def __init__(self, **kwargs):
config = Config(required={
"comparator": ("dbpath", "series_timeout", "cleanup_time")
})
self.comparator_config = config["comparator"]
self.comparator_config["dbpath"] = os.path.expanduser(
self.comparator_config["dbpath"])
dbpath_dirname = os.path.dirname(self.comparator_config["dbpath"])
try:
os.makedirs(dbpath_dirname, 0o700)
except OSError:
pass
super(Comparator, self).__init__(**kwargs)
# store dir doesn't exist, stop comparator
if not os.path.isdir(dbpath_dirname):
raise Exception(
'store dir does not exist, stop comparator, path:',
self.comparator_config["dbpath"])
# store directory exists, but it has no rights to write
if not os.access(dbpath_dirname, os.W_OK):
raise Exception(
'stop comparator, remember to set the rights'
' for user, which runs comparator, path:',
self.comparator_config["dbpath"])
self.state = ComparatorState(
int(self.comparator_config["cleanup_time"]))
self.db = ComparatorDataWrapper(self.comparator_config["dbpath"])
def __init__(self, **kwargs):
config = Config(required={"aggregator": ("dbpath", "time_tolerance")})
self.aggregator_config = config["aggregator"]
self.aggregator_config["dbpath"] = os.path.expanduser(
self.aggregator_config["dbpath"])
dbpath_dirname = os.path.dirname(self.aggregator_config["dbpath"])
try:
os.makedirs(dbpath_dirname, 0700)
except OSError:
pass
super(Aggregator, self).__init__(**kwargs)
# store dir doesn't exist, stop aggregator
if not os.path.isdir(dbpath_dirname):
raise Exception(
'store dir does not exist, stop aggregator, path:',
self.aggregator_config["dbpath"])
# store directory exists, but it has no rights to write
if not os.access(dbpath_dirname, os.W_OK):
raise Exception(
'stop aggregator, remember to set the rights'
' for user, which runs aggregator, path:',
self.aggregator_config["dbpath"])
self.db = AggregatorDataWrapper(
self.aggregator_config["dbpath"],
int(self.aggregator_config["time_tolerance"]))
self.timeout_id = None # id of the 'tick' timeout that executes source cleanup
def __make_converter__list_of_template_loc():
def convert_template_loc(loc):
if loc.startswith('@'):
try:
loc_parts = loc[1:].split(':', 1)
try:
package_name, package_subdir_path = loc_parts
except ValueError:
raise ValueError('the colon is missing')
if isabs(package_subdir_path):
raise ValueError(
'the subdirectory part is not a relative path')
except ValueError as exc:
raise ValueError(
'{!a} is not a valid package-based template '
'location ({})'.format(loc, ascii_str(exc)))
converted_loc = package_name, package_subdir_path
else:
converted_loc = expanduser(loc)
if not isabs(converted_loc):
raise ValueError(
'{!a} is not a valid absolute-path-based template '
'location (is not an absolute path)'.format(loc))
return converted_loc
return Config.make_list_converter(convert_template_loc,
name='list_of_template_loc')
def __init__(self, config=None):
"""
Args:
config: dict containing: mongohost, mongoport, mongodb,
count_try_connection, time_sleep_between_try_connect, uri
"""
if config is None:
config = Config(required={"archiveraw": ("mongohost",
"mongoport",
"mongodb",
"count_try_connection",
"time_sleep_between_try_connect",
"uri")})
self.config = config["archiveraw"]
else:
self.config = config
self.host = self.config['mongohost']
self.port = int(self.config['mongoport'])
self._currdb = self.config['mongodb']
self.uri = self.config["uri"]
self.connection = None
self._currcoll = None
self.conn_gridfs = None
self.time_sleep_between_try_connect = int(self.config['time_sleep_between_try_connect'])
self.count_try_connection = int(self.config['count_try_connection'])
self.indexes_store = []
self.backup_msg = '.backup_msg'
self.backup_msg_data = None
self.backup_msg_headers = None
def __init__(self, **kwargs):
LOGGER.info("Recorder Start")
config = Config(required={"recorder": ("uri", )})
self.config = config["recorder"]
self.record_dict = None
self.records = None
self.routing_key = None
self.session_db = self._setup_db()
self.dict_map_fun = {
"event.filtered": (RecordDict.from_json, self.new_event),
"bl-new.filtered": (BLRecordDict.from_json, self.blacklist_new),
"bl-change.filtered":
(BLRecordDict.from_json, self.blacklist_change),
"bl-delist.filtered":
(BLRecordDict.from_json, self.blacklist_delist),
"bl-expire.filtered":
(BLRecordDict.from_json, self.blacklist_expire),
"bl-update.filtered":
(BLRecordDict.from_json, self.blacklist_update),
"suppressed.filtered": (RecordDict.from_json,
self.suppressed_update),
}
