~bzr-pqm/bzr/bzr.dev

1 by mbp at sourcefrog
import from baz patch-364
1
#! /usr/bin/env python
2
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.
7
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.
12
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
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
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
802 by Martin Pool
- Remove XMLMixin class in favour of simple pack_xml, unpack_xml functions
25
try:
1227 by Martin Pool
- methods to deserialize objects from strings
26
    from util.cElementTree import (ElementTree, SubElement, Element,
27
                                   XMLTreeBuilder)
802 by Martin Pool
- Remove XMLMixin class in favour of simple pack_xml, unpack_xml functions
28
except ImportError:
1227 by Martin Pool
- methods to deserialize objects from strings
29
    from util.elementtree.ElementTree import (ElementTree, SubElement,
30
                                              Element, XMLTreeBuilder)
802 by Martin Pool
- Remove XMLMixin class in favour of simple pack_xml, unpack_xml functions
31
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
32
from bzrlib.inventory import ROOT_ID, Inventory, InventoryEntry
1182 by Martin Pool
- more disentangling of xml storage format from objects
33
from bzrlib.revision import Revision, RevisionReference        
1183 by Martin Pool
- implement version 5 xml storage, and tests
34
from bzrlib.errors import BzrError
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
35
36
37
class Serializer(object):
38
    """Abstract object serialize/deserialize"""
39
    def write_inventory(self, inv, f):
40
        """Write inventory to a file"""
41
        elt = self._pack_inventory(inv)
42
        self._write_element(elt, f)
43
1227 by Martin Pool
- methods to deserialize objects from strings
44
    def read_inventory_from_string(self, xml_string):
45
        return self._unpack_inventory(self._parse_string(xml_string))
46
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
47
    def read_inventory(self, f):
48
        return self._unpack_inventory(self._read_element(f))
49
1182 by Martin Pool
- more disentangling of xml storage format from objects
50
    def write_revision(self, rev, f):
51
        self._write_element(self._pack_revision(rev), f)
52
53
    def read_revision(self, f):
54
        return self._unpack_revision(self._read_element(f))
55
1227 by Martin Pool
- methods to deserialize objects from strings
56
    def read_revision_from_string(self, xml_string):
57
        return self._unpack_revision(self._parse_string(xml_string))
58
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
59
    def _write_element(self, elt, f):
60
        ElementTree(elt).write(f, 'utf-8')
61
        f.write('\n')
62
63
    def _read_element(self, f):
64
        return ElementTree().parse(f)
65
1227 by Martin Pool
- methods to deserialize objects from strings
66
    def _parse_string(self, xml_string):
67
        parser = XMLTreeBuilder()
68
        parser.feed(xml_string)
69
        return parser.close()
70
        
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
71
72
73
class _Serializer_v4(Serializer):
1183 by Martin Pool
- implement version 5 xml storage, and tests
74
    """Version 0.0.4 serializer
75
76
    You should use the serialzer_v4 singleton."""
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
77
    
