ruby_parser 3.0.0.a8 → 3.0.0.a9
Sign up to get free protection for your applications and to get access to all the features.
Potentially problematic release.
This version of ruby_parser might be problematic. Click here for more details.
- data.tar.gz.sig +0 -0
- data/.autotest +6 -0
- data/History.txt +40 -0
- data/README.txt +6 -11
- data/Rakefile +5 -0
- data/bin/ruby_parse_extract_error +23 -6
- data/lib/ruby18_parser.rb +18 -8
- data/lib/ruby18_parser.y +18 -8
- data/lib/ruby19_parser.rb +67 -35
- data/lib/ruby19_parser.y +62 -32
- data/lib/ruby_lexer.rb +84 -95
- data/lib/ruby_parser_extras.rb +140 -20
- data/test/test_ruby_lexer.rb +58 -3
- data/test/test_ruby_parser.rb +193 -2
- data/test/test_ruby_parser_extras.rb +109 -0
- metadata +17 -22
- metadata.gz.sig +5 -2
data/lib/ruby19_parser.y
CHANGED
@@ -239,7 +239,7 @@ rule
|
|
239
239
|
| block_command
|
240
240
|
|
241
241
|
block_command: block_call
|
242
|
-
| block_call tDOT operation2 command_args
|
242
|
+
| block_call tDOT operation2 command_args # TODO: dot_or_colon
|
243
243
|
{
|
244
244
|
result = new_call val[0], val[2], val[3]
|
245
245
|
}
|
@@ -271,9 +271,8 @@ rule
|
|
271
271
|
{
|
272
272
|
result = new_call nil, val[0].to_sym, val[1]
|
273
273
|
if val[2] then
|
274
|
-
|
275
|
-
|
276
|
-
end
|
274
|
+
block_dup_check result, val[2]
|
275
|
+
|
277
276
|
result, operation = val[2], result
|
278
277
|
result.insert 1, operation
|
279
278
|
end
|
@@ -284,7 +283,13 @@ rule
|
|
284
283
|
}
|
285
284
|
| primary_value tDOT operation2 command_args cmd_brace_block
|
286
285
|
{
|
287
|
-
|
286
|
+
recv, _, msg, args, block = val
|
287
|
+
call = new_call recv, msg.to_sym, args
|
288
|
+
|
289
|
+
block_dup_check call, block
|
290
|
+
|
291
|
+
block.insert 1, call
|
292
|
+
result = block
|
288
293
|
}
|
289
294
|
| primary_value tCOLON2 operation2 command_args =tLOWEST
|
290
295
|
{
|
@@ -292,14 +297,13 @@ rule
|
|
292
297
|
}
|
293
298
|
| primary_value tCOLON2 operation2 command_args cmd_brace_block
|
294
299
|
{
|
295
|
-
|
296
|
-
|
297
|
-
|
298
|
-
|
299
|
-
|
300
|
-
|
301
|
-
|
302
|
-
end
|
300
|
+
recv, _, msg, args, block = val
|
301
|
+
call = new_call recv, msg.to_sym, args
|
302
|
+
|
303
|
+
block_dup_check call, block
|
304
|
+
|
305
|
+
block.insert 1, call
|
306
|
+
result = block
|
303
307
|
}
|
304
308
|
| kSUPER command_args
|
305
309
|
{
|
@@ -838,7 +842,7 @@ rule
|
|
838
842
|
| block_arg
|
839
843
|
|
840
844
|
command_args: {
|
841
|
-
result = lexer.cmdarg.stack.dup
|
845
|
+
result = lexer.cmdarg.stack.dup # TODO: smell?
