pgen.py
47.9 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
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
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
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
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
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
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
"""A Python implementation of CPython's parser
This module is largely based on Jonathan Riehl's PyPgen included in
the Basil framework (http://code.google.com/p/basil).
"""
from snakes import SnakesError
from snakes.lang import ast
import tokenize, string, pprint, warnings, inspect, os.path
from snakes.compat import *
def warn (message) :
"""Issue a warning message.
"""
warnings.warn(message, stacklevel=2)
class Token (str) :
"""A token from the lexer.
Behaves as a string that is either the token value (if not empty),
or the token name. Additional attributes allow to extract various
information:
- self.kind: token number (like tokenize.ENDMARKER)
also available as int(self)
- self.text: token text
- self.srow: start row
- self.scol: start column
- self.erow: end row
- self.ecol: end column
- self.line: full line of text from which the token comes
- self.name: token name (ie, tokenize.tok_name[self.kind])
- self.lexer: the Tokenizer instance than produced this token
- self.filename: input file from which the token comes (or
'<string>' if None available)
"""
def __new__ (cls, token, lexer) :
"""Create a new instance.
__new__ is used instead of __init__ because str is a
non-mutable object and this __init__ could not assign a str
content. For more information see:
http://docs.python.org/reference/datamodel.html#object.__new__
"""
kind = token[0]
text = token[1]
name = lexer.tok_name[kind]
self = str.__new__(cls, text or name)
self.kind = kind
self.text = text
self.srow, self.scol = token[2]
self.erow, self.ecol = token[3]
self.line = token[4]
self.name = name
self.lexer = lexer
try :
self.filename = lexer.infile.name
except :
self.filename = "<string>"
return self
def __int__ (self) :
"""Coercion to int (return self.kind).
"""
return self.kind
class Location (str) :
"""A position in a parsed file
Used to aggregate the positions of all the tokens is a parse
(sub)tree. The following attributes are available:
- self.srow: start row
- self.scol: start column
- self.erow: end row
- self.ecol: end column
- self.filename: input file from which the token comes (or
'<string>' if None available)
"""
def __new__ (cls, first, last) :
"""Create a new instance
Expected arguments:
- first: the first Token or Location instance in the region
- last: the last one
"""
self = str.__new__(cls, "%s[%s:%s-%s:%s]"
% (first.filename, first.srow, first.scol,
last.erow, last.ecol))
self.srow, self.scol = first.srow, first.scol
self.erow, self.ecol = last.erow, last.ecol
self.filename = first.filename
self.lexer = first.lexer
return self
class ParseError (SnakesError) :
"""Exception raised when parsing fails.
It's better not to use SyntaxError because this soes not allows to
distinguish when the text being parser has a syntax error from
when because the parser itself has a syntax error.
"""
def __init__ (self, token, expected=None, reason=None) :
"""Initialize a new instance.
Expected arguments are:
- token: the erroneous token (a Token instance)
- expected: either a Token instance or a token kind (like
tokenize.NAME) to indicated what was expected instead
"""
self.token = token
if expected is not None :
expected = int(expected)
self.expected = expected
if token is None :
pos = ""
else :
pos = "%s[%s:%s]: " % (token.filename, token.srow, token.scol)
if reason is not None :
msg = reason
elif self.expected is not None :
msg = ("expected %s but found %r" %
(token.lexer.tok_name[expected], token))
else :
msg = "unexpected token %r" % token
SnakesError.__init__(self, pos + msg)
class Tokenizer (object) :
"""A simple lexical analyser based on Python's tokenize module.
The differences with tokenize module are:
- new simple tokens may be added (just strings, no regexps)
- Python's token may be not included
- tokens may be automatically skipped (removed from the output)
- tokenize.OP kind is refined (eg, ':' gets kind tokenize.COLON)
This class replaces two PyPgen elements:
- module basil.lang.python.TokenUtils
- module basil.lang.python.StdTokenizer
"""
_pyopmap = {
'(' : tokenize.LPAR,
')' : tokenize.RPAR,
'[' : tokenize.LSQB,
']' : tokenize.RSQB,
':' : tokenize.COLON,
',' : tokenize.COMMA,
';' : tokenize.SEMI,
'+' : tokenize.PLUS,
'+=' : tokenize.PLUSEQUAL,
'-' : tokenize.MINUS,
'-=' : tokenize.MINEQUAL,
'*' : tokenize.STAR,
'**' : tokenize.DOUBLESTAR,
'**=' : tokenize.DOUBLESTAREQUAL,
'*=' : tokenize.STAREQUAL,
'/' : tokenize.SLASH,
'//' : tokenize.DOUBLESLASH,
'//=' : tokenize.DOUBLESLASHEQUAL,
'/=' : tokenize.SLASHEQUAL,
'|' : tokenize.VBAR,
'|=' : tokenize.VBAREQUAL,
'&' : tokenize.AMPER,
'&=' : tokenize.AMPEREQUAL,
'<' : tokenize.LESS,
'<=' : tokenize.LESSEQUAL,
'<<' : tokenize.LEFTSHIFT,
'<<=' : tokenize.LEFTSHIFTEQUAL,
'>' : tokenize.GREATER,
'>=' : tokenize.GREATEREQUAL,
'>>' : tokenize.RIGHTSHIFT,
'>>=' : tokenize.RIGHTSHIFTEQUAL,
'=' : tokenize.EQUAL,
'==' : tokenize.EQEQUAL,
'.' : tokenize.DOT,
'%' : tokenize.PERCENT,
'%=' : tokenize.PERCENTEQUAL,
'{' : tokenize.LBRACE,
'}' : tokenize.RBRACE,
'^' : tokenize.CIRCUMFLEX,
'^=' : tokenize.CIRCUMFLEXEQUAL,
'~' : tokenize.TILDE,
'!=' : tokenize.NOTEQUAL,
'<>' : tokenize.NOTEQUAL,
'@' : tokenize.AT
}
def __init__ (self, python=True, opmap={}, skip=None, **extra) :
"""Initialize a new instance.
