def run_test(self):
if not _HAVE_DNSPYTHON:
raise unittest.SkipTest("DNS tests require the dnspython module")
uri = test_case['uri']
seeds = test_case['seeds']
hosts = test_case['hosts']
options = test_case.get('options')
if seeds:
seeds = split_hosts(','.join(seeds))
if hosts:
hosts = frozenset(split_hosts(','.join(hosts)))
if options:
for key, value in options.items():
# Convert numbers to strings for comparison
options[key] = str(value)
if seeds:
result = parse_uri(uri, validate=False)
self.assertEqual(sorted(result['nodelist']), sorted(seeds))
if options:
self.assertEqual(result['options'], options)
hostname = next(iter(client_context.client.nodes))[0]
# The replica set members must be configured as 'localhost'.
if hostname == 'localhost':
client = MongoClient(uri, **_SSL_OPTS)
# Force server selection
client.admin.command('ismaster')
wait_until(
lambda: hosts == client.nodes,
'match test hosts to client nodes')
else:
self.assertRaises(
ConfigurationError, parse_uri, uri, validate=False)
def run_test(self):
if not _HAVE_DNSPYTHON:
raise unittest.SkipTest("DNS tests require the dnspython module")
uri = test_case['uri']
seeds = test_case['seeds']
hosts = test_case['hosts']
options = test_case.get('options')
if seeds:
seeds = split_hosts(','.join(seeds))
if hosts:
hosts = frozenset(split_hosts(','.join(hosts)))
if options:
for key, value in options.items():
# Convert numbers / booleans to strings for comparison
if isinstance(value, bool):
options[key] = 'true' if value else 'false'
elif isinstance(value, (int, float)):
options[key] = str(value)
if seeds:
result = parse_uri(uri, validate=False)
self.assertEqual(sorted(result['nodelist']), sorted(seeds))
if options:
opts = result['options']
if 'readpreferencetags' in opts:
rpts = validate_read_preference_tags(
'readPreferenceTags', opts.pop('readpreferencetags'))
opts['readPreferenceTags'] = rpts
self.assertEqual(result['options'], options)
hostname = next(iter(client_context.client.nodes))[0]
# The replica set members must be configured as 'localhost'.
if hostname == 'localhost':
copts = client_context.default_client_options.copy()
if client_context.ssl is True:
# Our test certs don't support the SRV hosts used in these tests.
copts['ssl_match_hostname'] = False
client = MongoClient(uri, **copts)
# Force server selection
client.admin.command('ismaster')
wait_until(
lambda: hosts == client.nodes,
'match test hosts to client nodes')
else:
try:
parse_uri(uri)
except (ConfigurationError, ValueError):
pass
else:
self.fail("failed to raise an exception")
def run_test(self):
if not _HAVE_DNSPYTHON:
raise unittest.SkipTest("DNS tests require the dnspython module")
uri = test_case['uri']
seeds = test_case['seeds']
hosts = test_case['hosts']
options = test_case.get('options')
if seeds:
seeds = split_hosts(','.join(seeds))
if hosts:
hosts = frozenset(split_hosts(','.join(hosts)))
if options:
for key, value in options.items():
# Convert numbers / booleans to strings for comparison
if isinstance(value, bool):
options[key] = 'true' if value else 'false'
elif isinstance(value, (int, float)):
options[key] = str(value)
if seeds:
result = parse_uri(uri, validate=False)
self.assertEqual(sorted(result['nodelist']), sorted(seeds))
if options:
opts = result['options']
if 'readpreferencetags' in opts:
rpts = validate_read_preference_tags(
'readPreferenceTags', opts.pop('readpreferencetags'))
opts['readPreferenceTags'] = rpts
self.assertEqual(result['options'], options)
hostname = next(iter(client_context.client.nodes))[0]
# The replica set members must be configured as 'localhost'.
if hostname == 'localhost':
copts = client_context.default_client_options.copy()
if client_context.ssl is True:
# Our test certs don't support the SRV hosts used in these tests.
copts['ssl_match_hostname'] = False
client = MongoClient(uri, **copts)
# Force server selection
client.admin.command('ismaster')
wait_until(
lambda: hosts == client.nodes,
'match test hosts to client nodes')
else:
try:
parse_uri(uri)
except (ConfigurationError, ValueError):
pass
else:
self.fail("failed to raise an exception")
def test_split_hosts(self):
self.assertRaises(ConfigurationError, split_hosts, "localhost:27017,")
self.assertRaises(ConfigurationError, split_hosts, ",localhost:27017")
self.assertRaises(ConfigurationError, split_hosts, "localhost:27017,,localhost:27018")
self.assertEqual([("localhost", 27017), ("example.com", 27017)], split_hosts("localhost,example.com"))
self.assertEqual(
[("localhost", 27018), ("example.com", 27019)], split_hosts("localhost:27018,example.com:27019")
)
self.assertRaises(ConfigurationError, split_hosts, "::1", 27017)
self.assertRaises(ConfigurationError, split_hosts, "[::1:27017")
self.assertRaises(ConfigurationError, split_hosts, "::1")
self.assertRaises(ConfigurationError, split_hosts, "::1]:27017")
self.assertEqual([("::1", 27017)], split_hosts("[::1]:27017"))
self.assertEqual([("::1", 27017)], split_hosts("[::1]"))
def test_split_hosts(self):
self.assertRaises(ConfigurationError, split_hosts, 'localhost:27017,')
self.assertRaises(ConfigurationError, split_hosts, ',localhost:27017')
self.assertRaises(ConfigurationError, split_hosts,
'localhost:27017,,localhost:27018')
self.assertEqual([('localhost', 27017), ('example.com', 27017)],
split_hosts('localhost,example.com'))
self.assertEqual([('localhost', 27018), ('example.com', 27019)],
split_hosts('localhost:27018,example.com:27019'))
self.assertRaises(ConfigurationError, split_hosts, '::1', 27017)
self.assertRaises(ConfigurationError, split_hosts, '[::1:27017')
self.assertRaises(ConfigurationError, split_hosts, '::1')
self.assertRaises(ConfigurationError, split_hosts, '::1]:27017')
self.assertEqual([('::1', 27017)], split_hosts('[::1]:27017'))
self.assertEqual([('::1', 27017)], split_hosts('[::1]'))
def __init__(self, host=None, port=None, document_class=dict,
tz_aware=False, connect=True, _store=None, read_preference=None,
**kwargs):
if host:
self.host = host[0] if isinstance(host, (list, tuple)) else host
else:
self.host = self.HOST
self.port = port or self.PORT
self._tz_aware = tz_aware
self._database_accesses = {}
self._store = _store or ServerStore()
self._id = next(self._CONNECTION_ID)
self._document_class = document_class
if read_preference is not None:
read_preferences.ensure_read_preference_type('read_preference', read_preference)
self._read_preference = read_preference or _READ_PREFERENCE_PRIMARY
dbase = None
if '://' in self.host:
res = parse_uri(self.host, default_port=self.port, warn=True)
self.host, self.port = res['nodelist'][0]
dbase = res['database']
else:
self.host, self.port = split_hosts(self.host, default_port=self.port)[0]
self.__default_datebase_name = dbase
def get_mongo_creds(self):
properties = get_properties_from_xml_file(cfg['maven']['environment'],
self.settings_xml_file)
mongo_host = split_hosts(properties['eva.mongo.host'])[0][0]
mongo_user = properties['eva.mongo.user']
mongo_pass = properties['eva.mongo.passwd']
return mongo_host, mongo_user, mongo_pass
def test_split_hosts(self):
self.assertRaises(ConfigurationError, split_hosts,
'localhost:27017,')
self.assertRaises(ConfigurationError, split_hosts,
',localhost:27017')
self.assertRaises(ConfigurationError, split_hosts,
'localhost:27017,,localhost:27018')
self.assertEqual([('localhost', 27017), ('example.com', 27017)],
split_hosts('localhost,example.com'))
self.assertEqual([('localhost', 27018), ('example.com', 27019)],
split_hosts('localhost:27018,example.com:27019'))
self.assertRaises(ConfigurationError, split_hosts, '::1', 27017)
self.assertRaises(ConfigurationError, split_hosts, '[::1:27017')
self.assertRaises(ConfigurationError, split_hosts, '::1')
self.assertRaises(ConfigurationError, split_hosts, '::1]:27017')
self.assertEqual([('::1', 27017)], split_hosts('[::1]:27017'))
self.assertEqual([('::1', 27017)], split_hosts('[::1]'))
def test_split_hosts(self):
self.assertRaises(ConfigurationError, split_hosts,
'localhost:27017,')
self.assertRaises(ConfigurationError, split_hosts,
',localhost:27017')
self.assertRaises(ConfigurationError, split_hosts,
'localhost:27017,,localhost:27018')
self.assertEqual([('localhost', 27017), ('example.com', 27017)],
split_hosts('localhost,example.com'))
self.assertEqual([('localhost', 27018), ('example.com', 27019)],
split_hosts('localhost:27018,example.com:27019'))
self.assertEqual([('/tmp/mongodb-27017.sock', None)],
split_hosts('/tmp/mongodb-27017.sock'))
self.assertEqual([('/tmp/mongodb-27017.sock', None),
('example.com', 27017)],
split_hosts('/tmp/mongodb-27017.sock,'
'example.com:27017'))
self.assertEqual([('example.com', 27017),
('/tmp/mongodb-27017.sock', None)],
split_hosts('example.com:27017,'
'/tmp/mongodb-27017.sock'))
self.assertRaises(ValueError, split_hosts, '::1', 27017)
self.assertRaises(ValueError, split_hosts, '[::1:27017')
self.assertRaises(ValueError, split_hosts, '::1')
self.assertRaises(ValueError, split_hosts, '::1]:27017')
self.assertEqual([('::1', 27017)], split_hosts('[::1]:27017'))
self.assertEqual([('::1', 27017)], split_hosts('[::1]'))
def test_split_hosts(self):
self.assertRaises(ConfigurationError, split_hosts, 'localhost:27017,')
self.assertRaises(ConfigurationError, split_hosts, ',localhost:27017')
self.assertRaises(ConfigurationError, split_hosts,
'localhost:27017,,localhost:27018')
self.assertEqual([('localhost', 27017), ('example.com', 27017)],
split_hosts('localhost,example.com'))
self.assertEqual([('localhost', 27018), ('example.com', 27019)],
split_hosts('localhost:27018,example.com:27019'))
self.assertEqual([('/tmp/mongodb-27017.sock', None)],
split_hosts('/tmp/mongodb-27017.sock'))
self.assertEqual([('/tmp/mongodb-27017.sock', None),
('example.com', 27017)],
split_hosts('/tmp/mongodb-27017.sock,'
'example.com:27017'))
self.assertEqual([('example.com', 27017),
('/tmp/mongodb-27017.sock', None)],
split_hosts('example.com:27017,'
'/tmp/mongodb-27017.sock'))
self.assertRaises(ValueError, split_hosts, '::1', 27017)
self.assertRaises(ValueError, split_hosts, '[::1:27017')
self.assertRaises(ValueError, split_hosts, '::1')
self.assertRaises(ValueError, split_hosts, '::1]:27017')
self.assertEqual([('::1', 27017)], split_hosts('[::1]:27017'))
self.assertEqual([('::1', 27017)], split_hosts('[::1]'))
def get_primary_mongo_creds_for_profile(profile_name: str,
settings_xml_file: str):
"""
Gets primary host, username, and password for mongo database.
Useful for filling properties files, for connection purposes it is preferable to use
`mongo_utils.get_mongo_connection_handle` as that will handle multiple hosts appropriately.
"""
properties = get_properties_from_xml_file(profile_name, settings_xml_file)
# Use the primary mongo host from configuration:
# https://github.com/EBIvariation/configuration/blob/master/eva-maven-settings.xml#L111
# TODO: revisit once accessioning/variant pipelines can support multiple hosts
try:
mongo_host = split_hosts(properties['eva.mongo.host'])[1][0]
except IndexError: # some profiles have only one host
mongo_host = split_hosts(properties['eva.mongo.host'])[0][0]
mongo_user = properties['eva.mongo.user']
mongo_pass = properties['eva.mongo.passwd']
return mongo_host, mongo_user, mongo_pass
def main():
module = AnsibleModule(
argument_spec={
'login_user': {'aliases': ['user']},
'login_password': {'aliases': ['password'], no_log: True},
'login_host': {'aliases': ['host'], 'default': 'localhost'},
'login_port': {'aliases': ['port'], 'default': 27017},
'hosts': {'aliases': ['members'], 'required': True},
'replica_set': {'aliases': ['replset'], 'required': True}
},
required_together=[['login_host', 'login_port']]
)
if not pymongo_found:
module.fail_json(msg='Python module "pymongo" must be installed.')
user, password, host, port, hosts, replset = (
module.params[k] for k in ['login_user', 'login_password', 'login_host',
'login_port', 'hosts', 'replica_set'])
nodes = split_hosts(hosts)
if not user and not password:
user, password = read_mongocnf_creds()
initiated = False
try:
try:
client = MongoReplicaSetClient(hosts, replicaSet=replset,
read_preference=ReadPreference.PRIMARY)
initiated = True
except ConfigurationError, e:
if 'is not a member of replica set' in e.message:
client = MongoClient(host, int(port), read_preference=ReadPreference.SECONDARY)
else:
module.fail_json(msg="Unable to connect: %s" % e)
if user and password:
try:
client.admin.authenticate(user, password)
except OperationFailure, e:
pass # try to continue, maybe admin account is not set yet
def _parse_any_host(host, default_port=27017):
if isinstance(host, tuple):
return _parse_any_host(host[0], host[1])
if '://' in host:
return parse_uri(host, warn=True)['nodelist']
return split_hosts(host, default_port=default_port)
def __init__(self, hosts_or_uri=None, max_pool_size=10,
document_class=dict, tz_aware=False, **kwargs):
"""Create a new connection to a MongoDB replica set.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `hosts_or_uri` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a string of `host:port` pairs (e.g. 'host1:port1,host2:port2').
If `hosts_or_uri` is None 'localhost:27017' will be used.
:Parameters:
- `hosts_or_uri` (optional): A MongoDB URI or string of `host:port`
pairs. If a host is an IPv6 literal it must be enclosed in '[' and
']' characters following the RFC2732 URL syntax (e.g. '[::1]' for
localhost)
- `max_pool_size` (optional): The maximum size limit for
each connection pool.
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`Connection` will be timezone
aware (otherwise they will be naive)
- `replicaSet`: (required) The name of the replica set to connect to.
