTRING_DEND, matched if brace_nest < 0
|
515
|
+
end
|
516
|
+
|
517
|
+
# matching compare/parse26.y:8099
|
518
|
+
cond.pop
|
519
|
+
cmdarg.pop
|
520
|
+
|
521
|
+
case matched
|
522
|
+
when "}" then
|
523
|
+
self.lex_state = ruby24minus? ? EXPR_ENDARG : EXPR_END
|
508
524
|
return :tRCURLY, matched
|
509
525
|
when "]" then
|
510
526
|
self.paren_nest -= 1
|
@@ -573,6 +589,12 @@ class RubyLexer
|
|
573
589
|
end
|
574
590
|
end
|
575
591
|
|
592
|
+
def process_dots text
|
593
|
+
tokens = ruby27plus? && expr_beg? ? BTOKENS : TOKENS
|
594
|
+
|
595
|
+
result EXPR_BEG, tokens[text], text
|
596
|
+
end
|
597
|
+
|
576
598
|
def process_float text
|
577
599
|
rb_compile_error "Invalid numeric format" if text =~ /__/
|
578
600
|
|
@@ -605,7 +627,7 @@ class RubyLexer
|
|
605
627
|
end
|
606
628
|
|
607
629
|
def process_label text
|
608
|
-
symbol = possibly_escape_string text,
|
630
|
+
symbol = possibly_escape_string text, /^\"/
|
609
631
|
|
610
632
|
result EXPR_LAB, :tLABEL, [symbol, self.lineno]
|
611
633
|
end
|
@@ -619,7 +641,7 @@ class RubyLexer
|
|
619
641
|
text = text[0..-2]
|
620
642
|
end
|
621
643
|
|
622
|
-
result EXPR_END, :tSTRING, text[1..-2].gsub(/\\\\/, "\\").gsub(
|
644
|
+
result EXPR_END, :tSTRING, text[1..-2].gsub(/\\\\/, "\\").gsub(/\\\'/, "\'")
|
623
645
|
end
|
624
646
|
|
625
647
|
def process_lchevron text
|
@@ -791,12 +813,22 @@ class RubyLexer
|
|
791
813
|
c = if scan(/\\/) then
|
792
814
|
self.read_escape
|
793
815
|
else
|
794
|
-
|
816
|
+
getch
|
795
817
|
end
|
796
818
|
|
797
819
|
result EXPR_END, :tSTRING, c
|
798
820
|
end
|
799
821
|
|
822
|
+
def process_simple_string text
|
823
|
+
replacement = text[1..-2].gsub(ESC) {
|
824
|
+
unescape($1).b.force_encoding Encoding::UTF_8
|
825
|
+
}
|
826
|
+
|
827
|
+
replacement = replacement.b unless replacement.valid_encoding?
|
828
|
+
|
829
|
+
result EXPR_END, :tSTRING, replacement
|
830
|
+
end
|
831
|
+
|
800
832
|
def process_slash text
|
801
833
|
if is_beg? then
|
802
834
|
string STR_REGEXP
|
@@ -870,16 +902,16 @@ class RubyLexer
|
|
870
902
|
|
871
903
|
if [:tSTRING_END, :tREGEXP_END, :tLABEL_END].include? token_type then
|
872
904
|
self.lex_strterm = nil
|
873
|
-
self.lex_state = (token_type == :tLABEL_END) ? EXPR_PAR :
|
905
|
+
self.lex_state = (token_type == :tLABEL_END) ? EXPR_PAR : EXPR_LIT
|
874
906
|
end
|
875
907
|
|
876
908
|
return token
|
877
909
|
end
|
878
910
|
|
879
911
|
def process_symbol text
|
880
|
-
symbol = possibly_escape_string text,
|
912
|
+
symbol = possibly_escape_string text, /^:\"/ # stupid emacs
|
881
913
|
|
882
|
-
result
|
914
|
+
result EXPR_LIT, :tSYMBOL, symbol
|
883
915
|
end
|
884
916
|
|
885
917
|
def process_token text
|
@@ -928,6 +960,8 @@ class RubyLexer
|
|
928
960
|
EXPR_END
|
929
961
|
end
|
930
962
|
|
963
|
+
tok_id = :tIDENTIFIER if tok_id == :tCONSTANT && is_local_id(token)
|
964
|
+
|
931
965
|
if last_state !~ EXPR_DOT|EXPR_FNAME and
|
932
966
|
(tok_id == :tIDENTIFIER) and # not EXPR_FNAME, not attrasgn
|
933
967
|
lvar_defined?(token) then
|
@@ -951,18 +985,16 @@ class RubyLexer
|
|
951
985
|
self.command_start = true if lex_state =~ EXPR_BEG
|
952
986
|
|
953
987
|
case
|
954
|
-
when keyword.id0 == :kDO then
|
988
|
+
when keyword.id0 == :kDO then # parse26.y line 7591
|
955
989
|
case
|
956
990
|
when lambda_beginning? then
|
957
991
|
self.lpar_beg = nil # lambda_beginning? == FALSE in the body of "-> do ... end"
|
958
|
-
self.paren_nest -= 1
|
992
|
+
self.paren_nest -= 1 # TODO: question this?
