src/rdflib/compat.py
1"""
2Utility functions and objects to ease Python 2/3 compatibility,
3and different versions of support libraries.
4"""
5
6from __future__ import annotations
7
8import codecs
9import re
10import warnings
11from typing import Match
12
13
14def cast_bytes(s, enc="utf-8"):
15 if isinstance(s, str):
16 return s.encode(enc)
17 return s
18
19
20def ascii(stream):
21 return codecs.getreader("ascii")(stream)
22
23
24def bopen(*args, **kwargs):
25 # type error: No overload variant of "open" matches argument types "Tuple[Any, ...]", "str", "Dict[str, Any]"
26 return open(*args, mode="rb", **kwargs) # type: ignore[call-overload]
27
28
29long_type = int
30
31
32def sign(n):
33 if n < 0:
34 return -1
35 if n > 0:
36 return 1
37 return 0
38
39
40r_unicodeEscape = re.compile(r"(\\u[0-9A-Fa-f]{4}|\\U[0-9A-Fa-f]{8})") # noqa: N816
41
42
43def _unicodeExpand(s): # noqa: N802
44 return r_unicodeEscape.sub(lambda m: chr(int(m.group(0)[2:], 16)), s)
45
46
47def decodeStringEscape(s): # noqa: N802
48 warnings.warn(
49 DeprecationWarning(
50 "rdflib.compat.decodeStringEscape() is deprecated, "
51 "it will be removed in rdflib 7.0.0. "
52 "This function is not used anywhere in rdflib anymore "
53 "and the utility that it does provide is not implemented correctly."
54 )
55 )
56 r"""
57 s is byte-string - replace \ escapes in string
58 """
59
60 s = s.replace("\\t", "\t")
61 s = s.replace("\\n", "\n")
62 s = s.replace("\\r", "\r")
63 s = s.replace("\\b", "\b")
64 s = s.replace("\\f", "\f")
65 s = s.replace('\\"', '"')
66 s = s.replace("\\'", "'")
67 s = s.replace("\\\\", "\\")
68
69 return s
70 # return _unicodeExpand(s) # hmm - string escape doesn't do unicode escaping
71
72
73_string_escape_map = {
74 "t": "\t",
75 "b": "\b",
76 "n": "\n",
77 "r": "\r",
78 "f": "\f",
79 '"': '"',
80 "'": "'",
81 "\\": "\\",
82}
83
84
85def _turtle_escape_subber(match: Match[str]) -> str:
86 smatch, umatch = match.groups()
87 if smatch is not None:
88 return _string_escape_map[smatch]
89 else:
90 return chr(int(umatch[1:], 16))
91
92
93_turtle_escape_pattern = re.compile(
94 r"""\\(?:([tbnrf"'\\])|(u[0-9A-Fa-f]{4}|U[0-9A-Fa-f]{8}))""",
95)
96
97
98def decodeUnicodeEscape(escaped: str) -> str: # noqa: N802
99 if "\\" not in escaped:
100 # Most of times, there are no backslashes in strings.
101 return escaped
102 return _turtle_escape_pattern.sub(_turtle_escape_subber, escaped)