The driver will verify that each host it connects to is a member of
this replica set. Can be passed as a keyword argument or as a
MongoDB URI option.
Other optional parameters can be passed as keyword arguments:
- `safe`: Use getlasterror for each write operation?
- `j` or `journal`: Block until write operations have been commited
to the journal. Ignored if the server is running without
journaling. Implies safe=True.
- `w`: (integer or string) If this is a replica set write operations
won't return until they have been replicated to the specified
number or tagged set of servers.
Implies safe=True.
- `wtimeoutMS`: Used in conjunction with `j` and/or `w`. Wait this
many milliseconds for journal acknowledgement and/or write
replication. Implies safe=True.
- `fsync`: Force the database to fsync all files before returning
When used with `j` the server awaits the next group commit before
returning. Implies safe=True.
- `socketTimeoutMS`: How long a send or receive on a socket can take
before timing out.
- `connectTimeoutMS`: How long a connection can take to be opened
before timing out.
- `ssl`: If True, create the connection to the servers using SSL.
- `read_preference`: The read preference for this connection.
See :class:`~pymongo.ReadPreference` for available options.
- `slave_okay` or `slaveOk` (deprecated): Use `read_preference`
instead.
.. versionadded:: 2.1
"""
self.__max_pool_size = max_pool_size
self.__document_class = document_class
self.__tz_aware = tz_aware
self.__opts = {}
self.__seeds = set()
self.__hosts = None
self.__arbiters = set()
self.__writer = None
self.__readers = []
self.__pools = {}
self.__index_cache = {}
self.__auth_credentials = {}
username = None
db_name = None
if hosts_or_uri is None:
self.__seeds.add(('localhost', 27017))
elif '://' in hosts_or_uri:
res = uri_parser.parse_uri(hosts_or_uri)
self.__seeds.update(res['nodelist'])
username = res['username']
password = res['password']
db_name = res['database']
self.__opts = res['options']
else:
self.__seeds.update(uri_parser.split_hosts(hosts_or_uri))
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
self.__opts[option] = value
self.__name = self.__opts.get('replicaset')
if not self.__name:
raise ConfigurationError("the replicaSet "
"keyword parameter is required.")
self.__net_timeout = self.__opts.get('sockettimeoutms')
self.__conn_timeout = self.__opts.get('connecttimeoutms')
self.__use_ssl = self.__opts.get('ssl', False)
if self.__use_ssl and not pool.have_ssl:
raise ConfigurationError("The ssl module is not available. If you "
"are using a python version previous to "
"2.6 you must install the ssl package "
"from PyPI.")
super(ReplicaSetConnection, self).__init__(**self.__opts)
if self.slave_okay:
warnings.warn("slave_okay is deprecated. Please "
"use read_preference instead.", DeprecationWarning)
self.refresh()
monitor_thread = Monitor(self)
monitor_thread.setName("ReplicaSetMonitorThread")
monitor_thread.setDaemon(True)
monitor_thread.start()
if db_name and username is None:
warnings.warn("must provide a username and password "
"to authenticate to %s" % (db_name,))
if username:
db_name = db_name or 'admin'
if not self[db_name].authenticate(username, password):
raise ConfigurationError("authentication failed")
def _parse_any_host(host):
if '://' in host:
return parse_uri(host, warn=True)['nodelist']
return split_hosts(host)
def __init__(self, host=None, port=None, max_pool_size=10,
network_timeout=None, document_class=dict,
tz_aware=False, _connect=True, **kwargs):
"""Create a new connection to a single MongoDB instance at *host:port*.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`TypeError` if port is not an instance of
``int``. Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `host` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a simple hostname. It can also be a list of hostnames or
URIs. Any port specified in the host string(s) will override
the `port` parameter. If multiple mongodb URIs containing
database or auth information are passed, the last database,
username, and password present will be used. For username and
passwords reserved characters like ':', '/', '+' and '@' must be
escaped following RFC 2396.
:Parameters:
- `host` (optional): hostname or IP address of the
instance to connect to, or a mongodb URI, or a list of
hostnames / mongodb URIs. If `host` is an IPv6 literal
it must be enclosed in '[' and ']' characters following
the RFC2732 URL syntax (e.g. '[::1]' for localhost)
- `port` (optional): port number on which to connect
- `max_pool_size` (optional): The maximum size limit for
the connection pool.
- `network_timeout` (optional): timeout (in seconds) to use
for socket operations - default is no timeout
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`Connection` will be timezone
aware (otherwise they will be naive)
Other optional parameters can be passed as keyword arguments:
- `safe`: Use getlasterror for each write operation?
- `j` or `journal`: Block until write operations have been committed
to the journal. Ignored if the server is running without journaling.
Implies safe=True.
- `w`: (integer or string) If this is a replica set write operations
won't return until they have been replicated to the specified
number or tagged set of servers. `w` always includes the replica set
primary (e.g. w=3 means write to the primary and wait until replicated
to **two** secondaries). Implies safe=True.
- `wtimeout`: Used in conjunction with `j` and/or `w`. Wait this many
milliseconds for journal acknowledgement and/or write replication.
Implies safe=True.
- `fsync`: Force the database to fsync all files before returning
When used with `j` the server awaits the next group commit before
returning.
Implies safe=True.
- `replicaSet`: The name of the replica set to connect to. The driver
will verify that the replica set it connects to matches this name.
Implies that the hosts specified are a seed list and the driver should
attempt to find all members of the set.
- `socketTimeoutMS`: How long (in milliseconds) a send or receive on a
socket can take before timing out.
- `connectTimeoutMS`: How long (in milliseconds) a connection can take
to be opened before timing out.
- `ssl`: If True, create the connection to the server using SSL.
- `read_preference`: The read preference for this connection.
See :class:`~pymongo.read_preferences.ReadPreference` for available
options.
- `auto_start_request`: If True (the default), each thread that
accesses this Connection has a socket allocated to it for the
thread's lifetime. This ensures consistent reads, even if you read
after an unsafe write.
- `use_greenlets` (optional): if ``True``, :meth:`start_request()`
will ensure that the current greenlet uses the same socket for all
operations until :meth:`end_request()`
- `slave_okay` or `slaveOk` (deprecated): Use `read_preference`
instead.
.. seealso:: :meth:`end_request`
.. versionchanged:: 2.3
Added support for failover between mongos seed list members.
.. versionchanged:: 2.2
Added `auto_start_request` option back. Added `use_greenlets`
option.
.. versionchanged:: 2.1
Support `w` = integer or string.
Added `ssl` option.
DEPRECATED slave_okay/slaveOk.
.. versionchanged:: 2.0
`slave_okay` is a pure keyword argument. Added support for safe,
and getlasterror options as keyword arguments.
.. versionchanged:: 1.11
Added `max_pool_size`. Completely removed previously deprecated
`pool_size`, `auto_start_request` and `timeout` parameters.
.. versionchanged:: 1.8
The `host` parameter can now be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition
to a simple hostname. It can also be a list of hostnames or
URIs.
.. versionadded:: 1.8
The `tz_aware` parameter.
.. versionadded:: 1.7
The `document_class` parameter.
.. versionadded:: 1.1
The `network_timeout` parameter.
.. mongodoc:: connections
"""
if host is None:
host = self.HOST
if isinstance(host, basestring):
host = [host]
if port is None:
port = self.PORT
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
seeds = set()
username = None
password = None
db = None
options = {}
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port)
seeds.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
db = res["database"] or db
options = res["options"]
else:
idx = entity.find("://")
raise InvalidURI("Invalid URI scheme: "
"%s" % (entity[:idx],))
else:
seeds.update(uri_parser.split_hosts(entity, port))
if not seeds:
raise ConfigurationError("need to specify at least one host")
self.__nodes = seeds
self.__host = None
self.__port = None
self.__is_primary = False
self.__is_mongos = False
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
options[option] = value
self.__max_pool_size = common.validate_positive_integer(
'max_pool_size', max_pool_size)
self.__cursor_manager = CursorManager(self)
self.__repl = options.get('replicaset')
if len(seeds) == 1 and not self.__repl:
self.__direct = True
else:
self.__direct = False
self.__nodes = set()
if network_timeout is not None:
if (not isinstance(network_timeout, (int, float)) or
network_timeout <= 0):
raise ConfigurationError("network_timeout must "
"be a positive integer")
self.__net_timeout = (network_timeout or
options.get('sockettimeoutms'))
self.__conn_timeout = options.get('connecttimeoutms')
self.__use_ssl = options.get('ssl', False)
if self.__use_ssl and not pool.have_ssl:
raise ConfigurationError("The ssl module is not available. If you "
"are using a python version previous to "
"2.6 you must install the ssl package "
"from PyPI.")
if options.get('use_greenlets', False):
if not pool.have_greenlet:
raise ConfigurationError(
"The greenlet module is not available. "
"Install the greenlet package from PyPI."
)
self.pool_class = pool.GreenletPool
else:
self.pool_class = pool.Pool
self.__pool = self.pool_class(
None,
self.__max_pool_size,
self.__net_timeout,
self.__conn_timeout,
self.__use_ssl
)
self.__document_class = document_class
self.__tz_aware = common.validate_boolean('tz_aware', tz_aware)
self.__auto_start_request = options.get('auto_start_request', True)
# cache of existing indexes used by ensure_index ops
self.__index_cache = {}
self.__auth_credentials = {}
super(Connection, self).__init__(**options)
if self.slave_okay:
warnings.warn("slave_okay is deprecated. Please "
"use read_preference instead.", DeprecationWarning)
if _connect:
try:
self.__find_node(seeds)
except AutoReconnect, e:
# ConnectionFailure makes more sense here than AutoReconnect
raise ConnectionFailure(str(e))
def __init__(self,
hosts_or_uri=None,
max_pool_size=10,
document_class=dict,
tz_aware=False,
**kwargs):
"""Create a new connection to a MongoDB replica set.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `hosts_or_uri` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a string of `host:port` pairs (e.g. 'host1:port1,host2:port2').
If `hosts_or_uri` is None 'localhost:27017' will be used.
:Parameters:
- `hosts_or_uri` (optional): A MongoDB URI or string of `host:port`
pairs. If a host is an IPv6 literal it must be enclosed in '[' and
']' characters following the RFC2732 URL syntax (e.g. '[::1]' for
localhost)
- `max_pool_size` (optional): The maximum size limit for
each connection pool.
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`Connection` will be timezone
aware (otherwise they will be naive)
- `replicaSet`: (required) The name of the replica set to connect to.
The driver will verify that each host it connects to is a member of
this replica set. Can be passed as a keyword argument or as a
MongoDB URI option.
Other optional parameters can be passed as keyword arguments:
- `safe`: Use getlasterror for each write operation?
- `j` or `journal`: Block until write operations have been commited
to the journal. Ignored if the server is running without
journaling. Implies safe=True.
- `w`: (integer or string) If this is a replica set write operations
won't return until they have been replicated to the specified
number or tagged set of servers.
Implies safe=True.
- `wtimeoutMS`: Used in conjunction with `j` and/or `w`. Wait this
many milliseconds for journal acknowledgement and/or write
replication. Implies safe=True.
- `fsync`: Force the database to fsync all files before returning
When used with `j` the server awaits the next group commit before
returning. Implies safe=True.
- `socketTimeoutMS`: How long a send or receive on a socket can take
before timing out.
- `connectTimeoutMS`: How long a connection can take to be opened
before timing out.
- `ssl`: If True, create the connection to the servers using SSL.
- `read_preference`: The read preference for this connection.
See :class:`~pymongo.ReadPreference` for available options.
- `slave_okay` or `slaveOk` (deprecated): Use `read_preference`
instead.
.. versionadded:: 2.1
"""
self.__max_pool_size = max_pool_size
self.__document_class = document_class
self.__tz_aware = tz_aware
self.__opts = {}
self.__seeds = set()
self.__hosts = None
self.__arbiters = set()
self.__writer = None
self.__readers = []
self.__pools = {}
self.__index_cache = {}
self.__auth_credentials = {}
username = None
db_name = None
if hosts_or_uri is None:
self.__seeds.add(('localhost', 27017))
elif '://' in hosts_or_uri:
res = uri_parser.parse_uri(hosts_or_uri)
self.__seeds.update(res['nodelist'])
username = res['username']
password = res['password']
db_name = res['database']
self.__opts = res['options']
else:
self.__seeds.update(uri_parser.split_hosts(hosts_or_uri))
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
self.__opts[option] = value
self.__name = self.__opts.get('replicaset')
if not self.__name:
raise ConfigurationError("the replicaSet "
"keyword parameter is required.")
self.__net_timeout = self.__opts.get('sockettimeoutms')
self.__conn_timeout = self.__opts.get('connecttimeoutms')
self.__use_ssl = self.__opts.get('ssl', False)
if self.__use_ssl and not pool.have_ssl:
raise ConfigurationError("The ssl module is not available. If you "
"are using a python version previous to "
"2.6 you must install the ssl package "
"from PyPI.")
super(ReplicaSetConnection, self).__init__(**self.__opts)
if self.slave_okay:
warnings.warn(
"slave_okay is deprecated. Please "
"use read_preference instead.", DeprecationWarning)
self.refresh()
monitor_thread = Monitor(self)
monitor_thread.setName("ReplicaSetMonitorThread")
monitor_thread.setDaemon(True)
monitor_thread.start()
if db_name and username is None:
warnings.warn("must provide a username and password "
"to authenticate to %s" % (db_name, ))
if username:
db_name = db_name or 'admin'
if not self[db_name].authenticate(username, password):
raise ConfigurationError("authentication failed")
def run_test(self):
if not _HAVE_DNSPYTHON:
raise unittest.SkipTest("DNS tests require the dnspython module")
uri = test_case['uri']
seeds = test_case['seeds']
hosts = test_case['hosts']
options = test_case.get('options', {})
if 'ssl' in options:
options['tls'] = options.pop('ssl')
parsed_options = test_case.get('parsed_options')
# See DRIVERS-1324, unless tls is explicitly set to False we need TLS.
needs_tls = not (options and (options.get('ssl') == False
or options.get('tls') == False))
if needs_tls and not client_context.tls:
self.skipTest('this test requires a TLS cluster')
if not needs_tls and client_context.tls:
self.skipTest('this test requires a non-TLS cluster')
if seeds:
seeds = split_hosts(','.join(seeds))
if hosts:
hosts = frozenset(split_hosts(','.join(hosts)))
if seeds:
result = parse_uri(uri, validate=True)
self.assertEqual(sorted(result['nodelist']), sorted(seeds))
if options:
opts = result['options']
if 'readpreferencetags' in opts:
rpts = validate_read_preference_tags(
'readPreferenceTags', opts.pop('readpreferencetags'))
opts['readPreferenceTags'] = rpts
self.assertEqual(result['options'], options)
if parsed_options:
for opt, expected in parsed_options.items():
if opt == 'user':
self.assertEqual(result['username'], expected)
elif opt == 'password':
self.assertEqual(result['password'], expected)
elif opt == 'auth_database' or opt == 'db':
self.assertEqual(result['database'], expected)
hostname = next(iter(client_context.client.nodes))[0]
# The replica set members must be configured as 'localhost'.
if hostname == 'localhost':
copts = client_context.default_client_options.copy()
# Remove tls since SRV parsing should add it automatically.
copts.pop('tls', None)
if client_context.tls:
# Our test certs don't support the SRV hosts used in these
# tests.
copts['tlsAllowInvalidHostnames'] = True
# The SRV spec tests assume drivers auto discover replica set
# members. This should be removed during PYTHON-2679.
if not self.load_balanced and ('directconnection'
not in result['options']):
copts['directConnection'] = False
client = MongoClient(uri, **copts)
wait_until(lambda: hosts == client.nodes,
'match test hosts to client nodes')
else:
try:
parse_uri(uri)
except (ConfigurationError, ValueError):
pass
else:
self.fail("failed to raise an exception")
def __init__(
self,
host=None,
port=None,
max_pool_size=10,
network_timeout=None,
document_class=dict,
tz_aware=False,
_connect=True,
**kwargs
):
"""Create a new connection to a single MongoDB instance at *host:port*.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`TypeError` if port is not an instance of
``int``. Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `host` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a simple hostname. It can also be a list of hostnames or
URIs. Any port specified in the host string(s) will override
the `port` parameter. If multiple mongodb URIs containing
database or auth information are passed, the last database,
username, and password present will be used.