|
959
993
|
result lex_state, :kDO_LAMBDA, value
|
960
994
|
when cond.is_in_state then
|
961
995
|
result lex_state, :kDO_COND, value
|
962
996
|
when cmdarg.is_in_state && state != EXPR_CMDARG then
|
963
997
|
result lex_state, :kDO_BLOCK, value
|
964
|
-
when state =~ EXPR_BEG|EXPR_ENDARG then
|
965
|
-
result lex_state, :kDO_BLOCK, value
|
966
998
|
else
|
967
999
|
result lex_state, :kDO, value
|
968
1000
|
end
|
@@ -979,9 +1011,9 @@ class RubyLexer
|
|
979
1011
|
ss.unscan # put back "_"
|
980
1012
|
|
981
1013
|
if beginning_of_line? && scan(/\__END__(\r?\n|\Z)/) then
|
982
|
-
|
983
|
-
elsif scan(
|
984
|
-
|
1014
|
+
[RubyLexer::EOF, RubyLexer::EOF]
|
1015
|
+
elsif scan(/#{IDENT_CHAR}+/) then
|
1016
|
+
process_token matched
|
985
1017
|
end
|
986
1018
|
end
|
987
1019
|
|
@@ -1018,7 +1050,7 @@ class RubyLexer
|
|
1018
1050
|
when scan(/x([0-9a-fA-F]{1,2})/) then # hex constant
|
1019
1051
|
# TODO: force encode everything to UTF-8?
|
1020
1052
|
ss[1].to_i(16).chr.force_encoding Encoding::UTF_8
|
1021
|
-
when check(/M
|
1053
|
+
when check(/M-\\./) then
|
1022
1054
|
scan(/M-\\/) # eat it
|
1023
1055
|
c = self.read_escape
|
1024
1056
|
c[0] = (c[0].ord | 0x80).chr
|
@@ -1032,6 +1064,11 @@ class RubyLexer
|
|
1032
1064
|
c = self.read_escape
|
1033
1065
|
c[0] = (c[0].ord & 0x9f).chr
|
1034
1066
|
c
|
1067
|
+
when check(/(C-|c)\\(?!u|\\)/) then
|
1068
|
+
scan(/(C-|c)\\/) # eat it
|
1069
|
+
c = read_escape
|
1070
|
+
c[0] = (c[0].ord & 0x9f).chr
|
1071
|
+
c
|
1035
1072
|
when scan(/C-\?|c\?/) then
|
1036
1073
|
127.chr
|
1037
1074
|
when scan(/(C-|c)(.)/) then
|
@@ -1040,17 +1077,25 @@ class RubyLexer
|
|
1040
1077
|
c
|
1041
1078
|
when scan(/^[89]/i) then # bad octal or hex... MRI ignores them :(
|
1042
1079
|
matched
|
1043
|
-
when scan(/u(
|
1044
|
-
[ss[1].
|
1045
|
-
when scan(/u(
|
1080
|
+
when scan(/u(\h{4})/) then
|
1081
|
+
[ss[1].to_i(16)].pack("U")
|
1082
|
+
when scan(/u(\h{1,3})/) then
|
1046
1083
|
rb_compile_error "Invalid escape character syntax"
|
1084
|
+
when scan(/u\{(\h+(?:\s+\h+)*)\}/) then
|
1085
|
+
ss[1].split.map { |s| s.to_i(16) }.pack("U*")
|
1047
1086
|
when scan(/[McCx0-9]/) || end_of_stream? then
|
1048
1087
|
rb_compile_error("Invalid escape character syntax")
|
1049
1088
|
else
|
1050
|
-
|
1089
|
+
getch
|
1051
1090
|
end.dup
|
1052
1091
|
end
|
1053
1092
|
|
1093
|
+
def getch
|
1094
|
+
c = ss.getch
|
1095
|
+
c = ss.getch if c == "\r" && ss.peek(1) == "\n"
|
1096
|
+
c
|
1097
|
+
end
|
1098
|
+
|
1054
1099
|
def regx_options # TODO: rewrite / remove
|
1055
1100
|
good, bad = [], []
|
1056
1101
|
|
@@ -1106,6 +1151,10 @@ class RubyLexer
|
|
1106
1151
|
parser.class.version <= 24
|
1107
1152
|
end
|
1108
1153
|
|
1154
|
+
def ruby27plus?
