1
# Copyright (c) 2012 Canonical Ltd
3
# This program is free software: you can redistribute it and/or modify
4
# it under the terms of the GNU Affero General Public License as published by
5
# the Free Software Foundation, either version 3 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 Affero General Public License for more details.
13
# You should have received a copy of the GNU Affero General Public
14
# License along with this program. If not, see
15
# <http://www.gnu.org/licenses/>.
20
from email.utils import parsedate_tz
26
from pycassa.system_manager import (
31
from pycassa.util import convert_time_to_uuid
33
from grackle.cassandra import workaround_1779
36
def create_schema(host, keyspace, clobber=False, create_keyspace=False):
37
mgr = SystemManager(host)
40
mgr.create_keyspace(keyspace, replication_factor=1)
43
for cf in mgr.get_keyspace_column_families(keyspace):
44
mgr.drop_column_family(keyspace, cf)
48
mgr.create_column_family, keyspace, 'message',
49
key_validation_class=LEXICAL_UUID_TYPE)
51
mgr.create_column_family, keyspace, 'archive_message',
52
comparator_type=TIME_UUID_TYPE,
53
default_validation_class=LEXICAL_UUID_TYPE)
59
def _parse_message(message):
60
"""Get a date and dict of an RFC822 message."""
61
parsed = email.parser.Parser().parsestr(message)
64
for key in ('from', 'to', 'subject', 'message-id'):
65
value = parsed.get(key, None)
67
message_dict[key] = value
69
date = parsed.get('date')
72
pdate = parsedate_tz(date)
73
date = datetime.datetime(
75
tzinfo=dateutil.tz.tzoffset('', pdate[9]))
78
message_dict['date'] = date.isoformat() if date is not None else None
80
return date, message_dict
83
def _utc_datetime(dt):
84
return dt.astimezone(dateutil.tz.tzutc())
87
def _utc_timestamp(dt):
88
return time.mktime(_utc_datetime(dt).timetuple()) - time.timezone
91
def _utc_timeuuid(dt, lowest_val=True):
92
return convert_time_to_uuid(_utc_timestamp(dt), lowest_val)
95
def _cmp_timeuuid(a, b):
97
return cmp(a.time, b.time)
101
def _bound_timeuuid(a, b, max=False):
102
if a == '' or _cmp_timeuuid(b, a) == (1 if max else -1):
107
class CassandraConnection(object):
109
def __init__(self, keyspace, host):
110
self._keyspace = keyspace
112
self._connection = self._connect()
113
self.messages = self._column_family('message')
114
self.archive_messages = self._column_family('archive_message')
117
return pycassa.connect(self._keyspace, self._host)
119
def _column_family(self, name):
120
return pycassa.ColumnFamily(self._connection, name)
122
def add_message(self, archive_uuid, message):
123
message_uuid = uuid.uuid4()
124
message_date, message_dict = _parse_message(message)
125
message_dict['content'] = message
126
message_dict['date_created'] = (
127
datetime.datetime.utcnow().isoformat() + 'Z')
128
self.messages.insert(message_uuid, message_dict)
129
self.archive_messages.insert(
131
{_utc_timestamp(message_date): message_uuid})
133
'Imported %s into %s'
134
% (message_dict.get('message-id', None), archive_uuid))
137
def _format_message(self, message):
139
'date': message.get('date'),
140
'from': message.get('from'),
141
'subject': message.get('subject'),
142
'message-id': message.get('message-id'),
145
def _trim(self, sequence, end):
146
"""Return the sequence with one of the ends trimmed.
148
:param end: if true, remove the last element. otherwise remove
156
def get_messages(self, archive_uuid, order, count, memo, backward=False,
157
start_date=None, finish_date=None):
158
if order in ("date", "-date"):
159
reversed = order[0] == '-'
161
raise AssertionError("Unsupported order.")
163
memo = uuid.UUID(memo)
170
if start_date is not None:
171
start = _bound_timeuuid(
172
start, _utc_timeuuid(start_date, lowest_val=False), max=True)
173
if finish_date is not None:
174
finish = _bound_timeuuid(
175
finish, _utc_timeuuid(finish_date, lowest_val=False))
177
# Get up to n+1 messages from the memo: the last item of the
178
# previous batch (because that's where the memo starts) + this
180
pairs = self.archive_messages.get(
181
archive_uuid, column_count=count + 1, column_start=start,
182
column_finish=finish, column_reversed=reversed).items()
184
if len(pairs) and memo and pairs[0][0] <= memo:
185
# The memo (from the previous batch) was included in the result.
187
pairs = self._trim(pairs, False ^ backward)
188
elif len(pairs) > count:
189
# There was no memo in the result, so the n+1th element is
190
# unnecessary. Kill it.
191
pairs = self._trim(pairs, True ^ backward)
194
return (None, [], None)
196
assert 0 < len(pairs) <= count
198
# We've narrowed down the message references. Fetch the messages.
199
ids = [v for k, v in pairs]
200
messages = self.messages.multiget(
201
ids, columns=['date', 'from', 'subject', 'message-id'])
205
[self._format_message(messages[id]) for id in ids],