~bzr-pqm/bzr/bzr.dev

4763.2.4 by John Arbash Meinel
merge bzr.2.1 in preparation for NEWS entry.
1
# Copyright (C) 2005-2010 Canonical Ltd
1887.1.1 by Adeodato Simó
Do not separate paragraphs in the copyright statement with blank lines,
2
#
1 by mbp at sourcefrog
import from baz patch-364
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.
1887.1.1 by Adeodato Simó
Do not separate paragraphs in the copyright statement with blank lines,
7
#
1 by mbp at sourcefrog
import from baz patch-364
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.
1887.1.1 by Adeodato Simó
Do not separate paragraphs in the copyright statement with blank lines,
12
#
1 by mbp at sourcefrog
import from baz patch-364
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
4183.7.1 by Sabin Iacob
update FSF mailing address
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
1 by mbp at sourcefrog
import from baz patch-364
16
17
"""XML externalization support."""
18
48 by Martin Pool
witty comment
19
# "XML is like violence: if it doesn't solve your problem, you aren't
20
# using enough of it." -- various
21
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
22
# importing this module is fairly slow because it has to load several
23
# ElementTree bits
24
4237.3.1 by Jelmer Vernooij
Add new module with generic serializer information; keep XML-specific bits in
25
from bzrlib.serializer import Serializer
5121.2.4 by Jelmer Vernooij
Remove more unused imports.
26
from bzrlib.trace import mutter
1248 by Martin Pool
- new weave based cleanup [broken]
27
802 by Martin Pool
- Remove XMLMixin class in favour of simple pack_xml, unpack_xml functions
28
try:
2039.2.1 by Martin Pool
Load python2.5's ElementTree if present
29
    try:
30
        # it's in this package in python2.5
31
        from xml.etree.cElementTree import (ElementTree, SubElement, Element,
32
            XMLTreeBuilder, fromstring, tostring)
33
        import xml.etree as elementtree
34
    except ImportError:
35
        from cElementTree import (ElementTree, SubElement, Element,
36
                                  XMLTreeBuilder, fromstring, tostring)
3475.1.2 by John Arbash Meinel
Fix missing import
37
        import elementtree.ElementTree
2029.2.1 by Marien Zwart
Handle the different exception (non-c)ElementTree raises.
38
    ParseError = SyntaxError
802 by Martin Pool
- Remove XMLMixin class in favour of simple pack_xml, unpack_xml functions
39
except ImportError:
1185.33.68 by Martin Pool
Emit warning to trace file only if using cElementTree.
40
    mutter('WARNING: using slower ElementTree; consider installing cElementTree'
41
           " and make sure it's on your PYTHONPATH")
2039.2.1 by Martin Pool
Load python2.5's ElementTree if present
42
    # this copy is shipped with bzr
1227 by Martin Pool
- methods to deserialize objects from strings
43
    from util.elementtree.ElementTree import (ElementTree, SubElement,
1248 by Martin Pool
- new weave based cleanup [broken]
44
                                              Element, XMLTreeBuilder,
45
                                              fromstring, tostring)
1772.1.1 by mbp at sourcefrog
Fix up loading of fallback ElementTree
46
    import util.elementtree as elementtree
2029.2.1 by Marien Zwart
Handle the different exception (non-c)ElementTree raises.
47
    from xml.parsers.expat import ExpatError as ParseError
802 by Martin Pool
- Remove XMLMixin class in favour of simple pack_xml, unpack_xml functions
48
1910.2.31 by Aaron Bentley
Fix bugs in basis inventory handling, change filename
49
from bzrlib import errors
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
50
51
4237.3.1 by Jelmer Vernooij
Add new module with generic serializer information; keep XML-specific bits in
52
class XMLSerializer(Serializer):
53
    """Abstract XML object serialize/deserialize"""
1248 by Martin Pool
- new weave based cleanup [broken]
54
4416.5.1 by Jelmer Vernooij
Move squashing of XML-invalid characters to XMLSerializer.
55
    squashes_xml_invalid_characters = True
56
3882.6.23 by John Arbash Meinel
Change the XMLSerializer.read_inventory_from_string api.
57
    def read_inventory_from_string(self, xml_string, revision_id=None,
4849.4.2 by John Arbash Meinel
Change from being a per-serializer attribute to being a per-repo attribute.
