1# -*- coding: utf-8 -*-
2# Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
3# See https://llvm.org/LICENSE.txt for license information.
4# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
5""" This module implements basic shell escaping/unescaping methods. """
6
7import re
8import shlex
9
10__all__ = ['encode', 'decode']
11
12
13def encode(command):
14    """ Takes a command as list and returns a string. """
15
16    def needs_quote(word):
17        """ Returns true if arguments needs to be protected by quotes.
18
19        Previous implementation was shlex.split method, but that's not good
20        for this job. Currently is running through the string with a basic
21        state checking. """
22
23        reserved = {' ', '$', '%', '&', '(', ')', '[', ']', '{', '}', '*', '|',
24                    '<', '>', '@', '?', '!'}
25        state = 0
26        for current in word:
27            if state == 0 and current in reserved:
28                return True
29            elif state == 0 and current == '\\':
30                state = 1
31            elif state == 1 and current in reserved | {'\\'}:
32                state = 0
33            elif state == 0 and current == '"':
34                state = 2
35            elif state == 2 and current == '"':
36                state = 0
37            elif state == 0 and current == "'":
38                state = 3
39            elif state == 3 and current == "'":
40                state = 0
41        return state != 0
42
43    def escape(word):
44        """ Do protect argument if that's needed. """
45
46        table = {'\\': '\\\\', '"': '\\"'}
47        escaped = ''.join([table.get(c, c) for c in word])
48
49        return '"' + escaped + '"' if needs_quote(word) else escaped
50
51    return " ".join([escape(arg) for arg in command])
52
53
54def decode(string):
55    """ Takes a command string and returns as a list. """
56
57    def unescape(arg):
58        """ Gets rid of the escaping characters. """
59
60        if len(arg) >= 2 and arg[0] == arg[-1] and arg[0] == '"':
61            arg = arg[1:-1]
62            return re.sub(r'\\(["\\])', r'\1', arg)
63        return re.sub(r'\\([\\ $%&\(\)\[\]\{\}\*|<>@?!])', r'\1', arg)
64
65    return [unescape(arg) for arg in shlex.split(string)]
66