← Back to branch summary

~bzr-pqm/bzr/bzr.dev

« back to all changes in this revision

Viewing changes to bzrlib/cmdline.py

  • Committer: Canonical.com Patch Queue Manager
  • Date: 2010-02-18 03:02:28 UTC
  • mfrom: (5037.3.2 integration)
  • Revision ID: pqm@pqm.ubuntu.com-20100218030228-dtbmt8pdtf21ndo7
(mbp, for doxx) better win32 cmdline splitter

Show diffs side-by-side

added added

removed removed

Lines of Context:
bzrlib/cmdline.py
 
1
# Copyright (C) 2010 Canonical Ltd
 
2
#
 
3
# This program is free software; you can redistribute it and/or modify
 
4
# it under the terms of the GNU General Public License as published by
 
5
# the Free Software Foundation; either version 2 of the License, or
 
6
# (at your option) any later version.
 
7
#
 
8
# This program is distributed in the hope that it will be useful,
 
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
 
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 
11
# GNU General Public License for more details.
 
12
#
 
13
# You should have received a copy of the GNU General Public License
 
14
# along with this program; if not, write to the Free Software
 
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 
16
 
 
17
"""Unicode-compatible command-line splitter for all platforms."""
 
18
 
 
19
import re
 
20
 
 
21
 
 
22
_whitespace_match = re.compile(u'\s', re.UNICODE).match
 
23
 
 
24
 
 
25
class _PushbackSequence(object):
 
26
    def __init__(self, orig):
 
27
        self._iter = iter(orig)
 
28
        self._pushback_buffer = []
 
29
        
 
30
    def next(self):
 
31
        if len(self._pushback_buffer) > 0:
 
32
            return self._pushback_buffer.pop()
 
33
        else:
 
34
            return self._iter.next()
 
35
    
 
36
    def pushback(self, char):
 
37
        self._pushback_buffer.append(char)
 
38
        
 
39
    def __iter__(self):
 
40
        return self
 
41
 
 
42
 
 
43
class _Whitespace(object):
 
44
    def process(self, next_char, context):
 
45
        if _whitespace_match(next_char):
 
46
            if len(context.token) > 0:
 
47
                return None
 
48
            else:
 
49
                return self
 
50
        elif next_char in context.allowed_quote_chars:
 
51
            context.quoted = True
 
52
            return _Quotes(next_char, self)
 
53
        elif next_char == u'\\':
 
54
            return _Backslash(self)
 
55
        else:
 
56
            context.token.append(next_char)
 
57
            return _Word()
 
58
 
 
59
 
 
60
class _Quotes(object):
 
61
    def __init__(self, quote_char, exit_state):
 
62
        self.quote_char = quote_char
 
63
        self.exit_state = exit_state
 
64
 
 
65
    def process(self, next_char, context):
 
66
        if next_char == u'\\':
 
67
            return _Backslash(self)
 
68
        elif next_char == self.quote_char:
 
69
            return self.exit_state
 
70
        else:
 
71
            context.token.append(next_char)
 
72
            return self
 
73
 
 
74
 
 
75
class _Backslash(object):
 
76
    # See http://msdn.microsoft.com/en-us/library/bb776391(VS.85).aspx
 
77
    def __init__(self, exit_state):
 
78
        self.exit_state = exit_state
 
79
        self.count = 1
 
80
        
 
81
    def process(self, next_char, context):
 
82
        if next_char == u'\\':
 
83
            self.count += 1
 
84
            return self
 
85
        elif next_char in context.allowed_quote_chars:
 
86
            # 2N backslashes followed by a quote are N backslashes
 
87
            context.token.append(u'\\' * (self.count/2))
 
88
            # 2N+1 backslashes follwed by a quote are N backslashes followed by
 
89
            # the quote which should not be processed as the start or end of
 
90
            # the quoted arg
 
91
            if self.count % 2 == 1:
 
92
                # odd number of \ escapes the quote
 
93
                context.token.append(next_char)
 
94
            else:
 
95
                # let exit_state handle next_char
 
96
                context.seq.pushback(next_char)
 
97
            self.count = 0
 
98
            return self.exit_state
 
99
        else:
 
100
            # N backslashes not followed by a quote are just N backslashes
 
101
            if self.count > 0:
 
102
                context.token.append(u'\\' * self.count)
 
103
                self.count = 0
 
104
            # let exit_state handle next_char
 
105
            context.seq.pushback(next_char)
 
106
            return self.exit_state
 
107
    
 
108
    def finish(self, context):
 
109
        if self.count > 0:
 
110
            context.token.append(u'\\' * self.count)
 
111
 
 
112
 
 
113
class _Word(object):
 
114
    def process(self, next_char, context):
 
115
        if _whitespace_match(next_char):
 
116
            return None
 
117
        elif next_char in context.allowed_quote_chars:
 
118
            return _Quotes(next_char, self)
 
119
        elif next_char == u'\\':
 
120
            return _Backslash(self)
 
121
        else:
 
122
            context.token.append(next_char)
 
123
            return self
 
124
 
 
125
 
 
126
class Splitter(object):
 
127
    def __init__(self, command_line, single_quotes_allowed):
 
128
        self.seq = _PushbackSequence(command_line)
 
129
        self.allowed_quote_chars = u'"'
 
130
        if single_quotes_allowed:
 
131
            self.allowed_quote_chars += u"'"
 
132
    
 
133
    def __iter__(self):
 
134
        return self
 
135
    
 
136
    def next(self):
 
137
        quoted, token = self._get_token()
 
138
        if token is None:
 
139
            raise StopIteration
 
140
        return quoted, token
 
141
    
 
142
    def _get_token(self):
 
143
        self.quoted = False
 
144
        self.token = []
 
145
        state = _Whitespace()
 
146
        for next_char in self.seq:
 
147
            state = state.process(next_char, self)
 
148
            if state is None:
 
149
                break
 
150
        if not state is None and not getattr(state, 'finish', None) is None:
 
151
            state.finish(self)
 
152
        result = u''.join(self.token)
 
153
        if not self.quoted and result == '':
 
154
            result = None
 
155
        return self.quoted, result
 
156
 
 
157
 
 
158
def split(unsplit, single_quotes_allowed=True):
 
159
    splitter = Splitter(unsplit, single_quotes_allowed=single_quotes_allowed)
 
160
    return [arg for quoted, arg in splitter]