1
# Copyright (C) 2008, 2009 Canonical Ltd
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.
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.
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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
17
"""Serializer object for CHK based inventory storage."""
23
revision as _mod_revision,
29
def _validate_properties(props, _decode=cache_utf8._utf8_decode):
30
# TODO: we really want an 'isascii' check for key
31
# Cast the utf8 properties into Unicode 'in place'
32
for key, value in props.iteritems():
33
props[key] = _decode(value)[0]
37
def _is_format_10(value):
39
raise ValueError('Format number was not recognized, expected 10 got %d'
44
class BEncodeRevisionSerializer1(object):
45
"""Simple revision serializer based around bencode.
48
squashes_xml_invalid_characters = False
50
# Maps {key:(Revision attribute, bencode_type, validator)}
51
# This tells us what kind we expect bdecode to create, what variable on
52
# Revision we should be using, and a function to call to validate/transform
54
# TODO: add a 'validate_utf8' for things like revision_id and file_id
55
# and a validator for parent-ids
56
_schema = {'format': (None, int, _is_format_10),
57
'committer': ('committer', str, cache_utf8.decode),
58
'timezone': ('timezone', int, None),
59
'timestamp': ('timestamp', str, float),
60
'revision-id': ('revision_id', str, None),
61
'parent-ids': ('parent_ids', list, None),
62
'inventory-sha1': ('inventory_sha1', str, None),
63
'message': ('message', str, cache_utf8.decode),
64
'properties': ('properties', dict, _validate_properties),
67
def write_revision_to_string(self, rev):
68
encode_utf8 = cache_utf8._utf8_encode
69
# Use a list of tuples rather than a dict
70
# This lets us control the ordering, so that we are able to create
74
("committer", encode_utf8(rev.committer)[0]),
76
if rev.timezone is not None:
77
ret.append(("timezone", rev.timezone))
78
# For bzr revisions, the most common property is just 'branch-nick'
79
# which changes infrequently.
81
for key, value in rev.properties.iteritems():
82
revprops[key] = encode_utf8(value)[0]
83
ret.append(('properties', revprops))
85
("timestamp", "%.3f" % rev.timestamp),
86
("revision-id", rev.revision_id),
87
("parent-ids", rev.parent_ids),
88
("inventory-sha1", rev.inventory_sha1),
89
("message", encode_utf8(rev.message)[0]),
91
return bencode.bencode(ret)
93
def write_revision(self, rev, f):
94
f.write(self.write_revision_to_string(rev))
96
def read_revision_from_string(self, text):
97
# TODO: consider writing a Revision decoder, rather than using the
98
# generic bencode decoder
99
# However, to decode all 25k revisions of bzr takes approx 1.3s
100
# If we remove all extra validation that goes down to about 1.2s.
101
# Of that time, probably 0.6s is spend in bencode.bdecode().
102
# Regardless 'time bzr log' of everything is 7+s, so 1.3s to
103
# extract revision texts isn't a majority of time.
104
ret = bencode.bdecode(text)
105
if not isinstance(ret, list):
106
raise ValueError("invalid revision text")
107
schema = self._schema
108
# timezone is allowed to be missing, but should be set
109
bits = {'timezone': None}
110
for key, value in ret:
111
# Will raise KeyError if not a valid part of the schema, or an
112
# entry is given 2 times.
113
var_name, expected_type, validator = schema[key]
114
if value.__class__ is not expected_type:
115
raise ValueError('key %s did not conform to the expected type'
117
% (key, expected_type, type(value)))
118
if validator is not None:
119
value = validator(value)
120
bits[var_name] = value
121
if len(bits) != len(schema):
122
missing = [key for key, (var_name, _, _) in schema.iteritems()
123
if var_name not in bits]
124
raise ValueError('Revision text was missing expected keys %s.'
125
' text %r' % (missing, text))
126
del bits[None] # Get rid of 'format' since it doesn't get mapped
127
rev = _mod_revision.Revision(**bits)
130
def read_revision(self, f):
131
return self.read_revision_from_string(f.read())
134
class CHKSerializerSubtree(BEncodeRevisionSerializer1, xml7.Serializer_v7):
135
"""A CHKInventory based serializer that supports tree references"""
137
supported_kinds = set(['file', 'directory', 'symlink', 'tree-reference'])
139
revision_format_num = None
140
support_altered_by_hack = False
142
def _unpack_entry(self, elt):
144
if not kind in self.supported_kinds:
145
raise AssertionError('unsupported entry kind %s' % kind)
146
if kind == 'tree-reference':
147
file_id = elt.attrib['file_id']
148
name = elt.attrib['name']
149
parent_id = elt.attrib['parent_id']
150
revision = elt.get('revision')
151
reference_revision = elt.get('reference_revision')
152
return inventory.TreeReference(file_id, name, parent_id, revision,
155
return xml7.Serializer_v7._unpack_entry(self, elt)
157
def __init__(self, node_size, search_key_name):
158
self.maximum_size = node_size
159
self.search_key_name = search_key_name
162
class CHKSerializer(xml6.Serializer_v6):
163
"""A CHKInventory based serializer with 'plain' behaviour."""
166
revision_format_num = None
167
support_altered_by_hack = False
169
def __init__(self, node_size, search_key_name):
170
self.maximum_size = node_size
171
self.search_key_name = search_key_name
174
chk_serializer_255_bigpage = CHKSerializer(65536, 'hash-255-way')
177
class CHKBEncodeSerializer(BEncodeRevisionSerializer1, CHKSerializer):
178
"""A CHKInventory and BEncode based serializer with 'plain' behaviour."""
183
chk_bencode_serializer = CHKBEncodeSerializer(65536, 'hash-255-way')