|
842
846
|
lexer.cmdarg.push true
|
843
847
|
}
|
844
848
|
call_args
|
@@ -970,7 +974,7 @@ rule
|
|
970
974
|
}
|
971
975
|
| kNOT tLPAREN2 rparen
|
972
976
|
{
|
973
|
-
raise "no3
|
977
|
+
raise "no3\non#{val.inspect}"
|
974
978
|
}
|
975
979
|
| operation brace_block
|
976
980
|
{
|
@@ -984,7 +988,8 @@ rule
|
|
984
988
|
| method_call brace_block
|
985
989
|
{
|
986
990
|
call, iter = val[0], val[1]
|
987
|
-
|
991
|
+
block_dup_check call, iter
|
992
|
+
iter.insert 1, call # FIX
|
988
993
|
result = iter
|
989
994
|
}
|
990
995
|
| tLAMBDA lambda
|
@@ -1212,35 +1217,35 @@ rule
|
|
1212
1217
|
}
|
1213
1218
|
| f_marg_list tCOMMA tSTAR f_norm_arg
|
1214
1219
|
{
|
1215
|
-
|
1220
|
+
result = block_var val[0], val[3], nil
|
1216
1221
|
}
|
1217
1222
|
| f_marg_list tCOMMA tSTAR f_norm_arg tCOMMA f_marg_list
|
1218
1223
|
{
|
1219
|
-
raise "no10: #{val.inspect}"
|
1224
|
+
raise "no10\non: #{val.inspect}"
|
1220
1225
|
}
|
1221
1226
|
| f_marg_list tCOMMA tSTAR
|
1222
1227
|
{
|
1223
|
-
raise "no11: #{val.inspect}"
|
1228
|
+
raise "no11\non: #{val.inspect}"
|
1224
1229
|
}
|
1225
1230
|
| f_marg_list tCOMMA tSTAR tCOMMA f_marg_list
|
1226
1231
|
{
|
1227
|
-
raise "no12: #{val.inspect}"
|
1232
|
+
raise "no12\non: #{val.inspect}"
|
1228
1233
|
}
|
1229
1234
|
| tSTAR f_norm_arg
|
1230
1235
|
{
|
1231
|
-
raise "no13: #{val.inspect}"
|
1236
|
+
raise "no13\non: #{val.inspect}"
|
1232
1237
|
}
|
1233
1238
|
| tSTAR f_norm_arg tCOMMA f_marg_list
|
1234
1239
|
{
|
1235
|
-
raise "no14: #{val.inspect}"
|
1240
|
+
raise "no14\non: #{val.inspect}"
|
1236
1241
|
}
|
1237
1242
|
| tSTAR
|
1238
1243
|
{
|
1239
|
-
raise "no15: #{val.inspect}"
|
1244
|
+
raise "no15\non: #{val.inspect}"
|
1240
1245
|
}
|
1241
1246
|
| tSTAR tCOMMA f_marg_list
|
1242
1247
|
{
|
1243
|
-
raise "no16: #{val.inspect}"
|
1248
|
+
raise "no16\non: #{val.inspect}"
|
1244
1249
|
}
|
1245
1250
|
|
1246
1251
|
block_param: f_arg tCOMMA f_block_optarg tCOMMA f_rest_arg opt_f_block_arg
|
@@ -1253,7 +1258,12 @@ rule
|
|
1253
1258
|
}
|
1254
1259
|
| f_arg tCOMMA f_block_optarg opt_f_block_arg
|
1255
1260
|
{
|
1256
|
-
|
1261
|
+
arg, _, opt, block = val
|
1262
|
+
|
1263
|
+
result = arg
|
1264
|
+
result.concat opt[1..-1].map { |s| s[1] }
|
1265
|
+
result << "&#{block.last}".to_sym if block
|
1266
|
+
result << opt
|
1257
1267
|
}
|
1258
1268
|
| f_arg tCOMMA f_block_optarg tCOMMA f_arg opt_f_block_arg
|
1259
1269
|
{
|
@@ -1285,7 +1295,12 @@ rule
|
|
1285
1295
|
}
|
1286
1296
|
| f_block_optarg opt_f_block_arg
|
1287
1297
|
{
|
1288
|
-
|
1298
|
+
opt, block = val
|
1299
|
+
|
1300
|
+
result = s(:args)
|
1301
|
+
result.concat opt[1..-1].map { |s| s[1] }
|
1302
|
+
result << "&#{block.last}".to_sym if block
|
1303
|
+
result << opt
|
1289
1304
|
}
|
1290
1305
|
| f_block_optarg tCOMMA f_arg opt_f_block_arg