:Parameters:
- `host` (optional): hostname or IP address of the
instance to connect to, or a mongodb URI, or a list of
hostnames / mongodb URIs. If `host` is an IPv6 literal
it must be enclosed in '[' and ']' characters following
the RFC2732 URL syntax (e.g. '[::1]' for localhost)
- `port` (optional): port number on which to connect
- `max_pool_size` (optional): The maximum size limit for
the connection pool.
- `network_timeout` (optional): timeout (in seconds) to use
for socket operations - default is no timeout
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`Connection` will be timezone
aware (otherwise they will be naive)
Other optional parameters can be passed as keyword arguments:
- `slave_okay` or `slaveok`: Is it OK to perform queries if
this connection is to a secondary?
- `safe`: Use getlasterror for each write operation?
- `j` or `journal`: Block until write operations have been commited
to the journal. Ignored if the server is running without journaling.
Implies safe=True.
- `w`: (integer or string) If this is a replica set write operations
won't return until they have been replicated to the specified
number or tagged set of servers.
Implies safe=True.
- `wtimeout`: Used in conjunction with `j` and/or `w`. Wait this many
milliseconds for journal acknowledgement and/or write replication.
Implies safe=True.
- `fsync`: Force the database to fsync all files before returning
When used with `j` the server awaits the next group commit before
returning.
Implies safe=True.
- `replicaset`: The name of the replica set to connect to. The driver
will verify that the replica set it connects to matches this name.
Implies that the hosts specified are a seed list and the driver should
attempt to find all members of the set.
.. seealso:: :meth:`end_request`
.. versionchanged:: 2.0.1+
Support `w` = integer or string.
.. versionchanged:: 2.0
`slave_okay` is a pure keyword argument. Added support for safe,
and getlasterror options as keyword arguments.
.. versionchanged:: 1.11
Added `max_pool_size`. Completely removed previously deprecated
`pool_size`, `auto_start_request` and `timeout` parameters.
.. versionchanged:: 1.8
The `host` parameter can now be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition
to a simple hostname. It can also be a list of hostnames or
URIs.
.. versionadded:: 1.8
The `tz_aware` parameter.
.. versionadded:: 1.7
The `document_class` parameter.
.. versionchanged:: 1.4
DEPRECATED The `pool_size`, `auto_start_request`, and `timeout`
parameters.
.. versionadded:: 1.1
The `network_timeout` parameter.
.. mongodoc:: connections
"""
super(Connection, self).__init__(**kwargs)
if host is None:
host = self.HOST
if isinstance(host, basestring):
host = [host]
if port is None:
port = self.PORT
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
nodes = set()
username = None
password = None
db = None
options = {}
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port)
nodes.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
db = res["database"] or db
options = res["options"]
else:
idx = entity.find("://")
raise InvalidURI("Invalid URI scheme: " "%s" % (entity[:idx],))
else:
nodes.update(uri_parser.split_hosts(entity, port))
if not nodes:
raise ConfigurationError("need to specify at least one host")
self.__nodes = nodes
self.__host = None
self.__port = None
if options:
super(Connection, self)._set_options(**options)
assert isinstance(max_pool_size, int), "max_pool_size must be an int"
self.__max_pool_size = options.get("maxpoolsize") or max_pool_size
if self.__max_pool_size < 0:
raise ValueError("the maximum pool size must be >= 0")
self.__cursor_manager = CursorManager(self)
self.__repl = options.get("replicaset", kwargs.get("replicaset"))
self.__network_timeout = network_timeout
self.__pool = _Pool(self.__max_pool_size, self.__network_timeout)
self.__last_checkout = time.time()
self.__document_class = document_class
self.__tz_aware = tz_aware
# cache of existing indexes used by ensure_index ops
self.__index_cache = {}
self.__auth_credentials = {}
if _connect:
self.__find_node()
if db and username is None:
warnings.warn("must provide a username and password " "to authenticate to %s" % (db,))
if username:
db = db or "admin"
if not self[db].authenticate(username, password):
raise ConfigurationError("authentication failed")
def __init__(
self,
host=None,
port=None,
document_class=dict,
tz_aware=False,
connect=True,
**kwargs):
"""Client for a MongoDB instance, a replica set, or a set of mongoses.
The client object is thread-safe and has connection-pooling built in.
If an operation fails because of a network error,
:class:`~pymongo.errors.ConnectionFailure` is raised and the client
reconnects in the background. Application code should handle this
exception (recognizing that the operation failed) and then continue to
execute.
The `host` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a simple hostname. It can also be a list of hostnames or
URIs. Any port specified in the host string(s) will override
the `port` parameter. If multiple mongodb URIs containing
database or auth information are passed, the last database,
username, and password present will be used. For username and
passwords reserved characters like ':', '/', '+' and '@' must be
escaped following RFC 2396.
:Parameters:
- `host` (optional): hostname or IP address of the
instance to connect to, or a mongodb URI, or a list of
hostnames / mongodb URIs. If `host` is an IPv6 literal
it must be enclosed in '[' and ']' characters following
the RFC2732 URL syntax (e.g. '[::1]' for localhost)
- `port` (optional): port number on which to connect
- `document_class` (optional): default class to use for
documents returned from queries on this client
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`MongoClient` will be timezone
aware (otherwise they will be naive)
- `connect` (optional): if ``True`` (the default), immediately
begin connecting to MongoDB in the background. Otherwise connect
on the first operation.
| **Other optional parameters can be passed as keyword arguments:**
- `maxPoolSize` (optional): The maximum number of connections
that the pool will open simultaneously. If this is set, operations
will block if there are `maxPoolSize` outstanding connections
from the pool. Defaults to 100. Cannot be 0.
- `socketTimeoutMS`: (integer or None) Controls how long (in
milliseconds) the driver will wait for a response after sending an
ordinary (non-monitoring) database operation before concluding that
a network error has occurred. Defaults to ``None`` (no timeout).
- `connectTimeoutMS`: (integer or None) Controls how long (in
milliseconds) the driver will wait during server monitoring when
connecting a new socket to a server before concluding the server
is unavailable. Defaults to ``20000`` (20 seconds).
- `serverSelectionTimeoutMS`: (integer) Controls how long (in
milliseconds) the driver will wait to find an available,
appropriate server to carry out a database operation; while it is
waiting, multiple server monitoring operations may be carried out,
each controlled by `connectTimeoutMS`. Defaults to ``30000`` (30
seconds).
- `waitQueueTimeoutMS`: (integer or None) How long (in milliseconds)
a thread will wait for a socket from the pool if the pool has no
free sockets. Defaults to ``None`` (no timeout).
- `waitQueueMultiple`: (integer or None) Multiplied by maxPoolSize
to give the number of threads allowed to wait for a socket at one
time. Defaults to ``None`` (no limit).
- `socketKeepAlive`: (boolean) Whether to send periodic keep-alive
packets on connected sockets. Defaults to ``False`` (do not send
keep-alive packets).
| **Write Concern options:**
| (Only set if passed. No default values.)
- `w`: (integer or string) If this is a replica set, write operations
will block until they have been replicated to the specified number
or tagged set of servers. `w=<int>` always includes the replica set
primary (e.g. w=3 means write to the primary and wait until
replicated to **two** secondaries). Passing w=0 **disables write
acknowledgement** and all other write concern options.
- `wtimeout`: (integer) Used in conjunction with `w`. Specify a value
in milliseconds to control how long to wait for write propagation
to complete. If replication does not complete in the given
timeframe, a timeout exception is raised.
- `j`: If ``True`` block until write operations have been committed
to the journal. Cannot be used in combination with `fsync`. Prior
to MongoDB 2.6 this option was ignored if the server was running
without journaling. Starting with MongoDB 2.6 write operations will
fail with an exception if this option is used when the server is
running without journaling.
- `fsync`: If ``True`` and the server is running without journaling,
blocks until the server has synced all data files to disk. If the
server is running with journaling, this acts the same as the `j`
option, blocking until write operations have been committed to the
journal. Cannot be used in combination with `j`.
| **Replica set keyword arguments for connecting with a replica set
- either directly or via a mongos:**
- `replicaSet`: (string or None) The name of the replica set to
connect to. The driver will verify that all servers it connects to
match this name. Implies that the hosts specified are a seed list
and the driver should attempt to find all members of the set.
Defaults to ``None``.
- `read_preference`: The read preference for this client. If
connecting directly to a secondary then a read preference mode
*other* than PRIMARY is required - otherwise all queries will throw
:class:`~pymongo.errors.AutoReconnect` "not master".
See :class:`~pymongo.read_preferences.ReadPreference` for all
available read preference options. Defaults to ``PRIMARY``.
| **SSL configuration:**
- `ssl`: If ``True``, create the connection to the server using SSL.
Defaults to ``False``.
- `ssl_keyfile`: The private keyfile used to identify the local
connection against mongod. If included with the ``certfile`` then
only the ``ssl_certfile`` is needed. Implies ``ssl=True``.
Defaults to ``None``.
- `ssl_certfile`: The certificate file used to identify the local
connection against mongod. Implies ``ssl=True``. Defaults to
``None``.
- `ssl_cert_reqs`: Specifies whether a certificate is required from
the other side of the connection, and whether it will be validated
if provided. It must be one of the three values ``ssl.CERT_NONE``
(certificates ignored), ``ssl.CERT_OPTIONAL``
(not required, but validated if provided), or ``ssl.CERT_REQUIRED``
(required and validated). If the value of this parameter is not
``ssl.CERT_NONE`` and a value is not provided for ``ssl_ca_certs``
PyMongo will attempt to load system provided CA certificates.
If the python version in use does not support loading system CA
certificates then the ``ssl_ca_certs`` parameter must point
to a file of CA certificates. Implies ``ssl=True``. Defaults to
``ssl.CERT_REQUIRED`` if not provided and ``ssl=True``.
- `ssl_ca_certs`: The ca_certs file contains a set of concatenated
"certification authority" certificates, which are used to validate
certificates passed from the other end of the connection.
Implies ``ssl=True``. Defaults to ``None``.
- `ssl_match_hostname`: If ``True`` (the default), and
`ssl_cert_reqs` is not ``ssl.CERT_NONE``, enables hostname
verification using the :func:`~ssl.match_hostname` function from
python's :mod:`~ssl` module. Think very carefully before setting
this to ``False`` as that could make your application vulnerable to
man-in-the-middle attacks.
.. mongodoc:: connections
.. versionchanged:: 3.0
:class:`~pymongo.mongo_client.MongoClient` is now the one and only
client class for a standalone server, mongos, or replica set.
It includes the functionality that had been split into
:class:`~pymongo.mongo_client.MongoReplicaSetClient`: it can connect
to a replica set, discover all its members, and monitor the set for
stepdowns, elections, and reconfigs.
The :class:`~pymongo.mongo_client.MongoClient` constructor no
longer blocks while connecting to the server or servers, and it no
longer raises :class:`~pymongo.errors.ConnectionFailure` if they
are unavailable, nor :class:`~pymongo.errors.ConfigurationError`
if the user's credentials are wrong. Instead, the constructor
returns immediately and launches the connection process on
background threads.
Therefore the ``alive`` method is removed since it no longer
provides meaningful information; even if the client is disconnected,
it may discover a server in time to fulfill the next operation.
In PyMongo 2.x, :class:`~pymongo.MongoClient` accepted a list of
standalone MongoDB servers and used the first it could connect to::
MongoClient(['host1.com:27017', 'host2.com:27017'])
A list of multiple standalones is no longer supported; if multiple
servers are listed they must be members of the same replica set, or
mongoses in the same sharded cluster.
The behavior for a list of mongoses is changed from "high
availability" to "load balancing". Before, the client connected to
the lowest-latency mongos in the list, and used it until a network
error prompted it to re-evaluate all mongoses' latencies and
reconnect to one of them. In PyMongo 3, the client monitors its
network latency to all the mongoses continuously, and distributes
operations evenly among those with the lowest latency. See
:ref:`mongos-load-balancing` for more information.
The ``connect`` option is added.
The ``start_request``, ``in_request``, and ``end_request`` methods
are removed, as well as the ``auto_start_request`` option.
The ``copy_database`` method is removed, see the
:doc:`copy_database examples </examples/copydb>` for alternatives.
The :meth:`MongoClient.disconnect` method is removed; it was a
synonym for :meth:`~pymongo.MongoClient.close`.