Expected arguments are:
- python: a bool to indicate whether to include or not
Python's tokens (default to True)
- opmap: a dict to map litteral tokens (given as '...' in the
grammar) to token kinds (default to {}). This parameter is
useful only to redefine Python's mapping
- skip: a collection of tokens that the tokenizer will
automatically skip (default to [COMMENT, NL])
- additional keywords arguments allow to define new tokens,
for instance, providing DOLLAR='$' defines a new token
called 'DOLLAR' (its kind will be automatically computed)
An instance of Tokenizer has the following attributes:
- self.opmap: a dict mapping operators token literals to the
corresponding kind, for instance, ':' is mapped to
tokenize.COLON (this can be overridden using argument
opmap)
- self.tok_name: a replacement of tokenize.tok_name that also
include the user-defined tokens
- for each token called FOO (including user-defined ones), an
attribute self.FOO hols the corresponding kind
"""
self._python = python
self._opmap = opmap.copy()
if python :
self.opmap = self._pyopmap.copy()
self.opmap.update(opmap)
else :
self.opmap = opmap.copy()
self.tok_name = {}
self._extra = {}
if python :
for kind, name in tokenize.tok_name.items() :
self.tok_name[kind] = name
setattr(self, name, kind)
if not hasattr(self, "NT_OFFSET") :
self.NT_OFFSET = 256
last = max(n for n in self.tok_name if n != self.NT_OFFSET)
for shift, (name, txt) in enumerate(sorted(extra.items())) :
#WARNING: sorted above is required to guaranty that extra
# tokens will always get the same number (dict order is
# not guaranteed)
kind = last + shift
if kind >= self.NT_OFFSET :
raise TypeError("too many new tokens")
self.tok_name[kind] = name
setattr(self, name, kind)
self._extra[txt] = kind
self.opmap.update(self._extra)
if skip is None :
skip = [self.COMMENT, self.NL]
self._skip = set(skip)
def __repr__ (self) :
"""Encodes an instance as Python source code.
Non-default arguments provided to the constructor are included
so that exactly the same Tokenizer instance can be recovered
from the returned source code.
>>> print repr(Tokenizer())
Tokenizer()
>>> print repr(Tokenizer(DOLLAR='$'))
Tokenizer(DOLLAR='$')
>>> print repr(Tokenizer(skip=[], DOLLAR='$'))
Tokenizer(skip=[], DOLLAR='$')
"""
args = []
if not self._python :
args.append("python=%s" % self._python)
if self._opmap :
args.append("opmap=%r" % self._opmap)
if self._skip != set([self.COMMENT, self.NL]) :
args.append("skip=%r" % list(self._skip))
args.extend("%s=%r" % (self.tok_name[kind], txt) for txt, kind
in self._extra.items())
return "%s(%s)" % (self.__class__.__name__, ", ".join(args))
def tokenize (self, stream) :
"""Break an input stream into tokens.
Expected argument is:
- stream: a file-like object (with a method readline)
Return a generator of Token instances, ParseError is raised
whenever an erroneous token is encountered.
This is basically the same as tokenize.generate_tokens but:
- the appropriate tokens are skipped
- OP kind is converted according to self.opmap
- user-defined tokens are handled
During the iteration, two more attributes can be used:
- self.last: last recognized token (ie, last yielded)
- self.infile: the input stream passed to method tokenize
"""
self.infile = stream
self.last = None
self.lines = []
def readline () :
self.lines.append(stream.readline())
return self.lines[-1]
err = self.ERRORTOKEN
for token in tokenize.generate_tokens(readline) :
if token[0] == err :
try :
token = (self._extra[token[1]],) + token[1:]
except :
raise ParseError(Token(token, self))
elif token[0] in self._skip :
try:
self.skip_token(Token(token, self))
except :
pass
continue
elif token[0] == self.OP :
token = (self.opmap[token[1]],) + token[1:]
self.last = Token(token, self)
yield self.last
def skip_token (self, token) :
pass
try :
Tokenizer._pyopmap['`'] = tokenize.BACKQUOTE
except AttributeError :
pass
class PgenParser (object) :
"""A parser for pgen files.