|
1291
1306
|
{
|
@@ -1297,7 +1312,11 @@ rule
|
|
1297
1312
|
}
|
1298
1313
|
| f_rest_arg tCOMMA f_arg opt_f_block_arg
|
1299
1314
|
{
|
1300
|
-
|
1315
|
+
rest, _, args, block = val
|
1316
|
+
|
1317
|
+
result = args
|
1318
|
+
result[1,0] = rest
|
1319
|
+
result << "&#{block.last}".to_sym if block
|
1301
1320
|
}
|
1302
1321
|
| f_block_arg
|
1303
1322
|
{
|
@@ -1329,10 +1348,13 @@ rule
|
|
1329
1348
|
}
|
1330
1349
|
|
1331
1350
|
bv_decls: bvar
|
1351
|
+
{
|
1352
|
+
result = [val[0]]
|
1353
|
+
}
|
1332
1354
|
| bv_decls tCOMMA bvar
|
1333
1355
|
{
|
1334
|
-
result = val[0]
|
1335
|
-
raise "no18: #{val.inspect}"
|
1356
|
+
result = val[0].concat val[2]
|
1357
|
+
raise "no18\non: #{val.inspect}"
|
1336
1358
|
}
|
1337
1359
|
|
1338
1360
|
bvar: tIDENTIFIER
|
@@ -1393,8 +1415,12 @@ rule
|
|
1393
1415
|
|
1394
1416
|
block_call: command do_block
|
1395
1417
|
{
|
1396
|
-
|
1397
|
-
|
1418
|
+
# TODO:
|
1419
|
+
# if (nd_type($1) == NODE_YIELD) {
|
1420
|
+
# compile_error(PARSER_ARG "block given to yield");
|
1421
|
+
|
1422
|
+
syntax_error "Both block arg and actual block given." if
|
1423
|
+
val[0].block_pass?
|
1398
1424
|
|
1399
1425
|
result = val[1]
|
1400
1426
|
result.insert 1, val[0]
|
@@ -1924,9 +1950,13 @@ keyword_variable: kNIL { result = s(:nil) }
|
|
1924
1950
|
}
|
1925
1951
|
|
1926
1952
|
f_block_optarg: f_block_opt
|
1953
|
+
{
|
1954
|
+
result = s(:block, val[0])
|
1955
|
+
}
|
1927
1956
|
| f_block_optarg tCOMMA f_block_opt
|
1928
1957
|
{
|
1929
|
-
|
1958
|
+
result = val[0]
|
1959
|
+
result << val[2]
|
1930
1960
|
}
|
1931
1961
|
|
1932
1962
|
f_optarg: f_opt
|
data/lib/ruby_lexer.rb
CHANGED
@@ -2,6 +2,8 @@
|
|
2
2
|
|
3
3
|
class RubyLexer
|
4
4
|
|
5
|
+
RUBY19 = "".respond_to? :encoding
|
6
|
+
|
5
7
|
IDENT_CHAR_RE = case RUBY_VERSION
|
6
8
|
when /^1\.8/ then
|
7
9
|
/[\w\x80-\xFF]/
|
@@ -115,7 +117,7 @@ class RubyLexer
|
|
115
117
|
end
|
116
118
|
|
117
119
|
def fix_arg_lex_state
|
118
|
-
self.lex_state = if
|
120
|
+
self.lex_state = if in_lex_state? :expr_fname, :expr_dot then
|
119
121
|
:expr_arg
|
120
122
|
else
|
121
123
|
:expr_beg
|
@@ -235,6 +237,10 @@ class RubyLexer
|
|
235
237
|
end
|
236
238
|
end
|
237
239
|
|
240
|
+
def in_lex_state?(*states)
|
241
|
+
states.include? lex_state
|
242
|
+
end
|
243
|
+
|
238
244
|
def initialize v = 18
|
239
245
|
self.version = v
|
240
246
|
self.cond = RubyParserStuff::StackState.new(:cond)
|
@@ -273,13 +279,13 @@ class RubyLexer
|
|
273
279
|
self.lex_state = :expr_end
|
274
280
|
|
275
281
|
case
|
276
|
-
when src.scan(/[+-]?0[
|
282
|
+
when src.scan(/[+-]?0[xXbBdD]\b/) then
|
277
283
|
rb_compile_error "Invalid numeric format"
|
278
284
|
when src.scan(/[+-]?0x[a-f0-9_]+/i) then
|
279
285
|
int_with_base(16)
|
280
|
-
when src.scan(/[+-]?
