← Back to branch summary

~bzr-pqm/bzr/bzr.dev

« back to all changes in this revision

Viewing changes to bzrlib/xml_serializer.py

  • Committer: Vincent Ladeuil
  • Date: 2007-07-18 09:43:41 UTC
  • mto: (2778.5.1 vila)
  • mto: This revision was merged to the branch mainline in revision 2789.
  • Revision ID: v.ladeuil+lp@free.fr-20070718094341-edmgsog3el06yqow
Add performance analysis of missing.

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
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):
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
 
        """
 
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):
73
62
        try:
74
 
            return self._unpack_inventory(fromstring(xml_string), revision_id,
75
 
                                          entry_cache=entry_cache)
 
63
            return self._unpack_inventory(fromstring(xml_string))
76
64
        except ParseError, e:
77
65
            raise errors.UnexpectedInventoryFormat(e)
78
66
 
79
 
    def read_inventory(self, f, revision_id=None):
 
67
    def read_inventory(self, f):
80
68
        try:
81
 
            return self._unpack_inventory(self._read_element(f),
82
 
                revision_id=None)
 
69
            return self._unpack_inventory(self._read_element(f))
83
70
        except ParseError, e:
84
71
            raise errors.UnexpectedInventoryFormat(e)
85
72
 
117
104
    }
118
105
def _escape_replace(match, map=escape_map):
119
106
    return map[match.group()]
120
 
 
 
107
 
121
108
def _escape_attrib(text, encoding=None, replace=None):
122
109
    # escape attribute value
123
110
    try:
148
135
    }
149
136
def _escape_cdata_replace(match, map=escape_cdata_map):
150
137
    return map[match.group()]
151
 
 
 
138
 
152
139
def _escape_cdata(text, encoding=None, replace=None):
153
140
    # escape character data
154
141
    try:
168
155
        elementtree.ElementTree._raise_serialization_error(text)
169
156
 
170
157
elementtree.ElementTree._escape_cdata = _escape_cdata
171
 
 
172
 
 
173
 
def escape_invalid_chars(message):
174
 
    """Escape the XML-invalid characters in a commit message.
175
 
 
176
 
    :param message: Commit message to escape
177
 
    :return: tuple with escaped message and number of characters escaped
178
 
    """
179
 
    if message is None:
180
 
        return None, 0
181
 
    # Python strings can include characters that can't be
182
 
    # represented in well-formed XML; escape characters that
183
 
    # aren't listed in the XML specification
184
 
    # (http://www.w3.org/TR/REC-xml/#NT-Char).
185
 
    return re.subn(u'[^\x09\x0A\x0D\u0020-\uD7FF\uE000-\uFFFD]+',
186
 
            lambda match: match.group(0).encode('unicode_escape'),
187
 
            message)