← Back to branch summary

~bzr-pqm/bzr/bzr.dev

« back to all changes in this revision

Viewing changes to bzrlib/xml_serializer.py

  • Committer: John Arbash Meinel
  • Date: 2009-07-30 23:54:26 UTC
  • mto: This revision was merged to the branch mainline in revision 4580.
  • Revision ID: john@arbash-meinel.com-20090730235426-o8h73swbh7seqaf7
Update the breakin support to support CTRL-BREAK on Windows.

The signal handling code is very similar, but the testing code got a bit clumsy.

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
22
22
# importing this module is fairly slow because it has to load several
23
23
# ElementTree bits
24
24
 
25
 
import re
26
 
 
27
25
from bzrlib.serializer import Serializer
28
 
from bzrlib.trace import mutter
 
26
from bzrlib.trace import mutter, warning
29
27
 
30
28
try:
31
29
    try:
33
31
        from xml.etree.cElementTree import (ElementTree, SubElement, Element,
34
32
            XMLTreeBuilder, fromstring, tostring)
35
33
        import xml.etree as elementtree
36
 
        # Also import ElementTree module so monkey-patching below always works
37
 
        import xml.etree.ElementTree
38
34
    except ImportError:
39
35
        from cElementTree import (ElementTree, SubElement, Element,
40
36
                                  XMLTreeBuilder, fromstring, tostring)
59
55
    squashes_xml_invalid_characters = True
60
56
 
61
57
    def read_inventory_from_string(self, xml_string, revision_id=None,
62
 
                                   entry_cache=None, return_from_cache=False):
 
58
                                   entry_cache=None):
63
59
        """Read xml_string into an inventory object.
64
60
 
65
61
        :param xml_string: The xml to read.
73
69
        :param entry_cache: An optional cache of InventoryEntry objects. If
74
70
            supplied we will look up entries via (file_id, revision_id) which
75
71
            should map to a valid InventoryEntry (File/Directory/etc) object.
76
 
        :param return_from_cache: Return entries directly from the cache,
77
 
            rather than copying them first. This is only safe if the caller
78
 
            promises not to mutate the returned inventory entries, but it can
79
 
            make some operations significantly faster.
80
72
        """
81
73
        try:
82
74
            return self._unpack_inventory(fromstring(xml_string), revision_id,
83
 
                                          entry_cache=entry_cache,
84
 
                                          return_from_cache=return_from_cache)
 
75
                                          entry_cache=entry_cache)
85
76
        except ParseError, e:
86
77
            raise errors.UnexpectedInventoryFormat(e)
87
78
 
88
79
    def read_inventory(self, f, revision_id=None):
89
80
        try:
90
 
            try:
91
 
                return self._unpack_inventory(self._read_element(f),
92
 
                    revision_id=None)
93
 
            finally:
94
 
                f.close()
 
81
            return self._unpack_inventory(self._read_element(f),
 
82
                revision_id=None)
95
83
        except ParseError, e:
96
84
            raise errors.UnexpectedInventoryFormat(e)
97
85
 
115
103
        return ElementTree().parse(f)
116
104
 
117
105
 
 
106
# performance tuning for elementree's serialiser. This should be
 
107
# sent upstream - RBC 20060523.
 
108
# the functions here are patched into elementtree at runtime.
 
109
import re
 
110
escape_re = re.compile("[&'\"<>]")
 
111
escape_map = {
 
112
    "&":'&amp;',
 
113
    "'":"&apos;", # FIXME: overkill
 
114
    "\"":"&quot;",
 
115
    "<":"&lt;",
 
116
    ">":"&gt;",
 
117
    }
 
118
def _escape_replace(match, map=escape_map):
 
119
    return map[match.group()]
 
120
 
 
121
def _escape_attrib(text, encoding=None, replace=None):
 
122
    # escape attribute value
 
123
    try:
 
124
        if encoding:
 
125
            try:
 
126
                text = elementtree.ElementTree._encode(text, encoding)
 
127
            except UnicodeError:
 
128
                return elementtree.ElementTree._encode_entity(text)
 
129
        if replace is None:
 
130
            return escape_re.sub(_escape_replace, text)
 
131
        else:
 
132
            text = replace(text, "&", "&amp;")
 
133
            text = replace(text, "'", "&apos;") # FIXME: overkill
 
134
            text = replace(text, "\"", "&quot;")
 
135
            text = replace(text, "<", "&lt;")
 
136
            text = replace(text, ">", "&gt;")
 
137
            return text
 
138
    except (TypeError, AttributeError):
 
139
        elementtree.ElementTree._raise_serialization_error(text)
 
140
 
 
141
elementtree.ElementTree._escape_attrib = _escape_attrib
 
142
 
 
143
escape_cdata_re = re.compile("[&<>]")
 
144
escape_cdata_map = {
 
145
    "&":'&amp;',
 
146
    "<":"&lt;",
 
147
    ">":"&gt;",
 
148
    }
 
149
def _escape_cdata_replace(match, map=escape_cdata_map):
 
150
    return map[match.group()]
 
151
 
 
152
def _escape_cdata(text, encoding=None, replace=None):
 
153
    # escape character data
 
154
    try:
 
155
        if encoding:
 
156
            try:
 
157
                text = elementtree.ElementTree._encode(text, encoding)
 
158
            except UnicodeError:
 
159
                return elementtree.ElementTree._encode_entity(text)
 
160
        if replace is None:
 
161
            return escape_cdata_re.sub(_escape_cdata_replace, text)
 
162
        else:
 
163
            text = replace(text, "&", "&amp;")
 
164
            text = replace(text, "<", "&lt;")
 
165
            text = replace(text, ">", "&gt;")
 
166
            return text
 
167
    except (TypeError, AttributeError):
 
168
        elementtree.ElementTree._raise_serialization_error(text)
 
169
 
 
170
elementtree.ElementTree._escape_cdata = _escape_cdata
 
171
 
 
172
 
118
173
def escape_invalid_chars(message):
119
174
    """Escape the XML-invalid characters in a commit message.
120
175