|
286
|
+
when src.scan(/[+-]?0[Bb][01_]+/) then
|
281
287
|
int_with_base(2)
|
282
|
-
when src.scan(/[+-]?
|
288
|
+
when src.scan(/[+-]?0[Dd][0-9_]+/) then
|
283
289
|
int_with_base(10)
|
284
290
|
when src.scan(/[+-]?0[Oo]?[0-7_]*[89]/) then
|
285
291
|
rb_compile_error "Illegal octal digit."
|
@@ -344,7 +350,7 @@ class RubyLexer
|
|
344
350
|
[:tSYMBEG, STR_SSYM]
|
345
351
|
end
|
346
352
|
|
347
|
-
rb_compile_error "Bad %string type. Expected [
|
353
|
+
rb_compile_error "Bad %string type. Expected [Qq\Wwxrs], found '#{c}'." if
|
348
354
|
token_type.nil?
|
349
355
|
|
350
356
|
self.lex_strterm = [:strterm, string_type, nnd, beg]
|
@@ -414,7 +420,7 @@ class RubyLexer
|
|
414
420
|
|
415
421
|
def rb_compile_error msg
|
416
422
|
msg += ". near line #{self.lineno}: #{src.rest[/^.*/].inspect}"
|
417
|
-
raise SyntaxError, msg
|
423
|
+
raise RubyParser::SyntaxError, msg
|
418
424
|
end
|
419
425
|
|
420
426
|
def read_escape # 51 lines
|
@@ -614,23 +620,24 @@ class RubyLexer
|
|
614
620
|
return c
|
615
621
|
end
|
616
622
|
|
617
|
-
|
623
|
+
ESCAPES = {
|
624
|
+
"a" => "\007",
|
625
|
+
"b" => "\010",
|
626
|
+
"e" => "\033",
|
627
|
+
"f" => "\f",
|
628
|
+
"n" => "\n",
|
629
|
+
"r" => "\r",
|
630
|
+
"s" => " ",
|
631
|
+
"t" => "\t",
|
632
|
+
"v" => "\13",
|
633
|
+
"\\" => '\\',
|
634
|
+
"\n" => "",
|
635
|
+
"C-\?" => 127.chr,
|
636
|
+
"c\?" => 127.chr,
|
637
|
+
}
|
618
638
|
|
619
|
-
|
620
|
-
|
621
|
-
"b" => "\010",
|
622
|
-
"e" => "\033",
|
623
|
-
"f" => "\f",
|
624
|
-
"n" => "\n",
|
625
|
-
"r" => "\r",
|
626
|
-
"s" => " ",
|
627
|
-
"t" => "\t",
|
628
|
-
"v" => "\13",
|
629
|
-
"\\" => '\\',
|
630
|
-
"\n" => "",
|
631
|
-
"C-\?" => 127.chr,
|
632
|
-
"c\?" => 127.chr,
|
633
|
-
}[s]
|
639
|
+
def unescape s
|
640
|
+
r = ESCAPES[s]
|
634
641
|
|
635
642
|
return r if r
|
636
643
|
|
@@ -684,25 +691,26 @@ class RubyLexer
|
|
684
691
|
self.lineno = nil
|
685
692
|
c = src.matched
|
686
693
|
if c == '#' then
|
687
|
-
# TODO: add magic comment handling?
|
688
|
-
|
689
694
|
src.pos -= 1
|
690
695
|
|
691
696
|
while src.scan(/\s*#.*(\n+|\z)/) do
|
692
697
|
@comments << src.matched.gsub(/^ +#/, '#').gsub(/^ +$/, '')
|
693
698
|
end
|
694
699
|
|
695
|
-
if src.eos?