The following grammar is used:
mstart : ( rule | NEWLINE | newtok )* ENDMARKER
newtok : '$' NAME STRING NEWLINE
rule : NAME COLON rhs NEWLINE
rhs := alt ( VBAR alt )*
alt : item+
item : LSQB rhs RSQB | atom ( STAR | PLUS )?
atom : LPAR rhs RPAR | NAME | STRING
With respect to PyPgen, an additional rule 'newtok' has been added
to allow for user-defined tokens.
This class is adapted from module basil.parsing.PgenParser, it has
attributes MSTART, ..., provinding to the symbol numbers for the
corresponding rules.
"""
MSTART = 256
RULE = 257
RHS = 258
ALT = 259
ITEM = 260
ATOM = 261
NEWTOK = 262
def __init__ (self) :
self.lexer = Tokenizer(NEWTOK='$')
def expect (self, expected, found) :
if expected != found.kind :
raise ParseError(found, expected=expected)
@classmethod
def parse (cls, filename) :
"""Parse a pgen file.
Expected argument is:
- filename: path of pgen file to be parsed
Return a 2-tuple (G, T) where:
- G is the grammar's syntax tree
- T is a Tokenizer instance to be used with the parser
generated from this grammar (ie, including all the required
user-defined tokens)
This is basically the only method that is needed:
>>> mygrammar = PgenParser.parse('myfile.pgen')
"""
return cls().parse_file(filename)
def parse_file (self, filename) :
"""Parse a pgen file.
Expected argument is:
- filename: path of pgen file to be parsed
Return a 2-tuple (g, t):
- g: is the grammar's syntax tree
- t: is a Tokenizer instance to be used with the parser
generated from this grammar (ie, including all the required
user-defined tokens)
Recognize mstart : ( rule | NEWLINE )* ENDMARKER
Like in PyPgen, the parser is recursive descendent, each rule
'X' being recognized by a dedicated method 'handleX' (except
for 'mstart'). Each such method expects a current token (or
None if it has to be fetched from the tokenizer) and returns a
2-tuple (R, T) where:
- R is the resulting syntax tree
- T is the new current token (or None)
"""
self.infile = open(filename)
self.tokens = self.lexer.tokenize(self.infile)
extra = {}
children = []
current = next(self.tokens)
while current.kind != self.lexer.ENDMARKER :
if current.kind == self.lexer.NEWLINE :
children.append((current, []))
current = None
elif current.kind == self.lexer.NEWTOK :
name, text = self.handleNewtok(current)
current = None
extra[name] = text
else :
ruleResult, current = self.handleRule(current)
children.append(ruleResult)
if current is None :
current = next(self.tokens)
children.append((current, []))
return (self.MSTART, children), Tokenizer(**extra)
def handleNewtok (self, current=None) :
"""Recognize newtok : '$' NAME STRING NEWLINE
Unlike the other 'handleX' methods, this one does not return a
syntax tree because it implements a parsing directive.
Instead, it returns a 2-tuple (N, S) where:
- N is the user-defined token name
- S is the token string value
"""
if current is None :
current = next(self.tokens)
self.expect(self.lexer.NEWTOK, current)
name = next(self.tokens)
self.expect(self.lexer.NAME, name)
text = next(self.tokens)
self.expect(self.lexer.STRING, text)
nl = next(self.tokens)
self.expect(self.lexer.NEWLINE, nl)
return name, compile(text, "<string>", "eval",
ast.PyCF_ONLY_AST).body.s
def handleRule (self, current=None) :
"""Recognize rule : NAME COLON rhs NEWLINE
"""
children = []
if current is None :
current = next(self.tokens)
self.expect(self.lexer.NAME, current)
children.append((current, []))
current = next(self.tokens)
self.expect(self.lexer.COLON, current)
children.append((current, []))
rhsResult, current = self.handleRhs()
children.append(rhsResult)
if current is None :
current = next(self.tokens)
self.expect(self.lexer.NEWLINE, current)
children.append((current, []))
result = (self.RULE, children)
return result, None
def handleRhs (self, current=None) :
"""Recognize rhs : alt ( VBAR alt )*
"""
children = []
altResult, current = self.handleAlt(current)
children.append(altResult)
if current is None :
current = next(self.tokens)
while current.kind == self.lexer.VBAR :
children.append((current, []))
altResult, current = self.handleAlt()
children.append(altResult)
if current is None :
current = next(self.tokens)
result = (self.RHS, children)
return result, current
def handleAlt (self, current=None) :
""" Recognize alt : item+
"""
children = []
itemResult, current = self.handleItem(current)
children.append(itemResult)
if current is None :
current = next(self.tokens)
while current.kind in (self.lexer.LSQB, self.lexer.LPAR,
self.lexer.NAME, self.lexer.STRING) :
itemResult, current = self.handleItem(current)
children.append(itemResult)
if current is None :
current = next(self.tokens)
return (self.ALT, children), current
def handleItem (self, current=None) :
"""Recognize item : LSQB rhs RSQB | atom ( STAR | PLUS )?