58
                                   entry_cache=None, return_from_cache=False):
2889.1.1 by Robert Collins
* The class ``bzrlib.repofmt.knitrepo.KnitRepository3`` has been folded into
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
3169.2.2 by Robert Collins
Add a test to Repository.deserialise_inventory that the resulting ivnentory is the one asked for, and update relevant tests. Also tweak the model 1 to 2 regenerate inventories logic to use the revision trees parent marker which is more accurate in some cases.
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).
3882.6.23 by John Arbash Meinel
Change the XMLSerializer.read_inventory_from_string api.
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.
4849.4.2 by John Arbash Meinel
Change from being a per-serializer attribute to being a per-repo attribute.
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.
2889.1.1 by Robert Collins
* The class ``bzrlib.repofmt.knitrepo.KnitRepository3`` has been folded into
76
        """
1910.2.31 by Aaron Bentley
Fix bugs in basis inventory handling, change filename
77
        try:
3882.6.23 by John Arbash Meinel
Change the XMLSerializer.read_inventory_from_string api.
78
            return self._unpack_inventory(fromstring(xml_string), revision_id,
4849.4.2 by John Arbash Meinel
Change from being a per-serializer attribute to being a per-repo attribute.
79
                                          entry_cache=entry_cache,
80
                                          return_from_cache=return_from_cache)
2029.2.1 by Marien Zwart
Handle the different exception (non-c)ElementTree raises.
81
        except ParseError, e:
1910.2.31 by Aaron Bentley
Fix bugs in basis inventory handling, change filename
82
            raise errors.UnexpectedInventoryFormat(e)
1227 by Martin Pool
- methods to deserialize objects from strings
83
2889.1.1 by Robert Collins
* The class ``bzrlib.repofmt.knitrepo.KnitRepository3`` has been folded into
84
    def read_inventory(self, f, revision_id=None):
1910.2.31 by Aaron Bentley
Fix bugs in basis inventory handling, change filename
85
        try:
4708.2.1 by Martin
Ensure all files opened by bazaar proper are explicitly closed
86
            try:
87
                return self._unpack_inventory(self._read_element(f),
88
                    revision_id=None)
89
            finally:
90
                f.close()
2029.2.1 by Marien Zwart
Handle the different exception (non-c)ElementTree raises.
91
        except ParseError, e:
1910.2.31 by Aaron Bentley
Fix bugs in basis inventory handling, change filename
92
            raise errors.UnexpectedInventoryFormat(e)
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
93
1182 by Martin Pool
- more disentangling of xml storage format from objects
94
    def write_revision(self, rev, f):
95
        self._write_element(self._pack_revision(rev), f)
96
1248 by Martin Pool
- new weave based cleanup [broken]
97
    def write_revision_to_string(self, rev):
1185.16.123 by Martin Pool
Fix syntax of serializer_v5.pack_revision_to_string
98
        return tostring(self._pack_revision(rev)) + '\n'
1248 by Martin Pool
- new weave based cleanup [broken]
99
1182 by Martin Pool
- more disentangling of xml storage format from objects
100
    def read_revision(self, f):
101
        return self._unpack_revision(self._read_element(f))
102
1227 by Martin Pool
- methods to deserialize objects from strings
103
    def read_revision_from_string(self, xml_string):
1248 by Martin Pool
- new weave based cleanup [broken]
104
        return self._unpack_revision(fromstring(xml_string))
1227 by Martin Pool
- methods to deserialize objects from strings
105
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
106
    def _write_element(self, elt, f):
107
        ElementTree(elt).write(f, 'utf-8')
108
        f.write('\n')
109
110
    def _read_element(self, f):
111
        return ElementTree().parse(f)
1713.1.12 by Robert Collins
Improve serialisation of xml performance by overriding elementree's escape routines.
112
113
1772.1.1 by mbp at sourcefrog
Fix up loading of fallback ElementTree
114
# performance tuning for elementree's serialiser. This should be
1713.1.14 by Robert Collins
Review feedback.
115
# sent upstream - RBC 20060523.
1759.2.1 by Jelmer Vernooij
Fix some types (found using aspell).
116
# the functions here are patched into elementtree at runtime.
1713.1.12 by Robert Collins
Improve serialisation of xml performance by overriding elementree's escape routines.
