← Back to branch summary

~bzr-pqm/bzr/bzr.dev

« back to all changes in this revision

Viewing changes to bzrlib/cmdline.py

merge merge tweaks from aaron, which includes latest .dev

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
 
from __future__ import absolute_import
24
 
 
25
 
import re
26
 
 
27
 
 
28
 
_whitespace_match = re.compile(u'\s', re.UNICODE).match
29
 
 
30
 
 
31
 
class _PushbackSequence(object):
32
 
    def __init__(self, orig):
33
 
        self._iter = iter(orig)
34
 
        self._pushback_buffer = []
35
 
 
36
 
    def next(self):
37
 
        if len(self._pushback_buffer) > 0:
38
 
            return self._pushback_buffer.pop()
39
 
        else:
40
 
            return self._iter.next()
41
 
 
42
 
    def pushback(self, char):
43
 
        self._pushback_buffer.append(char)
44
 
 
45
 
    def __iter__(self):
46
 
        return self
47
 
 
48
 
 
49
 
class _Whitespace(object):
50
 
    def process(self, next_char, context):
51
 
        if _whitespace_match(next_char):
52
 
            if len(context.token) > 0:
53
 
                return None
54
 
            else:
55
 
                return self
56
 
        elif next_char in context.allowed_quote_chars:
57
 
            context.quoted = True
58
 
            return _Quotes(next_char, self)
59
 
        elif next_char == u'\\':
60
 
            return _Backslash(self)
61
 
        else:
62
 
            context.token.append(next_char)
63
 
            return _Word()
64
 
 
65
 
 
66
 
class _Quotes(object):
67
 
    def __init__(self, quote_char, exit_state):
68
 
        self.quote_char = quote_char
69
 
        self.exit_state = exit_state
70
 
 
71
 
    def process(self, next_char, context):
72
 
        if next_char == u'\\':
73
 
            return _Backslash(self)
74
 
        elif next_char == self.quote_char:
75
 
            return self.exit_state
76
 
        else:
77
 
            context.token.append(next_char)
78
 
            return self
79
 
 
80
 
 
81
 
class _Backslash(object):
82
 
    # See http://msdn.microsoft.com/en-us/library/bb776391(VS.85).aspx
83
 
    def __init__(self, exit_state):
84
 
        self.exit_state = exit_state
85
 
        self.count = 1
86
 
 
87
 
    def process(self, next_char, context):
88
 
        if next_char == u'\\':
89
 
            self.count += 1
90
 
            return self
91
 
        elif next_char in context.allowed_quote_chars:
92
 
            # 2N backslashes followed by a quote are N backslashes
93
 
            context.token.append(u'\\' * (self.count/2))
94
 
            # 2N+1 backslashes follwed by a quote are N backslashes followed by
95
 
            # the quote which should not be processed as the start or end of
96
 
            # the quoted arg
97
 
            if self.count % 2 == 1:
98
 
                # odd number of \ escapes the quote
99
 
                context.token.append(next_char)
100
 
            else:
101
 
                # let exit_state handle next_char
102
 
                context.seq.pushback(next_char)
103
 
            self.count = 0
104
 
            return self.exit_state
105
 
        else:
106
 
            # N backslashes not followed by a quote are just N backslashes
107
 
            if self.count > 0:
108
 
                context.token.append(u'\\' * self.count)
109
 
                self.count = 0
110
 
            # let exit_state handle next_char
111
 
            context.seq.pushback(next_char)
112
 
            return self.exit_state
113
 
 
114
 
    def finish(self, context):
115
 
        if self.count > 0:
116
 
            context.token.append(u'\\' * self.count)
117
 
 
118
 
 
119
 
class _Word(object):
120
 
    def process(self, next_char, context):
121
 
        if _whitespace_match(next_char):
122
 
            return None
123
 
        elif next_char in context.allowed_quote_chars:
124
 
            return _Quotes(next_char, self)
125
 
        elif next_char == u'\\':
126
 
            return _Backslash(self)
127
 
        else:
128
 
            context.token.append(next_char)
129
 
            return self
130
 
 
131
 
 
132
 
class Splitter(object):
133
 
    def __init__(self, command_line, single_quotes_allowed):
134
 
        self.seq = _PushbackSequence(command_line)
135
 
        self.allowed_quote_chars = u'"'
136
 
        if single_quotes_allowed:
137
 
            self.allowed_quote_chars += u"'"
138
 
 
139
 
    def __iter__(self):
140
 
        return self
141
 
 
142
 
    def next(self):
143
 
        quoted, token = self._get_token()
144
 
        if token is None:
145
 
            raise StopIteration
146
 
        return quoted, token
147
 
 
148
 
    def _get_token(self):
149
 
        self.quoted = False
150
 
        self.token = []
151
 
        state = _Whitespace()
152
 
        for next_char in self.seq:
153
 
            state = state.process(next_char, self)
154
 
            if state is None:
155
 
                break
156
 
        if not state is None and not getattr(state, 'finish', None) is None:
157
 
            state.finish(self)
158
 
        result = u''.join(self.token)
159
 
        if not self.quoted and result == '':
160
 
            result = None
161
 
        return self.quoted, result
162
 
 
163
 
 
164
 
def split(unsplit, single_quotes_allowed=True):
165
 
    splitter = Splitter(unsplit, single_quotes_allowed=single_quotes_allowed)
166
 
    return [arg for quoted, arg in splitter]