"""
children = []
if current is None :
current = next(self.tokens)
if current.kind == self.lexer.LSQB :
children.append((current, []))
rhsResult, current = self.handleRhs()
children.append(rhsResult)
if current is None :
current = next(self.tokens)
self.expect(self.lexer.RSQB, current)
children.append((current, []))
current = None
else :
atomResult, current = self.handleAtom(current)
children.append(atomResult)
if current is None :
current = next(self.tokens)
if current.kind in (self.lexer.STAR, self.lexer.PLUS) :
children.append((current, []))
current = None
return (self.ITEM, children), current
def handleAtom (self, current=None) :
"""Recognize atom : LPAR rhs RPAR | NAME | STRING
"""
children = []
if current is None :
current = next(self.tokens)
tokType = current.kind
if tokType == self.lexer.LPAR :
children.append((current, []))
rhsResult, current = self.handleRhs()
children.append(rhsResult)
if current is None :
current = next(self.tokens)
self.expect(self.lexer.RPAR, current)
children.append((current, []))
elif tokType == self.lexer.STRING :
children.append((current, []))
else :
self.expect(self.lexer.NAME, current)
children.append((current, []))
return (self.ATOM, children), None
class Parser (object) :
"""A LL1 parser for a generated grammar.
This class aggregates two elements from PyPgen:
- module basil.lang.python.DFAParser
- class basil.parsing.PyPgen.PyPgenParser
The main differences are:
- simplified interface
- adapt to handle Token instances instead of 3-tuples (kind,
text, lineno)
- remove functions arguments that can now be retreived as
instance attributes
- use Python warnings instead of print statements
- many minor code edits (for my own understanding)
Docstrings are provided only for methods that did not have an
equivalent in PyPgen or that have been changed in a significant
way.
"""
def __init__ (self, grammar, tokenizer) :
"""Initialize a new instance.
Expected arguments are:
- grammar: the gramar object as returned by PyPgen.grammar()
- tokenizer: a Tokenizer instance suitable for this grammar
(eg, that passed to PyPgen's constructor)
"""
self.grammar = grammar
self.start = grammar[2]
self.stringMap = {}
for dfa in self.grammar[0] :
dfaType, dfaName = dfa[:2]
self.stringMap[dfaName] = dfaType
self.symbolMap = {}
for dfa in self.grammar[0] :
dfaType, dfaName = dfa[:2]
self.symbolMap[dfaType] = dfaName
self.tokenizer = tokenizer
self.addAccelerators()
def parseTokens (self, tokens, start=None) :
"""Parse a series of tokens.
Expected arguments:
- tokens: a generator of Token instances
- start: the start symbol to be recognized (or None to use
the default one)
The token generator should provide only tokens that are
compatible with the tokenizer passed to the Parser's
constructor. Otherwise, ParseError will be raised as unknown
tokens will be issued.
"""
self.tokens = tokens
return self._parse(start)
def parseFile (self, filename, start=None) :
"""Parse a text file provided by its path.
Expected arguments:
- filename: a file name
- start: the start symbol to be recognized (or None to use
the default one)
The start symbol may be provided by its number (int) or its
name (str) as specified in the grammar.
"""
self.tokens = self.tokenizer.tokenize(open(filename))
return self._parse(start)
def parseStream (self, stream, start=None) :
"""Parse text from an opened file.
Expected arguments:
- stream: a file-like object (with a method readline)
- start: the start symbol to be recognized (or None to use
the default one)
The start symbol may be provided by its number (int) or its
name (str) as specified in the grammar.
"""
self.tokens = self.tokenizer.tokenize(stream)
return self._parse(start)
def parseString (self, text, start=None, filename="<string>") :
"""Parse text given as a string.
Expected arguments:
- text: a string-like object
- start: the start symbol to be recognized (or None to use
the default one)
The start symbol may be provided by its number (int) or its
name (str) as specified in the grammar.
"""
data = io.StringIO(text)
data.name = filename
self.tokens = self.tokenizer.tokenize(data)
return self._parse(start)
def _parse (self, start=None) :
"""Main parsing method.
Expected argument:
- start: the start symbol to be recognized (or None to use
the default one)
The start symbol may be provided by its number (int) or its
name (str) as specified in the grammar.
"""
if start is None :
start = self.start
elif start in self.stringMap :
start = self.stringMap[start]
elif start not in self.symbolMap :
raise ValueError("unknown start symbol %r" % start)
tokens = self.tokens
# initialize the parsing stack
rootNode = ((start, None, 0), [])
dfa = self.findDFA(start)
self.stack = [(dfa[3][dfa[2]], dfa, rootNode)]
# parse all of it
result = self._LL1_OK
while result == self._LL1_OK :
result, expected = self.addToken(next(tokens))
if result == self._LL1_DONE :
return self._fix_locations(rootNode)
elif result == self._LL1_SYNTAX :
raise ParseError(self.tokenizer.last, expected=expected)
def _tostrings (self, st) :
"""Substitute symbol numbers by strings in a syntax tree.
Expected argument:
- st: a syntax tree as returned by the parser
"""
(kind, token, lineno), children = st
if kind >= self.tokenizer.NT_OFFSET :
name = self.symbolMap
else :
name = self.tokenizer.tok_name
return ((name[kind], token, lineno),
[self._tostrings(c) for c in children])
def _fix_locations (self, st) :
"""Replaces None in non-terminal nodes by a Location instance.