78
    __slots__ = []
79
    
80
    def _pack_inventory(self, inv):
81
        """Convert to XML Element"""
82
        e = Element('inventory')
83
        e.text = '\n'
84
        if inv.root.file_id not in (None, ROOT_ID):
85
            e.set('file_id', inv.root.file_id)
86
        for path, ie in inv.iter_entries():
87
            e.append(self._pack_entry(ie))
88
        return e
89
90
91
    def _pack_entry(self, ie):
92
        """Convert InventoryEntry to XML element"""
93
        e = Element('entry')
94
        e.set('name', ie.name)
95
        e.set('file_id', ie.file_id)
96
        e.set('kind', ie.kind)
97
98
        if ie.text_size != None:
99
            e.set('text_size', '%d' % ie.text_size)
100
101
        for f in ['text_id', 'text_sha1']:
102
            v = getattr(ie, f)
103
            if v != None:
104
                e.set(f, v)
105
106
        # to be conservative, we don't externalize the root pointers
107
        # for now, leaving them as null in the xml form.  in a future
108
        # version it will be implied by nested elements.
109
        if ie.parent_id != ROOT_ID:
110
            assert isinstance(ie.parent_id, basestring)
111
            e.set('parent_id', ie.parent_id)
112
113
        e.tail = '\n'
114
115
        return e
116
117
118
    def _unpack_inventory(self, elt):
119
        """Construct from XML Element
120
        """
121
        assert elt.tag == 'inventory'
122
        root_id = elt.get('file_id') or ROOT_ID
123
        inv = Inventory(root_id)
124
        for e in elt:
125
            ie = self._unpack_entry(e)
126
            if ie.parent_id == ROOT_ID:
127
                ie.parent_id = root_id
128
            inv.add(ie)
129
        return inv
130
131
132
    def _unpack_entry(self, elt):
133
        assert elt.tag == 'entry'
134
135
        ## original format inventories don't have a parent_id for
136
        ## nodes in the root directory, but it's cleaner to use one
137
        ## internally.
138
        parent_id = elt.get('parent_id')
139
        if parent_id == None:
140
            parent_id = ROOT_ID
141
142
        ie = InventoryEntry(elt.get('file_id'),
1189 by Martin Pool
- BROKEN: partial support for commit into weave
143
                            elt.get('name'),
144
                            elt.get('kind'),
145
                            parent_id)
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
146
        ie.text_id = elt.get('text_id')
147
        ie.text_sha1 = elt.get('text_sha1')
148
149
        ## mutter("read inventoryentry: %r" % (elt.attrib))
150
151
        v = elt.get('text_size')
152
        ie.text_size = v and int(v)
153
154
        return ie
155
156
1182 by Martin Pool
- more disentangling of xml storage format from objects
157
    def _pack_revision(self, rev):
158
        """Revision object -> xml tree"""
159
        root = Element('revision',
160
                       committer = rev.committer,
161
                       timestamp = '%.9f' % rev.timestamp,
162
                       revision_id = rev.revision_id,
163
                       inventory_id = rev.inventory_id,
164
                       inventory_sha1 = rev.inventory_sha1,
165
                       )
166
        if rev.timezone:
167
            root.set('timezone', str(rev.timezone))
168
        root.text = '\n'
169
170
        msg = SubElement(root, 'message')
171
        msg.text = rev.message
172
        msg.tail = '\n'
173
174
        if rev.parents:
175
            pelts = SubElement(root, 'parents')
176
            pelts.tail = pelts.text = '\n'
177
            for rr in rev.parents:
178
                assert isinstance(rr, RevisionReference)
179
                p = SubElement(pelts, 'revision_ref')
180
                p.tail = '\n'
181
                assert rr.revision_id
182
                p.set('revision_id', rr.revision_id)
183
                if rr.revision_sha1:
184
                    p.set('revision_sha1', rr.revision_sha1)
185
186
        return root
187
188
    
189
    def _unpack_revision(self, elt):
190
        """XML Element -> Revision object"""
191
        
192
        # <changeset> is deprecated...
193
        if elt.tag not in ('revision', 'changeset'):
1183 by Martin Pool
- implement version 5 xml storage, and tests
194
            raise BzrError("unexpected tag in revision file: %r" % elt)
1182 by Martin Pool
- more disentangling of xml storage format from objects
195
196
        rev = Revision(committer = elt.get('committer'),
197
                       timestamp = float(elt.get('timestamp')),
198
                       revision_id = elt.get('revision_id'),
199
                       inventory_id = elt.get('inventory_id'),
200
                       inventory_sha1 = elt.get('inventory_sha1')
201
                       )
202
203
        precursor = elt.get('precursor')
204
        precursor_sha1 = elt.get('precursor_sha1')
205
206
        pelts = elt.find('parents')
207
208
        if pelts:
209
            for p in pelts:
210
                assert p.tag == 'revision_ref', \
211
                       "bad parent node tag %r" % p.tag
212
                rev_ref = RevisionReference(p.get('revision_id'),
213
                                            p.get('revision_sha1'))
214
                rev.parents.append(rev_ref)
215
216
            if precursor:
217
                # must be consistent
218
                prec_parent = rev.parents[0].revision_id
219
                assert prec_parent == precursor
220
        elif precursor:
221
            # revisions written prior to 0.0.5 have a single precursor
222
            # give as an attribute
223
            rev_ref = RevisionReference(precursor, precursor_sha1)
224
            rev.parents.append(rev_ref)
225
226
        v = elt.get('timezone')
227
        rev.timezone = v and int(v)
228
229
        rev.message = elt.findtext('message') # text of <message>
230
        return rev
231
232
233
1183 by Martin Pool
- implement version 5 xml storage, and tests
234
1180 by Martin Pool
- start splitting code for xml (de)serialization away from objects
235
"""singleton instance"""
236
serializer_v4 = _Serializer_v4()
1183 by Martin Pool
- implement version 5 xml storage, and tests
237