:class:`~pymongo.mongo_client.MongoClient` no longer returns an
instance of :class:`~pymongo.database.Database` for attribute names
with leading underscores. You must use dict-style lookups instead::
client['__my_database__']
Not::
client.__my_database__
"""
if host is None:
host = self.HOST
if isinstance(host, string_type):
host = [host]
if port is None:
port = self.PORT
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
seeds = set()
username = None
password = None
dbase = None
opts = {}
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port, False)
seeds.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
dbase = res["database"] or dbase
opts = res["options"]
else:
idx = entity.find("://")
raise InvalidURI("Invalid URI scheme: "
"%s" % (entity[:idx],))
else:
seeds.update(uri_parser.split_hosts(entity, port))
if not seeds:
raise ConfigurationError("need to specify at least one host")
# _pool_class, _monitor_class, and _condition_class are for deep
# customization of PyMongo, e.g. Motor.
pool_class = kwargs.pop('_pool_class', None)
monitor_class = kwargs.pop('_monitor_class', None)
condition_class = kwargs.pop('_condition_class', None)
kwargs['document_class'] = document_class
kwargs['tz_aware'] = tz_aware
opts.update(kwargs)
self.__options = options = ClientOptions(
username, password, dbase, opts)
self.__default_database_name = dbase
self.__lock = threading.Lock()
self.__cursor_manager = CursorManager(self)
self.__kill_cursors_queue = []
# Cache of existing indexes used by ensure_index ops.
self.__index_cache = {}
super(MongoClient, self).__init__(options.codec_options,
options.read_preference,
options.write_concern)
self.__all_credentials = {}
creds = options.credentials
if creds:
self._cache_credentials(creds.source, creds)
self._topology_settings = TopologySettings(
seeds=seeds,
replica_set_name=options.replica_set_name,
pool_class=pool_class,
pool_options=options.pool_options,
monitor_class=monitor_class,
condition_class=condition_class,
local_threshold_ms=options.local_threshold_ms,
server_selection_timeout=options.server_selection_timeout)
self._topology = Topology(self._topology_settings)
if connect:
self._topology.open()
def target():
client = self_ref()
if client is None:
return False # Stop the executor.
MongoClient._process_kill_cursors_queue(client)
return True
executor = periodic_executor.PeriodicExecutor(
condition_class=self._topology_settings.condition_class,
interval=common.KILL_CURSOR_FREQUENCY,
min_interval=0,
target=target)
# We strongly reference the executor and it weakly references us via
# this closure. When the client is freed, stop the executor soon.
self_ref = weakref.ref(self, executor.close)
self._kill_cursors_executor = executor
executor.open()
def __init__(self, host=None, port=None, max_pool_size=10,
document_class=dict, tz_aware=False, _connect=True, **kwargs):
"""Create a new connection to a single MongoDB instance at *host:port*.
The resultant client object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`TypeError` if port is not an instance of
``int``. Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `host` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a simple hostname. It can also be a list of hostnames or
URIs. Any port specified in the host string(s) will override
the `port` parameter. If multiple mongodb URIs containing
database or auth information are passed, the last database,
username, and password present will be used. For username and
passwords reserved characters like ':', '/', '+' and '@' must be
escaped following RFC 2396.
:Parameters:
- `host` (optional): hostname or IP address of the
instance to connect to, or a mongodb URI, or a list of
hostnames / mongodb URIs. If `host` is an IPv6 literal
it must be enclosed in '[' and ']' characters following
the RFC2732 URL syntax (e.g. '[::1]' for localhost)
- `port` (optional): port number on which to connect
- `max_pool_size` (optional): The maximum number of idle connections
to keep open in the pool for future use
- `document_class` (optional): default class to use for
documents returned from queries on this client
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`MongoClient` will be timezone
aware (otherwise they will be naive)
| **Other optional parameters can be passed as keyword arguments:**
- `socketTimeoutMS`: (integer) How long (in milliseconds) a send or
receive on a socket can take before timing out.
- `connectTimeoutMS`: (integer) How long (in milliseconds) a
connection can take to be opened before timing out.
- `auto_start_request`: If ``True``, each thread that accesses
this :class:`MongoClient` has a socket allocated to it for the
thread's lifetime. This ensures consistent reads, even if you
read after an unacknowledged write. Defaults to ``False``
- `use_greenlets`: If ``True``, :meth:`start_request()` will ensure
that the current greenlet uses the same socket for all
operations until :meth:`end_request()`
| **Write Concern options:**
- `w`: (integer or string) If this is a replica set, write operations
will block until they have been replicated to the specified number
or tagged set of servers. `w=<int>` always includes the replica set
primary (e.g. w=3 means write to the primary and wait until
replicated to **two** secondaries). Passing w=0 **disables write
acknowledgement** and all other write concern options.
- `wtimeout`: (integer) Used in conjunction with `w`. Specify a value
in milliseconds to control how long to wait for write propagation
to complete. If replication does not complete in the given
timeframe, a timeout exception is raised.
- `j`: If ``True`` block until write operations have been committed
to the journal. Ignored if the server is running without journaling.
- `fsync`: If ``True`` force the database to fsync all files before
returning. When used with `j` the server awaits the next group
commit before returning.
| **Replica set keyword arguments for connecting with a replica set
- either directly or via a mongos:**
| (ignored by standalone mongod instances)
- `replicaSet`: (string) The name of the replica set to connect to.
The driver will verify that the replica set it connects to matches
this name. Implies that the hosts specified are a seed list and the
driver should attempt to find all members of the set. *Ignored by
mongos*.
- `read_preference`: The read preference for this client. If
connecting to a secondary then a read preference mode *other* than
PRIMARY is required - otherwise all queries will throw
:class:`~pymongo.errors.AutoReconnect` "not master".
See :class:`~pymongo.read_preferences.ReadPreference` for all
available read preference options.
- `tag_sets`: Ignored unless connecting to a replica set via mongos.
Specify a priority-order for tag sets, provide a list of
tag sets: ``[{'dc': 'ny'}, {'dc': 'la'}, {}]``. A final, empty tag
set, ``{}``, means "read from any member that matches the mode,
ignoring tags.
| **SSL configuration:**
- `ssl`: If ``True``, create the connection to the server using SSL.
- `ssl_keyfile`: The private keyfile used to identify the local
connection against mongod. If included with the ``certfile` then
only the ``ssl_certfile`` is needed. Implies ``ssl=True``.
- `ssl_certfile`: The certificate file used to identify the local
connection against mongod. Implies ``ssl=True``.
- `ssl_cert_reqs`: Specifies whether a certificate is required from
the other side of the connection, and whether it will be validated
if provided. It must be one of the three values ``ssl.CERT_NONE``
(certificates ignored), ``ssl.CERT_OPTIONAL``
(not required, but validated if provided), or ``ssl.CERT_REQUIRED``
(required and validated). If the value of this parameter is not
``ssl.CERT_NONE``, then the ``ssl_ca_certs`` parameter must point
to a file of CA certificates. Implies ``ssl=True``.
- `ssl_ca_certs`: The ca_certs file contains a set of concatenated
"certification authority" certificates, which are used to validate
certificates passed from the other end of the connection.
Implies ``ssl=True``.
.. seealso:: :meth:`end_request`
.. mongodoc:: connections
.. versionchanged:: 2.5
Added additional ssl options
.. versionadded:: 2.4
"""
if host is None:
host = self.HOST
if isinstance(host, basestring):
host = [host]
if port is None:
port = self.PORT
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
seeds = set()
username = None
password = None
db_name = None
opts = {}
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port)
seeds.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
db_name = res["database"] or db_name
opts = res["options"]
else:
idx = entity.find("://")
raise InvalidURI("Invalid URI scheme: "
"%s" % (entity[:idx],))
else:
seeds.update(uri_parser.split_hosts(entity, port))
if not seeds:
raise ConfigurationError("need to specify at least one host")
self.__nodes = seeds
self.__host = None
self.__port = None
self.__is_primary = False
self.__is_mongos = False
# _pool_class option is for deep customization of PyMongo, e.g. Motor.
# SHOULD NOT BE USED BY DEVELOPERS EXTERNAL TO 10GEN.
pool_class = kwargs.pop('_pool_class', pool.Pool)
options = {}
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
options[option] = value
options.update(opts)
self.__max_pool_size = common.validate_positive_integer(
'max_pool_size', max_pool_size)
self.__cursor_manager = CursorManager(self)
self.__repl = options.get('replicaset')
if len(seeds) == 1 and not self.__repl:
self.__direct = True
else:
self.__direct = False
self.__nodes = set()
self.__net_timeout = options.get('sockettimeoutms')
self.__conn_timeout = options.get('connecttimeoutms')
self.__use_ssl = options.get('ssl', None)
self.__ssl_keyfile = options.get('ssl_keyfile', None)
self.__ssl_certfile = options.get('ssl_certfile', None)
self.__ssl_cert_reqs = options.get('ssl_cert_reqs', None)
self.__ssl_ca_certs = options.get('ssl_ca_certs', None)
ssl_kwarg_keys = [k for k in kwargs.keys() if k.startswith('ssl_')]
if self.__use_ssl == False and ssl_kwarg_keys:
raise ConfigurationError("ssl has not been enabled but the "
"following ssl parameters have been set: "
"%s. Please set `ssl=True` or remove."
% ', '.join(ssl_kwarg_keys))
if self.__ssl_cert_reqs and not self.__ssl_ca_certs:
raise ConfigurationError("If `ssl_cert_reqs` is not "
"`ssl.CERT_NONE` then you must "
"include `ssl_ca_certs` to be able "
"to validate the server.")
if ssl_kwarg_keys and self.__use_ssl is None:
# ssl options imply ssl = True
self.__use_ssl = True
if self.__use_ssl and not HAS_SSL:
raise ConfigurationError("The ssl module is not available. If you "
"are using a python version previous to "
"2.6 you must install the ssl package "
"from PyPI.")
self.__use_greenlets = options.get('use_greenlets', False)
self.__pool = pool_class(
None,
self.__max_pool_size,
self.__net_timeout,
self.__conn_timeout,
self.__use_ssl,
use_greenlets=self.__use_greenlets,
ssl_keyfile=self.__ssl_keyfile,
ssl_certfile=self.__ssl_certfile,
ssl_cert_reqs=self.__ssl_cert_reqs,
ssl_ca_certs=self.__ssl_ca_certs)
self.__document_class = document_class
self.__tz_aware = common.validate_boolean('tz_aware', tz_aware)
self.__auto_start_request = options.get('auto_start_request', False)
# cache of existing indexes used by ensure_index ops
self.__index_cache = {}
self.__auth_credentials = {}
super(MongoClient, self).__init__(**options)
if self.slave_okay:
warnings.warn("slave_okay is deprecated. Please "
"use read_preference instead.", DeprecationWarning,
stacklevel=2)
if _connect:
try:
self.__find_node(seeds)
except AutoReconnect, e:
# ConnectionFailure makes more sense here than AutoReconnect
raise ConnectionFailure(str(e))
def __init__(self,
host=None,
port=None,
max_pool_size=10,
document_class=dict,
tz_aware=False,
_connect=True,
**kwargs):
"""Create a new connection to a single MongoDB instance at *host:port*.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`TypeError` if port is not an instance of
``int``. Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `host` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a simple hostname. It can also be a list of hostnames or
URIs. Any port specified in the host string(s) will override
the `port` parameter. If multiple mongodb URIs containing
database or auth information are passed, the last database,
username, and password present will be used. For username and
passwords reserved characters like ':', '/', '+' and '@' must be
escaped following RFC 2396.
:Parameters:
- `host` (optional): hostname or IP address of the
instance to connect to, or a mongodb URI, or a list of
hostnames / mongodb URIs. If `host` is an IPv6 literal
it must be enclosed in '[' and ']' characters following
the RFC2732 URL syntax (e.g. '[::1]' for localhost)
- `port` (optional): port number on which to connect
- `max_pool_size` (optional): The maximum number of idle connections
to keep open in the pool for future use
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`MongoClient` will be timezone
aware (otherwise they will be naive)
**Other optional parameters can be passed as keyword arguments:**
- `w`: (integer or string) If this is a replica set, write operations
will block until they have been replicated to the specified number
or tagged set of servers. `w=<int>` always includes the replica set
primary (e.g. w=3 means write to the primary and wait until
replicated to **two** secondaries). **Passing w=0 disables write
acknowledgement and all other write concern options.**
- `wtimeout`: (integer) Used in conjunction with `w`. Specify a value
in milliseconds to control how long to wait for write propagation
to complete. If replication does not complete in the given
timeframe, a timeout exception is raised.
- `j`: If ``True`` block until write operations have been committed
to the journal. Ignored if the server is running without journaling.
- `fsync`: If ``True`` force the database to fsync all files before
returning. When used with `j` the server awaits the next group
commit before returning.
- `replicaSet`: (string) The name of the replica set to connect to.
The driver will verify that the replica set it connects to matches
this name. Implies that the hosts specified are a seed list and the
driver should attempt to find all members of the set.
- `socketTimeoutMS`: (integer) How long (in milliseconds) a send or
receive on a socket can take before timing out.
- `connectTimeoutMS`: (integer) How long (in milliseconds) a
connection can take to be opened before timing out.
- `ssl`: If ``True``, create the connection to the server using SSL.
- `read_preference`: The read preference for this connection.
See :class:`~pymongo.read_preferences.ReadPreference` for available
options.
- `auto_start_request`: If ``True``, each thread that accesses
this :class:`MongoClient` has a socket allocated to it for the
thread's lifetime. This ensures consistent reads, even if you
read after an unacknowledged write. Defaults to ``False``
- `use_greenlets`: If ``True``, :meth:`start_request()` will ensure
that the current greenlet uses the same socket for all
operations until :meth:`end_request()`
.. seealso:: :meth:`end_request`
.. mongodoc:: connections
.. versionadded:: 2.4
"""
if host is None:
host = self.HOST
if isinstance(host, basestring):
host = [host]
if port is None:
port = self.PORT
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
seeds = set()
username = None
password = None
db = None
opts = {}
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port)
seeds.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
db = res["database"] or db
opts = res["options"]
else:
idx = entity.find("://")
raise InvalidURI("Invalid URI scheme: "
"%s" % (entity[:idx], ))
else:
seeds.update(uri_parser.split_hosts(entity, port))
if not seeds:
raise ConfigurationError("need to specify at least one host")
self.__nodes = seeds
self.__host = None
self.__port = None
self.__is_primary = False
self.__is_mongos = False
options = {}
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
options[option] = value
options.update(opts)
self.__max_pool_size = common.validate_positive_integer(
'max_pool_size', max_pool_size)
self.__cursor_manager = CursorManager(self)
self.__repl = options.get('replicaset')
if len(seeds) == 1 and not self.__repl:
self.__direct = True
else:
self.__direct = False
self.__nodes = set()
self.__net_timeout = options.get('sockettimeoutms')
self.__conn_timeout = options.get('connecttimeoutms')
self.__use_ssl = options.get('ssl', False)
if self.__use_ssl and not pool.have_ssl:
raise ConfigurationError("The ssl module is not available. If you "
"are using a python version previous to "
"2.6 you must install the ssl package "
"from PyPI.")
if options.get('use_greenlets', False):
if not pool.have_greenlet:
raise ConfigurationError(
"The greenlet module is not available. "
"Install the greenlet package from PyPI.")
self.pool_class = pool.GreenletPool
else:
self.pool_class = pool.Pool
self.__pool = self.pool_class(None, self.__max_pool_size,
self.__net_timeout, self.__conn_timeout,
self.__use_ssl)
self.__document_class = document_class
self.__tz_aware = common.validate_boolean('tz_aware', tz_aware)
self.__auto_start_request = options.get('auto_start_request', False)
# cache of existing indexes used by ensure_index ops
self.__index_cache = {}
self.__auth_credentials = {}
super(MongoClient, self).__init__(**options)
if self.slave_okay:
warnings.warn(
"slave_okay is deprecated. Please "
"use read_preference instead.", DeprecationWarning)
if _connect:
try:
self.__find_node(seeds)
except AutoReconnect, e:
# ConnectionFailure makes more sense here than AutoReconnect
raise ConnectionFailure(str(e))
def __init__(self, hosts_or_uri, max_pool_size=10,
document_class=dict, tz_aware=False, **kwargs):
"""Create a new connection to a MongoDB replica set.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `hosts_or_uri` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a string of `host:port` pairs (e.g. 'host1:port1,host2:port2').
