1#!/usr/bin/env python
2
3"""List all those Python files that require a coding directive
4
5Usage: nocoding.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 >>sys.stderr, ("The pysource module is not available; "
32                         "no sophisticated Python source file search will be done.")
33
34
35decl_re = re.compile(r'^[ \t\f]*#.*?coding[:=][ \t]*([-\w.]+)')
36blank_re = re.compile(r'^[ \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        unicode(text, codec)
47    except UnicodeDecodeError:
48        return False
49    else:
50        return True
51
52def needs_declaration(fullpath):
53    try:
54        infile = open(fullpath, 'rU')
55    except IOError: # Oops, the file was removed - ignore it
56        return None
57
58    line1 = infile.readline()
59    line2 = infile.readline()
60
61    if (get_declaration(line1) or
62        blank_re.match(line1) and get_declaration(line2)):
63        # the file does have an encoding declaration, so trust it
64        infile.close()
65        return False
66
67    # check the whole file for non-ASCII characters
68    rest = infile.read()
69    infile.close()
70
71    if has_correct_encoding(line1+line2+rest, "ascii"):
72        return False
73
74    return True
75
76
77usage = """Usage: %s [-cd] paths...
78    -c: recognize Python source files trying to compile them
79    -d: debug output""" % sys.argv[0]
80
81try:
82    opts, args = getopt.getopt(sys.argv[1:], 'cd')
83except getopt.error, msg:
84    print >>sys.stderr, msg
85    print >>sys.stderr, usage
86    sys.exit(1)
87
88is_python = pysource.looks_like_python
89debug = False
90
91for o, a in opts:
92    if o == '-c':
93        is_python = pysource.can_be_compiled
94    elif o == '-d':
95        debug = True
96
97if not args:
98    print >>sys.stderr, usage
99    sys.exit(1)
100
101for fullpath in pysource.walk_python_files(args, is_python):
102    if debug:
103        print "Testing for coding: %s" % fullpath
104    result = needs_declaration(fullpath)
105    if result:
106        print fullpath
107