1
# Copyright (C) 2005-2010 Canonical Ltd
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
5
5
# the Free Software Foundation; either version 2 of the License, or
6
6
# (at your option) any later version.
8
8
# This program is distributed in the hope that it will be useful,
9
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
11
# GNU General Public License for more details.
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
17
17
"""XML externalization support."""
22
22
# importing this module is fairly slow because it has to load several
27
from bzrlib.serializer import Serializer
28
from bzrlib.trace import mutter
25
from bzrlib.trace import mutter, warning
32
# it's in this package in python2.5
33
from xml.etree.cElementTree import (ElementTree, SubElement, Element,
34
XMLTreeBuilder, fromstring, tostring)
35
import xml.etree as elementtree
36
# Also import ElementTree module so monkey-patching below always works
37
import xml.etree.ElementTree
39
from cElementTree import (ElementTree, SubElement, Element,
40
XMLTreeBuilder, fromstring, tostring)
41
import elementtree.ElementTree
42
ParseError = SyntaxError
28
from util.cElementTree import (ElementTree, SubElement, Element,
29
XMLTreeBuilder, fromstring, tostring)
43
30
except ImportError:
44
mutter('WARNING: using slower ElementTree; consider installing cElementTree'
45
" and make sure it's on your PYTHONPATH")
46
# this copy is shipped with bzr
31
warning('using slower ElementTree; consider installing cElementTree')
47
32
from util.elementtree.ElementTree import (ElementTree, SubElement,
48
33
Element, XMLTreeBuilder,
49
34
fromstring, tostring)
50
import util.elementtree as elementtree
51
from xml.parsers.expat import ExpatError as ParseError
53
from bzrlib import errors
56
class XMLSerializer(Serializer):
57
"""Abstract XML object serialize/deserialize"""
59
squashes_xml_invalid_characters = True
61
def read_inventory_from_string(self, xml_string, revision_id=None,
62
entry_cache=None, return_from_cache=False):
63
"""Read xml_string into an inventory object.
65
:param xml_string: The xml to read.
66
:param revision_id: If not-None, the expected revision id of the
67
inventory. Some serialisers use this to set the results' root
68
revision. This should be supplied for deserialising all
69
from-repository inventories so that xml5 inventories that were
70
serialised without a revision identifier can be given the right
71
revision id (but not for working tree inventories where users can
72
edit the data without triggering checksum errors or anything).
73
:param entry_cache: An optional cache of InventoryEntry objects. If
74
supplied we will look up entries via (file_id, revision_id) which
75
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.
82
return self._unpack_inventory(fromstring(xml_string), revision_id,
83
entry_cache=entry_cache,
84
return_from_cache=return_from_cache)
86
raise errors.UnexpectedInventoryFormat(e)
88
def read_inventory(self, f, revision_id=None):
91
return self._unpack_inventory(self._read_element(f),
96
raise errors.UnexpectedInventoryFormat(e)
36
from bzrlib.inventory import ROOT_ID, Inventory, InventoryEntry
37
from bzrlib.revision import Revision, RevisionReference
38
from bzrlib.errors import BzrError
41
class Serializer(object):
42
"""Abstract object serialize/deserialize"""
43
def write_inventory(self, inv, f):
44
"""Write inventory to a file"""
45
elt = self._pack_inventory(inv)
46
self._write_element(elt, f)
48
def write_inventory_to_string(self, inv):
49
return tostring(self._pack_inventory(inv))
51
def read_inventory_from_string(self, xml_string):
52
return self._unpack_inventory(fromstring(xml_string))
54
def read_inventory(self, f):
55
return self._unpack_inventory(self._read_element(f))
98
57
def write_revision(self, rev, f):
99
58
self._write_element(self._pack_revision(rev), f)
101
60
def write_revision_to_string(self, rev):
102
return tostring(self._pack_revision(rev)) + '\n'
61
return tostring(self._pack_revision(rev), f)
104
63
def read_revision(self, f):
105
64
return self._unpack_revision(self._read_element(f))
115
74
return ElementTree().parse(f)
118
def escape_invalid_chars(message):
119
"""Escape the XML-invalid characters in a commit message.
121
:param message: Commit message to escape
122
:return: tuple with escaped message and number of characters escaped
126
# Python strings can include characters that can't be
127
# represented in well-formed XML; escape characters that
128
# aren't listed in the XML specification
129
# (http://www.w3.org/TR/REC-xml/#NT-Char).
130
return re.subn(u'[^\x09\x0A\x0D\u0020-\uD7FF\uE000-\uFFFD]+',
131
lambda match: match.group(0).encode('unicode_escape'),
77
class _Serializer_v4(Serializer):
78
"""Version 0.0.4 serializer
80
You should use the serialzer_v4 singleton."""
