Mercurial > repos > shellac > guppy_basecaller

comparison env/lib/python3.7/site-packages/chardet/cli/chardetect.py @ 5:9b1c78e6ba9c draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload commit 6c0a8142489327ece472c84e558c47da711a9142"
author shellac
date Mon, 01 Jun 2020 08:59:25 -0400
parents 79f47841a781
children
comparison
equal deleted inserted replaced
4:79f47841a781 5:9b1c78e6ba9c
1 #!/usr/bin/env python
2 """
3 Script which takes one or more file paths and reports on their detected
4 encodings
5
6 Example::
7
8 % chardetect somefile someotherfile
9 somefile: windows-1252 with confidence 0.5
10 someotherfile: ascii with confidence 1.0
11
12 If no paths are provided, it takes its input from stdin.
13
14 """
15
16 from __future__ import absolute_import, print_function, unicode_literals
17
18 import argparse
19 import sys
20
21 from chardet import __version__
22 from chardet.compat import PY2
23 from chardet.universaldetector import UniversalDetector
24
25
26 def description_of(lines, name='stdin'):
27 """
28 Return a string describing the probable encoding of a file or
29 list of strings.
30
31 :param lines: The lines to get the encoding of.
32 :type lines: Iterable of bytes
33 :param name: Name of file or collection of lines
34 :type name: str
35 """
36 u = UniversalDetector()
37 for line in lines:
38 line = bytearray(line)
39 u.feed(line)
40 # shortcut out of the loop to save reading further - particularly useful if we read a BOM.
41 if u.done:
42 break
43 u.close()
44 result = u.result
45 if PY2:
46 name = name.decode(sys.getfilesystemencoding(), 'ignore')
47 if result['encoding']:
48 return '{0}: {1} with confidence {2}'.format(name, result['encoding'],
49 result['confidence'])
50 else:
51 return '{0}: no result'.format(name)
52
53
54 def main(argv=None):
55 """
56 Handles command line arguments and gets things started.
57
58 :param argv: List of arguments, as if specified on the command-line.
59 If None, ``sys.argv[1:]`` is used instead.
60 :type argv: list of str
61 """
62 # Get command line arguments
63 parser = argparse.ArgumentParser(
64 description="Takes one or more file paths and reports their detected \
65 encodings")
66 parser.add_argument('input',
67 help='File whose encoding we would like to determine. \
68 (default: stdin)',
69 type=argparse.FileType('rb'), nargs='*',
70 default=[sys.stdin if PY2 else sys.stdin.buffer])
71 parser.add_argument('--version', action='version',
72 version='%(prog)s {0}'.format(__version__))
73 args = parser.parse_args(argv)
74
75 for f in args.input:
76 if f.isatty():
77 print("You are running chardetect interactively. Press " +
78 "CTRL-D twice at the start of a blank line to signal the " +
79 "end of your input. If you want help, run chardetect " +
80 "--help\n", file=sys.stderr)
81 print(description_of(f, f.name))
82
83
84 if __name__ == '__main__':
85 main()