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
|
import re
import codecs
from typing import AnyStr, Optional
def always_bytes(str_or_bytes: Optional[AnyStr], *encode_args) -> Optional[bytes]:
if isinstance(str_or_bytes, bytes) or str_or_bytes is None:
return str_or_bytes
elif isinstance(str_or_bytes, str):
return str_or_bytes.encode(*encode_args)
else:
raise TypeError("Expected str or bytes, but got {}.".format(type(str_or_bytes).__name__))
def always_str(str_or_bytes: Optional[AnyStr], *decode_args) -> Optional[str]:
"""
Returns,
str_or_bytes unmodified, if
"""
if isinstance(str_or_bytes, str) or str_or_bytes is None:
return str_or_bytes
elif isinstance(str_or_bytes, bytes):
return str_or_bytes.decode(*decode_args)
else:
raise TypeError("Expected str or bytes, but got {}.".format(type(str_or_bytes).__name__))
def replace_surrogates(text: str, errors='replace') -> str:
"""Convert surrogates to replacement characters (e.g., "\udc80" becomes "�")
by applying a different error handler.
Uses the "replace" error handler by default, but any input
error handler may be specified.
For an introduction to surrogateescape, see https://www.python.org/dev/peps/pep-0383/.
"""
return text.encode('utf-8', 'surrogateescape').decode('utf-8', errors)
# Translate control characters to "safe" characters. This implementation initially
# replaced them with the matching control pictures (http://unicode.org/charts/PDF/U2400.pdf),
# but that turned out to render badly with monospace fonts. We are back to "." therefore.
_control_char_trans = {
x: ord(".") # x + 0x2400 for unicode control group pictures
for x in range(32)
}
_control_char_trans[127] = ord(".") # 0x2421
_control_char_trans_newline = _control_char_trans.copy()
for x in ("\r", "\n", "\t"):
del _control_char_trans_newline[ord(x)]
_control_char_trans = str.maketrans(_control_char_trans)
_control_char_trans_newline = str.maketrans(_control_char_trans_newline)
def escape_control_characters(text: str, keep_spacing=True) -> str:
"""
Replace all unicode C1 control characters from the given text with a single "."
Args:
keep_spacing: If True, tabs and newlines will not be replaced.
"""
if not isinstance(text, str):
raise ValueError("text type must be unicode but is {}".format(type(text).__name__))
trans = _control_char_trans_newline if keep_spacing else _control_char_trans
return text.translate(trans)
def bytes_to_escaped_str(data, keep_spacing=False, escape_single_quotes=False):
"""
Take bytes and return a safe string that can be displayed to the user.
Single quotes are always escaped, double quotes are never escaped:
"'" + bytes_to_escaped_str(...) + "'"
gives a valid Python string.
Args:
keep_spacing: If True, tabs and newlines will not be escaped.
"""
if not isinstance(data, bytes):
raise ValueError("data must be bytes, but is {}".format(data.__class__.__name__))
# We always insert a double-quote here so that we get a single-quoted string back
# https://stackoverflow.com/questions/29019340/why-does-python-use-different-quotes-for-representing-strings-depending-on-their
ret = repr(b'"' + data).lstrip("b")[2:-1]
if not escape_single_quotes:
ret = re.sub(r"(?<!\\)(\\\\)*\\'", lambda m: (m.group(1) or "") + "'", ret)
if keep_spacing:
ret = re.sub(
r"(?<!\\)(\\\\)*\\([nrt])",
lambda m: (m.group(1) or "") + dict(n="\n", r="\r", t="\t")[m.group(2)],
ret
)
return ret
def escaped_str_to_bytes(data):
"""
Take an escaped string and return the unescaped bytes equivalent.
Raises:
ValueError, if the escape sequence is invalid.
"""
if not isinstance(data, str):
raise ValueError("data must be str, but is {}".format(data.__class__.__name__))
# This one is difficult - we use an undocumented Python API here
# as per http://stackoverflow.com/a/23151714/934719
return codecs.escape_decode(data)[0]
def is_mostly_bin(s: bytes) -> bool:
if not s or len(s) == 0:
return False
return sum(
i < 9 or 13 < i < 32 or 126 < i
for i in s[:100]
) / len(s[:100]) > 0.3
def is_xml(s: bytes) -> bool:
return s.strip().startswith(b"<")
def clean_hanging_newline(t):
"""
Many editors will silently add a newline to the final line of a
document (I'm looking at you, Vim). This function fixes this common
problem at the risk of removing a hanging newline in the rare cases
where the user actually intends it.
"""
if t and t[-1] == "\n":
return t[:-1]
return t
def hexdump(s):
"""
Returns:
A generator of (offset, hex, str) tuples
"""
for i in range(0, len(s), 16):
offset = "{:0=10x}".format(i)
part = s[i:i + 16]
x = " ".join("{:0=2x}".format(i) for i in part)
x = x.ljust(47) # 16*2 + 15
part_repr = always_str(escape_control_characters(
part.decode("ascii", "replace").replace(u"\ufffd", u"."),
False
))
yield (offset, x, part_repr)
|