Expected argument:
- st: a syntax tree as returned by the parser
"""
(kind, token, lineno), children = st
children = [self._fix_locations(c) for c in children]
if kind >= self.tokenizer.NT_OFFSET :
token = Location(children[0][0][1], children[-1][0][1])
return ((kind, token, lineno), children)
def pprint (self, st) :
"""Return a human-readable representation of a syntax tree.
Expected argument:
- st: a syntax tree as returned by the parser
All symbol numbers are substituted by the corresponding names
and the text is indented appropriately.
"""
return pprint.pformat(self._tostrings(st))
# the rest of the class has not changed too much
def addAccelerators (self) :
if self.grammar[-1] : # already has accelerators
return
dfas, labels, start, accel = self.grammar
def handleState (state) :
arcs, accel, accept = state
accept = 0
labelCount = len(labels)
accelArray = [-1] * labelCount
for arc in arcs :
labelIndex, arrow = arc
kind = labels[labelIndex][0]
if arrow >= 128 :
warn("too many states (%d >= 128)!" % arrow)
continue
if kind >= self.tokenizer.NT_OFFSET :
targetFirstSet = self.findDFA(kind)[4]
if kind - self.tokenizer.NT_OFFSET >= 128 :
warn("nonterminal too high (%d >= %d)!" %
(kind, 128 + self.tokenizer.NT_OFFSET))
continue
for ibit in range(labelCount) :
if self.testbit(targetFirstSet, ibit) :
accelVal = (arrow | 128 |
((kind - self.tokenizer.NT_OFFSET) << 8))
oldVal = accelArray[ibit]
if oldVal != -1 :
# XXX Make this error reporting more better.
oldType = oldVal >> 8
# FIXME: bug in the original source
#warn("ambiguity at bit %d (for %d: was to %x,"
# " now to %x)."
# % (ibit, states.index(state),
# oldVal, accelVal))
warn("ambiguity at bit %d" % ibit)
accelArray[ibit] = (arrow | 128 |
((kind - self.tokenizer.NT_OFFSET) << 8))
elif labelIndex == 0 :
accept = 1
elif labelIndex >= 0 and labelIndex < labelCount :
accelArray[labelIndex] = arrow
# Now compute the upper and lower bounds.
accelUpper = labelCount
while accelUpper > 0 and accelArray[accelUpper-1] == -1 :
accelUpper -= 1
accelLower = 0
while accelLower < accelUpper and accelArray[accelLower] == -1 :
accelLower += 1
accelArray = accelArray[accelLower:accelUpper]
return (arcs, (accelUpper, accelLower, accelArray), accept)
def handleDFA (dfa) :
kind, name, initial, states, first = dfa
return (kind, name, initial, list(map(handleState, states)))
self.grammar = (list(map(handleDFA, dfas)), labels, start, 1)
_LL1_OK = 0 # replaced E_ prefix with _LL1_ to prevent potential
_LL1_DONE = 1 # conflicts with grammar symbols
_LL1_SYNTAX = 2
def testbit (self, bitstr, ibit) :
return (ord(bitstr[ibit >> 3]) & (1 << (ibit & 0x7))) != 0
def classify (self, token) :
labels = self.grammar[1]
if token.kind == self.tokenizer.NAME :
for i, label in enumerate(labels) :
if (token.kind, token) == label :
return i
for i, label in enumerate(labels) :
if (token.kind == label[0]) and (label[1] is None) :
return i
return -1
def findDFA (self, start) :
return self.grammar[0][start - self.tokenizer.NT_OFFSET]
def addToken (self, token) :
stack = self.stack
ilabel = self.classify(token)
while True :
state, dfa, parent = stack[-1]
# Perform accelerator
arcs, (accelUpper, accelLower, accelTable), accept = state
if accelLower <= ilabel < accelUpper :
accelResult = accelTable[ilabel - accelLower]
if accelResult != -1 :
# Handle accelerator result
if accelResult & 128 :
# Push non-terminal
nt = (accelResult >> 8) + self.tokenizer.NT_OFFSET
arrow = accelResult & 127
nextDFA = self.findDFA(nt)
# INLINE PUSH
newAstNode = ((nt, None, token.srow), [])
parent[1].append(newAstNode)
stack[-1] = (dfa[3][arrow], dfa, parent)
stack.append((nextDFA[3][nextDFA[2]], nextDFA,
newAstNode))
continue
# INLINE SHIFT
parent[1].append(((token.kind, token, token.srow), []))
nextState = dfa[3][accelResult]
stack[-1] = (nextState, dfa, parent)
state = nextState
while state[2] and len(state[0]) == 1 :
# INLINE POP
stack.pop(-1)
if not stack :
return self._LL1_DONE, None
else :
state, dfa, parent = stack[-1]
return self._LL1_OK, None
if accept :
stack.pop(-1)
if not stack :
return self._LL1_SYNTAX, self.tokenizer.ENDMARKER
continue
if ((accelUpper < accelLower) and
(self.grammar[1][accelLower][1] is not None)) :
expected = self.grammar[1][accelLower][1]
else :
expected = None
return self._LL1_SYNTAX, expected
class PyPgen (object) :
"""A grammar generator.