|
1155
|
+
parser.class.version >= 27
|
1156
|
+
end
|
1157
|
+
|
1109
1158
|
def scan re
|
1110
1159
|
ss.scan re
|
1111
1160
|
end
|
@@ -1274,10 +1323,12 @@ class RubyLexer
|
|
1274
1323
|
s
|
1275
1324
|
when /^[McCx0-9]/ then
|
1276
1325
|
rb_compile_error("Invalid escape character syntax")
|
1277
|
-
when /u(
|
1326
|
+
when /u(\h{4})/ then
|
1278
1327
|
[$1.delete("{}").to_i(16)].pack("U")
|
1279
|
-
when /u(
|
1328
|
+
when /u(\h{1,3})/ then
|
1280
1329
|
rb_compile_error("Invalid escape character syntax")
|
1330
|
+
when /u\{(\h+(?:\s+\h+)*)\}/ then
|
1331
|
+
$1.split.map { |s| s.to_i(16) }.pack("U*")
|
1281
1332
|
else
|
1282
1333
|
s
|
1283
1334
|
end
|
@@ -1355,11 +1406,11 @@ class RubyLexer
|
|
1355
1406
|
# extra fake lex_state names to make things a bit cleaner
|
1356
1407
|
|
1357
1408
|
EXPR_LAB = EXPR_ARG|EXPR_LABELED
|
1358
|
-
|
1409
|
+
EXPR_LIT = EXPR_END|EXPR_ENDARG
|
1359
1410
|
EXPR_PAR = EXPR_BEG|EXPR_LABEL
|
1360
1411
|
EXPR_PAD = EXPR_BEG|EXPR_LABELED
|
1361
1412
|
|
1362
|
-
|
1413
|
+
EXPR_NUM = EXPR_LIT
|
1363
1414
|
|
1364
1415
|
expr_names.merge!(EXPR_NONE => "EXPR_NONE",
|
1365
1416
|
EXPR_BEG => "EXPR_BEG",
|
data/lib/ruby_lexer.rex
CHANGED
@@ -6,9 +6,9 @@ class RubyLexer
|
|
6
6
|
|
7
7
|
macro
|
8
8
|
|
9
|
-
|
9
|
+
IDENT_CHAR /[a-zA-Z0-9_[:^ascii:]]/
|
10
10
|
|
11
|
-
ESC /\\((?>[0-7]{1,3}|x
|
11
|
+
ESC /\\((?>[0-7]{1,3}|x\h{1,2}|M-[^\\]|(C-|c)[^\\]|u\h{1,4}|u\{\h+(?:\s+\h+)*\}|[^0-7xMCc]))/
|
12
12
|
SIMPLE_STRING /((#{ESC}|\#(#{ESC}|[^\{\#\@\$\"\\])|[^\"\\\#])*)/o
|
13
13
|
SSTRING /((\\.|[^\'])*)/
|
14
14
|
|
@@ -48,7 +48,7 @@ rule
|
|
48
48
|
| /\![=~]?/ { result :arg_state, TOKENS[text], text }
|
49
49
|
|
50
50
|
: /\./
|
51
|
-
| /\.\.\.?/
|
51
|
+
| /\.\.\.?/ process_dots
|
52
52
|
| /\.\d/ { rb_compile_error "no .<digit> floating literal anymore put 0 before dot" }
