← Back to branch summary

~bzr-pqm/bzr/bzr.dev

« back to all changes in this revision

Viewing changes to bzrlib/cmdline.py

  • Committer: Ian Clatworthy
  • Date: 2007-11-30 04:28:32 UTC
  • mto: (3054.1.1 ianc-integration)
  • mto: This revision was merged to the branch mainline in revision 3055.
  • Revision ID: ian.clatworthy@internode.on.net-20071130042832-6prruj0kzg3fodm8
chapter 2 tweaks

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
 
            context.token.append(u'')
76
 
            return self.exit_state
77
 
        else:
78
 
            context.token.append(next_char)
79
 
            return self
80
 
 
81
 
 
82
 
class _Backslash(object):
83
 
    # See http://msdn.microsoft.com/en-us/library/bb776391(VS.85).aspx
84
 
    def __init__(self, exit_state):
85
 
        self.exit_state = exit_state
86
 
        self.count = 1
87
 
 
88
 
    def process(self, next_char, context):
89
 
        if next_char == u'\\':
90
 
            self.count += 1
91
 
            return self
92
 
        elif next_char in context.allowed_quote_chars:
93
 
            # 2N backslashes followed by a quote are N backslashes
94
 
            context.token.append(u'\\' * (self.count/2))
95
 
            # 2N+1 backslashes follwed by a quote are N backslashes followed by
96
 
            # the quote which should not be processed as the start or end of
97
 
            # the quoted arg
98
 
            if self.count % 2 == 1:
99
 
                # odd number of \ escapes the quote
100
 
                context.token.append(next_char)
101
 
            else:
102
 
                # let exit_state handle next_char
103
 
                context.seq.pushback(next_char)
104
 
            self.count = 0
105
 
            return self.exit_state
106
 
        else:
107
 
            # N backslashes not followed by a quote are just N backslashes
108
 
            if self.count > 0:
109
 
                context.token.append(u'\\' * self.count)
110
 
                self.count = 0
111
 
            # let exit_state handle next_char
112
 
            context.seq.pushback(next_char)
113
 
            return self.exit_state
114
 
 
115
 
    def finish(self, context):
116
 
        if self.count > 0:
117
 
            context.token.append(u'\\' * self.count)
118
 
 
119
 
 
120
 
class _Word(object):
121
 
    def process(self, next_char, context):
122
 
        if _whitespace_match(next_char):
123
 
            return None
124
 
        elif next_char in context.allowed_quote_chars:
125
 
            return _Quotes(next_char, self)
126
 
        elif next_char == u'\\':
127
 
            return _Backslash(self)
128
 
        else:
129
 
            context.token.append(next_char)
130
 
            return self
131
 
 
132
 
 
133
 
class Splitter(object):
134
 
    def __init__(self, command_line, single_quotes_allowed):
135
 
        self.seq = _PushbackSequence(command_line)
136
 
        self.allowed_quote_chars = u'"'
137
 
        if single_quotes_allowed:
138
 
            self.allowed_quote_chars += u"'"
139
 
 
140
 
    def __iter__(self):
141
 
        return self
142
 
 
143
 
    def next(self):
144
 
        quoted, token = self._get_token()
145
 
        if token is None:
146
 
            raise StopIteration
147
 
        return quoted, token
148
 
 
149
 
    def _get_token(self):
150
 
        self.quoted = False
151
 
        self.token = []
152
 
        state = _Whitespace()
153
 
        for next_char in self.seq:
154
 
            state = state.process(next_char, self)
155
 
            if state is None:
156
 
                break
157
 
        if not state is None and not getattr(state, 'finish', None) is None:
158
 
            state.finish(self)
159
 
        result = u''.join(self.token)
160
 
        if not self.quoted and result == '':
161
 
            result = None
162
 
        return self.quoted, result
163
 
 
164
 
 
165
 
def split(unsplit, single_quotes_allowed=True):
166
 
    splitter = Splitter(unsplit, single_quotes_allowed=single_quotes_allowed)
167
 
    return [arg for quoted, arg in splitter]