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
|
import codecs
import re
from typing import (IO, Iterator, Match, NamedTuple, Optional, # noqa:F401
Pattern, Sequence, Tuple)
def make_regex(string: str, extra_flags: int = 0) -> Pattern[str]:
return re.compile(string, re.UNICODE | extra_flags)
_newline = make_regex(r"(\r\n|\n|\r)")
_multiline_whitespace = make_regex(r"\s*", extra_flags=re.MULTILINE)
_whitespace = make_regex(r"[^\S\r\n]*")
_export = make_regex(r"(?:export[^\S\r\n]+)?")
_single_quoted_key = make_regex(r"'([^']+)'")
_unquoted_key = make_regex(r"([^=\#\s]+)")
_equal_sign = make_regex(r"(=[^\S\r\n]*)")
_single_quoted_value = make_regex(r"'((?:\\'|[^'])*)'")
_double_quoted_value = make_regex(r'"((?:\\"|[^"])*)"')
_unquoted_value = make_regex(r"([^\r\n]*)")
_comment = make_regex(r"(?:[^\S\r\n]*#[^\r\n]*)?")
_end_of_line = make_regex(r"[^\S\r\n]*(?:\r\n|\n|\r|$)")
_rest_of_line = make_regex(r"[^\r\n]*(?:\r|\n|\r\n)?")
_double_quote_escapes = make_regex(r"\\[\\'\"abfnrtv]")
_single_quote_escapes = make_regex(r"\\[\\']")
class Original(NamedTuple):
string: str
line: int
class Binding(NamedTuple):
key: Optional[str]
value: Optional[str]
original: Original
error: bool
class Position:
def __init__(self, chars: int, line: int) -> None:
self.chars = chars
self.line = line
@classmethod
def start(cls) -> "Position":
return cls(chars=0, line=1)
def set(self, other: "Position") -> None:
self.chars = other.chars
self.line = other.line
def advance(self, string: str) -> None:
self.chars += len(string)
self.line += len(re.findall(_newline, string))
class Error(Exception):
pass
class Reader:
def __init__(self, stream: IO[str]) -> None:
self.string = stream.read()
self.position = Position.start()
self.mark = Position.start()
def has_next(self) -> bool:
return self.position.chars < len(self.string)
def set_mark(self) -> None:
self.mark.set(self.position)
def get_marked(self) -> Original:
return Original(
string=self.string[self.mark.chars:self.position.chars],
line=self.mark.line,
)
def peek(self, count: int) -> str:
return self.string[self.position.chars:self.position.chars + count]
def read(self, count: int) -> str:
result = self.string[self.position.chars:self.position.chars + count]
if len(result) < count:
raise Error("read: End of string")
self.position.advance(result)
return result
def read_regex(self, regex: Pattern[str]) -> Sequence[str]:
match = regex.match(self.string, self.position.chars)
if match is None:
raise Error("read_regex: Pattern not found")
self.position.advance(self.string[match.start():match.end()])
return match.groups()
def decode_escapes(regex: Pattern[str], string: str) -> str:
def decode_match(match: Match[str]) -> str:
return codecs.decode(match.group(0), 'unicode-escape') # type: ignore
return regex.sub(decode_match, string)
def parse_key(reader: Reader) -> Optional[str]:
char = reader.peek(1)
if char == "#":
return None
elif char == "'":
(key,) = reader.read_regex(_single_quoted_key)
else:
(key,) = reader.read_regex(_unquoted_key)
return key
def parse_unquoted_value(reader: Reader) -> str:
(part,) = reader.read_regex(_unquoted_value)
return re.sub(r"\s+#.*", "", part).rstrip()
def parse_value(reader: Reader) -> str:
char = reader.peek(1)
if char == u"'":
(value,) = reader.read_regex(_single_quoted_value)
return decode_escapes(_single_quote_escapes, value)
elif char == u'"':
(value,) = reader.read_regex(_double_quoted_value)
return decode_escapes(_double_quote_escapes, value)
elif char in (u"", u"\n", u"\r"):
return u""
else:
return parse_unquoted_value(reader)
def parse_binding(reader: Reader) -> Binding:
reader.set_mark()
try:
reader.read_regex(_multiline_whitespace)
if not reader.has_next():
return Binding(
key=None,
value=None,
original=reader.get_marked(),
error=False,
)
reader.read_regex(_export)
key = parse_key(reader)
reader.read_regex(_whitespace)
if reader.peek(1) == "=":
reader.read_regex(_equal_sign)
value: Optional[str] = parse_value(reader)
else:
value = None
reader.read_regex(_comment)
reader.read_regex(_end_of_line)
return Binding(
key=key,
value=value,
original=reader.get_marked(),
error=False,
)
except Error:
reader.read_regex(_rest_of_line)
return Binding(
key=None,
value=None,
original=reader.get_marked(),
error=True,
)
def parse_stream(stream: IO[str]) -> Iterator[Binding]:
reader = Reader(stream)
while reader.has_next():
yield parse_binding(reader)
|