117
import re
1713.1.14 by Robert Collins
Review feedback.
118
escape_re = re.compile("[&'\"<>]")
1713.1.12 by Robert Collins
Improve serialisation of xml performance by overriding elementree's escape routines.
119
escape_map = {
120
    "&":'&amp;',
121
    "'":"&apos;", # FIXME: overkill
122
    "\"":"&quot;",
123
    "<":"&lt;",
124
    ">":"&gt;",
125
    }
126
def _escape_replace(match, map=escape_map):
127
    return map[match.group()]
3943.8.1 by Marius Kruger
remove all trailing whitespace from bzr source
128
1713.1.12 by Robert Collins
Improve serialisation of xml performance by overriding elementree's escape routines.
129
def _escape_attrib(text, encoding=None, replace=None):
130
    # escape attribute value
131
    try:
132
        if encoding:
133
            try:
134
                text = elementtree.ElementTree._encode(text, encoding)
135
            except UnicodeError:
136
                return elementtree.ElementTree._encode_entity(text)
137
        if replace is None:
138
            return escape_re.sub(_escape_replace, text)
139
        else:
140
            text = replace(text, "&", "&amp;")
141
            text = replace(text, "'", "&apos;") # FIXME: overkill
142
            text = replace(text, "\"", "&quot;")
143
            text = replace(text, "<", "&lt;")
144
            text = replace(text, ">", "&gt;")
145
            return text
146
    except (TypeError, AttributeError):
147
        elementtree.ElementTree._raise_serialization_error(text)
148
149
elementtree.ElementTree._escape_attrib = _escape_attrib
150
1713.1.14 by Robert Collins
Review feedback.
151
escape_cdata_re = re.compile("[&<>]")
1713.1.12 by Robert Collins
Improve serialisation of xml performance by overriding elementree's escape routines.
152
escape_cdata_map = {
153
    "&":'&amp;',
154
    "<":"&lt;",
155
    ">":"&gt;",
156
    }
157
def _escape_cdata_replace(match, map=escape_cdata_map):
158
    return map[match.group()]
3943.8.1 by Marius Kruger
remove all trailing whitespace from bzr source
159
1713.1.12 by Robert Collins
Improve serialisation of xml performance by overriding elementree's escape routines.
160
def _escape_cdata(text, encoding=None, replace=None):
161
    # escape character data
162
    try:
163
        if encoding:
164
            try:
165
                text = elementtree.ElementTree._encode(text, encoding)
166
            except UnicodeError:
167
                return elementtree.ElementTree._encode_entity(text)
168
        if replace is None:
169
            return escape_cdata_re.sub(_escape_cdata_replace, text)
170
        else:
171
            text = replace(text, "&", "&amp;")
172
            text = replace(text, "<", "&lt;")
173
            text = replace(text, ">", "&gt;")
174
            return text
175
    except (TypeError, AttributeError):
176
        elementtree.ElementTree._raise_serialization_error(text)
177
178
elementtree.ElementTree._escape_cdata = _escape_cdata
2520.4.101 by Aaron Bentley
Use a registry to look up xml serializers by format
179
180
4222.1.1 by Jelmer Vernooij
Make function for escaping invalid XML characters public.
181
def escape_invalid_chars(message):
182
    """Escape the XML-invalid characters in a commit message.
183
184
    :param message: Commit message to escape
4354.1.1 by Jelmer Vernooij
Fix docstring for bzrlib.xml_serializer.escape_invalid_chars.
185
    :return: tuple with escaped message and number of characters escaped
4222.1.1 by Jelmer Vernooij
Make function for escaping invalid XML characters public.
186
    """
4416.5.1 by Jelmer Vernooij
Move squashing of XML-invalid characters to XMLSerializer.
187
    if message is None:
188
        return None, 0
4222.1.1 by Jelmer Vernooij
Make function for escaping invalid XML characters public.
189
    # Python strings can include characters that can't be
190
    # represented in well-formed XML; escape characters that
191
    # aren't listed in the XML specification
192
    # (http://www.w3.org/TR/REC-xml/#NT-Char).
193
    return re.subn(u'[^\x09\x0A\x0D\u0020-\uD7FF\uE000-\uFFFD]+',
194
            lambda match: match.group(0).encode('unicode_escape'),
195
            message)