# keys in each of the tuples being values of `dict_map_fun`
self.FROM_JSON = 0
self.HANDLE_EVENT = 1
super(Recorder, self).__init__(**kwargs)
def from_predefined(cls,
default_template_name=None,
autoescape=None,
settings=None,
**kwargs):
config = Config.section(
cls.config_spec,
settings=settings,
custom_converters={
'list_of_template_loc':
cls.__make_converter__list_of_template_loc(),
})
env = Environment(loader=PackageOrFileSystemLoader(
locations=config['template_locations'],
encoding=config['template_encoding'],
fs_followlinks=config['follow_symlinks']),
autoescape=(autoescape if autoescape is not None else
DEFAULT_AUTOESCAPE))
jinja_extensions = sorted(
set(config['jinja_extensions'] +
list(kwargs.pop('jinja_extensions', []))))
return cls(jinja_env=env,
jinja_extensions=jinja_extensions,
default_template_name=default_template_name,
**kwargs)
def __init__(self, **kwargs):
LOGGER.info("Recorder Start")
config = Config(required={"recorder": ("uri", "echo")})
self.config = config["recorder"]
self.rows = None
self.record_dict = None
self.source = None
self.dir_name = None
self.wait_timeout = int(self.config.get("wait_timeout", 28800))
engine = create_engine(self.config["uri"], echo=bool((int(self.config["echo"]))))
self.session_db = N6DataBackendAPI.configure_db_session(engine)
self.set_session_wait_timeout()
self.records = None
self.routing_key = None
self.dict_map_fun = {
"event.filtered": (RecordDict.from_json, self.new_event),
"bl-new.filtered": (BLRecordDict.from_json, self.blacklist_new),
"bl-change.filtered": (BLRecordDict.from_json, self.blacklist_change),
"bl-delist.filtered": (BLRecordDict.from_json, self.blacklist_delist),
"bl-expire.filtered": (BLRecordDict.from_json, self.blacklist_expire),
"bl-update.filtered": (BLRecordDict.from_json, self.blacklist_update),
"suppressed.filtered": (RecordDict.from_json, self.suppressed_update),
}
# keys in each of the tuples being values of `dict_map_fun`
self.FROM_JSON = 0
self.HANDLE_EVENT = 1
super(Recorder, self).__init__(**kwargs)
def get_amqp_connection_params_dict(rabbitmq_config_section='rabbitmq'):
"""
Get the AMQP connection parameters (as a dict) from config.
Returns:
A dict that can be used as **kwargs for pika.ConnectionParameters.
"""
# Config is imported here to avoid circular dependency
from n6lib.config import Config
config_spec = RABBITMQ_CONFIG_SPEC_PATTERN.format(
rabbitmq_config_section=rabbitmq_config_section)
queue_conf = Config.section(config_spec)
params_dict = dict(
host=queue_conf["host"],
port=queue_conf["port"],
ssl=queue_conf["ssl"],
ssl_options={},
heartbeat_interval=queue_conf['heartbeat_interval'],
)
if params_dict['ssl']:
params_dict['credentials'] = pika.credentials.ExternalCredentials()
params_dict['ssl_options'].update(
ca_certs=os.path.expanduser(queue_conf["ssl_ca_certs"]),
certfile=os.path.expanduser(queue_conf["ssl_certfile"]),
keyfile=os.path.expanduser(queue_conf["ssl_keyfile"]),
cert_reqs=ssl.CERT_REQUIRED,
)
return params_dict
def get_connection_params_dict(cls):
params_dict = super(BaseParser, cls).get_connection_params_dict()
config = Config(required={
cls.rabbitmq_config_section: ("heartbeat_interval_parsers", )
})
queue_conf = config[cls.rabbitmq_config_section]
params_dict['heartbeat_interval'] = int(
queue_conf['heartbeat_interval_parsers'])
return params_dict
def __init__(self, **fmt_args):
self.config = Config.section(self.general_config_spec)
if self.is_config_spec_or_group_declared():
self.config.update(self.get_config_section(**fmt_args))
self.parameters = self.config.copy()
config_ttl = self.parameters.get(self.config_ttl_opt_name)
if config_ttl:
self.parameters[self.parameter_ttl_opt_name] = config_ttl
del self.parameters[self.config_ttl_opt_name]
if self.unused_options:
self.parameters.update({key: None for key in self.unused_options})
if self.default_none_options:
self.parameters.update({key: None for key in self.default_none_options
if key not in self.parameters})
def __init__(self, **kwargs):
self.gi_asn = None
self.gi_cc = None
self._resolver = None
config = Config(
required={
"enrich": ("dnshost", "dnsport", "geoippath",
"asndatabasefilename", "citydatabasefilename")
})
self._enrich_config = config["enrich"]
self.excluded_ips = self._get_excluded_ips()
self._setup_geodb()
self._setup_dnsresolver(self._enrich_config["dnshost"],
int(self._enrich_config["dnsport"]))
super(Enricher, self).__init__(**kwargs)
def get_amqp_connection_params_dict(rabbitmq_config_section='rabbitmq'):
"""
Get the AMQP connection parameters (as a dict) from config.