This class aggregates two elements from PyPgen:
- class basil.parsing.PyPgen.PyPgen
- function basil.parsing.PyPgen.buildParser
- parts of function basil.parsing.PyPgen.parserMain
The main differences are:
- simplified interface
- adapt to handle Token instances instead of 3-tuples (kind,
text, lineno)
- remove functions arguments that can now be retreived as
instance attributes
- use Python warnings instead of print statements
- many minor code edits (for my own understanding)
Docstrings are provided only for methods that did not have an
equivalent in PyPgen or that have been changed in a significant
way.
"""
def __init__ (self, gst, tokenizer) :
"""Initialize a new instance.
Expected arguments are:
- gst: the grammar's syntax tree as returned by
PgenParser.parse()
- tokenizer: a Tokenizer instance suitable for this grammar,
also returned by PgenParser.parse()
"""
self.tokenizer = tokenizer
self.EMPTY = self.tokenizer.ENDMARKER
self.gst = gst
self.nfaGrammar = self.dfaGrammar = None
self.nfa = None
self.crntKind = self.tokenizer.NT_OFFSET
self.operatorMap = tokenizer.opmap
def grammar (self) :
"""Generate and return the grammar object.
"""
nfaGrammar = self.handleStart(self.gst)
grammar = self.generateDfaGrammar(nfaGrammar)
self.translateLabels(grammar)
self.generateFirstSets(grammar)
grammar[0] = list(map(tuple, grammar[0]))
# Trick to add accelerators at generation time: it's easier to
# do it this way than to extract the required elements from
# class Parser.
return Parser(tuple(grammar), self.tokenizer).grammar
def python (self, pgen="pgen", inline=False) :
"""Build and return Python code for parsing module.
Expected arguments are:
- pgen: the name of module pgen in the generated source
(default to 'pgen')
- inline: a bool value to indicate whether to import or
inline pgen module in the generated code
If inline=True, the generated code is much bigger but does not
depend on any non-standard module.
"""
pysrc = ("%(pgen)s"
"tokenizer = %(prefix)s%(tokenizer)r\n"
"grammar = %(grammar)s\n"
"parser = %(prefix)sParser(grammar, tokenizer)\n\n"
"if __name__ == '__main__' :\n"
" # just for test purpose\n"
" import sys, pprint\n"
" st = parser.parseStream(sys.stdin)\n"
" print(parser.pprint(st))\n")
format = {"grammar" : pprint.pformat(self.grammar()),
"prefix" : pgen + ".",
"tokenizer" : self.tokenizer,
"pgen" : "import tokenize, %s\n\n" % pgen,
"inline" : "",
}
if inline :
format["prefix"] = ""
source = inspect.getsource(inspect.getmodule(self))
source = source.rsplit("if __name__ == '__main__' :", 1)[0]
format["pgen"] = ("### module '%s.py' inlined\n"
"%s\n### end of '%s.py'\n\n"
% (pgen, source.rstrip(), pgen))
return pysrc % format
@classmethod
def translate (cls, src, tgt=None, pgen="pgen", inline=False) :
"""Translate a pgen file to a Python file that implements the
corresponding parser.
Expected arguments are:
- src: path of the pgen file
- tgt: path of target Python file, if None, its name is
derived from src (replacing its extension by .py)
- pgen, inline: like in PyPgen.python()
Warning: the output file is silently overwritten if it already
exist.
"""
if tgt is None :
tgt = os.path.splitext(src)[0] + ".py"
gst, tokenizer = PgenParser.parse(src)
self = PyPgen(gst, tokenizer)
outfile = open(tgt, "w")
outfile.write(("# this file has been automatically generated running:\n"
"# %s\n\n") % " ".join(sys.argv))
outfile.write(self.python(pgen, inline))
outfile.close()
# the rest of the class has not changed too much
def addLabel (self, labelList, tokKind, tokName) :
labelTup = (tokKind, tokName)
if labelTup in labelList :
return labelList.index(labelTup)
labelIndex = len(labelList)
labelList.append(labelTup)
return labelIndex
def handleStart (self, gst) :
self.nfaGrammar = [[],[(self.tokenizer.ENDMARKER, "EMPTY")]]
self.crntKind = self.tokenizer.NT_OFFSET
kind, children = gst
for child in children :
if int(child[0]) == PgenParser.RULE :
self.handleRule(child)
return self.nfaGrammar
def handleRule (self, gst) :
# NFA := [ type : Int, name : String, [ STATE ], start : Int,
# finish : Int ]
# STATE := [ ARC ]
# ARC := ( labelIndex : Int, stateIndex : Int )
####
# build the NFA shell
self.nfa = [self.crntKind, None, [], -1, -1]
self.crntKind += 1
# work on the AST node
kind, children = gst
