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
|
import itertools
import os
import re
import sys
def get_characters():
"""Find every Unicode character that is valid in a Python `identifier`_ but
is not matched by the regex ``\\w`` group.
``\\w`` matches some characters that aren't valid in identifiers, but
:meth:`str.isidentifier` will catch that later in lexing.
All start characters are valid continue characters, so we only test for
continue characters.
_identifier: https://docs.python.org/3/reference/lexical_analysis.html#identifiers
"""
for cp in range(sys.maxunicode + 1):
s = chr(cp)
if ("a" + s).isidentifier() and not re.match(r"\w", s):
yield s
def collapse_ranges(data):
"""Given a sorted list of unique characters, generate ranges representing
sequential code points.
Source: https://stackoverflow.com/a/4629241/400617
"""
for _, b in itertools.groupby(enumerate(data), lambda x: ord(x[1]) - x[0]):
b = list(b)
yield b[0][1], b[-1][1]
def build_pattern(ranges):
"""Output the regex pattern for ranges of characters.
One and two character ranges output the individual characters.
"""
out = []
for a, b in ranges:
if a == b: # single char
out.append(a)
elif ord(b) - ord(a) == 1: # two chars, range is redundant
out.append(a)
out.append(b)
else:
out.append(f"{a}-{b}")
return "".join(out)
def main():
"""Build the regex pattern and write it to
``jinja2/_identifier.py``.
"""
pattern = build_pattern(collapse_ranges(get_characters()))
filename = os.path.abspath(
os.path.join(os.path.dirname(__file__), "..", "src", "jinja2", "_identifier.py")
)
with open(filename, "w", encoding="utf8") as f:
f.write("import re\n\n")
f.write("# generated by scripts/generate_identifier_pattern.py\n")
f.write("pattern = re.compile(\n")
f.write(f' r"[\\w{pattern}]+" # noqa: B950\n')
f.write(")\n")
if __name__ == "__main__":
main()
|
