1 # -*- coding: utf-8 -*- 2 # The LLVM Compiler Infrastructure 3 # 4 # This file is distributed under the University of Illinois Open Source 5 # License. See LICENSE.TXT for details. 6 """ This module implements basic shell escaping/unescaping methods. """ 7 8 import re 9 import shlex 10 11 __all__ = ['encode', 'decode'] 12 13 14 def encode(command): 15 """ Takes a command as list and returns a string. """ 16 17 def needs_quote(word): 18 """ Returns true if arguments needs to be protected by quotes. 19 20 Previous implementation was shlex.split method, but that's not good 21 for this job. Currently is running through the string with a basic 22 state checking. """ 23 24 reserved = {' ', '$', '%', '&', '(', ')', '[', ']', '{', '}', '*', '|', 25 '<', '>', '@', '?', '!'} 26 state = 0 27 for current in word: 28 if state == 0 and current in reserved: 29 return True 30 elif state == 0 and current == '\\': 31 state = 1 32 elif state == 1 and current in reserved | {'\\'}: 33 state = 0 34 elif state == 0 and current == '"': 35 state = 2 36 elif state == 2 and current == '"': 37 state = 0 38 elif state == 0 and current == "'": 39 state = 3 40 elif state == 3 and current == "'": 41 state = 0 42 return state != 0 43 44 def escape(word): 45 """ Do protect argument if that's needed. """ 46 47 table = {'\\': '\\\\', '"': '\\"'} 48 escaped = ''.join([table.get(c, c) for c in word]) 49 50 return '"' + escaped + '"' if needs_quote(word) else escaped 51 52 return " ".join([escape(arg) for arg in command]) 53 54 55 def decode(string): 56 """ Takes a command string and returns as a list. """ 57 58 def unescape(arg): 59 """ Gets rid of the escaping characters. """ 60 61 if len(arg) >= 2 and arg[0] == arg[-1] and arg[0] == '"': 62 arg = arg[1:-1] 63 return re.sub(r'\\(["\\])', r'\1', arg) 64 return re.sub(r'\\([\\ $%&\(\)\[\]\{\}\*|<>@?!])', r'\1', arg) 65 66 return [unescape(arg) for arg in shlex.split(string)] 67