|
53
53
|
| /\./ { self.lex_state = EXPR_BEG; result EXPR_DOT, :tDOT, "." }
|
54
54
|
|
@@ -62,7 +62,7 @@ rule
|
|
62
62
|
| /\=(?=begin\b)/ { result arg_state, TOKENS[text], text }
|
63
63
|
|
64
64
|
ruby22_label? /\"#{SIMPLE_STRING}\":/o process_label
|
65
|
-
/\"(#{SIMPLE_STRING})\"/o
|
65
|
+
/\"(#{SIMPLE_STRING})\"/o process_simple_string
|
66
66
|
/\"/ { string STR_DQUOTE; result nil, :tSTRING_BEG, text }
|
67
67
|
|
68
68
|
/\@\@?\d/ { rb_compile_error "`#{text}` is not allowed as a variable name" }
|
@@ -164,13 +164,12 @@ was_label? /\'#{SSTRING}\':?/o process_label_or_string
|
|
164
164
|
| in_fname? /\$([1-9]\d*)/ process_gvar
|
165
165
|
| /\$([1-9]\d*)/ process_nthref
|
166
166
|
| /\$0/ process_gvar
|
167
|
-
|
|
167
|
+
| /\$#{IDENT_CHAR}+/ process_gvar
|
168
168
|
| /\$\W|\$\z/ process_gvar_oddity
|
169
|
-
| /\$\w+/ process_gvar
|
170
169
|
|
171
170
|
/\_/ process_underscore
|
172
171
|
|
173
|
-
/#{
|
172
|
+
/#{IDENT_CHAR}+/o process_token
|
174
173
|
|
175
174
|
/\004|\032|\000|\Z/ { [RubyLexer::EOF, RubyLexer::EOF] }
|
176
175
|
|
data/lib/ruby_lexer.rex.rb
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
# encoding: UTF-8
|
2
2
|
#--
|
3
3
|
# This file is automatically generated. Do not modify it.
|
4
|
-
# Generated by: oedipus_lex version 2.5.
|
4
|
+
# Generated by: oedipus_lex version 2.5.3.
|
5
5
|
# Source: lib/ruby_lexer.rex
|
6
6
|
#++
|
7
7
|
|
@@ -16,8 +16,8 @@ class RubyLexer
|
|
16
16
|
require 'strscan'
|
17
17
|
|
18
18
|
# :stopdoc:
|
19
|
-
|
20
|
-
ESC = /\\((?>[0-7]{1,3}|x
|
19
|
+
IDENT_CHAR = /[a-zA-Z0-9_[:^ascii:]]/
|
20
|
+
ESC = /\\((?>[0-7]{1,3}|x\h{1,2}|M-[^\\]|(C-|c)[^\\]|u\h{1,4}|u\{\h+(?:\s+\h+)*\}|[^0-7xMCc]))/
|
21
21
|
SIMPLE_STRING = /((#{ESC}|\#(#{ESC}|[^\{\#\@\$\"\\])|[^\"\\\#])*)/o
|
22
22
|
SSTRING = /((\\.|[^\'])*)/
|
23
23
|
INT_DEC = /[+]?(?:(?:[1-9][\d_]*|0)(?!\.\d)(ri|r|i)?\b|0d[0-9_]+)(ri|r|i)?/i
|
@@ -138,7 +138,7 @@ class RubyLexer
|
|
138
138
|
when ss.match?(/\./) then
|
139
139
|
case
|
140
140
|
when text = ss.scan(/\.\.\.?/) then
|
141
|
-
|
141
|
+
process_dots text
|
142
142
|
when ss.skip(/\.\d/) then
|
143
143
|
action { rb_compile_error "no .<digit> floating literal anymore put 0 before dot" }
|
144
144
|
when ss.skip(/\./) then
|
@@ -160,7 +160,7 @@ class RubyLexer
|
|
160
160
|
when ruby22_label? && (text = ss.scan(/\"#{SIMPLE_STRING}\":/o)) then
|
161
161
|
process_label text
|
162
162
|
when text = ss.scan(/\"(#{SIMPLE_STRING})\"/o) then
|
163
|
-
|
163
|
+
process_simple_string text
|
164
164
|
when text = ss.scan(/\"/) then
|
165
165
|
action { string STR_DQUOTE; result nil, :tSTRING_BEG, text }
|
166
166
|
when text = ss.scan(/\@\@?\d/) then
|
@@ -328,16 +328,14 @@ class RubyLexer
|
|
328
328
|
process_nthref text
|
329
329
|
when text = ss.scan(/\$0/) then
|
330
330
|
process_gvar text
|
331
|
-
when text = ss.scan(
|
331
|
+
when text = ss.scan(/\$#{IDENT_CHAR}+/) then
|
332
332
|
process_gvar text
|
333
333
|
when text = ss.scan(/\$\W|\$\z/) then
|
334
334
|
process_gvar_oddity text
|
335
|
-
when text = ss.scan(/\$\w+/) then
|
336
|
-
process_gvar text
|
337
335
|
end # group /\$/
|
338
336
|
when text = ss.scan(/\_/) then
|
339
337
|
process_underscore text
|
340
|
-
when text = ss.scan(/#{
|
338
|
+
when text = ss.scan(/#{IDENT_CHAR}+/o) then
|
341
339
|
process_token text
|
342
340
|
when ss.skip(/\004|\032|\000|\Z/) then
|
343
341
|
action { [RubyLexer::EOF, RubyLexer::EOF] }
|
data/lib/ruby_parser.rb
CHANGED