|
696
|
-
return RubyLexer::EOF
|
697
|
-
end
|
700
|
+
return RubyLexer::EOF if src.eos?
|
698
701
|
end
|
699
702
|
|
700
703
|
# Replace a string of newlines with a single one
|
701
704
|
src.scan(/\n+/)
|
702
705
|
|
703
|
-
if
|
704
|
-
|
705
|
-
|
706
|
+
next if in_lex_state?(:expr_beg, :expr_fname, :expr_dot, :expr_class,
|
707
|
+
:expr_value)
|
708
|
+
|
709
|
+
if src.scan(/([\ \t\r\f\v]*)\./) then
|
710
|
+
self.space_seen = true unless src[1].empty?
|
711
|
+
|
712
|
+
src.pos -= 1
|
713
|
+
next unless src.check(/\.\./)
|
706
714
|
end
|
707
715
|
|
708
716
|
self.command_start = true
|
@@ -778,7 +786,7 @@ class RubyLexer
|
|
778
786
|
|
779
787
|
return process_token(command_state)
|
780
788
|
elsif src.scan(/\:\:/) then
|
781
|
-
if is_beg? ||
|
789
|
+
if is_beg? || in_lex_state?(:expr_class) || is_space_arg? then
|
782
790
|
self.lex_state = :expr_beg
|
783
791
|
self.yacc_value = "::"
|
784
792
|
return :tCOLON3
|
@@ -787,7 +795,7 @@ class RubyLexer
|
|
787
795
|
self.lex_state = :expr_dot
|
788
796
|
self.yacc_value = "::"
|
789
797
|
return :tCOLON2
|
790
|
-
elsif ! is_end? && src.scan(/:([a-zA-Z_]#{IDENT_CHAR_RE}*(?:[?!]|=(
|
798
|
+
elsif ! is_end? && src.scan(/:([a-zA-Z_]#{IDENT_CHAR_RE}*(?:[?!]|=(?==>)|=(?![=>]))?)/) then
|
791
799
|
# scanning shortcut to symbols
|
792
800
|
self.yacc_value = src[1]
|
793
801
|
self.lex_state = :expr_end
|
@@ -816,7 +824,7 @@ class RubyLexer
|
|
816
824
|
elsif src.scan(/\[/) then
|
817
825
|
result = src.matched
|
818
826
|
|
819
|
-
if
|
827
|
+
if in_lex_state? :expr_fname, :expr_dot then
|
820
828
|
self.lex_state = :expr_arg
|
821
829
|
case
|
822
830
|
when src.scan(/\]\=/) then
|
@@ -831,7 +839,7 @@ class RubyLexer
|
|
831
839
|
elsif is_beg? then
|
832
840
|
self.tern.push false
|
833
841
|
result = :tLBRACK
|
834
|
-
elsif
|
842
|
+
elsif is_arg? && space_seen then
|
835
843
|
self.tern.push false
|
836
844
|
result = :tLBRACK
|
837
845
|
else
|
@@ -870,9 +878,9 @@ class RubyLexer
|
|
870
878
|
return :tLAMBEG
|
871
879
|
end
|
872
880
|
|
873
|
-
result = if
|
881
|
+
result = if is_arg? || in_lex_state?(:expr_end) then
|
874
882
|
:tLCURLY # block (primary)
|
875
|
-
elsif
|
883
|
+
elsif in_lex_state?(:expr_endarg) then
|
876
884
|
:tLBRACE_ARG # block (expr)