name, colon, rhs, newline = children
self.nfa[1] = name[0]
if (self.tokenizer.NAME, name[0]) not in self.nfaGrammar[1] :
self.nfaGrammar[1].append((self.tokenizer.NAME, name[0]))
start, finish = self.handleRhs(rhs)
self.nfa[3] = start
self.nfa[4] = finish
# append the NFA to the grammar
self.nfaGrammar[0].append(self.nfa)
def handleRhs (self, gst) :
kind, children = gst
start, finish = self.handleAlt(children[0])
if len(children) > 1 :
cStart = start
cFinish = finish
start = len(self.nfa[2])
self.nfa[2].append([(self.EMPTY, cStart)])
finish = len(self.nfa[2])
self.nfa[2].append([])
self.nfa[2][cFinish].append((self.EMPTY, finish))
for child in children[2:] :
if int(child[0]) == PgenParser.ALT :
cStart, cFinish = self.handleAlt(child)
self.nfa[2][start].append((self.EMPTY, cStart))
self.nfa[2][cFinish].append((self.EMPTY, finish))
return start, finish
def handleAlt (self, gst) :
kind, children = gst
start, finish = self.handleItem(children[0])
if len(children) > 1 :
for child in children[1:] :
cStart, cFinish = self.handleItem(child)
self.nfa[2][finish].append((self.EMPTY, cStart))
finish = cFinish
return start, finish
def handleItem (self, gst) :
nodeKind, children = gst
if int(children[0][0]) == PgenParser.ATOM :
start, finish = self.handleAtom(children[0])
if len(children) > 1 :
# Short out the child NFA
self.nfa[2][finish].append((self.EMPTY, start))
if children[1][0].kind == self.tokenizer.STAR :
finish = start
else :
start = len(self.nfa[2])
finish = start + 1
self.nfa[2].append([(self.EMPTY, finish)])
self.nfa[2].append([])
cStart, cFinish = self.handleRhs(children[1])
self.nfa[2][start].append((self.EMPTY, cStart))
self.nfa[2][cFinish].append((self.EMPTY, finish))
return start, finish
def handleAtom (self, gst) :
nodeKind, children = gst
tok = children[0][0]
if tok.kind == self.tokenizer.LPAR :
start, finish = self.handleRhs(children[1])
elif tok.kind in (self.tokenizer.STRING, self.tokenizer.NAME) :
start = len(self.nfa[2])
finish = start + 1
labelIndex = self.addLabel(self.nfaGrammar[1], tok.kind, tok)
self.nfa[2].append([(labelIndex, finish)])
self.nfa[2].append([])
return start, finish
def generateDfaGrammar (self, nfaGrammar, start=None) :
# See notes in pgen.lang.python.DFAParser for output schema.
dfas = []
for nfa in nfaGrammar[0] :
dfas.append(self.nfaToDfa(nfa))
kind = dfas[0][0]
if start is not None :
found = False
for dfa in dfas :
if dfa[1] == start :
kind = dfa[0]
found = True
break
if not found :
warn("couldn't find nonterminal %r, "
"using %r instead." % (start, dfas[0][1]))
return [dfas, nfaGrammar[1][:], kind, 0]
def addClosure (self, stateList, nfa, istate) :
stateList[istate] = True
arcs = nfa[2][istate]
for label, arrow in arcs :
if label == self.EMPTY :
self.addClosure(stateList, nfa, arrow)
def nfaToDfa (self, nfa) :
tempStates = []
crntTempState = [[False] * len(nfa[2]), [], False]
self.addClosure(crntTempState[0], nfa, nfa[3])
crntTempState[2] = crntTempState[0][nfa[4]]
if crntTempState[2] :
warn("nonterminal %r may produce empty." % nfa[1])
tempStates.append(crntTempState)
index = 0
while index < len(tempStates) :
crntTempState = tempStates[index]
for componentState in range(len(nfa[2])) :
if not crntTempState[0][componentState] :
continue
nfaArcs = nfa[2][componentState]
for label, nfaArrow in nfaArcs :
if label == self.EMPTY :
continue
foundTempArc = False
for tempArc in crntTempState[1] :
if tempArc[0] == label :
foundTempArc = True
break
if not foundTempArc :
tempArc = [label, -1, [False] * len(nfa[2])]
crntTempState[1].append(tempArc)
self.addClosure(tempArc[2], nfa, nfaArrow)
for arcIndex in range(len(crntTempState[1])) :
label, arrow, targetStateList = crntTempState[1][arcIndex]
targetFound = False
arrow = 0
for destTempState in tempStates :
if targetStateList == destTempState[0] :
targetFound = True
break
arrow += 1
if not targetFound :
assert arrow == len(tempStates)
tempState = [targetStateList[:], [],
targetStateList[nfa[4]]]
tempStates.append(tempState)
# Write arrow value back to the arc
crntTempState[1][arcIndex][1] = arrow
index += 1
tempStates = self.simplifyTempDfa(nfa, tempStates)
return self.tempDfaToDfa(nfa, tempStates)
def sameState (self, s1, s2) :
if len(s1[1]) != len(s2[1]) or s1[2] != s2[2] :
return False
for arcIndex in range(len(s1[1])) :
arc1 = s1[1][arcIndex]