:Parameters:
- `hosts_or_uri`: A MongoDB URI or string of `host:port` pairs.
If a host is an IPv6 literal it must be enclosed in '[' and ']'
characters following the RFC2732 URL syntax (e.g. '[::1]' for
localhost)
- `max_pool_size` (optional): The maximum size limit for
each connection pool.
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`Connection` will be timezone
aware (otherwise they will be naive)
- `replicaset`: (required) The name of the replica set to connect to.
The driver will verify that each host it connects to is a member
of this replica set.
Other optional parameters can be passed as keyword arguments:
- `safe`: Use getlasterror for each write operation?
- `j` or `journal`: Block until write operations have been commited
to the journal. Ignored if the server is running without
journaling. Implies safe=True.
- `w`: (integer or string) If this is a replica set write operations
won't return until they have been replicated to the specified
number or tagged set of servers.
Implies safe=True.
- `wtimeout`: Used in conjunction with `j` and/or `w`. Wait this many
milliseconds for journal acknowledgement and/or write replication.
Implies safe=True.
- `fsync`: Force the database to fsync all files before returning
When used with `j` the server awaits the next group commit before
returning. Implies safe=True.
- `socketTimeoutMS`: How long a send or receive on a socket can take
before timing out.
- `connectTimeoutMS`: How long a connection can take to be opened
before timing out.
.. versionadded:: 2.0.1+
"""
warnings.warn("ReplicaSetConnection IS NOT READY "
"FOR PRODUCTION USE. USE AT YOUR OWN RISK!")
self.__max_pool_size = max_pool_size
self.__document_class = document_class
self.__tz_aware = tz_aware
self.__opts = {}
self.__hosts = set()
self.__arbiters = set()
self.__writer = None
self.__readers = []
self.__reader_pools = {}
self.__index_cache = {}
self.__auth_credentials = {}
self.__read_pref = SECONDARY
username = None
db_name = None
if '://' in hosts_or_uri:
res = uri_parser.parse_uri(hosts_or_uri)
self.__seeds = res['nodes']
username = res['username']
password = res['password']
db_name = res['database']
self.__opts = res['options']
else:
self.__seeds = uri_parser.split_hosts(hosts_or_uri)
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
self.__opts[option] = value
self.__name = self.__opts.get('replicaset')
if not self.__name:
raise ConfigurationError("You must provide a replica set name.")
self.__net_timeout = self.__opts.get('sockettimeoutms')
self.__conn_timeout = self.__opts.get('connecttimeoutms')
super(ReplicaSetConnection, self).__init__(**self.__opts)
self.slave_okay = True
if db_name and username is None:
warnings.warn("must provide a username and password "
"to authenticate to %s" % (db_name,))
if username:
db_name = db_name or 'admin'
if not self[db_name].authenticate(username, password):
raise ConfigurationError("authentication failed")
def __init__(
self, host=None, port=None, max_pool_size=10, document_class=dict, tz_aware=False, _connect=True, **kwargs
):
"""Create a new connection to a single MongoDB instance at *host:port*.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`TypeError` if port is not an instance of
``int``. Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `host` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a simple hostname. It can also be a list of hostnames or
URIs. Any port specified in the host string(s) will override
the `port` parameter. If multiple mongodb URIs containing
database or auth information are passed, the last database,
username, and password present will be used. For username and
passwords reserved characters like ':', '/', '+' and '@' must be
escaped following RFC 2396.
:Parameters:
- `host` (optional): hostname or IP address of the
instance to connect to, or a mongodb URI, or a list of
hostnames / mongodb URIs. If `host` is an IPv6 literal
it must be enclosed in '[' and ']' characters following
the RFC2732 URL syntax (e.g. '[::1]' for localhost)
- `port` (optional): port number on which to connect
- `max_pool_size` (optional): The maximum number of idle connections
to keep open in the pool for future use
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`MongoClient` will be timezone
aware (otherwise they will be naive)
**Other optional parameters can be passed as keyword arguments:**
- `w`: (integer or string) If this is a replica set, write operations
will block until they have been replicated to the specified number
or tagged set of servers. `w=<int>` always includes the replica set
primary (e.g. w=3 means write to the primary and wait until
replicated to **two** secondaries). **Passing w=0 disables write
acknowledgement and all other write concern options.**
- `wtimeout`: (integer) Used in conjunction with `w`. Specify a value
in milliseconds to control how long to wait for write propagation
to complete. If replication does not complete in the given
timeframe, a timeout exception is raised.
- `j`: If ``True`` block until write operations have been committed
to the journal. Ignored if the server is running without journaling.
- `fsync`: If ``True`` force the database to fsync all files before
returning. When used with `j` the server awaits the next group
commit before returning.
- `replicaSet`: (string) The name of the replica set to connect to.
The driver will verify that the replica set it connects to matches
this name. Implies that the hosts specified are a seed list and the
driver should attempt to find all members of the set.
- `socketTimeoutMS`: (integer) How long (in milliseconds) a send or
receive on a socket can take before timing out.
- `connectTimeoutMS`: (integer) How long (in milliseconds) a
connection can take to be opened before timing out.
- `ssl`: If ``True``, create the connection to the server using SSL.
- `read_preference`: The read preference for this connection.
See :class:`~pymongo.read_preferences.ReadPreference` for available
options.
- `auto_start_request`: If ``True``, each thread that accesses
this :class:`MongoClient` has a socket allocated to it for the
thread's lifetime. This ensures consistent reads, even if you
read after an unacknowledged write. Defaults to ``False``
- `use_greenlets`: If ``True``, :meth:`start_request()` will ensure
that the current greenlet uses the same socket for all
operations until :meth:`end_request()`
.. seealso:: :meth:`end_request`
.. mongodoc:: connections
.. versionadded:: 2.4
"""
if host is None:
host = self.HOST
if isinstance(host, basestring):
host = [host]
if port is None:
port = self.PORT
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
seeds = set()
username = None
password = None
db = None
opts = {}
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port)
seeds.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
db = res["database"] or db
opts = res["options"]
else:
idx = entity.find("://")
raise InvalidURI("Invalid URI scheme: " "%s" % (entity[:idx],))
else:
seeds.update(uri_parser.split_hosts(entity, port))
if not seeds:
raise ConfigurationError("need to specify at least one host")
self.__nodes = seeds
self.__host = None
self.__port = None
self.__is_primary = False
self.__is_mongos = False
options = {}
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
options[option] = value
options.update(opts)
self.__max_pool_size = common.validate_positive_integer("max_pool_size", max_pool_size)
self.__cursor_manager = CursorManager(self)
self.__repl = options.get("replicaset")
if len(seeds) == 1 and not self.__repl:
self.__direct = True
else:
self.__direct = False
self.__nodes = set()
self.__net_timeout = options.get("sockettimeoutms")
self.__conn_timeout = options.get("connecttimeoutms")
self.__use_ssl = options.get("ssl", False)
if self.__use_ssl and not pool.have_ssl:
raise ConfigurationError(
"The ssl module is not available. If you "
"are using a python version previous to "
"2.6 you must install the ssl package "
"from PyPI."
)
use_greenlets = options.get("use_greenlets", False)
self.__pool = pool.Pool(
None, self.__max_pool_size, self.__net_timeout, self.__conn_timeout, self.__use_ssl, use_greenlets
)
self.__document_class = document_class
self.__tz_aware = common.validate_boolean("tz_aware", tz_aware)
self.__auto_start_request = options.get("auto_start_request", False)
# cache of existing indexes used by ensure_index ops
self.__index_cache = {}
self.__auth_credentials = {}
super(MongoClient, self).__init__(**options)
if self.slave_okay:
warnings.warn("slave_okay is deprecated. Please " "use read_preference instead.", DeprecationWarning)
if _connect:
try:
self.__find_node(seeds)
except AutoReconnect, e:
# ConnectionFailure makes more sense here than AutoReconnect
raise ConnectionFailure(str(e))
def __init__(self,
host=None,
port=None,
max_pool_size=10,
network_timeout=None,
document_class=dict,
tz_aware=False,
_connect=True,
**kwargs):
"""Create a new connection to a single MongoDB instance at *host:port*.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`TypeError` if port is not an instance of
``int``. Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `host` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a simple hostname. It can also be a list of hostnames or
URIs. Any port specified in the host string(s) will override
the `port` parameter. If multiple mongodb URIs containing
database or auth information are passed, the last database,
username, and password present will be used.
:Parameters:
- `host` (optional): hostname or IP address of the
instance to connect to, or a mongodb URI, or a list of
hostnames / mongodb URIs. If `host` is an IPv6 literal
it must be enclosed in '[' and ']' characters following
the RFC2732 URL syntax (e.g. '[::1]' for localhost)
- `port` (optional): port number on which to connect
- `max_pool_size` (optional): The maximum size limit for
the connection pool.
- `network_timeout` (optional): timeout (in seconds) to use
for socket operations - default is no timeout
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`Connection` will be timezone
aware (otherwise they will be naive)
Other optional parameters can be passed as keyword arguments:
- `slave_okay` or `slaveok`: Is it OK to perform queries if
this connection is to a secondary?
- `safe`: Use getlasterror for each write operation?
- `j`: Block until write operations have been commited to the
journal. Ignored if the server is running without journaling.
Implies safe=True.
- `w`: If this is a replica set the server won't return until
write operations have replicated to this many set members.
Implies safe=True.
- `wtimeout`: Used in conjunction with `j` and/or `w`. Wait this many
milliseconds for journal acknowledgement and/or write replication.
Implies safe=True.
- `fsync`: Force the database to fsync all files before returning
When used with `j` the server awaits the next group commit before
returning.
Implies safe=True.
.. seealso:: :meth:`end_request`
.. versionchanged:: 1.11+
`slave_okay` is a pure keyword argument. Added support for safe,
and getlasterror options as keyword arguments.
.. versionchanged:: 1.11
Added `max_pool_size`. Completely removed previously deprecated
`pool_size`, `auto_start_request` and `timeout` parameters.
.. versionchanged:: 1.8
The `host` parameter can now be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition
to a simple hostname. It can also be a list of hostnames or
URIs.
.. versionadded:: 1.8
The `tz_aware` parameter.
.. versionadded:: 1.7
The `document_class` parameter.
.. versionchanged:: 1.4
DEPRECATED The `pool_size`, `auto_start_request`, and `timeout`
parameters.
.. versionadded:: 1.1
The `network_timeout` parameter.
.. mongodoc:: connections
"""
super(Connection, self).__init__(**kwargs)
if host is None:
host = self.HOST
if isinstance(host, basestring):
host = [host]
if port is None:
port = self.PORT
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
nodes = set()
username = None
password = None
db = None
coll = None
options = {}
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port)
nodes.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
db = res["database"] or db
coll = res["collection"] or coll
options = res["options"]
else:
idx = entity.find("://")
raise InvalidURI("Invalid URI scheme: "
"%s" % (entity[:idx], ))
else:
nodes.update(uri_parser.split_hosts(entity, port))
if not nodes:
raise ConfigurationError("need to specify at least one host")
self.__nodes = nodes
self.__host = None
self.__port = None
if options:
super(Connection, self)._set_options(**options)
assert isinstance(max_pool_size, int), "max_pool_size must be an int"
self.__max_pool_size = options.get("maxpoolsize") or max_pool_size
if self.__max_pool_size < 0:
raise ValueError("the maximum pool size must be >= 0")
# TODO - Support using other options like w and fsync from URI
self.__options = options
# TODO - Support setting the collection from URI like the Java driver
self.__collection = coll
self.__cursor_manager = CursorManager(self)
self.__network_timeout = network_timeout
self.__pool = _Pool(self.__max_pool_size, self.__network_timeout)
self.__last_checkout = time.time()
self.__document_class = document_class
self.__tz_aware = tz_aware
# cache of existing indexes used by ensure_index ops
self.__index_cache = {}
if _connect:
self.__find_node()
if db and username is None:
warnings.warn("must provide a username and password "
"to authenticate to %s" % (db, ))
if username:
db = db or "admin"
if not self[db].authenticate(username, password):
raise ConfigurationError("authentication failed")
def mongo_client(*args, **kwargs):
""" MongoClient factory supporting instrumentation for fast-fail behavior.