@@ -78,10 +78,14 @@ require "ruby23_parser"
|
|
78
78
|
require "ruby24_parser"
|
79
79
|
require "ruby25_parser"
|
80
80
|
require "ruby26_parser"
|
81
|
+
require "ruby27_parser"
|
82
|
+
require "ruby30_parser"
|
81
83
|
|
82
84
|
class RubyParser # HACK
|
83
85
|
VERSIONS.clear # also a HACK caused by racc namespace issues
|
84
86
|
|
87
|
+
class V30 < ::Ruby30Parser; end
|
88
|
+
class V27 < ::Ruby27Parser; end
|
85
89
|
class V26 < ::Ruby26Parser; end
|
86
90
|
class V25 < ::Ruby25Parser; end
|
87
91
|
class V24 < ::Ruby24Parser; end
|
data/lib/ruby_parser.yy
CHANGED
@@ -14,6 +14,10 @@ class Ruby24Parser
|
|
14
14
|
class Ruby25Parser
|
15
15
|
#elif V == 26
|
16
16
|
class Ruby26Parser
|
17
|
+
#elif V == 27
|
18
|
+
class Ruby27Parser
|
19
|
+
#elif V == 30
|
20
|
+
class Ruby30Parser
|
17
21
|
#else
|
18
22
|
fail "version not specified or supported on code generation"
|
19
23
|
#endif
|
@@ -44,6 +48,9 @@ token kCLASS kMODULE kDEF kUNDEF kBEGIN kRESCUE kENSURE kEND kIF kUNLESS
|
|
44
48
|
#if V >= 23
|
45
49
|
tLONELY
|
46
50
|
#endif
|
51
|
+
#if V >= 26
|
52
|
+
tBDOT2 tBDOT3
|
53
|
+
#endif
|
47
54
|
|
48
55
|
preclow
|
49
56
|
nonassoc tLOWEST
|
@@ -55,7 +62,7 @@ preclow
|
|
55
62
|
right tEQL tOP_ASGN
|
56
63
|
left kRESCUE_MOD
|
57
64
|
right tEH tCOLON
|
58
|
-
nonassoc tDOT2 tDOT3
|
65
|
+
nonassoc tDOT2 tDOT3 tBDOT2 tBDOT3
|
59
66
|
left tOROP
|
60
67
|
left tANDOP
|
61
68
|
nonassoc tCMP tEQ tEQQ tNEQ tMATCH tNMATCH
|
@@ -78,6 +85,9 @@ rule
|
|
78
85
|
top_compstmt
|
79
86
|
{
|
80
87
|
result = new_compstmt val
|
88
|
+
|
89
|
+
lexer.cond.pop # local_pop
|
90
|
+
lexer.cmdarg.pop
|
81
91
|
}
|
82
92
|
|
83
93
|
top_compstmt: top_stmts opt_terms
|
@@ -286,13 +296,15 @@ rule
|
|
286
296
|
}
|
287
297
|
| primary_value tCOLON2 tCONSTANT tOP_ASGN command_rhs
|
288
298
|
{
|
289
|
-
|
290
|
-
|
299
|
+
lhs1, _, lhs2, op, rhs = val
|
300
|
+
|
301
|
+
result = s(:op_asgn, lhs1, rhs, lhs2.to_sym, op.to_sym)
|
291
302
|
}
|
292
303
|
| primary_value tCOLON2 tIDENTIFIER tOP_ASGN command_rhs
|
293
304
|
{
|
294
|
-
|
295
|
-
|
305
|
+
lhs1, _, lhs2, op, rhs = val
|
306
|
+
|
307
|
+
result = s(:op_asgn, lhs1, rhs, lhs2.to_sym, op.to_sym)
|
296
308
|
}
|
297
309
|
| backref tOP_ASGN command_rhs
|
298
310
|
{
|
@@ -852,6 +864,24 @@ rule
|
|
852
864
|
result = s(:dot3, v1, v2).line v1.line
|
853
865
|
}
|
854
866
|
#endif
|
867
|
+
|
868
|
+
#if V >= 27
|
869
|
+
| tBDOT2 arg
|
870
|
+
{
|
871
|
+
_, v2, = val
|
872
|
+
v1 = nil
|
873
|
+
|
874
|
+
result = s(:dot2, v1, v2).line v2.line
|
875
|
+
}
|
876
|
+
| tBDOT3 arg
|
877
|
+
{
|
878
|
+
_, v2 = val
|
879
|
+
v1 = nil
|
880
|
+
|
881
|
+
result = s(:dot3, v1, v2).line v2.line
|
882
|
+
}
|
883
|
+
#endif
|
884
|
+
|
855
885
|
| arg tPLUS arg
|
856
886
|
{
|
857
887
|
result = new_call val[0], :+, argl(val[2])
|
@@ -1036,6 +1066,26 @@ rule
|
|
1036
1066
|
_, args, _ = val
|
1037
1067
|
result = args
|
1038
1068
|
}
|
1069
|
+
#if V >= 27
|
1070
|
+
| tLPAREN2 args tCOMMA args_forward rparen
|
1071
|
+
{
|
1072
|
+
yyerror "Unexpected ..." unless
|
1073
|
+
self.lexer.is_local_id(:"*") &&
|
1074
|
+
self.lexer.is_local_id(:"**") &&
|
1075
|
+
self.lexer.is_local_id(:"&")
|
1076
|
+
|
1077
|
+
result = call_args val
|
1078
|
+
}
|
1079
|
+
| tLPAREN2 args_forward rparen
|
1080
|
+
{
|
1081
|
+
yyerror "Unexpected ..." unless
|
1082
|
+
self.lexer.is_local_id(:"*") &&
|
1083
|
+
self.lexer.is_local_id(:"**") &&
|
1084
|
+
self.lexer.is_local_id(:"&")