84
def _pack_inventory(self, inv):
85
"""Convert to XML Element"""
86
e = Element('inventory')
88
if inv.root.file_id not in (None, ROOT_ID):
89
e.set('file_id', inv.root.file_id)
90
for path, ie in inv.iter_entries():
91
e.append(self._pack_entry(ie))
95
def _pack_entry(self, ie):
96
"""Convert InventoryEntry to XML element"""
98
e.set('name', ie.name)
99
e.set('file_id', ie.file_id)
100
e.set('kind', ie.kind)
102
if ie.text_size != None:
103
e.set('text_size', '%d' % ie.text_size)
105
for f in ['text_id', 'text_sha1']:
110
# to be conservative, we don't externalize the root pointers
111
# for now, leaving them as null in the xml form. in a future
112
# version it will be implied by nested elements.
113
if ie.parent_id != ROOT_ID:
114
assert isinstance(ie.parent_id, basestring)
115
e.set('parent_id', ie.parent_id)
122
def _unpack_inventory(self, elt):
123
"""Construct from XML Element
125
assert elt.tag == 'inventory'
126
root_id = elt.get('file_id') or ROOT_ID
127
inv = Inventory(root_id)
129
ie = self._unpack_entry(e)
130
if ie.parent_id == ROOT_ID:
131
ie.parent_id = root_id
136
def _unpack_entry(self, elt):
137
assert elt.tag == 'entry'
139
## original format inventories don't have a parent_id for
140
## nodes in the root directory, but it's cleaner to use one
142
parent_id = elt.get('parent_id')
143
if parent_id == None:
146
ie = InventoryEntry(elt.get('file_id'),
150
ie.text_id = elt.get('text_id')
151
ie.text_sha1 = elt.get('text_sha1')
153
## mutter("read inventoryentry: %r" % (elt.attrib))
155
v = elt.get('text_size')
156
ie.text_size = v and int(v)
161
def _pack_revision(self, rev):
162
"""Revision object -> xml tree"""
163
root = Element('revision',
164
committer = rev.committer,
165
timestamp = '%.9f' % rev.timestamp,
166
revision_id = rev.revision_id,
167
inventory_id = rev.inventory_id,
168
inventory_sha1 = rev.inventory_sha1,
171
root.set('timezone', str(rev.timezone))
174
msg = SubElement(root, 'message')
175
msg.text = rev.message
179
pelts = SubElement(root, 'parents')
180
pelts.tail = pelts.text = '\n'
181
for rr in rev.parents:
182
assert isinstance(rr, RevisionReference)
183
p = SubElement(pelts, 'revision_ref')
185
assert rr.revision_id
186
p.set('revision_id', rr.revision_id)
188
p.set('revision_sha1', rr.revision_sha1)
193
def _unpack_revision(self, elt):
194
"""XML Element -> Revision object"""
196
# <changeset> is deprecated...
197
if elt.tag not in ('revision', 'changeset'):
198
raise BzrError("unexpected tag in revision file: %r" % elt)
200
rev = Revision(committer = elt.get('committer'),
201
timestamp = float(elt.get('timestamp')),
202
revision_id = elt.get('revision_id'),
203
inventory_id = elt.get('inventory_id'),
204
inventory_sha1 = elt.get('inventory_sha1')
207
precursor = elt.get('precursor')
208
precursor_sha1 = elt.get('precursor_sha1')
210
pelts = elt.find('parents')
214
assert p.tag == 'revision_ref', \
215
"bad parent node tag %r" % p.tag
216
rev_ref = RevisionReference(p.get('revision_id'),
217
p.get('revision_sha1'))
218
rev.parents.append(rev_ref)
222
prec_parent = rev.parents[0].revision_id
223
assert prec_parent == precursor
225
# revisions written prior to 0.0.5 have a single precursor
226
# give as an attribute
227
rev_ref = RevisionReference(precursor, precursor_sha1)
228
rev.parents.append(rev_ref)
230
v = elt.get('timezone')
231
rev.timezone = v and int(v)
233
rev.message = elt.findtext('message') # text of <message>
239
"""singleton instance"""
240
serializer_v4 = _Serializer_v4()
added
removed
bzrlib/xml.py
