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
|
# Copyright 2004-2005 Elemental Security, Inc. All Rights Reserved.
# Licensed to PSF under a Contributor Agreement.
# Modifications:
# Copyright David Halter and Contributors
# Modifications are dual-licensed: MIT and PSF.
from typing import Optional, Iterator, Tuple, List
from parso.python.tokenize import tokenize
from parso.utils import parse_version_string
from parso.python.token import PythonTokenTypes
class NFAArc:
def __init__(self, next_: 'NFAState', nonterminal_or_string: Optional[str]):
self.next: NFAState = next_
self.nonterminal_or_string: Optional[str] = nonterminal_or_string
def __repr__(self):
return '<%s: %s>' % (self.__class__.__name__, self.nonterminal_or_string)
class NFAState:
def __init__(self, from_rule: str):
self.from_rule: str = from_rule
self.arcs: List[NFAArc] = []
def add_arc(self, next_, nonterminal_or_string=None):
assert nonterminal_or_string is None or isinstance(nonterminal_or_string, str)
assert isinstance(next_, NFAState)
self.arcs.append(NFAArc(next_, nonterminal_or_string))
def __repr__(self):
return '<%s: from %s>' % (self.__class__.__name__, self.from_rule)
class GrammarParser:
"""
The parser for Python grammar files.
"""
def __init__(self, bnf_grammar: str):
self._bnf_grammar = bnf_grammar
self.generator = tokenize(
bnf_grammar,
version_info=parse_version_string('3.9')
)
self._gettoken() # Initialize lookahead
def parse(self) -> Iterator[Tuple[NFAState, NFAState]]:
# grammar: (NEWLINE | rule)* ENDMARKER
while self.type != PythonTokenTypes.ENDMARKER:
while self.type == PythonTokenTypes.NEWLINE:
self._gettoken()
# rule: NAME ':' rhs NEWLINE
self._current_rule_name = self._expect(PythonTokenTypes.NAME)
self._expect(PythonTokenTypes.OP, ':')
a, z = self._parse_rhs()
self._expect(PythonTokenTypes.NEWLINE)
yield a, z
def _parse_rhs(self):
# rhs: items ('|' items)*
a, z = self._parse_items()
if self.value != "|":
return a, z
else:
aa = NFAState(self._current_rule_name)
zz = NFAState(self._current_rule_name)
while True:
# Add the possibility to go into the state of a and come back
# to finish.
aa.add_arc(a)
z.add_arc(zz)
if self.value != "|":
break
self._gettoken()
a, z = self._parse_items()
return aa, zz
def _parse_items(self):
# items: item+
a, b = self._parse_item()
while self.type in (PythonTokenTypes.NAME, PythonTokenTypes.STRING) \
or self.value in ('(', '['):
c, d = self._parse_item()
# Need to end on the next item.
b.add_arc(c)
b = d
return a, b
def _parse_item(self):
# item: '[' rhs ']' | atom ['+' | '*']
if self.value == "[":
self._gettoken()
a, z = self._parse_rhs()
self._expect(PythonTokenTypes.OP, ']')
# Make it also possible that there is no token and change the
# state.
a.add_arc(z)
return a, z
else:
a, z = self._parse_atom()
value = self.value
if value not in ("+", "*"):
return a, z
self._gettoken()
# Make it clear that we can go back to the old state and repeat.
z.add_arc(a)
if value == "+":
return a, z
else:
# The end state is the same as the beginning, nothing must
# change.
return a, a
def _parse_atom(self):
# atom: '(' rhs ')' | NAME | STRING
if self.value == "(":
self._gettoken()
a, z = self._parse_rhs()
self._expect(PythonTokenTypes.OP, ')')
return a, z
elif self.type in (PythonTokenTypes.NAME, PythonTokenTypes.STRING):
a = NFAState(self._current_rule_name)
z = NFAState(self._current_rule_name)
# Make it clear that the state transition requires that value.
a.add_arc(z, self.value)
self._gettoken()
return a, z
else:
self._raise_error("expected (...) or NAME or STRING, got %s/%s",
self.type, self.value)
def _expect(self, type_, value=None):
if self.type != type_:
self._raise_error("expected %s, got %s [%s]",
type_, self.type, self.value)
if value is not None and self.value != value:
self._raise_error("expected %s, got %s", value, self.value)
value = self.value
self._gettoken()
return value
def _gettoken(self):
tup = next(self.generator)
self.type, self.value, self.begin, prefix = tup
def _raise_error(self, msg, *args):
if args:
try:
msg = msg % args
except:
msg = " ".join([msg] + list(map(str, args)))
line = self._bnf_grammar.splitlines()[self.begin[0] - 1]
raise SyntaxError(msg, ('<grammar>', self.begin[0],
self.begin[1], line))
|
