← Back to branch summary

~bzr-pqm/bzr/bzr.dev

« back to all changes in this revision

Viewing changes to bzrlib/cmdline.py

Merge bzr.dev to resolve conflicts

Show diffs side-by-side

added added

removed removed

Lines of Context:
bzrlib/cmdline.py
 
1
# Copyright (C) 2010-2011 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
The user-visible behaviour of this module is described in
 
20
configuring_bazaar.txt.
 
21
"""
 
22
 
 
23
import re
 
24
 
 
25
 
 
26
_whitespace_match = re.compile(u'\s', re.UNICODE).match
 
27
 
 
28
 
 
29
class _PushbackSequence(object):
 
30
    def __init__(self, orig):
 
31
        self._iter = iter(orig)
 
32
        self._pushback_buffer = []
 
33
 
 
34
    def next(self):
 
35
        if len(self._pushback_buffer) > 0:
 
36
            return self._pushback_buffer.pop()
 
37
        else:
 
38
            return self._iter.next()
 
39
 
 
40
    def pushback(self, char):
 
41
        self._pushback_buffer.append(char)
 
42
 
 
43
    def __iter__(self):
 
44
        return self
 
45
 
 
46
 
 
47
class _Whitespace(object):
 
48
    def process(self, next_char, context):
 
49
        if _whitespace_match(next_char):
 
50
            if len(context.token) > 0:
 
51
                return None
 
52
            else:
 
53
                return self
 
54
        elif next_char in context.allowed_quote_chars:
 
55
            context.quoted = True
 
56
            return _Quotes(next_char, self)
 
57
        elif next_char == u'\\':
 
58
            return _Backslash(self)
 
59
        else:
 
60
            context.token.append(next_char)
 
61
            return _Word()
 
62
 
 
63
 
 
64
class _Quotes(object):
 
65
    def __init__(self, quote_char, exit_state):
 
66
        self.quote_char = quote_char
 
67
        self.exit_state = exit_state
 
68
 
 
69
    def process(self, next_char, context):
 
70
        if next_char == u'\\':
 
71
            return _Backslash(self)
 
72
        elif next_char == self.quote_char:
 
73
            return self.exit_state
 
74
        else:
 
75
            context.token.append(next_char)
 
76
            return self
 
77
 
 
78
 
 
79
class _Backslash(object):
 
80
    # See http://msdn.microsoft.com/en-us/library/bb776391(VS.85).aspx
 
81
    def __init__(self, exit_state):
 
82
        self.exit_state = exit_state
 
83
        self.count = 1
 
84
 
 
85
    def process(self, next_char, context):
 
86
        if next_char == u'\\':
 
87
            self.count += 1
 
88
            return self
 
89
        elif next_char in context.allowed_quote_chars:
 
90
            # 2N backslashes followed by a quote are N backslashes
 
91
            context.token.append(u'\\' * (self.count/2))
 
92
            # 2N+1 backslashes follwed by a quote are N backslashes followed by
 
93
            # the quote which should not be processed as the start or end of
 
94
            # the quoted arg
 
95
            if self.count % 2 == 1:
 
96
                # odd number of \ escapes the quote
 
97
                context.token.append(next_char)
 
98
            else:
 
99
                # let exit_state handle next_char
 
100
                context.seq.pushback(next_char)
 
101
            self.count = 0
 
102
            return self.exit_state
 
103
        else:
 
104
            # N backslashes not followed by a quote are just N backslashes
 
105
            if self.count > 0:
 
106
                context.token.append(u'\\' * self.count)
 
107
                self.count = 0
 
108
            # let exit_state handle next_char
 
109
            context.seq.pushback(next_char)
 
110
            return self.exit_state
 
111
 
 
112
    def finish(self, context):
 
113
        if self.count > 0:
 
114
            context.token.append(u'\\' * self.count)
 
115
 
 
116
 
 
117
class _Word(object):
 
118
    def process(self, next_char, context):
 
119
        if _whitespace_match(next_char):
 
120
            return None
 
121
        elif next_char in context.allowed_quote_chars:
 
122
            return _Quotes(next_char, self)
 
123
        elif next_char == u'\\':
 
124
            return _Backslash(self)
 
125
        else:
 
126
            context.token.append(next_char)
 
127
            return self
 
128
 
 
129
 
 
130
class Splitter(object):
 
131
    def __init__(self, command_line, single_quotes_allowed):
 
132
        self.seq = _PushbackSequence(command_line)
 
133
        self.allowed_quote_chars = u'"'
 
134
        if single_quotes_allowed:
 
135
            self.allowed_quote_chars += u"'"
 
136
 
 
137
    def __iter__(self):
 
138
        return self
 
139
 
 
140
    def next(self):
 
141
        quoted, token = self._get_token()
 
142
        if token is None:
 
143
            raise StopIteration
 
144
        return quoted, token
 
145
 
 
146
    def _get_token(self):
 
147
        self.quoted = False
 
148
        self.token = []
 
149
        state = _Whitespace()
 
150
        for next_char in self.seq:
 
151
            state = state.process(next_char, self)
 
152
            if state is None:
 
153
                break
 
154
        if not state is None and not getattr(state, 'finish', None) is None:
 
155
            state.finish(self)
 
156
        result = u''.join(self.token)
 
157
        if not self.quoted and result == '':
 
158
            result = None
 
159
        return self.quoted, result
 
160
 
 
161
 
 
162
def split(unsplit, single_quotes_allowed=True):
 
163
    splitter = Splitter(unsplit, single_quotes_allowed=single_quotes_allowed)
 
164
    return [arg for quoted, arg in splitter]