All parameters are the same as those you would pass to ``MongoClient``, with
the addition of the following keyword arguments:
:param bool fail_fast: provide "fail fast" semantics on connect (default
is True)
:param list state_selectors: list of server state selectors (see:
``SERVER_STATE_SELECTORS``)
:param list type_selectors: list of server type selectors (see:
``SERVER_TYPE_SELECTORS``)
:returns: An instance of ``MongoClient``
:raises: ``errors.ServerSelectionTryOnceTimeoutError``
*NOTE*: currently, ``fail_fast==True`` implies ``connect==True``
"""
# MongoClient default parameters
_args = ('host', 'port', 'document_class', 'tz_aware', 'connect')
_kwargs = dict(zip(_args, MongoClient.__init__.func_defaults))
# update default parameters with positional args if they were passed
for i, arg in enumerate(args):
_kwargs[_args[i]] = args[i]
# grab arguments to this factory function
fail_fast = kwargs.pop('fail_fast', True)
state_selectors = kwargs.pop('state_selectors', None)
type_selectors = kwargs.pop('type_selectors', None)
# updated kwargs with default parameters
for k, v in _kwargs.iteritems():
kwargs[k] = v
if fail_fast:
# extract the seed list from the host argument
seeds = set()
if kwargs['host'] is None:
kwargs['host'] = MongoClient.HOST
if kwargs['port'] is None:
kwargs['port'] = MongoClient.PORT
if isinstance(kwargs['host'], string_type):
kwargs['host'] = [kwargs['host']]
for host in kwargs['host']:
if '://' in host:
if host.startswith('mongodb://'):
seeds.update(parse_uri(host, kwargs['port'])['nodelist'])
else:
# let MongoClient raise the error
MongoClient(**kwargs)
else:
seeds.update(split_hosts(host, kwargs['port']))
# use pymongo to parse out connect_timeout
client_options = \
ClientOptions(
None, None, None,
dict([common.validate(k, v)
for k, v in filter(lambda x: x[0] not in _args,
kwargs.items())]))
# create our event listener
listener = \
ServerHeartbeatListener(
seeds, client_options.pool_options.connect_timeout,
state_selectors, type_selectors)
if 'event_listeners' not in kwargs:
kwargs['event_listeners'] = []
# add it to the list of listeners associated with
kwargs['event_listeners'].append(listener)
if LooseVersion(pymongo.__version__) < LooseVersion('3.3'):
# if we are on a version < 3.3 we have to deliver server heartbeat
# events ourselves so we wrap the monitor class and add it to the
# parameters being passed to MongoClient
from .monitor import Monitor
listener = kwargs['event_listeners'].pop()
kwargs['_monitor_class'] = functools.partial(Monitor, listener)
# XXX: always set connect == True if we are using "fail_fast", we
# should accommodate a lazy version of this down the road
kwargs['connect'] = True
# finally, create the client with our listener
c = MongoClient(**kwargs)
# wait for the seed list to update or throw an exception
listener.wait()
# there is at least one seed that is up and that satisfies at least on
# of the server selectors specified
return c
# if fail_fast is False, simply pass through arguments to MongoClient
return MongoClient(**kwargs)
def __init__(self,
host=None,
port=None,
max_pool_size=10,
document_class=dict,
tz_aware=False,
_connect=True,
**kwargs):
"""Create a new connection to a single MongoDB instance at *host:port*.
The resultant client object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`TypeError` if port is not an instance of
``int``. Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `host` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a simple hostname. It can also be a list of hostnames or
URIs. Any port specified in the host string(s) will override
the `port` parameter. If multiple mongodb URIs containing
database or auth information are passed, the last database,
username, and password present will be used. For username and
passwords reserved characters like ':', '/', '+' and '@' must be
escaped following RFC 2396.
:Parameters:
- `host` (optional): hostname or IP address of the
instance to connect to, or a mongodb URI, or a list of
hostnames / mongodb URIs. If `host` is an IPv6 literal
it must be enclosed in '[' and ']' characters following
the RFC2732 URL syntax (e.g. '[::1]' for localhost)
- `port` (optional): port number on which to connect
- `max_pool_size` (optional): The maximum number of idle connections
to keep open in the pool for future use
- `document_class` (optional): default class to use for
documents returned from queries on this client
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`MongoClient` will be timezone
aware (otherwise they will be naive)
| **Other optional parameters can be passed as keyword arguments:**
- `socketTimeoutMS`: (integer) How long (in milliseconds) a send or
receive on a socket can take before timing out.
- `connectTimeoutMS`: (integer) How long (in milliseconds) a
connection can take to be opened before timing out.
- `auto_start_request`: If ``True``, each thread that accesses
this :class:`MongoClient` has a socket allocated to it for the
thread's lifetime. This ensures consistent reads, even if you
read after an unacknowledged write. Defaults to ``False``
- `use_greenlets`: If ``True``, :meth:`start_request()` will ensure
that the current greenlet uses the same socket for all
operations until :meth:`end_request()`
| **Write Concern options:**
- `w`: (integer or string) If this is a replica set, write operations
will block until they have been replicated to the specified number
or tagged set of servers. `w=<int>` always includes the replica set
primary (e.g. w=3 means write to the primary and wait until
replicated to **two** secondaries). Passing w=0 **disables write
acknowledgement** and all other write concern options.
- `wtimeout`: (integer) Used in conjunction with `w`. Specify a value
in milliseconds to control how long to wait for write propagation
to complete. If replication does not complete in the given
timeframe, a timeout exception is raised.
- `j`: If ``True`` block until write operations have been committed
to the journal. Ignored if the server is running without journaling.
- `fsync`: If ``True`` force the database to fsync all files before
returning. When used with `j` the server awaits the next group
commit before returning.
| **Replica set keyword arguments for connecting with a replica set
- either directly or via a mongos:**
| (ignored by standalone mongod instances)
- `replicaSet`: (string) The name of the replica set to connect to.
The driver will verify that the replica set it connects to matches
this name. Implies that the hosts specified are a seed list and the
driver should attempt to find all members of the set. *Ignored by
mongos*.
- `read_preference`: The read preference for this client. If
connecting to a secondary then a read preference mode *other* than
PRIMARY is required - otherwise all queries will throw
:class:`~pymongo.errors.AutoReconnect` "not master".
See :class:`~pymongo.read_preferences.ReadPreference` for all
available read preference options.
- `tag_sets`: Ignored unless connecting to a replica set via mongos.
Specify a priority-order for tag sets, provide a list of
tag sets: ``[{'dc': 'ny'}, {'dc': 'la'}, {}]``. A final, empty tag
set, ``{}``, means "read from any member that matches the mode,
ignoring tags.
| **SSL configuration:**
- `ssl`: If ``True``, create the connection to the server using SSL.
- `ssl_keyfile`: The private keyfile used to identify the local
connection against mongod. If included with the ``certfile` then
only the ``ssl_certfile`` is needed. Implies ``ssl=True``.
- `ssl_certfile`: The certificate file used to identify the local
connection against mongod. Implies ``ssl=True``.
- `ssl_cert_reqs`: Specifies whether a certificate is required from
the other side of the connection, and whether it will be validated
if provided. It must be one of the three values ``ssl.CERT_NONE``
(certificates ignored), ``ssl.CERT_OPTIONAL``
(not required, but validated if provided), or ``ssl.CERT_REQUIRED``
(required and validated). If the value of this parameter is not
``ssl.CERT_NONE``, then the ``ssl_ca_certs`` parameter must point
to a file of CA certificates. Implies ``ssl=True``.
- `ssl_ca_certs`: The ca_certs file contains a set of concatenated
"certification authority" certificates, which are used to validate
certificates passed from the other end of the connection.
Implies ``ssl=True``.
.. seealso:: :meth:`end_request`
.. mongodoc:: connections
.. versionchanged:: 2.5
Added addtional ssl options
.. versionadded:: 2.4
"""
if host is None:
host = self.HOST
if isinstance(host, basestring):
host = [host]
if port is None:
port = self.PORT
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
seeds = set()
username = None
password = None
db_name = None
opts = {}
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port)
seeds.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
db_name = res["database"] or db_name
opts = res["options"]
else:
idx = entity.find("://")
raise InvalidURI("Invalid URI scheme: "
"%s" % (entity[:idx], ))
else:
seeds.update(uri_parser.split_hosts(entity, port))
if not seeds:
raise ConfigurationError("need to specify at least one host")
self.__nodes = seeds
self.__host = None
self.__port = None
self.__is_primary = False
self.__is_mongos = False
# _pool_class option is for deep customization of PyMongo, e.g. Motor.
# SHOULD NOT BE USED BY DEVELOPERS EXTERNAL TO 10GEN.
pool_class = kwargs.pop('_pool_class', pool.Pool)
options = {}
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
options[option] = value
options.update(opts)
self.__max_pool_size = common.validate_positive_integer(
'max_pool_size', max_pool_size)
self.__cursor_manager = CursorManager(self)
self.__repl = options.get('replicaset')
if len(seeds) == 1 and not self.__repl:
self.__direct = True
else:
self.__direct = False
self.__nodes = set()
self.__net_timeout = options.get('sockettimeoutms')
self.__conn_timeout = options.get('connecttimeoutms')
self.__use_ssl = options.get('ssl', None)
self.__ssl_keyfile = options.get('ssl_keyfile', None)
self.__ssl_certfile = options.get('ssl_certfile', None)
self.__ssl_cert_reqs = options.get('ssl_cert_reqs', None)
self.__ssl_ca_certs = options.get('ssl_ca_certs', None)
ssl_kwarg_keys = [k for k in kwargs.keys() if k.startswith('ssl_')]
if self.__use_ssl == False and ssl_kwarg_keys:
raise ConfigurationError("ssl has not been enabled but the "
"following ssl parameters have been set: "
"%s. Please set `ssl=True` or remove." %
', '.join(ssl_kwarg_keys))
if self.__ssl_cert_reqs and not self.__ssl_ca_certs:
raise ConfigurationError("If `ssl_cert_reqs` is not "
"`ssl.CERT_NONE` then you must "
"include `ssl_ca_certs` to be able "
"to validate the server.")
if ssl_kwarg_keys and self.__use_ssl is None:
# ssl options imply ssl = True
self.__use_ssl = True
if self.__use_ssl and not HAS_SSL:
raise ConfigurationError("The ssl module is not available. If you "
"are using a python version previous to "
"2.6 you must install the ssl package "
"from PyPI.")
self.__use_greenlets = options.get('use_greenlets', False)
self.__pool = pool_class(None,
self.__max_pool_size,
self.__net_timeout,
self.__conn_timeout,
self.__use_ssl,
use_greenlets=self.__use_greenlets,
ssl_keyfile=self.__ssl_keyfile,
ssl_certfile=self.__ssl_certfile,
ssl_cert_reqs=self.__ssl_cert_reqs,
ssl_ca_certs=self.__ssl_ca_certs)
self.__document_class = document_class
self.__tz_aware = common.validate_boolean('tz_aware', tz_aware)
self.__auto_start_request = options.get('auto_start_request', False)
# cache of existing indexes used by ensure_index ops
self.__index_cache = {}
self.__auth_credentials = {}
super(MongoClient, self).__init__(**options)
if self.slave_okay:
warnings.warn(
"slave_okay is deprecated. Please "
"use read_preference instead.",
DeprecationWarning,
stacklevel=2)
if _connect:
try:
self.__find_node(seeds)
except AutoReconnect, e:
# ConnectionFailure makes more sense here than AutoReconnect
raise ConnectionFailure(str(e))
def __init__(self, hosts_or_uri=None, max_pool_size=10,
document_class=dict, tz_aware=False, **kwargs):
"""Create a new connection to a MongoDB replica set.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `hosts_or_uri` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a string of `host:port` pairs (e.g. 'host1:port1,host2:port2').
If `hosts_or_uri` is None 'localhost:27017' will be used.
:Parameters:
- `hosts_or_uri` (optional): A MongoDB URI or string of `host:port`
pairs. If a host is an IPv6 literal it must be enclosed in '[' and
']' characters following the RFC2732 URL syntax (e.g. '[::1]' for
localhost)
- `max_pool_size` (optional): The maximum size limit for
each connection pool.
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`ReplicaSetConnection` will be timezone
aware (otherwise they will be naive)
- `replicaSet`: (required) The name of the replica set to connect to.
The driver will verify that each host it connects to is a member of
this replica set. Can be passed as a keyword argument or as a
MongoDB URI option.
Other optional parameters can be passed as keyword arguments:
- `safe`: Use getlasterror for each write operation?
- `j` or `journal`: Block until write operations have been commited
to the journal. Ignored if the server is running without
journaling. Implies safe=True.
- `w`: (integer or string) If this is a replica set write operations
won't return until they have been replicated to the specified
number or tagged set of servers.
Implies safe=True.
- `wtimeoutMS`: Used in conjunction with `j` and/or `w`. Wait this
many milliseconds for journal acknowledgement and/or write
replication. Implies safe=True.
- `fsync`: Force the database to fsync all files before returning
When used with `j` the server awaits the next group commit before
returning. Implies safe=True.
- `socketTimeoutMS`: How long a send or receive on a socket can take
before timing out.
- `connectTimeoutMS`: How long a connection can take to be opened
before timing out.
- `ssl`: If True, create the connection to the servers using SSL.
- `read_preference`: The read preference for this connection.
See :class:`~pymongo.ReadPreference` for available options.
- `auto_start_request`: If True (the default), each thread that
accesses this :class:`ReplicaSetConnection` has a socket allocated
to it for the thread's lifetime, for each member of the set. For
:class:`~pymongo.ReadPreference` PRIMARY, auto_start_request=True
ensures consistent reads, even if you read after an unsafe
write. For read preferences other than PRIMARY, there are no
consistency guarantees. (The semantics of auto_start_request,
:class:`~pymongo.ReadPreference`, and :class:`ReplicaSetConnection`
may change in future releases of PyMongo.)
- `use_greenlets` (optional): if ``True``, :meth:`start_request()`
will ensure that the current greenlet uses the same socket for all
operations until :meth:`end_request()`
- `slave_okay` or `slaveOk` (deprecated): Use `read_preference`
instead.
- `host`: For compatibility with connection.Connection. If both
`host` and `hosts_or_uri` are specified `host` takes precedence.
- `port`: For compatibility with connection.Connection. The default
port number to use for hosts.
- `network_timeout`: For compatibility with connection.Connection.
The timeout (in seconds) to use for socket operations - default
is no timeout. If both `network_timeout` and `socketTimeoutMS` are
are specified `network_timeout` takes precedence, matching
connection.Connection.
.. versionchanged:: 2.1.1+
Added `auto_start_request` option.
Added support for `host`, `port`, and `network_timeout` keyword
arguments for compatibility with connection.Connection.
.. versionadded:: 2.1
"""
self.__opts = {}
self.__seeds = set()
self.__hosts = None
self.__arbiters = set()
self.__writer = None
self.__readers = []
self.__pools = {}
self.__index_cache = {}
self.__auth_credentials = {}
self.__max_pool_size = common.validate_positive_integer(
'max_pool_size', max_pool_size)
self.__tz_aware = common.validate_boolean('tz_aware', tz_aware)
self.__document_class = document_class
# Compatibility with connection.Connection
host = kwargs.pop('host', hosts_or_uri)
port = kwargs.pop('port', 27017)
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
network_timeout = kwargs.pop('network_timeout', None)
if network_timeout is not None:
if (not isinstance(network_timeout, (int, float)) or
network_timeout <= 0):
raise ConfigurationError("network_timeout must "
"be a positive integer")
username = None
db_name = None
if host is None:
self.__seeds.add(('localhost', port))
elif '://' in host:
res = uri_parser.parse_uri(host, port)
self.__seeds.update(res['nodelist'])
username = res['username']
password = res['password']
db_name = res['database']
self.__opts = res['options']
else:
self.__seeds.update(uri_parser.split_hosts(host, port))
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
self.__opts[option] = value
if self.__opts.get('use_greenlets', False):
if not have_gevent:
raise ConfigurationError(
"The gevent module is not available. "
"Install the gevent package from PyPI."