Returns:
A dict that can be used as **kwargs for pika.ConnectionParameters.
"""
# Config is imported here to avoid circular dependency
from n6lib.config import Config
config_spec = RABBITMQ_CONFIG_SPEC_PATTERN.format(
rabbitmq_config_section=rabbitmq_config_section)
queue_conf = Config.section(config_spec)
return get_amqp_connection_params_dict_from_args(
host=queue_conf["host"],
port=queue_conf["port"],
heartbeat_interval=queue_conf["heartbeat_interval"],
ssl=queue_conf["ssl"],
ca_certs=queue_conf.get("ssl_ca_certs", None),
certfile=queue_conf.get("ssl_certfile", None),
keyfile=queue_conf.get("ssl_keyfile", None))
def get_pipeline_binding_states(pipeline_group,
pipeline_name,
pipeline_config_section='pipeline'):
"""
Get the list of "binding states" for the component, or its group,
from the pipeline config. Pipeline config for an individual
component has a priority over group's config.
Args:
`pipeline_group`:
A group which the component is bound to.
`pipeline_name`:
Name of the component in the pipeline config format,
which is, by default, a lowercase component's
class' name.
`pipeline_config_section`:
Name of the pipeline config section, "pipeline"
by default.
Returns:
The list of "binding states" for the component, or None,
if no config option could be found.
"""
from n6lib.config import Config
config_spec = PIPELINE_CONFIG_SPEC_PATTERN.format(
pipeline_config_section=pipeline_config_section)
pipeline_conf = Config.section(config_spec,
default_converter='list_of_str')
try:
return pipeline_conf[pipeline_name]
except KeyError:
pass
try:
return pipeline_conf[pipeline_group]
except KeyError:
return None
class TestBaseParser(unittest.TestCase):
def setUp(self):
self.mock = Mock(__class__=BaseParser, allow_empty_results=False)
self.meth = MethodProxy(BaseParser, self.mock)
def _asserts_of_proper__new__instance_adjustment(self, instance):
# BaseQueued.__new__() ensures that
self.assertIsNot(instance.input_queue, BaseParser.input_queue)
def _asserts_of_proper_preinit_hook_instance_adjustment(
self, instance, binding_key):
# for classes with `default_binding_key`
# BaseParser.preinit_hook() ensures that
self.assertEqual(
instance.input_queue, {
'exchange': 'raw',
'exchange_type': 'topic',
'queue_name': binding_key,
'binding_keys': [binding_key],
})
self.assertEqual(BaseParser.input_queue, {
'exchange': 'raw',
'exchange_type': 'topic',
})
def _basic_init_related_asserts(self, instance, subclass, super_mock,
super_cls_mock, expected_config,
expected_config_full):
# assert that an instance of the proper type has been returned
self.assertIsInstance(instance, subclass)
# assert that super used properly
super_mock.assert_called_once_with(BaseParser, instance)
super_cls_mock.__init__.assert_called_once_with(a=sentinel.a,
bb=sentinel.bb)
# assert that configuration stuff has been obtained properly
self.assertEqual(instance.config, expected_config)
self.assertIsInstance(instance.config, ConfigSection)
self.assertEqual(instance.config_full, expected_config_full)
self.assertIsInstance(instance.config_full, Config)
def test_basics(self):
self.assertTrue(issubclass(BaseParser, QueuedBase))
self.assertTrue(hasattr(BaseParser, 'default_binding_key'))
self.assertTrue(hasattr(BaseParser, 'config_spec_pattern'))
self.assertTrue(hasattr(BaseParser, 'constant_items'))
self.assertTrue(hasattr(BaseParser, 'record_dict_class'))
self.assertTrue(hasattr(BaseParser, 'event_type'))
def test_config_spec_pattern(self):
config_spec = BaseParser.config_spec_pattern.format(
parser_class_name='example_foo')
config_spec_parsed = parse_config_spec(config_spec)
prefetch_count_opt_spec = config_spec_parsed.get_opt_spec(
'example_foo.prefetch_count')
self.assertEqual(prefetch_count_opt_spec.name, 'prefetch_count')
self.assertEqual(prefetch_count_opt_spec.converter_spec, 'int')
def test_initialization_without_default_binding_key(self):
class SomeParser(BaseParser):
pass # no `default_binding_key` defined => it's an abstract class
with self.assertRaises(NotImplementedError):
SomeParser()
unready_instance = SomeParser.__new__(SomeParser)
self._asserts_of_proper__new__instance_adjustment(unready_instance)
