~unity-2d-team/unity-2d/Shell-MultiMonitor

# Copyright (c) 2012 Canonical Ltd

#

# This program is free software: you can redistribute it and/or modify

# it under the terms of the GNU Affero General Public License as published by

# the Free Software Foundation, either version 3 of the License, or

# (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU Affero General Public License for more details.

#

# You should have received a copy of the GNU Affero General Public

# License along with this program. If not, see

# <http://www.gnu.org/licenses/>.

import datetime

import dateutil.tz

import email.parser

from email.utils import parsedate_tz

import functools

import logging

import time

import uuid

import pycassa

from pycassa.system_manager import (

    LEXICAL_UUID_TYPE,

    SystemManager,

    TIME_UUID_TYPE,

    )

from pycassa.util import convert_time_to_uuid

from grackle.cassandra import workaround_1779

def create_schema(host, keyspace, clobber=False, create_keyspace=False):

    mgr = SystemManager(host)

    if create_keyspace:

        mgr.create_keyspace(keyspace, replication_factor=1)

    if clobber:

        for cf in mgr.get_keyspace_column_families(keyspace):

            mgr.drop_column_family(keyspace, cf)

    try:

        workaround_1779(

            mgr.create_column_family, keyspace, 'message',

            key_validation_class=LEXICAL_UUID_TYPE)

        workaround_1779(

            mgr.create_column_family, keyspace, 'archive_message',

            comparator_type=TIME_UUID_TYPE,

            default_validation_class=LEXICAL_UUID_TYPE)

        pass

    finally:

        mgr.close()

def _parse_message(message):

    """Get a date and dict of an RFC822 message."""

    parsed = email.parser.Parser().parsestr(message)

    message_dict = {}

    for key in ('from', 'to', 'subject', 'message-id'):

        value = parsed.get(key, None)

        if value is not None:

            message_dict[key] = value

    date = parsed.get('date')

    if date is not None:

        try:

            pdate = parsedate_tz(date)

            date = datetime.datetime(

                *pdate[:6],

                tzinfo=dateutil.tz.tzoffset('', pdate[9]))

        except ValueError:

            pass

    message_dict['date'] = date.isoformat() if date is not None else None

    return date, message_dict

def _utc_datetime(dt):

    return dt.astimezone(dateutil.tz.tzutc())

def _utc_timestamp(dt):

    return time.mktime(_utc_datetime(dt).timetuple()) - time.timezone

def _utc_timeuuid(dt, lowest_val=True):

    return convert_time_to_uuid(_utc_timestamp(dt), lowest_val)

def _cmp_timeuuid(a, b):

    if a.time != b.time:

        return cmp(a.time, b.time)

    return cmp(a, b)

def _bound_timeuuid(a, b, max=False):

    if a == '' or _cmp_timeuuid(b, a) == (1 if max else -1):

        return b

    return a

def _make_bounds(memo, range_start, range_finish, backward):

    start = finish = ''

    if memo != '':

        memo = uuid.UUID(memo)

    if backward:

        finish = memo

    else:

        start = memo

    if range_start is not None:

        start = _bound_timeuuid(

            start, _utc_timeuuid(range_start), max=True)

    if range_finish is not None:

        finish = _bound_timeuuid(

            finish, _utc_timeuuid(range_finish, lowest_val=False))

    return memo, start, finish

LEGAL_HEADERS = set([

    'date', 'from', 'subject', 'message-id',

    ])

def _format_message(message, headers=[], include_raw=False):

    data = {}

    if headers:

        assert not set(headers).difference(LEGAL_HEADERS)

        hdict = {}

        for header in headers:

            hdict[header] = message.get(header)

        data['headers'] = hdict

    if include_raw:

        data['raw'] = message['raw']

    return data

class CassandraConnection(object):

    def __init__(self, keyspace, host):

        self._keyspace = keyspace

        self._host = host

        self._connection = self._connect()

        self.messages = self._column_family('message')

        self.archive_messages = self._column_family('archive_message')

    def _connect(self):

        return pycassa.connect(self._keyspace, self._host)

    def _column_family(self, name):

        return pycassa.ColumnFamily(self._connection, name)

    def add_message(self, archive_uuid, message):

        message_uuid = uuid.uuid4()

        message_date, message_dict = _parse_message(message)

        message_dict['raw'] = message

        message_dict['date_created'] = (

            datetime.datetime.utcnow().isoformat() + 'Z')

        self.messages.insert(message_uuid, message_dict)

        self.archive_messages.insert(

            archive_uuid,

            {_utc_timestamp(message_date): message_uuid})

        logging.debug(

            'Imported %s into %s'

            % (message_dict.get('message-id', None), archive_uuid))

        return message_uuid

    def _trim(self, sequence, end):

        """Return the sequence with one of the ends trimmed.

        :param end: if true, remove the last element. otherwise remove

            the first.

        """

        if end:

            return sequence[:-1]

        else:

            return sequence[1:]

    def get_messages(self, archive_uuid, order, count, memo, backward=False,

                     start_date=None, finish_date=None, format='all',

                     headers=['from', 'date', 'subject', 'message-id']):

        if order in ("date", "-date"):

            reversed = order[0] == '-'

        else:

            raise AssertionError("Unsupported order.")

        memo, start, finish = _make_bounds(

            memo, start_date, finish_date, backward)

        # Get up to n+1 messages from the memo: the last item of the

        # previous batch (because that's where the memo starts) + this

        # batch.

        pairs = self.archive_messages.get(

            archive_uuid, column_count=count + 1, column_start=start,

            column_finish=finish, column_reversed=reversed).items()

        if len(pairs) and memo and pairs[0][0] <= memo:

            # The memo (from the previous batch) was included in the result.

            # Trim it.

            pairs = self._trim(pairs, False ^ backward)

        elif len(pairs) > count:

            # There was no memo in the result, so the n+1th element is

            # unnecessary. Kill it.

            pairs = self._trim(pairs, True ^ backward)

        if len(pairs) == 0:

            return (None, [], None)

        assert 0 < len(pairs) <= count

        # We've narrowed down the message references. Fetch the messages.

        ids = [v for k, v in pairs]

        formatter = functools.partial(

            _format_message, headers=headers, include_raw=True)

        # XXX: No need to get all columns. Restrict based on format.

        messages = self.messages.multiget(ids)

        return (

            str(pairs[0][0]),

            [formatter(messages[id]) for id in ids],

            str(pairs[-1][0]),

            )