← Back to branch summary

~bzr-pqm/bzr/bzr.dev

« back to all changes in this revision

Viewing changes to bzrlib/xml_serializer.py

  • Committer: Aaron Bentley
  • Date: 2007-06-21 23:43:17 UTC
  • mto: (2520.5.2 bzr.mpbundle)
  • mto: This revision was merged to the branch mainline in revision 2631.
  • Revision ID: abentley@panoramicfeedback.com-20070621234317-5w3h8h36oe90sups
Implement new merge directive format

Show diffs side-by-side

added added

removed removed

Lines of Context:
bzrlib/xml_serializer.py
1
 
# Copyright (C) 2005-2010 Canonical Ltd
 
1
# Copyright (C) 2005, 2006 Canonical Ltd
2
2
#
3
3
# This program is free software; you can redistribute it and/or modify
4
4
# it under the terms of the GNU General Public License as published by
12
12
#
13
13
# You should have received a copy of the GNU General Public License
14
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
 
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
16
16
 
17
17
"""XML externalization support."""
18
18
 
22
22
# importing this module is fairly slow because it has to load several
23
23
# ElementTree bits
24
24
 
25
 
from bzrlib.serializer import Serializer
26
25
from bzrlib.trace import mutter, warning
27
26
 
28
27
try:
34
33
    except ImportError:
35
34
        from cElementTree import (ElementTree, SubElement, Element,
36
35
                                  XMLTreeBuilder, fromstring, tostring)
37
 
        import elementtree.ElementTree
 
36
        import elementtree
38
37
    ParseError = SyntaxError
39
38
except ImportError:
40
39
    mutter('WARNING: using slower ElementTree; consider installing cElementTree'
49
48
from bzrlib import errors
50
49
 
51
50
 
52
 
class XMLSerializer(Serializer):
53
 
    """Abstract XML object serialize/deserialize"""
54
 
 
55
 
    squashes_xml_invalid_characters = True
56
 
 
57
 
    def read_inventory_from_string(self, xml_string, revision_id=None,
58
 
                                   entry_cache=None, return_from_cache=False):
59
 
        """Read xml_string into an inventory object.
60
 
 
61
 
        :param xml_string: The xml to read.
62
 
        :param revision_id: If not-None, the expected revision id of the
63
 
            inventory. Some serialisers use this to set the results' root
64
 
            revision. This should be supplied for deserialising all
65
 
            from-repository inventories so that xml5 inventories that were
66
 
            serialised without a revision identifier can be given the right
67
 
            revision id (but not for working tree inventories where users can
68
 
            edit the data without triggering checksum errors or anything).
69
 
        :param entry_cache: An optional cache of InventoryEntry objects. If
70
 
            supplied we will look up entries via (file_id, revision_id) which
71
 
            should map to a valid InventoryEntry (File/Directory/etc) object.
72
 
        :param return_from_cache: Return entries directly from the cache,
73
 
            rather than copying them first. This is only safe if the caller
74
 
            promises not to mutate the returned inventory entries, but it can
75
 
            make some operations significantly faster.
76
 
        """
 
51
class Serializer(object):
 
52
    """Abstract object serialize/deserialize"""
 
53
    def write_inventory(self, inv, f):
 
54
        """Write inventory to a file"""
 
55
        elt = self._pack_inventory(inv)
 
56
        self._write_element(elt, f)
 
57
 
 
58
    def write_inventory_to_string(self, inv):
 
59
        return tostring(self._pack_inventory(inv)) + '\n'
 
60
 
 
61
    def read_inventory_from_string(self, xml_string):
77
62
        try:
78
 
            return self._unpack_inventory(fromstring(xml_string), revision_id,
79
 
                                          entry_cache=entry_cache,
80
 
                                          return_from_cache=return_from_cache)
 
63
            return self._unpack_inventory(fromstring(xml_string))
81
64
        except ParseError, e:
82
65
            raise errors.UnexpectedInventoryFormat(e)
83
66
 
84
 
    def read_inventory(self, f, revision_id=None):
 
67
    def read_inventory(self, f):
85
68
        try:
86
 
            return self._unpack_inventory(self._read_element(f),
87
 
                revision_id=None)
 
69
            return self._unpack_inventory(self._read_element(f))
88
70
        except ParseError, e:
89
71
            raise errors.UnexpectedInventoryFormat(e)
90
72
 
122
104
    }
123
105
def _escape_replace(match, map=escape_map):
124
106
    return map[match.group()]
125
 
 
 
107
 
126
108
def _escape_attrib(text, encoding=None, replace=None):
127
109
    # escape attribute value
128
110
    try:
153
135
    }
154
136
def _escape_cdata_replace(match, map=escape_cdata_map):
155
137
    return map[match.group()]
156
 
 
 
138
 
157
139
def _escape_cdata(text, encoding=None, replace=None):
158
140
    # escape character data
159
141
    try:
173
155
        elementtree.ElementTree._raise_serialization_error(text)
174
156
 
175
157
elementtree.ElementTree._escape_cdata = _escape_cdata
176
 
 
177
 
 
178
 
def escape_invalid_chars(message):
179
 
    """Escape the XML-invalid characters in a commit message.
180
 
 
181
 
    :param message: Commit message to escape
182
 
    :return: tuple with escaped message and number of characters escaped
183
 
    """
184
 
    if message is None:
185
 
        return None, 0
186
 
    # Python strings can include characters that can't be
187
 
    # represented in well-formed XML; escape characters that
188
 
    # aren't listed in the XML specification
189
 
    # (http://www.w3.org/TR/REC-xml/#NT-Char).
190
 
    return re.subn(u'[^\x09\x0A\x0D\u0020-\uD7FF\uE000-\uFFFD]+',
191
 
            lambda match: match.group(0).encode('unicode_escape'),
192
 
            message)