|
1085
|
+
|
1086
|
+
result = call_args val
|
1087
|
+
}
|
1088
|
+
#endif
|
1039
1089
|
|
1040
1090
|
opt_paren_args: none
|
1041
1091
|
| paren_args
|
@@ -1078,12 +1128,39 @@ rule
|
|
1078
1128
|
}
|
1079
1129
|
|
1080
1130
|
command_args: {
|
1081
|
-
|
1131
|
+
# parse26.y line 2200
|
1132
|
+
|
1133
|
+
# If call_args starts with a open paren '(' or
|
1134
|
+
# '[', look-ahead reading of the letters calls
|
1135
|
+
# CMDARG_PUSH(0), but the push must be done
|
1136
|
+
# after CMDARG_PUSH(1). So this code makes them
|
1137
|
+
# consistent by first cancelling the premature
|
1138
|
+
# CMDARG_PUSH(0), doing CMDARG_PUSH(1), and
|
1139
|
+
# finally redoing CMDARG_PUSH(0).
|
1140
|
+
|
1141
|
+
result = yychar = self.last_token_type.first
|
1142
|
+
lookahead = [:tLPAREN, :tLPAREN_ARG, :tLPAREN2, :tLBRACK, :tLBRACK2].include?(yychar)
|
1143
|
+
lexer.cmdarg.pop if lookahead
|
1144
|
+
lexer.cmdarg.push true
|
1145
|
+
lexer.cmdarg.push false if lookahead
|
1082
1146
|
}
|
1083
1147
|
call_args
|
1084
1148
|
{
|
1085
|
-
|
1086
|
-
|
1149
|
+
yychar, args = val
|
1150
|
+
|
1151
|
+
# call_args can be followed by tLBRACE_ARG (that
|
1152
|
+
# does CMDARG_PUSH(0) in the lexer) but the push
|
1153
|
+
# must be done after CMDARG_POP() in the parser.
|
1154
|
+
# So this code does CMDARG_POP() to pop 0 pushed
|
1155
|
+
# by tLBRACE_ARG, CMDARG_POP() to pop 1 pushed
|
1156
|
+
# by command_args, and CMDARG_PUSH(0) to restore
|
1157
|
+
# back the flag set by tLBRACE_ARG.
|
1158
|
+
|
1159
|
+
lookahead = [:tLBRACE_ARG].include?(yychar)
|
1160
|
+
lexer.cmdarg.pop if lookahead
|
1161
|
+
lexer.cmdarg.pop
|
1162
|
+
lexer.cmdarg.push false if lookahead
|
1163
|
+
result = args
|
1087
1164
|
}
|
1088
1165
|
|
1089
1166
|
block_arg: tAMPER arg_value
|
@@ -1101,8 +1178,9 @@ rule
|
|
1101
1178
|
args: arg_value
|
1102
1179
|
{
|
1103
1180
|
arg, = val
|
1181
|
+
lineno = arg.line || lexer.lineno # HACK
|
1104
1182
|
|
1105
|
-
result = s(:array, arg).line
|
1183
|
+
result = s(:array, arg).line lineno
|
1106
1184
|
}
|
1107
1185
|
| tSTAR arg_value
|
1108
1186
|
{
|
@@ -1114,9 +1192,11 @@ rule
|
|
1114
1192
|
args, _, id = val
|
1115
1193
|
result = self.list_append args, id
|
1116
1194
|
}
|
1117
|
-
| args tCOMMA tSTAR
|
1195
|
+
| args tCOMMA tSTAR arg_value
|
1118
1196
|
{
|
1119
|
-
|
1197
|
+
# TODO: the line number from tSTAR has been dropped
|
1198
|
+
args, _, _, id = val
|
1199
|
+
line = lexer.lineno
|
1120
1200
|
result = self.list_append args, s(:splat, id).line(line)
|
1121
1201
|
}
|
1122
1202
|
|
@@ -1137,7 +1217,6 @@ rule
|
|
1137
1217
|
}
|
1138
1218
|
| args tCOMMA tSTAR arg_value
|
1139
1219
|
{
|
1140
|
-
# FIX: bad shift/reduce conflict with rhs' comma star prod
|
1141
1220
|
# TODO: make all tXXXX terminals include lexer.lineno
|
1142
1221
|
arg, _, _, splat = val
|
1143
1222
|
result = self.arg_concat arg, splat
|
@@ -1165,21 +1244,13 @@ rule
|
|
1165
1244
|
}
|
1166
1245
|
| k_begin
|
1167
1246
|
{
|
1247
|
+
lexer.cmdarg.push false
|
1168
1248
|
result = self.lexer.lineno
|
1169
|
-
# TODO:
|
1170
|
-
# $<val>1 = cmdarg_stack;
|
1171
|
-
# CMDARG_SET(0);
|
1172
1249
|
}
|
1173
1250
|
bodystmt k_end
|
1174
1251
|
{
|
1175
|
-
|
1176
|
-
|
1177
|
-
result = s(:nil)
|
1178
|
-
else
|
1179
|
-
result = s(:begin, val[2])
|
1180
|
-
end
|
1181
|
-
|
1182
|
-
result.line = val[1]
|
1252
|
+
lexer.cmdarg.pop
|
1253
|
+
result = new_begin val
|
1183
1254
|
}
|
1184
1255
|
| tLPAREN_ARG
|
1185
1256
|
{
|
@@ -1192,18 +1263,14 @@ rule
|
|
1192
1263
|
result = s(:begin).line line
|
1193
1264
|
}
|
1194
1265
|
| tLPAREN_ARG
|
1195
|
-
{
|
1196
|
-
result = lexer.cmdarg.store false
|
1197
|
-
}
|
1198
1266
|
stmt
|
1199
1267
|
{
|
1200
1268
|
lexer.lex_state = EXPR_ENDARG
|
1201
1269
|
}
|
1202
1270
|
rparen
|
1203
1271
|
{
|
1204
|
-
_,
|
1205
|
-
warning "(...) interpreted as grouped expression"
|
1206
|
-
lexer.cmdarg.restore cmdarg
|
1272
|
+
_, stmt, _, _, = val
|
1273
|
+
# warning "(...) interpreted as grouped expression"
|
1207
1274
|
result = stmt
|
1208
1275
|
}
|
1209
1276
|
| tLPAREN compstmt tRPAREN
|
@@ -1386,48 +1453,61 @@ rule
|
|
1386
1453
|
}
|
1387
1454
|
| k_def fname
|
1388
1455
|
{
|
1389
|
-
result =
|
1456
|
+
result = self.in_def
|
1390
1457
|
|
1391
|
-
self.
