1#!/usr/bin/env python3
2
3"""List all those Python files that require a coding directive
4
5Usage: findnocoding.py dir1 [dir2...]
6"""
7
8__author__ = "Oleg Broytmann, Georg Brandl"
9
10import sys, os, re, getopt
11
12# our pysource module finds Python source files
13try:
14    import pysource
15except ImportError:
16    # emulate the module with a simple os.walk
17    class pysource:
18        has_python_ext = looks_like_python = can_be_compiled = None
19        def walk_python_files(self, paths, *args, **kwargs):
20            for path in paths:
21                if os.path.isfile(path):
22                    yield path.endswith(".py")
23                elif os.path.isdir(path):
24                    for root, dirs, files in os.walk(path):
25                        for filename in files:
26                            if filename.endswith(".py"):
27                                yield os.path.join(root, filename)
28    pysource = pysource()
29
30
31    print("The pysource module is not available; "
32                         "no sophisticated Python source file search will be done.", file=sys.stderr)
33
34
35decl_re = re.compile(rb'^[ \t\f]*#.*?coding[:=][ \t]*([-\w.]+)')
36blank_re = re.compile(rb'^[ \t\f]*(?:[#\r\n]|$)')
37
38def get_declaration(line):
39    match = decl_re.match(line)
40    if match:
41        return match.group(1)
42    return b''
43
44def has_correct_encoding(text, codec):
45    try:
46        str(text, codec)
47    except UnicodeDecodeError:
48        return False
49    else:
50        return True
51
52def needs_declaration(fullpath):
53    try:
54        infile = open(fullpath, 'rb')
55    except IOError: # Oops, the file was removed - ignore it
56        return None
57
58    with infile:
59        line1 = infile.readline()
60        line2 = infile.readline()
61
62        if (get_declaration(line1) or
63            blank_re.match(line1) and get_declaration(line2)):
64            # the file does have an encoding declaration, so trust it
65            return False
66
67        # check the whole file for non utf-8 characters
68        rest = infile.read()
69
70    if has_correct_encoding(line1+line2+rest, "utf-8"):
71        return False
72
73    return True
74
75
76usage = """Usage: %s [-cd] paths...
77    -c: recognize Python source files trying to compile them
78    -d: debug output""" % sys.argv[0]
79
80if __name__ == '__main__':
81
82    try:
83        opts, args = getopt.getopt(sys.argv[1:], 'cd')
84    except getopt.error as msg:
85        print(msg, file=sys.stderr)
86        print(usage, file=sys.stderr)
87        sys.exit(1)
88
89    is_python = pysource.looks_like_python
90    debug = False
91
92    for o, a in opts:
93        if o == '-c':
94            is_python = pysource.can_be_compiled
95        elif o == '-d':
96            debug = True
97
98    if not args:
99        print(usage, file=sys.stderr)
100        sys.exit(1)
101
102    for fullpath in pysource.walk_python_files(args, is_python):
103        if debug:
104            print("Testing for coding: %s" % fullpath)
105        result = needs_declaration(fullpath)
106        if result:
107            print(fullpath)
108