|
877
885
|
else
|
878
886
|
self.tern.push false
|
@@ -895,7 +903,7 @@ class RubyLexer
|
|
895
903
|
[:tUMINUS, :tMINUS]
|
896
904
|
end
|
897
905
|
|
898
|
-
if
|
906
|
+
if in_lex_state? :expr_fname, :expr_dot then
|
899
907
|
self.lex_state = :expr_arg
|
900
908
|
if src.scan(/@/) then
|
901
909
|
self.yacc_value = "#{sign}@"
|
@@ -913,8 +921,8 @@ class RubyLexer
|
|
913
921
|
end
|
914
922
|
|
915
923
|
if (is_beg? ||
|
916
|
-
(
|
917
|
-
if
|
924
|
+
(is_arg? && space_seen && !src.check(/\s/))) then
|
925
|
+
if is_arg? then
|
918
926
|
arg_ambiguous
|
919
927
|
end
|
920
928
|
|
@@ -949,7 +957,7 @@ class RubyLexer
|
|
949
957
|
self.yacc_value = "*"
|
950
958
|
return :tOP_ASGN
|
951
959
|
elsif src.scan(/\*/) then
|
952
|
-
result = if
|
960
|
+
result = if is_arg? && space_seen && src.check(/\S/) then
|
953
961
|
warning("`*' interpreted as argument prefix")
|
954
962
|
:tSTAR
|
955
963
|
elsif is_beg? then
|
@@ -977,9 +985,9 @@ class RubyLexer
|
|
977
985
|
self.yacc_value = "\<\<"
|
978
986
|
return :tOP_ASGN
|
979
987
|
elsif src.scan(/\<\</) then
|
980
|
-
if (!
|
981
|
-
|
982
|
-
(!
|
988
|
+
if (! in_lex_state?(:expr_end, :expr_dot,
|
989
|
+
:expr_endarg, :expr_class) &&
|
990
|
+
(!is_arg? || space_seen)) then
|
983
991
|
tok = self.heredoc_identifier
|
984
992
|
if tok then
|
985
993
|
return tok
|
@@ -1030,8 +1038,9 @@ class RubyLexer
|
|
1030
1038
|
self.lex_strterm = [:strterm, STR_XQUOTE, '`', "\0"]
|
1031
1039
|
return :tXSTRING_BEG
|
1032
1040
|
elsif src.scan(/\?/) then
|
1033
|
-
|
1034
|
-
|
1041
|
+
|
1042
|
+
if is_end? then
|
1043
|
+
self.lex_state = ruby18 ? :expr_beg : :expr_value # HACK?
|
1035
1044
|
self.tern.push true
|
1036
1045
|
self.yacc_value = "?"
|
1037
1046
|
return :tEH
|
@@ -1042,7 +1051,7 @@ class RubyLexer
|
|
1042
1051
|
end
|
1043
1052
|
|
1044
1053
|
if src.check(/\s|\v/) then
|
1045
|
-
unless
|
1054
|
+
unless is_arg? then
|
1046
1055
|
c2 = { " " => 's',
|
1047
1056
|
"\n" => 'n',
|
1048
1057
|
"\t" => 't',
|
@@ -1056,7 +1065,7 @@ class RubyLexer
|
|
1056
1065
|
end
|
1057
1066
|
|
1058
1067
|
# ternary
|
1059
|
-
self.lex_state = :expr_beg
|
1068
|
+
self.lex_state = ruby18 ? :expr_beg : :expr_value # HACK?
|
1060
1069
|
self.tern.push true
|
1061
1070
|
self.yacc_value = "?"