|
1392
|
-
self.in_def = true
|
1458
|
+
self.in_def = true # group = local_push
|
1393
1459
|
self.env.extend
|
1394
|
-
|
1395
|
-
|
1396
|
-
|
1460
|
+
lexer.cmdarg.push false
|
1461
|
+
lexer.cond.push false
|
1462
|
+
|
1463
|
+
self.comments.push self.lexer.comments
|
1397
1464
|
}
|
1398
1465
|
f_arglist bodystmt { result = lexer.lineno } k_end
|
1399
1466
|
{
|
1400
|
-
in_def
|
1467
|
+
in_def = val[2]
|
1401
1468
|
|
1402
1469
|
result = new_defn val
|
1403
1470
|
|
1404
|
-
lexer.
|
1471
|
+
lexer.cond.pop # group = local_pop
|
1472
|
+
lexer.cmdarg.pop
|
1405
1473
|
self.env.unextend
|
1406
1474
|
self.in_def = in_def
|
1475
|
+
|
1407
1476
|
self.lexer.comments # we don't care about comments in the body
|
1408
1477
|
}
|
1409
1478
|
| k_def singleton dot_or_colon
|
1410
1479
|
{
|
1411
|
-
self.comments.push self.lexer.comments
|
1412
1480
|
lexer.lex_state = EXPR_FNAME
|
1413
1481
|
}
|
1414
1482
|
fname
|
1415
1483
|
{
|
1416
|
-
self.
|
1484
|
+
result = [self.in_def, lexer.lineno]
|
1485
|
+
|
1486
|
+
self.in_single += 1 # TODO: remove?
|
1487
|
+
|
1488
|
+
self.in_def = true # local_push
|
1417
1489
|
self.env.extend
|
1418
|
-
lexer.
|
1419
|
-
|
1420
|
-
|
1490
|
+
lexer.cmdarg.push false
|
1491
|
+
lexer.cond.push false
|
1492
|
+
|
1493
|
+
lexer.lex_state = EXPR_ENDFN|EXPR_LABEL
|
1494
|
+
self.comments.push self.lexer.comments
|
1421
1495
|
}
|
1422
1496
|
f_arglist bodystmt k_end
|
1423
1497
|
{
|
1424
|
-
_,
|
1425
|
-
result = new_defs val
|
1498
|
+
_, _recv, _, _, _name, (in_def, _lineno), _args, _body, _ = val
|
1426
1499
|
|
1427
|
-
|
1500
|
+
result = new_defs val
|
1428
1501
|
|
1502
|
+
lexer.cond.pop # group = local_pop
|
1503
|
+
lexer.cmdarg.pop
|
1429
1504
|
self.env.unextend
|
1505
|
+
self.in_def = in_def
|
1506
|
+
|
1430
1507
|
self.in_single -= 1
|
1508
|
+
|
1509
|
+
# TODO: restore cur_arg ? what's cur_arg?
|
1510
|
+
|
1431
1511
|
self.lexer.comments # we don't care about comments in the body
|
1432
1512
|
}
|
1433
1513
|
| kBREAK
|
@@ -1712,20 +1792,19 @@ opt_block_args_tail: tCOMMA block_args_tail
|
|
1712
1792
|
}
|
1713
1793
|
f_larglist
|
1714
1794
|
{
|
1715
|
-
|
1795
|
+
lexer.cmdarg.push false
|
1716
1796
|
}
|
1717
1797
|
lambda_body
|
1718
1798
|
{
|
1719
|
-
(line, lpar), args,
|
1799
|
+
(line, lpar), args, _cmdarg, body = val
|
1720
1800
|
lexer.lpar_beg = lpar
|
1721
1801
|
|
1722
|
-
lexer.cmdarg.
