1
# Copyright (C) 2005-2010 Canonical Ltd
2
# -*- coding: UTF-8 -*-
3
4
# This program is free software; you can redistribute it and/or modify
4
5
# it under the terms of the GNU General Public License as published by
5
6
# the Free Software Foundation; either version 2 of the License, or
6
7
# (at your option) any later version.
8
9
# This program is distributed in the hope that it will be useful,
9
10
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
11
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
12
# GNU General Public License for more details.
13
14
# You should have received a copy of the GNU General Public License
14
15
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
16
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
18
"""XML externalization support."""
22
23
# importing this module is fairly slow because it has to load several
25
from bzrlib.serializer import Serializer
26
from bzrlib.trace import mutter
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
27
from util.cElementTree import ElementTree, SubElement, Element
39
28
except ImportError:
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,
58
entry_cache=None, return_from_cache=False):
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.
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.
78
return self._unpack_inventory(fromstring(xml_string), revision_id,
79
entry_cache=entry_cache,
80
return_from_cache=return_from_cache)
82
raise errors.UnexpectedInventoryFormat(e)
84
def read_inventory(self, f, revision_id=None):
86
return self._unpack_inventory(self._read_element(f),
89
raise errors.UnexpectedInventoryFormat(e)
91
def write_revision(self, rev, f):
92
self._write_element(self._pack_revision(rev), f)
94
def write_revision_to_string(self, rev):
95
return tostring(self._pack_revision(rev)) + '\n'
97
def read_revision(self, f):
98
return self._unpack_revision(self._read_element(f))
100
def read_revision_from_string(self, xml_string):
101
return self._unpack_revision(fromstring(xml_string))
29
from util.elementtree.ElementTree import ElementTree, SubElement, Element
31
from bzrlib.inventory import ROOT_ID, Inventory, InventoryEntry
35
class Serializer(object):
36
"""Abstract object serialize/deserialize"""
37
def write_inventory(self, inv, f):
38
"""Write inventory to a file"""
39
elt = self._pack_inventory(inv)
40
self._write_element(elt, f)
42
def read_inventory(self, f):
43
return self._unpack_inventory(self._read_element(f))
103
45
def _write_element(self, elt, f):
104
46
ElementTree(elt).write(f, 'utf-8')
108
50
return ElementTree().parse(f)
111
# performance tuning for elementree's serialiser. This should be
112
# sent upstream - RBC 20060523.
113
# the functions here are patched into elementtree at runtime.
115
escape_re = re.compile("[&'\"<>]")
118
"'":"'", # FIXME: overkill
123
def _escape_replace(match, map=escape_map):
124
return map[match.group()]
126
def _escape_attrib(text, encoding=None, replace=None):
127
# escape attribute value
131
text = elementtree.ElementTree._encode(text, encoding)
133
return elementtree.ElementTree._encode_entity(text)
135
return escape_re.sub(_escape_replace, text)
137
text = replace(text, "&", "&")
138
text = replace(text, "'", "'") # FIXME: overkill
139
text = replace(text, "\"", """)
140
text = replace(text, "<", "<")
141
text = replace(text, ">", ">")
143
except (TypeError, AttributeError):
144
elementtree.ElementTree._raise_serialization_error(text)
146
elementtree.ElementTree._escape_attrib = _escape_attrib
148
escape_cdata_re = re.compile("[&<>]")
154
def _escape_cdata_replace(match, map=escape_cdata_map):
155
return map[match.group()]
157
def _escape_cdata(text, encoding=None, replace=None):
158
# escape character data
162
text = elementtree.ElementTree._encode(text, encoding)
164
return elementtree.ElementTree._encode_entity(text)
166
return escape_cdata_re.sub(_escape_cdata_replace, text)
168
text = replace(text, "&", "&")
169
text = replace(text, "<", "<")
170
text = replace(text, ">", ">")
172
except (TypeError, AttributeError):
173
elementtree.ElementTree._raise_serialization_error(text)
175
elementtree.ElementTree._escape_cdata = _escape_cdata
178
def escape_invalid_chars(message):
179
"""Escape the XML-invalid characters in a commit message.
181
:param message: Commit message to escape
182
:return: tuple with escaped message and number of characters escaped
54
class _Serializer_v4(Serializer):
55
"""Version 0.0.4 serializer"""
59
def _pack_inventory(self, inv):
60
"""Convert to XML Element"""
61
e = Element('inventory')
63
if inv.root.file_id not in (None, ROOT_ID):
64
e.set('file_id', inv.root.file_id)
65
for path, ie in inv.iter_entries():
66
e.append(self._pack_entry(ie))
70
def _pack_entry(self, ie):
71
"""Convert InventoryEntry to XML element"""
73
e.set('name', ie.name)
74
e.set('file_id', ie.file_id)
75
e.set('kind', ie.kind)
77
if ie.text_size != None:
78
e.set('text_size', '%d' % ie.text_size)
80
for f in ['text_id', 'text_sha1']:
85
# to be conservative, we don't externalize the root pointers
86
# for now, leaving them as null in the xml form. in a future
87
# version it will be implied by nested elements.
88
if ie.parent_id != ROOT_ID:
89
assert isinstance(ie.parent_id, basestring)
90
e.set('parent_id', ie.parent_id)
97
def _unpack_inventory(self, elt):
98
"""Construct from XML Element
100
assert elt.tag == 'inventory'
101
root_id = elt.get('file_id') or ROOT_ID
102
inv = Inventory(root_id)
104
ie = self._unpack_entry(e)
105
if ie.parent_id == ROOT_ID:
106
ie.parent_id = root_id
111
def _unpack_entry(self, elt):
112
assert elt.tag == 'entry'
114
## original format inventories don't have a parent_id for
115
## nodes in the root directory, but it's cleaner to use one
117
parent_id = elt.get('parent_id')
118
if parent_id == None:
121
ie = InventoryEntry(elt.get('file_id'),
125
ie.text_id = elt.get('text_id')
126
ie.text_sha1 = elt.get('text_sha1')
128
## mutter("read inventoryentry: %r" % (elt.attrib))
130
v = elt.get('text_size')
131
ie.text_size = v and int(v)
136
"""singleton instance"""
137
serializer_v4 = _Serializer_v4()
144
"""Write object o to file f as XML.
146
o must provide a to_element method.
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'),
148
ElementTree(o.to_element()).write(f, 'utf-8')
152
def unpack_xml(cls, f):
153
return cls.from_element(ElementTree().parse(f))
added
removed
bzrlib/xml.py