)
self.pool_class = pool.GreenletPool
else:
self.pool_class = pool.Pool
self.__auto_start_request = self.__opts.get('auto_start_request', True)
self.__in_request = self.__auto_start_request
self.__name = self.__opts.get('replicaset')
if not self.__name:
raise ConfigurationError("the replicaSet "
"keyword parameter is required.")
self.__net_timeout = (network_timeout or
self.__opts.get('sockettimeoutms'))
self.__conn_timeout = self.__opts.get('connecttimeoutms')
self.__use_ssl = self.__opts.get('ssl', False)
if self.__use_ssl and not pool.have_ssl:
raise ConfigurationError("The ssl module is not available. If you "
"are using a python version previous to "
"2.6 you must install the ssl package "
"from PyPI.")
super(ReplicaSetConnection, self).__init__(**self.__opts)
if self.slave_okay:
warnings.warn("slave_okay is deprecated. Please "
"use read_preference instead.", DeprecationWarning)
self.refresh()
if self.__opts.get('use_greenlets', False):
monitor = MonitorGreenlet(self)
else:
monitor = MonitorThread(self)
monitor.start()
if db_name and username is None:
warnings.warn("must provide a username and password "
"to authenticate to %s" % (db_name,))
if username:
db_name = db_name or 'admin'
if not self[db_name].authenticate(username, password):
raise ConfigurationError("authentication failed")
def __init__(self,
hosts_or_uri=None,
max_pool_size=10,
document_class=dict,
tz_aware=False,
**kwargs):
"""Create a new connection to a MongoDB replica set.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `hosts_or_uri` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a string of `host:port` pairs (e.g. 'host1:port1,host2:port2').
If `hosts_or_uri` is None 'localhost:27017' will be used.
.. note:: Instances of :class:`~ReplicaSetConnection` start a
background task to monitor the state of the replica set. This allows
it to quickly respond to changes in replica set configuration.
Before discarding an instance of :class:`~ReplicaSetConnection` make
sure you call :meth:`~close` to ensure that the monitor task is
cleanly shut down.
:Parameters:
- `hosts_or_uri` (optional): A MongoDB URI or string of `host:port`
pairs. If a host is an IPv6 literal it must be enclosed in '[' and
']' characters following the RFC2732 URL syntax (e.g. '[::1]' for
localhost)
- `max_pool_size` (optional): The maximum size limit for
each connection pool.
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`ReplicaSetConnection` will be timezone
aware (otherwise they will be naive)
- `replicaSet`: (required) The name of the replica set to connect to.
The driver will verify that each host it connects to is a member of
this replica set. Can be passed as a keyword argument or as a
MongoDB URI option.
Other optional parameters can be passed as keyword arguments:
- `safe`: Use getlasterror for each write operation?
- `j` or `journal`: Block until write operations have been commited
to the journal. Ignored if the server is running without
journaling. Implies safe=True.
- `w`: (integer or string) If this is a replica set write operations
won't return until they have been replicated to the specified
number or tagged set of servers.
Implies safe=True.
- `wtimeoutMS`: Used in conjunction with `j` and/or `w`. Wait this
many milliseconds for journal acknowledgement and/or write
replication. Implies safe=True.
- `fsync`: Force the database to fsync all files before returning
When used with `j` the server awaits the next group commit before
returning. Implies safe=True.
- `socketTimeoutMS`: How long a send or receive on a socket can take
before timing out.
- `connectTimeoutMS`: How long a connection can take to be opened
before timing out.
- `ssl`: If True, create the connection to the servers using SSL.
- `read_preference`: The read preference for this connection.
See :class:`~pymongo.ReadPreference` for available options.
- `auto_start_request`: If True (the default), each thread that
accesses this :class:`ReplicaSetConnection` has a socket allocated
to it for the thread's lifetime, for each member of the set. For
:class:`~pymongo.ReadPreference` PRIMARY, auto_start_request=True
ensures consistent reads, even if you read after an unsafe
write. For read preferences other than PRIMARY, there are no
consistency guarantees. (The semantics of auto_start_request,
:class:`~pymongo.ReadPreference`, and :class:`ReplicaSetConnection`
may change in future releases of PyMongo.)
- `use_greenlets` (optional): if ``True``, use a background Greenlet
instead of a background thread to monitor state of replica set.
:meth:`start_request()` will ensure that the current greenlet uses
the same socket for all operations until :meth:`end_request()`.
`use_greenlets` with ReplicaSetConnection requires `Gevent
<http://gevent.org/>`_ to be installed.
- `slave_okay` or `slaveOk` (deprecated): Use `read_preference`
instead.
- `host`: For compatibility with connection.Connection. If both
`host` and `hosts_or_uri` are specified `host` takes precedence.
- `port`: For compatibility with connection.Connection. The default
port number to use for hosts.
- `network_timeout`: For compatibility with connection.Connection.
The timeout (in seconds) to use for socket operations - default
is no timeout. If both `network_timeout` and `socketTimeoutMS` are
are specified `network_timeout` takes precedence, matching
connection.Connection.
.. versionchanged:: 2.2
Added `auto_start_request` and `use_greenlets` options.
Added support for `host`, `port`, and `network_timeout` keyword
arguments for compatibility with connection.Connection.
.. versionadded:: 2.1
"""
self.__opts = {}
self.__seeds = set()
self.__hosts = None
self.__arbiters = set()
self.__writer = None
self.__readers = []
self.__pools = {}
self.__index_cache = {}
self.__auth_credentials = {}
self.__max_pool_size = common.validate_positive_integer(
'max_pool_size', max_pool_size)
self.__tz_aware = common.validate_boolean('tz_aware', tz_aware)
self.__document_class = document_class
# Compatibility with connection.Connection
host = kwargs.pop('host', hosts_or_uri)
port = kwargs.pop('port', 27017)
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
network_timeout = kwargs.pop('network_timeout', None)
if network_timeout is not None:
if (not isinstance(network_timeout, (int, float))
or network_timeout <= 0):
raise ConfigurationError("network_timeout must "
"be a positive integer")
username = None
db_name = None
if host is None:
self.__seeds.add(('localhost', port))
elif '://' in host:
res = uri_parser.parse_uri(host, port)
self.__seeds.update(res['nodelist'])
username = res['username']
password = res['password']
db_name = res['database']
self.__opts = res['options']
else:
self.__seeds.update(uri_parser.split_hosts(host, port))
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
self.__opts[option] = value
if self.__opts.get('use_greenlets', False):
if not have_gevent:
raise ConfigurationError(
"The gevent module is not available. "
"Install the gevent package from PyPI.")
self.pool_class = pool.GreenletPool
else:
self.pool_class = pool.Pool
self.__auto_start_request = self.__opts.get('auto_start_request', True)
self.__in_request = self.__auto_start_request
self.__name = self.__opts.get('replicaset')
if not self.__name:
raise ConfigurationError("the replicaSet "
"keyword parameter is required.")
self.__net_timeout = (network_timeout
or self.__opts.get('sockettimeoutms'))
self.__conn_timeout = self.__opts.get('connecttimeoutms')
self.__use_ssl = self.__opts.get('ssl', False)
if self.__use_ssl and not pool.have_ssl:
raise ConfigurationError("The ssl module is not available. If you "
"are using a python version previous to "
"2.6 you must install the ssl package "
"from PyPI.")
super(ReplicaSetConnection, self).__init__(**self.__opts)
if self.slave_okay:
warnings.warn(
"slave_okay is deprecated. Please "
"use read_preference instead.", DeprecationWarning)
self.refresh()
if db_name and username is None:
warnings.warn("must provide a username and password "
"to authenticate to %s" % (db_name, ))
if username:
db_name = db_name or 'admin'
if not self[db_name].authenticate(username, password):
raise ConfigurationError("authentication failed")
# Start the monitor after we know the configuration is correct.
if self.__opts.get('use_greenlets', False):
self.__monitor = MonitorGreenlet(self)
else:
# NOTE: Don't ever make this a daemon thread in CPython. Daemon
# threads in CPython cause serious issues when the interpreter is
# torn down. Symptoms range from random exceptions to the
# interpreter dumping core.
self.__monitor = MonitorThread(self)
# Sadly, weakrefs aren't totally reliable in PyPy and Jython
# so use a daemon thread there.
if (sys.platform.startswith('java') or 'PyPy' in sys.version):
self.__monitor.setDaemon(True)
self.__monitor.start()
def __init__(self,
host=None,
port=None,
max_pool_size=10,
network_timeout=None,
document_class=dict,
tz_aware=False,
_connect=True,
**kwargs):
"""Create a new connection to a single MongoDB instance at *host:port*.
The resultant connection object has connection-pooling built
in. It also performs auto-reconnection when necessary. If an
operation fails because of a connection error,
:class:`~pymongo.errors.ConnectionFailure` is raised. If
auto-reconnection will be performed,
:class:`~pymongo.errors.AutoReconnect` will be
raised. Application code should handle this exception
(recognizing that the operation failed) and then continue to
execute.
Raises :class:`TypeError` if port is not an instance of
``int``. Raises :class:`~pymongo.errors.ConnectionFailure` if
the connection cannot be made.
The `host` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a simple hostname. It can also be a list of hostnames or
URIs. Any port specified in the host string(s) will override
the `port` parameter. If multiple mongodb URIs containing
database or auth information are passed, the last database,
username, and password present will be used. For username and
passwords reserved characters like ':', '/', '+' and '@' must be
escaped following RFC 2396.
:Parameters:
- `host` (optional): hostname or IP address of the
instance to connect to, or a mongodb URI, or a list of
hostnames / mongodb URIs. If `host` is an IPv6 literal
it must be enclosed in '[' and ']' characters following
the RFC2732 URL syntax (e.g. '[::1]' for localhost)
- `port` (optional): port number on which to connect
- `max_pool_size` (optional): The maximum size limit for
the connection pool.
- `network_timeout` (optional): timeout (in seconds) to use
for socket operations - default is no timeout
- `document_class` (optional): default class to use for
documents returned from queries on this connection
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`Connection` will be timezone
aware (otherwise they will be naive)
Other optional parameters can be passed as keyword arguments:
- `safe`: Use getlasterror for each write operation?
- `j` or `journal`: Block until write operations have been committed
to the journal. Ignored if the server is running without journaling.
Implies safe=True.
- `w`: (integer or string) If this is a replica set write operations
won't return until they have been replicated to the specified
number or tagged set of servers.
Implies safe=True.
- `wtimeout`: Used in conjunction with `j` and/or `w`. Wait this many
milliseconds for journal acknowledgement and/or write replication.
Implies safe=True.
- `fsync`: Force the database to fsync all files before returning
When used with `j` the server awaits the next group commit before
returning.
Implies safe=True.
- `replicaSet`: The name of the replica set to connect to. The driver
will verify that the replica set it connects to matches this name.
Implies that the hosts specified are a seed list and the driver should
attempt to find all members of the set.
- `socketTimeoutMS`: How long a send or receive on a socket can take
before timing out.
- `connectTimeoutMS`: How long a connection can take to be opened
before timing out.
- `ssl`: If True, create the connection to the server using SSL.
- `read_preference`: The read preference for this connection.
See :class:`~pymongo.read_preferences.ReadPreference` for available
options.
- `auto_start_request`: If True (the default), each thread that
accesses this Connection has a socket allocated to it for the
thread's lifetime. This ensures consistent reads, even if you read
after an unsafe write.
- `use_greenlets` (optional): if ``True``, :meth:`start_request()`
will ensure that the current greenlet uses the same socket for all
operations until :meth:`end_request()`
- `slave_okay` or `slaveOk` (deprecated): Use `read_preference`
instead.
.. seealso:: :meth:`end_request`
.. versionchanged:: 2.3
Added support for failover between mongos seed list members.
.. versionchanged:: 2.2
Added `auto_start_request` option back. Added `use_greenlets`
option.
.. versionchanged:: 2.1
Support `w` = integer or string.
Added `ssl` option.
DEPRECATED slave_okay/slaveOk.
.. versionchanged:: 2.0
`slave_okay` is a pure keyword argument. Added support for safe,
and getlasterror options as keyword arguments.
.. versionchanged:: 1.11
Added `max_pool_size`. Completely removed previously deprecated
`pool_size`, `auto_start_request` and `timeout` parameters.
.. versionchanged:: 1.8
The `host` parameter can now be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition
to a simple hostname. It can also be a list of hostnames or
URIs.
.. versionadded:: 1.8
The `tz_aware` parameter.
.. versionadded:: 1.7
The `document_class` parameter.
.. versionadded:: 1.1
The `network_timeout` parameter.
.. mongodoc:: connections
"""
if host is None:
host = self.HOST
if isinstance(host, basestring):
host = [host]
if port is None:
port = self.PORT
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
seeds = set()
username = None
password = None
db = None
options = {}
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port)
seeds.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
db = res["database"] or db
options = res["options"]
else:
idx = entity.find("://")
raise InvalidURI("Invalid URI scheme: "
"%s" % (entity[:idx], ))
else:
seeds.update(uri_parser.split_hosts(entity, port))
if not seeds:
raise ConfigurationError("need to specify at least one host")
self.__nodes = seeds
self.__host = None
self.__port = None
self.__is_primary = False
self.__is_mongos = False
for option, value in kwargs.iteritems():
option, value = common.validate(option, value)
options[option] = value
self.__max_pool_size = common.validate_positive_integer(
'max_pool_size', max_pool_size)
self.__cursor_manager = CursorManager(self)
self.__repl = options.get('replicaset')
if len(seeds) == 1 and not self.__repl:
self.__direct = True
else:
self.__direct = False
self.__nodes = set()
if network_timeout is not None:
if (not isinstance(network_timeout, (int, float))
or network_timeout <= 0):
raise ConfigurationError("network_timeout must "
"be a positive integer")
self.__net_timeout = (network_timeout
or options.get('sockettimeoutms'))
self.__conn_timeout = options.get('connecttimeoutms')
self.__use_ssl = options.get('ssl', False)
if self.__use_ssl and not pool.have_ssl:
raise ConfigurationError("The ssl module is not available. If you "
"are using a python version previous to "
"2.6 you must install the ssl package "
"from PyPI.")
if options.get('use_greenlets', False):
if not pool.have_greenlet:
raise ConfigurationError(
"The greenlet module is not available. "
"Install the greenlet package from PyPI.")
self.pool_class = pool.GreenletPool
else:
self.pool_class = pool.Pool
self.__pool = self.pool_class(None, self.__max_pool_size,
self.__net_timeout, self.__conn_timeout,
self.__use_ssl)
self.__document_class = document_class
self.__tz_aware = common.validate_boolean('tz_aware', tz_aware)
self.__auto_start_request = options.get('auto_start_request', True)
# cache of existing indexes used by ensure_index ops
self.__index_cache = {}
self.__auth_credentials = {}
super(Connection, self).__init__(**options)
if self.slave_okay:
warnings.warn(
"slave_okay is deprecated. Please "
"use read_preference instead.", DeprecationWarning)
if _connect:
self.__find_node(seeds)
if db and username is None:
warnings.warn("must provide a username and password "
"to authenticate to %s" % (db, ))
if username:
db = db or "admin"
if not self[db].authenticate(username, password):
raise ConfigurationError("authentication failed")
def __init__(self,
host=None,
port=None,
document_class=dict,
tz_aware=False,
connect=True,
**kwargs):
"""Client for a MongoDB instance, a replica set, or a set of mongoses.