arc2 = s2[1][arcIndex]
if arc1[:-1] != arc2[:-1] :
return False
return True
def simplifyTempDfa (self, nfa, tempStates) :
changes = True
deletedStates = []
while changes :
changes = False
for i in range(1, len(tempStates)) :
if i in deletedStates :
continue
for j in range(i) :
if j in deletedStates :
continue
if self.sameState(tempStates[i], tempStates[j]) :
deletedStates.append(i)
for k in range(len(tempStates)) :
if k in deletedStates :
continue
for arc in tempStates[k][1] :
if arc[1] == i :
arc[1] = j
changes = True
break
for stateIndex in deletedStates :
tempStates[stateIndex] = None
return tempStates
def tempDfaToDfa (self, nfa, tempStates) :
dfaStates = []
dfa = [nfa[0], nfa[1], 0, dfaStates, None]
stateMap = {}
tempIndex = 0
for tempState in tempStates :
if tempState is not None :
stateMap[tempIndex] = len(dfaStates)
dfaStates.append(([], (0,0,()), 0))
tempIndex += 1
for tempIndex in stateMap.keys() :
stateList, tempArcs, accepting = tempStates[tempIndex]
dfaStateIndex = stateMap[tempIndex]
dfaState = dfaStates[dfaStateIndex]
for tempArc in tempArcs :
dfaState[0].append((tempArc[0], stateMap[tempArc[1]]))
if accepting :
dfaState[0].append((self.EMPTY, dfaStateIndex))
return dfa
def translateLabels (self, grammar) :
tokenNames = list(self.tokenizer.tok_name.values())
# Recipe 252143 (remixed for laziness)
tokenValues = dict(([v, k] for k, v in
self.tokenizer.tok_name.items()))
labelList = grammar[1]
for labelIndex, (kind, name) in enumerate(labelList) :
if kind == self.tokenizer.NAME :
isNonTerminal = False
for dfa in grammar[0] :
if dfa[1] == name :
labelList[labelIndex] = (dfa[0], None)
isNonTerminal = True
break
if not isNonTerminal :
if name in tokenNames :
labelList[labelIndex] = (tokenValues[name], None)
else :
warn("can't translate NAME label '%s'" % name)
elif kind == self.tokenizer.STRING :
assert name[0] == name[-1]
sname = name[1:-1]
if (sname[0] in string.letters) or (sname[0] == "_") :
labelList[labelIndex] = (self.tokenizer.NAME, sname)
elif sname in self.operatorMap :
labelList[labelIndex] = (self.operatorMap[sname],
None)
else :
warn("can't translate STRING label %s" % name)
return grammar
def calcFirstSet (self, grammar, dfa) :
if dfa[4] == -1 :
warn("left-recursion for %r" % dfa[1])
return
if dfa[4] != None :
warn("re-calculating FIRST set for %r" % dfa[1])
dfa[4] = -1
symbols = []
result = 0
state = dfa[3][dfa[2]]
for arc in state[0] :
sym = arc[0]
if sym not in symbols :
symbols.append(sym)
kind = grammar[1][sym][0]
if kind >= self.tokenizer.NT_OFFSET :
# Nonterminal
ddfa = grammar[0][kind - self.tokenizer.NT_OFFSET]
if ddfa[4] == -1 :
warn("left recursion below %r" % dfa[1])
else :
if ddfa[4] == None :
self.calcFirstSet(grammar, ddfa)
result |= ddfa[4]
else :
result |= 1 << sym
dfa[4] = result
def generateFirstSets (self, grammar) :
dfas = grammar[0]
index = 0
while index < len(dfas) :
dfa = dfas[index]
if None == dfa[4] :
self.calcFirstSet(grammar, dfa)
index += 1
for dfa in dfas :
set = dfa[4]
result = []
while set > 0 :
crntBits = set & 0xff
result.append(chr(crntBits))
set >>= 8
properSize = (len(grammar[1]) / 8) + 1
if len(result) < properSize :
result.append('\x00' * (properSize - len(result)))
dfa[4] = "".join(result)
return grammar
if __name__ == '__main__' :
# a simple CLI
import sys, getopt
tgt, pgen, inline = None, "snakes.lang.pgen", False
try :
opts, args = getopt.getopt(sys.argv[1:], "h",
["help", "inline", "output=",
"pgen=", "start="])
if ("-h", "") in opts or ("--help", "") in opts :
opts = [("-h", "")]
args = [None]
elif not args :
raise getopt.GetoptError("no input file provided"
" (try -h to get help)")
elif len(args) > 1 :
raise getopt.GetoptError("more than one input file provided")
except getopt.GetoptError :
sys.stderr.write("%s: %s\n" % (__file__, sys.exc_info()[1]))
sys.exit(1)
for (flag, arg) in opts :
if flag in ("-h", "--help") :
print("""usage: %s [OPTIONS] INFILE
Options:
-h, --help print this help and exit
--inline inline 'pgen.py' in the generated file
--output=OUTPUT set output file
--pgen=PGEN name of 'pgen' module in output file""" % __file__)
sys.exit(0)
elif flag == "--inline" :
inline = True
elif flag == "--output" :
tgt = arg
elif flag == "--pgen" :
pgen = arg
PyPgen.translate(args[0], tgt=tgt, pgen=pgen, inline=inline)