|
1062
1071
|
return :tEH
|
@@ -1095,11 +1104,11 @@ class RubyLexer
|
|
1095
1104
|
self.lex_state = :expr_beg
|
1096
1105
|
return :tOP_ASGN
|
1097
1106
|
elsif src.scan(/&/) then
|
1098
|
-
result = if
|
1107
|
+
result = if is_arg? && space_seen &&
|
1099
1108
|
!src.check(/\s/) then
|
1100
1109
|
warning("`&' interpreted as argument prefix")
|
1101
1110
|
:tAMPER
|
1102
|
-
elsif
|
1111
|
+
elsif in_lex_state? :expr_beg, :expr_mid then
|
1103
1112
|
:tAMPER
|
1104
1113
|
else
|
1105
1114
|
:tAMPER2
|
@@ -1122,7 +1131,7 @@ class RubyLexer
|
|
1122
1131
|
return :tOP_ASGN
|
1123
1132
|
end
|
1124
1133
|
|
1125
|
-
if
|
1134
|
+
if is_arg? && space_seen then
|
1126
1135
|
unless src.scan(/\s/) then
|
1127
1136
|
arg_ambiguous
|
1128
1137
|
self.lex_strterm = [:strterm, STR_REGEXP, '/', "\0"]
|
@@ -1149,7 +1158,7 @@ class RubyLexer
|
|
1149
1158
|
self.yacc_value = ";"
|
1150
1159
|
return :tSEMI
|
1151
1160
|
elsif src.scan(/\~/) then
|
1152
|
-
if
|
1161
|
+
if in_lex_state? :expr_fname, :expr_dot then
|
1153
1162
|
src.scan(/@/)
|
1154
1163
|
end
|
1155
1164
|
|
@@ -1175,9 +1184,7 @@ class RubyLexer
|
|
1175
1184
|
return :tOP_ASGN
|
1176
1185
|
end
|
1177
1186
|
|
1178
|
-
if
|
1179
|
-
return parse_quote
|
1180
|
-
end
|
1187
|
+
return parse_quote if is_arg? && space_seen && ! src.check(/\s/)
|
1181
1188
|
|
1182
1189
|
self.fix_arg_lex_state
|
1183
1190
|
self.yacc_value = "%"
|
@@ -1258,12 +1265,12 @@ class RubyLexer
|
|
1258
1265
|
self.command_start = true
|
1259
1266
|
result = :tLPAREN2
|
1260
1267
|
|
1261
|
-
if
|
1268
|
+
if in_lex_state? :expr_beg, :expr_mid then
|
1262
1269
|
result = :tLPAREN
|
1263
1270
|
elsif space_seen then
|
1264
|
-
if
|
1271
|
+
if in_lex_state? :expr_cmdarg then
|
1265
1272
|
result = :tLPAREN_ARG
|
1266
|
-
elsif
|
1273
|
+
elsif in_lex_state? :expr_arg then
|
1267
1274
|
self.tern.push false
|
1268
1275
|
warning "don't put space before argument parentheses"
|
1269
1276
|
end
|
@@ -1275,20 +1282,15 @@ class RubyLexer
|
|
1275
1282
|
end
|
1276
1283
|
|
1277
1284
|
def is_end?
|
1278
|
-
|
1279
|
-
lex_state == :expr_endarg ||
|
1280
|
-
lex_state == :expr_endfn)
|
1285
|
+
in_lex_state? :expr_end, :expr_endarg, :expr_endfn
|
1281
1286
|
end
|
1282
1287
|
|
1283
1288
|
def is_arg?
|
1284
|
-
|
1289
|
+
in_lex_state? :expr_arg, :expr_cmdarg
|
1285
1290
|
end
|
1286
1291
|
|
1287
1292
|
def is_beg?
|
1288
|
-
|
1289
|
-
lex_state == :expr_mid ||
|
1290
|
-
lex_state == :expr_value ||
|
1291
|
-
lex_state == :expr_class)
|
1293
|
+
in_lex_state? :expr_beg, :expr_mid, :expr_value, :expr_class
|
1292
1294
|
end
|
1293
1295
|
|
1294
1296
|
def is_space_arg? c = "x"
|
@@ -1296,34 +1298,21 @@ class RubyLexer
|
|
1296
1298
|
end
|
1297
1299
|
|
1298
1300
|
def is_label_possible? command_state
|
1299
|
-
(
|
1301
|
+
(in_lex_state?(:expr_beg) && !command_state) || is_arg?