The client object is thread-safe and has connection-pooling built in.
If an operation fails because of a network error,
:class:`~pymongo.errors.ConnectionFailure` is raised and the client
reconnects in the background. Application code should handle this
exception (recognizing that the operation failed) and then continue to
execute.
The `host` parameter can be a full `mongodb URI
<http://dochub.mongodb.org/core/connections>`_, in addition to
a simple hostname. It can also be a list of hostnames or
URIs. Any port specified in the host string(s) will override
the `port` parameter. If multiple mongodb URIs containing
database or auth information are passed, the last database,
username, and password present will be used. For username and
passwords reserved characters like ':', '/', '+' and '@' must be
escaped following RFC 2396.
.. warning:: When using PyMongo in a multiprocessing context, please
read :ref:`multiprocessing` first.
:Parameters:
- `host` (optional): hostname or IP address of the
instance to connect to, or a mongodb URI, or a list of
hostnames / mongodb URIs. If `host` is an IPv6 literal
it must be enclosed in '[' and ']' characters following
the RFC2732 URL syntax (e.g. '[::1]' for localhost)
- `port` (optional): port number on which to connect
- `document_class` (optional): default class to use for
documents returned from queries on this client
- `tz_aware` (optional): if ``True``,
:class:`~datetime.datetime` instances returned as values
in a document by this :class:`MongoClient` will be timezone
aware (otherwise they will be naive)
- `connect` (optional): if ``True`` (the default), immediately
begin connecting to MongoDB in the background. Otherwise connect
on the first operation.
| **Other optional parameters can be passed as keyword arguments:**
- `maxPoolSize` (optional): The maximum number of connections
that the pool will open simultaneously. If this is set, operations
will block if there are `maxPoolSize` outstanding connections
from the pool. Defaults to 100. Cannot be 0.
- `socketTimeoutMS`: (integer or None) Controls how long (in
milliseconds) the driver will wait for a response after sending an
ordinary (non-monitoring) database operation before concluding that
a network error has occurred. Defaults to ``None`` (no timeout).
- `connectTimeoutMS`: (integer or None) Controls how long (in
milliseconds) the driver will wait during server monitoring when
connecting a new socket to a server before concluding the server
is unavailable. Defaults to ``20000`` (20 seconds).
- `serverSelectionTimeoutMS`: (integer) Controls how long (in
milliseconds) the driver will wait to find an available,
appropriate server to carry out a database operation; while it is
waiting, multiple server monitoring operations may be carried out,
each controlled by `connectTimeoutMS`. Defaults to ``30000`` (30
seconds).
- `waitQueueTimeoutMS`: (integer or None) How long (in milliseconds)
a thread will wait for a socket from the pool if the pool has no
free sockets. Defaults to ``None`` (no timeout).
- `waitQueueMultiple`: (integer or None) Multiplied by maxPoolSize
to give the number of threads allowed to wait for a socket at one
time. Defaults to ``None`` (no limit).
- `socketKeepAlive`: (boolean) Whether to send periodic keep-alive
packets on connected sockets. Defaults to ``False`` (do not send
keep-alive packets).
| **Write Concern options:**
| (Only set if passed. No default values.)
- `w`: (integer or string) If this is a replica set, write operations
will block until they have been replicated to the specified number
or tagged set of servers. `w=<int>` always includes the replica set
primary (e.g. w=3 means write to the primary and wait until
replicated to **two** secondaries). Passing w=0 **disables write
acknowledgement** and all other write concern options.
- `wtimeout`: (integer) Used in conjunction with `w`. Specify a value
in milliseconds to control how long to wait for write propagation
to complete. If replication does not complete in the given
timeframe, a timeout exception is raised.
- `j`: If ``True`` block until write operations have been committed
to the journal. Cannot be used in combination with `fsync`. Prior
to MongoDB 2.6 this option was ignored if the server was running
without journaling. Starting with MongoDB 2.6 write operations will
fail with an exception if this option is used when the server is
running without journaling.
- `fsync`: If ``True`` and the server is running without journaling,
blocks until the server has synced all data files to disk. If the
server is running with journaling, this acts the same as the `j`
option, blocking until write operations have been committed to the
journal. Cannot be used in combination with `j`.
| **Replica set keyword arguments for connecting with a replica set
- either directly or via a mongos:**
- `replicaSet`: (string or None) The name of the replica set to
connect to. The driver will verify that all servers it connects to
match this name. Implies that the hosts specified are a seed list
and the driver should attempt to find all members of the set.
Defaults to ``None``.
- `read_preference`: The read preference for this client. If
connecting directly to a secondary then a read preference mode
*other* than PRIMARY is required - otherwise all queries will throw
:class:`~pymongo.errors.AutoReconnect` "not master".
See :class:`~pymongo.read_preferences.ReadPreference` for all
available read preference options. Defaults to ``PRIMARY``.
| **SSL configuration:**
- `ssl`: If ``True``, create the connection to the server using SSL.
Defaults to ``False``.
- `ssl_keyfile`: The private keyfile used to identify the local
connection against mongod. If included with the ``certfile`` then
only the ``ssl_certfile`` is needed. Implies ``ssl=True``.
Defaults to ``None``.
- `ssl_certfile`: The certificate file used to identify the local
connection against mongod. Implies ``ssl=True``. Defaults to
``None``.
- `ssl_cert_reqs`: Specifies whether a certificate is required from
the other side of the connection, and whether it will be validated
if provided. It must be one of the three values ``ssl.CERT_NONE``
(certificates ignored), ``ssl.CERT_OPTIONAL``
(not required, but validated if provided), or ``ssl.CERT_REQUIRED``
(required and validated). If the value of this parameter is not
``ssl.CERT_NONE`` and a value is not provided for ``ssl_ca_certs``
PyMongo will attempt to load system provided CA certificates.
If the python version in use does not support loading system CA
certificates then the ``ssl_ca_certs`` parameter must point
to a file of CA certificates. Implies ``ssl=True``. Defaults to
``ssl.CERT_REQUIRED`` if not provided and ``ssl=True``.
- `ssl_ca_certs`: The ca_certs file contains a set of concatenated
"certification authority" certificates, which are used to validate
certificates passed from the other end of the connection.
Implies ``ssl=True``. Defaults to ``None``.
- `ssl_match_hostname`: If ``True`` (the default), and
`ssl_cert_reqs` is not ``ssl.CERT_NONE``, enables hostname
verification using the :func:`~ssl.match_hostname` function from
python's :mod:`~ssl` module. Think very carefully before setting
this to ``False`` as that could make your application vulnerable to
man-in-the-middle attacks.
.. mongodoc:: connections
.. versionchanged:: 3.0
:class:`~pymongo.mongo_client.MongoClient` is now the one and only
client class for a standalone server, mongos, or replica set.
It includes the functionality that had been split into
:class:`~pymongo.mongo_client.MongoReplicaSetClient`: it can connect
to a replica set, discover all its members, and monitor the set for
stepdowns, elections, and reconfigs.
The :class:`~pymongo.mongo_client.MongoClient` constructor no
longer blocks while connecting to the server or servers, and it no
longer raises :class:`~pymongo.errors.ConnectionFailure` if they
are unavailable, nor :class:`~pymongo.errors.ConfigurationError`
if the user's credentials are wrong. Instead, the constructor
returns immediately and launches the connection process on
background threads.
Therefore the ``alive`` method is removed since it no longer
provides meaningful information; even if the client is disconnected,
it may discover a server in time to fulfill the next operation.
In PyMongo 2.x, :class:`~pymongo.MongoClient` accepted a list of
standalone MongoDB servers and used the first it could connect to::
MongoClient(['host1.com:27017', 'host2.com:27017'])
A list of multiple standalones is no longer supported; if multiple
servers are listed they must be members of the same replica set, or
mongoses in the same sharded cluster.
The behavior for a list of mongoses is changed from "high
availability" to "load balancing". Before, the client connected to
the lowest-latency mongos in the list, and used it until a network
error prompted it to re-evaluate all mongoses' latencies and
reconnect to one of them. In PyMongo 3, the client monitors its
network latency to all the mongoses continuously, and distributes
operations evenly among those with the lowest latency. See
:ref:`mongos-load-balancing` for more information.
The ``connect`` option is added.
The ``start_request``, ``in_request``, and ``end_request`` methods
are removed, as well as the ``auto_start_request`` option.
The ``copy_database`` method is removed, see the
:doc:`copy_database examples </examples/copydb>` for alternatives.
The :meth:`MongoClient.disconnect` method is removed; it was a
synonym for :meth:`~pymongo.MongoClient.close`.
:class:`~pymongo.mongo_client.MongoClient` no longer returns an
instance of :class:`~pymongo.database.Database` for attribute names
with leading underscores. You must use dict-style lookups instead::
client['__my_database__']
Not::
client.__my_database__
"""
if host is None:
host = self.HOST
if isinstance(host, string_type):
host = [host]
if port is None:
port = self.PORT
if not isinstance(port, int):
raise TypeError("port must be an instance of int")
seeds = set()
username = None
password = None
dbase = None
opts = {}
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port, warn=True)
seeds.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
dbase = res["database"] or dbase
opts = res["options"]
else:
idx = entity.find("://")
raise InvalidURI("Invalid URI scheme: "
"%s" % (entity[:idx], ))
else:
seeds.update(uri_parser.split_hosts(entity, port))
if not seeds:
raise ConfigurationError("need to specify at least one host")
# _pool_class, _monitor_class, and _condition_class are for deep
# customization of PyMongo, e.g. Motor.
pool_class = kwargs.pop('_pool_class', None)
monitor_class = kwargs.pop('_monitor_class', None)
condition_class = kwargs.pop('_condition_class', None)
keyword_opts = kwargs
keyword_opts['document_class'] = document_class
keyword_opts['tz_aware'] = tz_aware
keyword_opts['connect'] = connect
# Validate all keyword options.
keyword_opts = dict(
common.validate(k, v) for k, v in keyword_opts.items())
opts.update(keyword_opts)
self.__options = options = ClientOptions(username, password, dbase,
opts)
self.__default_database_name = dbase
self.__lock = threading.Lock()
self.__cursor_manager = CursorManager(self)
self.__kill_cursors_queue = []
# Cache of existing indexes used by ensure_index ops.
self.__index_cache = {}
super(MongoClient,
self).__init__(options.codec_options, options.read_preference,
options.write_concern)
self.__all_credentials = {}
creds = options.credentials
if creds:
self._cache_credentials(creds.source, creds)
self._topology_settings = TopologySettings(
seeds=seeds,
replica_set_name=options.replica_set_name,
pool_class=pool_class,
pool_options=options.pool_options,
monitor_class=monitor_class,
condition_class=condition_class,
local_threshold_ms=options.local_threshold_ms,
server_selection_timeout=options.server_selection_timeout)
self._topology = Topology(self._topology_settings)
if connect:
self._topology.open()
def target():
client = self_ref()
if client is None:
return False # Stop the executor.
MongoClient._process_kill_cursors_queue(client)
return True
executor = periodic_executor.PeriodicExecutor(
condition_class=self._topology_settings.condition_class,
interval=common.KILL_CURSOR_FREQUENCY,
min_interval=0,
target=target,
name="pymongo_kill_cursors_thread")
# We strongly reference the executor and it weakly references us via
# this closure. When the client is freed, stop the executor soon.
self_ref = weakref.ref(self, executor.close)
self._kill_cursors_executor = executor
executor.open()
def __init__(self,
host=None,
port=27017,
mincached=0,
maxcached=0,
maxconnections=0,
maxusage=0,
dbname=None,
slave_okay=False,
*args, **kwargs):
if isinstance(host, basestring):
host = [host]
else:
assert isinstance(host, list)
assert isinstance(port, int)
assert isinstance(mincached, int)
assert isinstance(maxcached, int)
assert isinstance(maxconnections, int)
assert isinstance(maxusage, int)
assert isinstance(dbname, (str, unicode, None.__class__))
assert isinstance(slave_okay, bool)
if mincached and maxcached:
assert mincached <= maxcached
if maxconnections:
assert maxconnections >= maxcached
assert maxconnections >= mincached
self._args, self._kwargs = args, kwargs
self._maxusage = maxusage
self._mincached = mincached
self._maxcached = maxcached
self._maxconnections = maxconnections
self._idle_cache = [] # the actual connections that can be used
self._condition = Condition()
self._kwargs['slave_okay'] = self._slave_okay = slave_okay
self._connections = 0
nodes = set()
username = None # TODO: username/password ignored for now
password = None
for entity in host:
if "://" in entity:
if entity.startswith("mongodb://"):
res = uri_parser.parse_uri(entity, port)
nodes.update(res["nodelist"])
username = res["username"] or username
password = res["password"] or password
dbname = res["database"] or dbname
else:
idx = entity.find("://")
raise ProgrammingError("Invalid URI scheme: "
"%s" % (entity[:idx],))
else:
nodes.update(uri_parser.split_hosts(entity, port))
if not nodes:
raise ProgrammingError("Need to specify at least one host")
self._nodes = nodes
self._dbname = dbname
# Establish an initial number of idle database connections:
idle = [self.connection() for i in range(mincached)]
while idle:
self.cache(idle.pop())