# for classes without `default_binding_key`
# `queue_name` and `binding_keys` items are *not* added...
self.assertEqual(unready_instance.input_queue, BaseParser.input_queue)
self.assertEqual(BaseParser.input_queue, {
'exchange': 'raw',
'exchange_type': 'topic',
})
@foreach(
param(
mocked_conf_from_files={},
expected_config=ConfigSection('SomeParser', {'prefetch_count': 1}),
expected_config_full=Config.make(
{'SomeParser': {
'prefetch_count': 1
}}),
),
param(
mocked_conf_from_files={
'SomeParser': {
'prefetch_count': '42'
},
'another_section': {
'another_opt': '123.456'
},
},
expected_config=ConfigSection('SomeParser',
{'prefetch_count': 42}),
expected_config_full=Config.make(
{'SomeParser': {
'prefetch_count': 42
}}),
),
param(
custom_config_spec_pattern=concat_reducing_indent(
BaseParser.config_spec_pattern,
'''
some_opt = [-3, null] :: json
[another_section]
another_opt :: float
yet_another_opt = Foo Bar Spam Ham
''',
),
mocked_conf_from_files={
'SomeParser': {
'prefetch_count': '42'
},
'another_section': {
'another_opt': '123.456'
},
},
expected_config=ConfigSection('SomeParser', {
'prefetch_count': 42,
'some_opt': [-3, None],
}),
expected_config_full=Config.make({
'SomeParser': {
'prefetch_count': 42,
'some_opt': [-3, None],
},
'another_section': {
'another_opt': 123.456,
'yet_another_opt': 'Foo Bar Spam Ham',
},
}),
),
)
@foreach(
param(binding_key='foo.bar'),
param(binding_key='foo.bar.33'),
)
def test_initialization_with_default_binding_key(
self,
binding_key,
mocked_conf_from_files,
expected_config,
expected_config_full,
custom_config_spec_pattern=None):
class SomeParser(BaseParser):
default_binding_key = binding_key # => it's a concrete class
if custom_config_spec_pattern is not None:
SomeParser.config_spec_pattern = custom_config_spec_pattern
unready_instance = SomeParser.__new__(SomeParser)
self._asserts_of_proper__new__instance_adjustment(unready_instance)
self._asserts_of_proper_preinit_hook_instance_adjustment(
unready_instance, binding_key)
super_cls_mock = SimpleNamespace(__init__=Mock())
with patch_always('n6.parsers.generic.super',
return_value=super_cls_mock) as super_mock, \
patch('n6.parsers.generic.Config._load_n6_config_files',
return_value=mocked_conf_from_files):
# instantiation
instance = SomeParser(a=sentinel.a, bb=sentinel.bb)
self._asserts_of_proper__new__instance_adjustment(instance)
self._asserts_of_proper_preinit_hook_instance_adjustment(
instance, binding_key)
self._basic_init_related_asserts(instance, SomeParser, super_mock,
super_cls_mock, expected_config,
expected_config_full)
def test__make_binding_keys(self):
self.mock.default_binding_key = 'fooo.barr'
binding_keys = self.meth.make_binding_keys()
self.assertEqual(binding_keys, ['fooo.barr'])
self.assertEqual(self.mock.mock_calls, [])
def test__make_binding_keys_with_raw_format_version_tag(self):
self.mock.default_binding_key = 'fooo.barr.33'
binding_keys = self.meth.make_binding_keys()
self.assertEqual(binding_keys, ['fooo.barr.33'])
self.assertEqual(self.mock.mock_calls, [])
def test__get_script_init_kwargs(self):
self.assertIsInstance(
vars(BaseParser)['get_script_init_kwargs'], classmethod)
init_kwargs = BaseParser.get_script_init_kwargs.__func__(self.mock)
self.assertEqual(init_kwargs, {})
self.assertEqual(self.mock.mock_calls, [])
def test__run_handling__interrupted(self):
self.mock.configure_mock(**{'run.side_effect': KeyboardInterrupt})
self.meth.run_handling()
self.mock.run.assert_called_once_with()
self.mock.stop.assert_called_once_with()