|
1300
1302
|
end
|
1301
1303
|
|
1302
|
-
def yylex_paren19
|
1303
|
-
|
1304
|
-
|
1305
|
-
|
1306
|
-
|
1307
|
-
|
1308
|
-
|
1309
|
-
|
1310
|
-
|
1311
|
-
|
1312
|
-
# p :wtf_paren => [lex_state, space_seen, result]
|
1304
|
+
def yylex_paren19 # TODO: move or remove
|
1305
|
+
result =
|
1306
|
+
if is_beg? then
|
1307
|
+
:tLPAREN
|
1308
|
+
elsif is_space_arg? then
|
1309
|
+
:tLPAREN_ARG
|
1310
|
+
else
|
1311
|
+
:tLPAREN2 # plain '(' in parse.y
|
1312
|
+
end
|
1313
1313
|
|
1314
|
-
#
|
1314
|
+
# paren_nest++; # TODO
|
1315
1315
|
|
1316
|
-
# HACK: this is a mess, but it makes the tests pass, so suck it
|
1317
|
-
# (stolen from the 1.8 side)
|
1318
|
-
if lex_state == :expr_beg || lex_state == :expr_mid then
|
1319
|
-
# do nothing
|
1320
|
-
elsif space_seen then
|
1321
|
-
if lex_state == :expr_arg then
|
1322
|
-
self.tern.push false
|
1323
|
-
end
|
1324
|
-
else
|
1325
|
-
self.tern.push false
|
1326
|
-
end
|
1327
1316
|
result
|
1328
1317
|
end
|
1329
1318
|
|
@@ -1345,7 +1334,7 @@ class RubyLexer
|
|
1345
1334
|
if token =~ /[!?]$/ then
|
1346
1335
|
result = :tFID
|
1347
1336
|
else
|
1348
|
-
if
|
1337
|
+
if in_lex_state? :expr_fname then
|
1349
1338
|
# ident=, not =~ => == or followed by =>
|
1350
1339
|
# TODO test lexing of a=>b vs a==>b
|
1351
1340
|
if src.scan(/=(?:(?![~>=])|(?==>))/) then
|
@@ -1361,7 +1350,7 @@ class RubyLexer
|
|
1361
1350
|
end
|
1362
1351
|
end
|
1363
1352
|
|
1364
|
-
unless
|
1353
|
+
unless ruby18
|
1365
1354
|
if is_label_possible? command_state then
|
1366
1355
|
colon = src.scan(/:/)
|
1367
1356
|
|
@@ -1373,9 +1362,9 @@ class RubyLexer
|
|
1373
1362
|
|
1374
1363
|
src.unscan if colon
|
1375
1364
|
end
|
1376
|
-
end
|
1365
|
+
end
|
1377
1366
|
|
1378
|
-
unless
|
1367
|
+
unless in_lex_state? :expr_dot then
|
1379
1368
|
# See if it is a reserved word.
|
1380
1369
|
keyword = if ruby18 then # REFACTOR need 18/19 lexer subclasses
|
1381
1370
|
RubyParserStuff::Keyword.keyword18 token
|
@@ -1417,13 +1406,13 @@ class RubyLexer
|
|
1417
1406
|
# if (mb == ENC_CODERANGE_7BIT && lex_state != EXPR_DOT) {
|
1418
1407
|
|
1419
1408
|
self.lex_state =
|
1420
|
-
if is_beg? ||
|
1409
|
+
if is_beg? || in_lex_state?(:expr_dot) || is_arg? then
|
1421
1410
|
if command_state then
|
1422
1411
|
:expr_cmdarg
|
1423
1412
|
else
|
1424
1413
|
:expr_arg
|
1425
1414
|
end
|
1426
|
-
elsif ruby19 &&
|
1415
|
+
elsif ruby19 && in_lex_state?(:expr_fname) then
|
1427
1416
|
:expr_endfn
|
1428
1417
|
else
|
1429
1418
|
:expr_end
|