|
1723
|
-
lexer.cmdarg.lexpop
|
1802
|
+
lexer.cmdarg.pop
|
1724
1803
|
|
1725
1804
|
call = s(:lambda).line line
|
1726
1805
|
result = new_iter call, args, body
|
1727
1806
|
result.line = line
|
1728
|
-
self.env.unextend
|
1807
|
+
self.env.unextend # TODO: dynapush & dynapop
|
1729
1808
|
}
|
1730
1809
|
|
1731
1810
|
f_larglist: tLPAREN2 f_args opt_bv_decl rparen
|
@@ -1878,7 +1957,7 @@ opt_block_args_tail: tCOMMA block_args_tail
|
|
1878
1957
|
}
|
1879
1958
|
|
1880
1959
|
do_body: { self.env.extend :dynamic; result = self.lexer.lineno }
|
1881
|
-
{
|
1960
|
+
{ lexer.cmdarg.push false }
|
1882
1961
|
opt_block_param
|
1883
1962
|
#if V >= 25
|
1884
1963
|
bodystmt
|
@@ -1886,11 +1965,11 @@ opt_block_args_tail: tCOMMA block_args_tail
|
|
1886
1965
|
compstmt
|
1887
1966
|
#endif
|
1888
1967
|
{
|
1889
|
-
line,
|
1968
|
+
line, _cmdarg, param, cmpstmt = val
|
1890
1969
|
|
1891
1970
|
result = new_do_body param, cmpstmt, line
|
1971
|
+
lexer.cmdarg.pop
|
1892
1972
|
self.env.unextend
|
1893
|
-
lexer.cmdarg.restore cmdarg
|
1894
1973
|
}
|
1895
1974
|
|
1896
1975
|
case_body: k_when
|
@@ -1990,7 +2069,7 @@ opt_block_args_tail: tCOMMA block_args_tail
|
|
1990
2069
|
|
1991
2070
|
xstring: tXSTRING_BEG xstring_contents tSTRING_END
|
1992
2071
|
{
|
1993
|
-
result = new_xstring val
|
2072
|
+
result = new_xstring val
|
1994
2073
|
# TODO: dedent?!?! SERIOUSLY?!?
|
1995
2074
|
}
|
1996
2075
|
|
@@ -2132,12 +2211,13 @@ regexp_contents: none
|
|
2132
2211
|
result = [lexer.lex_strterm,
|
2133
2212
|
lexer.brace_nest,
|
2134
2213
|
lexer.string_nest, # TODO: remove
|
2135
|
-
lexer.cond.store,
|
2136
|
-
lexer.cmdarg.store,
|
2137
2214
|
lexer.lex_state,
|
2138
2215
|
lexer.lineno,
|
2139
2216
|
]
|
2140
2217
|
|
2218
|
+
lexer.cmdarg.push false
|
2219
|
+
lexer.cond.push false
|
2220
|
+
|
2141
2221
|
lexer.lex_strterm = nil
|
2142
2222
|
lexer.brace_nest = 0
|
2143
2223
|
lexer.string_nest = 0
|
@@ -2149,14 +2229,15 @@ regexp_contents: none
|
|
2149
2229
|
{
|
2150
2230
|
_, memo, stmt, _ = val
|
2151
2231
|
|
2152
|
-
lex_strterm, brace_nest, string_nest,
|
2232
|
+
lex_strterm, brace_nest, string_nest, oldlex_state, line = memo
|
2233
|
+
# TODO: heredoc_indent
|
2153
2234
|
|
2154
2235
|
lexer.lex_strterm = lex_strterm
|
2155
2236
|
lexer.brace_nest = brace_nest
|
2156
2237
|
lexer.string_nest = string_nest
|
2157
2238
|
|
2158
|
-
lexer.
|
2159
|
-
lexer.
|
2239
|
+
lexer.cmdarg.pop
|
2240
|
+
lexer.cond.pop
|
2160
2241
|
|
2161
2242
|
lexer.lex_state = oldlex_state
|
2162
2243
|
|
@@ -2305,6 +2386,22 @@ keyword_variable: kNIL { result = s(:nil).line lexer.lineno }
|
|
2305
2386
|
self.lexer.lex_state = EXPR_BEG
|
2306
2387
|
self.lexer.command_start = true
|
2307
2388
|
}
|
2389
|
+
#if V >= 27
|
2390
|
+
| tLPAREN2 f_arg tCOMMA args_forward rparen
|
2391
|
+
{
|
2392
|
+
result = args val
|
2393
|
+
|
2394
|
+
self.lexer.lex_state = EXPR_BEG
|
2395
|
+
self.lexer.command_start = true
|
2396
|
+
}
|
2397
|
+
| tLPAREN2 args_forward rparen
|
2398
|
+
{
|
2399
|
+
result = args val
|
2400
|
+
|
2401
|
+
self.lexer.lex_state = EXPR_BEG
|
2402
|
+
self.lexer.command_start = true
|
2403
|
+
}
|
2404
|
+
#endif
|
2308
2405
|
| {
|
2309
2406
|
result = self.in_kwarg
|
2310
2407
|
self.in_kwarg = true
|
@@ -2404,6 +2501,13 @@ keyword_variable: kNIL { result = s(:nil).line lexer.lineno }
|
|
2404
2501
|
result = args val
|
2405
2502
|
}
|
2406
2503
|
|
2504
|
+
#if V >= 27
|
2505
|
+
args_forward: tBDOT3
|
2506
|
+
{
|
2507
|
+
result = s(:forward_args).line lexer.lineno
|
2508
|
+
}
|
2509
|
+
#endif
|
2510
|
+
|
2407
2511
|
f_bad_arg: tCONSTANT
|
2408
2512
|
{
|
2409
2513
|
yyerror "formal argument cannot be a constant"
|
@@ -2552,6 +2656,7 @@ keyword_variable: kNIL { result = s(:nil).line lexer.lineno }
|
|
2552
2656
|
| kwrest_mark
|
2553
2657
|
{
|
2554
2658
|
result = :"**"
|
2659
|
+
self.env[result] = :lvar
|
2555
2660
|
}
|
2556
2661
|
|
2557
2662
|
#if V == 20
|