def test__run_handling__not_interrupted(self):
self.meth.run_handling()
self.mock.run.assert_called_once_with()
self.assertEqual(self.mock.stop.mock_calls, [])
@patch('n6.parsers.generic.FilePagedSequence')
def test__input_callback(self, FilePagedSequence_mock):
FilePagedSequence_mock.return_value = MagicMock()
FilePagedSequence_mock.return_value.__enter__.return_value = sentinel.working_seq
data = MagicMock(**{'get.return_value': sentinel.rid})
self.mock.configure_mock(
**{
'_fix_body.return_value':
sentinel.body,
'prepare_data.return_value':
data,
'setting_error_event_info':
MagicMock(),
'get_output_rk.return_value':
sentinel.output_rk,
'get_output_bodies.return_value':
[sentinel.output_body1, sentinel.output_body2],
})
self.meth.input_callback(sentinel.routing_key, sentinel.body,
sentinel.properties)
self.assertEqual(self.mock.mock_calls, [
call._fix_body(sentinel.body),
call.prepare_data(sentinel.routing_key, sentinel.body,
sentinel.properties),
call.prepare_data().get('properties.message_id'),
call.setting_error_event_info(sentinel.rid),
call.setting_error_event_info().__enter__(),
call.get_output_rk(data),
call.get_output_bodies(data, sentinel.working_seq),
call.publish_output(routing_key=sentinel.output_rk,
body=sentinel.output_body1),
call.publish_output(routing_key=sentinel.output_rk,
body=sentinel.output_body2),
call.setting_error_event_info().__exit__(None, None, None),
])
self.assertEqual(FilePagedSequence_mock.mock_calls, [
call(page_size=1000),
call().__enter__(),
call().__exit__(None, None, None),
])
def test__prepare_data(self):
data = self.meth.prepare_data(routing_key='ham.spam',
body=sentinel.body,
properties=SimpleNamespace(
foo=sentinel.foo,
bar=sentinel.bar,
timestamp=1389348840,
headers={'a': sentinel.a}))
self.assertEqual(
data, {
'a': sentinel.a,
'properties.foo': sentinel.foo,
'properties.bar': sentinel.bar,
'source': 'ham.spam',
'properties.timestamp': '2014-01-10 10:14:00',
'raw_format_version_tag': None,
'raw': sentinel.body,
})
def test__prepare_data__rk__with_raw_format_version_tag(self):
data = self.meth.prepare_data(routing_key='ham.spam.33',
body=sentinel.body,
properties=SimpleNamespace(
foo=sentinel.foo,
bar=sentinel.bar,
timestamp=1389348840,
headers={'a': sentinel.a}))
self.assertEqual(
data, {
'a': sentinel.a,
'properties.foo': sentinel.foo,
'properties.bar': sentinel.bar,
'source': 'ham.spam',
'properties.timestamp': '2014-01-10 10:14:00',
'raw_format_version_tag': '33',
'raw': sentinel.body,
})
def test__get_output_rk(self):
self.mock.configure_mock(**{
'event_type': 'foobar',
})
data = {'source': 'ham.spam'}
output_rk = self.meth.get_output_rk(data)
self.assertEqual(output_rk, 'foobar.parsed.ham.spam')
def test__get_output_bodies(self):
parsed = [
MagicMock(
**{
'__class__':
RecordDict,
'used_as_context_manager':
True,
'get_ready_json.return_value':
getattr(sentinel, 'output_body{}'.format(i))
}) for i in (1, 2)
]
self.mock.configure_mock(
**{
'parse.return_value':
parsed,
'get_output_message_id.side_effect': [
sentinel.msg_A,
sentinel.msg_B,
],
'setting_error_event_info':
MagicMock(),
'postprocess_parsed.side_effect': (
lambda data, parsed, total, item_no: parsed),
})
seq_mock = FilePagedSequence._instance_mock()
output_bodies = self.meth.get_output_bodies(sentinel.data, seq_mock)
self.assertIs(output_bodies, seq_mock)
self.assertEqual(seq_mock._list, [
sentinel.output_body1,
sentinel.output_body2,
])
self.assertEqual(parsed[0].mock_calls, [
call.__setitem__('id', sentinel.msg_A),
