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
|
#!/usr/bin/env python3
"""Convert nixfmt's 2-space indentation to tabs.
Usage: ./tabify-nix.py [FILE...] (default: every git-tracked .nix file)
Only lines that start in code are touched. Lines that start inside a string
('' or ") are left exactly as they are, because Nix only strips leading
*spaces* from '' strings, so tabs there would change the string's value.
"""
import subprocess
import sys
INDENT = 2
def lex_line(line, stack):
"""Advance the lexer state `stack` over one line (mutates it in place).
Stack entries: ["code", brace_depth], ["ind"] (''), ["dq"] ("), ["block"].
A "code" entry above a string entry is a ${ } interpolation.
"""
i, n = 0, len(line)
while i < n:
mode = stack[-1][0]
rest = line[i:]
if mode == "code":
c = line[i]
if c == "#":
return
if rest.startswith("/*"):
stack.append(["block"])
i += 2
elif c == '"':
stack.append(["dq"])
i += 1
elif rest.startswith("''"):
stack.append(["ind"])
i += 2
elif c == "{":
stack[-1][1] += 1
i += 1
elif c == "}":
if stack[-1][1] == 0 and len(stack) > 1:
stack.pop() # end of ${ } interpolation
else:
stack[-1][1] -= 1
i += 1
else:
i += 1
elif mode == "block":
if rest.startswith("*/"):
stack.pop()
i += 2
else:
i += 1
elif mode == "dq":
if rest.startswith("\\"):
i += 2
elif rest.startswith("$${"):
i += 3
elif rest.startswith("${"):
stack.append(["code", 0])
i += 2
elif rest.startswith('"'):
stack.pop()
i += 1
else:
i += 1
elif mode == "ind":
if rest.startswith("'''") or rest.startswith("''$"):
i += 3
elif rest.startswith("''\\"):
i += 4
elif rest.startswith("''"):
stack.pop()
i += 2
elif rest.startswith("$${"):
i += 3
elif rest.startswith("${"):
stack.append(["code", 0])
i += 2
else:
i += 1
def tabify(text):
stack = [["code", 0]]
out = []
for line in text.split("\n"):
if stack[-1][0] in ("code", "block"):
stripped = line.lstrip(" \t")
ws = line[: len(line) - len(stripped)]
if stripped:
tabs = ws.count("\t")
spaces = ws.count(" ")
line = "\t" * (tabs + spaces // INDENT) + " " * (spaces % INDENT) + stripped
else:
line = ""
out.append(line)
lex_line(line, stack)
if stack != [["code", 0]]:
raise ValueError(f"unbalanced lexer state at EOF: {stack}")
return "\n".join(out)
def main():
files = sys.argv[1:] or subprocess.run(
["git", "ls-files", "*.nix"], capture_output=True, text=True, check=True
).stdout.split()
failed = False
for path in files:
with open(path) as f:
src = f.read()
try:
new = tabify(src)
except ValueError as e:
print(f"{path}: {e}, skipped", file=sys.stderr)
failed = True
continue
if new != src:
with open(path, "w") as f:
f.write(new)
print(f"tabified {path}")
sys.exit(1 if failed else 0)
if __name__ == "__main__":
main()
|
