← Back to branch summary

~bzr-pqm/bzr/bzr.dev

« back to all changes in this revision

Viewing changes to bzrlib/xml_serializer.py

  • Committer: Daniel Watkins
  • Date: 2007-08-02 18:04:35 UTC
  • mto: This revision was merged to the branch mainline in revision 2705.
  • Revision ID: d.m.watkins@warwick.ac.uk-20070802180435-cs6xalg5tkze209b
tests.blackbox.test_aliases now uses internals where appropriate.

Show diffs side-by-side

added added

removed removed

Lines of Context:
bzrlib/xml_serializer.py
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
 
25
from bzrlib import registry
26
26
from bzrlib.trace import mutter, warning
27
27
 
28
28
try:
34
34
    except ImportError:
35
35
        from cElementTree import (ElementTree, SubElement, Element,
36
36
                                  XMLTreeBuilder, fromstring, tostring)
37
 
        import elementtree.ElementTree
 
37
        import elementtree
38
38
    ParseError = SyntaxError
39
39
except ImportError:
40
40
    mutter('WARNING: using slower ElementTree; consider installing cElementTree'
49
49
from bzrlib import errors
50
50
 
51
51
 
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
 
        """
 
52
class Serializer(object):
 
53
    """Abstract object serialize/deserialize"""
 
54
    def write_inventory(self, inv, f):
 
55
        """Write inventory to a file"""
 
56
        elt = self._pack_inventory(inv)
 
57
        self._write_element(elt, f)
 
58
 
 
59
    def write_inventory_to_string(self, inv):
 
60
        return tostring(self._pack_inventory(inv)) + '\n'
 
61
 
 
62
    def read_inventory_from_string(self, xml_string):
77
63
        try:
78
 
            return self._unpack_inventory(fromstring(xml_string), revision_id,
79
 
                                          entry_cache=entry_cache,
80
 
                                          return_from_cache=return_from_cache)
 
64
            return self._unpack_inventory(fromstring(xml_string))
81
65
        except ParseError, e:
82
66
            raise errors.UnexpectedInventoryFormat(e)
83
67
 
84
 
    def read_inventory(self, f, revision_id=None):
 
68
    def read_inventory(self, f):
85
69
        try:
86
 
            return self._unpack_inventory(self._read_element(f),
87
 
                revision_id=None)
 
70
            return self._unpack_inventory(self._read_element(f))
88
71
        except ParseError, e:
89
72
            raise errors.UnexpectedInventoryFormat(e)
90
73
 
122
105
    }
123
106
def _escape_replace(match, map=escape_map):
124
107
    return map[match.group()]
125
 
 
 
108
 
126
109
def _escape_attrib(text, encoding=None, replace=None):
127
110
    # escape attribute value
128
111
    try:
153
136
    }
154
137
def _escape_cdata_replace(match, map=escape_cdata_map):
155
138
    return map[match.group()]
156
 
 
 
139
 
157
140
def _escape_cdata(text, encoding=None, replace=None):
158
141
    # escape character data
159
142
    try:
175
158
elementtree.ElementTree._escape_cdata = _escape_cdata
176
159
 
177
160
 
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)
 
161
class SerializerRegistry(registry.Registry):
 
162
    """Registry for serializer objects"""
 
163
 
 
164
 
 
165
format_registry = SerializerRegistry()
 
166
format_registry.register_lazy('4', 'bzrlib.xml4', 'serializer_v4')
 
167
format_registry.register_lazy('5', 'bzrlib.xml5', 'serializer_v5')
 
168
format_registry.register_lazy('6', 'bzrlib.xml6', 'serializer_v6')
 
169
format_registry.register_lazy('7', 'bzrlib.xml7', 'serializer_v7')