2
# -*- coding: UTF-8 -*-
1
# Copyright (C) 2005, 2006 Canonical Ltd
4
3
# This program is free software; you can redistribute it and/or modify
5
4
# it under the terms of the GNU General Public License as published by
6
5
# the Free Software Foundation; either version 2 of the License, or
7
6
# (at your option) any later version.
9
8
# This program is distributed in the hope that it will be useful,
10
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
11
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12
11
# GNU General Public License for more details.
14
13
# You should have received a copy of the GNU General Public License
15
14
# along with this program; if not, write to the Free Software
16
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
18
17
"""XML externalization support."""
20
19
# "XML is like violence: if it doesn't solve your problem, you aren't
21
20
# using enough of it." -- various
22
# importing this module is fairly slow because it has to load several
24
__copyright__ = "Copyright (C) 2005 Canonical Ltd."
25
__author__ = "Martin Pool <mbp@canonical.com>"
25
from bzrlib.serializer import Serializer
26
from bzrlib.trace import mutter, warning
28
from cElementTree import Element, ElementTree, SubElement
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
35
from cElementTree import (ElementTree, SubElement, Element,
36
XMLTreeBuilder, fromstring, tostring)
37
import elementtree.ElementTree
38
ParseError = SyntaxError
29
39
except ImportError:
30
from elementtree.ElementTree import Element, ElementTree, SubElement
33
from trace import mutter
37
raise Exception("XMLMixin.to_element must be overridden in concrete classes")
39
def write_xml(self, f):
40
ElementTree(self.to_element()).write(f, 'utf-8')
40
mutter('WARNING: using slower ElementTree; consider installing cElementTree'
41
" and make sure it's on your PYTHONPATH")
42
# this copy is shipped with bzr
43
from util.elementtree.ElementTree import (ElementTree, SubElement,
44
Element, XMLTreeBuilder,
46
import util.elementtree as elementtree
47
from xml.parsers.expat import ExpatError as ParseError
49
from bzrlib import errors
52
class XMLSerializer(Serializer):
53
"""Abstract XML object serialize/deserialize"""
55
squashes_xml_invalid_characters = True
57
def read_inventory_from_string(self, xml_string, revision_id=None,
59
"""Read xml_string into an inventory object.
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.
74
return self._unpack_inventory(fromstring(xml_string), revision_id,
75
entry_cache=entry_cache)
77
raise errors.UnexpectedInventoryFormat(e)
79
def read_inventory(self, f, revision_id=None):
81
return self._unpack_inventory(self._read_element(f),
84
raise errors.UnexpectedInventoryFormat(e)
86
def write_revision(self, rev, f):
87
self._write_element(self._pack_revision(rev), f)
89
def write_revision_to_string(self, rev):
90
return tostring(self._pack_revision(rev)) + '\n'
92
def read_revision(self, f):
93
return self._unpack_revision(self._read_element(f))
95
def read_revision_from_string(self, xml_string):
96
return self._unpack_revision(fromstring(xml_string))
98
def _write_element(self, elt, f):
99
ElementTree(elt).write(f, 'utf-8')
44
return cls.from_element(ElementTree().parse(f))
46
read_xml = classmethod(read_xml)
102
def _read_element(self, f):
103
return ElementTree().parse(f)
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.
110
escape_re = re.compile("[&'\"<>]")
113
"'":"'", # FIXME: overkill
118
def _escape_replace(match, map=escape_map):
119
return map[match.group()]
121
def _escape_attrib(text, encoding=None, replace=None):
122
# escape attribute value
126
text = elementtree.ElementTree._encode(text, encoding)
128
return elementtree.ElementTree._encode_entity(text)
130
return escape_re.sub(_escape_replace, text)
132
text = replace(text, "&", "&")
133
text = replace(text, "'", "'") # FIXME: overkill
134
text = replace(text, "\"", """)
135
text = replace(text, "<", "<")
136
text = replace(text, ">", ">")
138
except (TypeError, AttributeError):
139
elementtree.ElementTree._raise_serialization_error(text)
141
elementtree.ElementTree._escape_attrib = _escape_attrib
143
escape_cdata_re = re.compile("[&<>]")
149
def _escape_cdata_replace(match, map=escape_cdata_map):
150
return map[match.group()]
152
def _escape_cdata(text, encoding=None, replace=None):
153
# escape character data
157
text = elementtree.ElementTree._encode(text, encoding)
159
return elementtree.ElementTree._encode_entity(text)
161
return escape_cdata_re.sub(_escape_cdata_replace, text)
163
text = replace(text, "&", "&")
164
text = replace(text, "<", "<")
165
text = replace(text, ">", ">")
167
except (TypeError, AttributeError):
168
elementtree.ElementTree._raise_serialization_error(text)
170
elementtree.ElementTree._escape_cdata = _escape_cdata
173
def escape_invalid_chars(message):
174
"""Escape the XML-invalid characters in a commit message.
176
:param message: Commit message to escape
177
:return: tuple with escaped message and number of characters escaped
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'),