call.get_ready_json(),
])
self.assertEqual(parsed[1].mock_calls, [
call.__setitem__('id', sentinel.msg_B),
call.get_ready_json(),
])
self.assertEqual(self.mock.mock_calls, [
call.parse(sentinel.data),
call.get_output_message_id(parsed[0]),
call.delete_too_long_address(parsed[0]),
call.get_output_message_id(parsed[1]),
call.delete_too_long_address(parsed[1]),
call.setting_error_event_info(parsed[0]),
call.setting_error_event_info().__enter__(),
call.postprocess_parsed(sentinel.data, parsed[0], 2, item_no=1),
call.setting_error_event_info().__exit__(None, None, None),
call.setting_error_event_info(parsed[1]),
call.setting_error_event_info().__enter__(),
call.postprocess_parsed(sentinel.data, parsed[1], 2, item_no=2),
call.setting_error_event_info().__exit__(None, None, None),
])
def test__get_output_bodies__record_dict_not_used_as_context_manager(self):
parsed = [
MagicMock(**{
'__class__': RecordDict,
'used_as_context_manager': False
}) for i in (1, 2)
]
self.mock.configure_mock(**{'parse.return_value': parsed})
with self.assertRaises(AssertionError):
self.meth.get_output_bodies(sentinel.data,
FilePagedSequence._instance_mock())
self.assertEqual(self.mock.method_calls, [
call.parse(sentinel.data),
])
def test__get_output_bodies__parse_yielded_no_items(self):
self.mock.configure_mock(**{'parse.return_value': iter([])})
with self.assertRaises(ValueError):
self.meth.get_output_bodies(sentinel.data,
FilePagedSequence._instance_mock())
self.assertEqual(self.mock.method_calls, [
call.parse(sentinel.data),
])
def test__get_output_bodies__parse_yielded_no_items__allow_empty_results(
self):
self.mock.configure_mock(**{
'parse.return_value': iter([]),
'allow_empty_results': True
})
seq_mock = FilePagedSequence._instance_mock()
output_bodies = self.meth.get_output_bodies(sentinel.data, seq_mock)
self.assertIs(output_bodies, seq_mock)
self.assertEqual(seq_mock._list, []) # just empty
self.assertEqual(self.mock.mock_calls, [
call.parse(sentinel.data),
])
def test__delete_too_long_address__address_is_ok(self):
parsed = RecordDict()
parsed['address'] = [{'ip': i + 1} for i in xrange(MAX_IPS_IN_ADDRESS)]
expected = RecordDict()
expected['address'] = [{
'ip': i + 1
} for i in xrange(MAX_IPS_IN_ADDRESS)]
self.meth.delete_too_long_address(parsed)
self.assertEqual(parsed, expected)
def test__delete_too_long_address__address_is_too_long(self):
ips = MAX_IPS_IN_ADDRESS + 1
parsed = RecordDict()
parsed['id'] = '0123456789abcdef0123456789abcdef'
parsed['address'] = [{'ip': i + 1} for i in xrange(ips)]
expected = RecordDict()
expected['id'] = '0123456789abcdef0123456789abcdef'
self.meth.delete_too_long_address(parsed)
self.assertEqual(parsed, expected)
def test__delete_too_long_address__address_is_empty(self):
parsed = RecordDict()
parsed.update({'source': 'foo.bar'})
expected = RecordDict()
expected.update({'source': 'foo.bar'})
self.meth.delete_too_long_address(parsed)
self.assertEqual(parsed, expected)
def test__get_output_message_id(self):
inputs_and_resultant_hash_bases = [
# basics
({
'source': 'foo.bar'
}, 'source,foo.bar'),
({
u'source': u'foo.bar'
}, 'source,foo.bar'),
# proper sorting of multiple values
({
'key1': 2,
'key2': ['value2', 'value3', 'value1']
}, 'key1,2\nkey2,value1,value2,value3'),
# ...and of keys + proper encoding of unicode keys/values
({
u'key2': [u'value3', u'value1', u'value2'],
u'key1': 2L
}, 'key1,2\nkey2,value1,value2,value3'),
# ...as well as proper int/long normalization/representation
({
u'key2': [30, 10, 20L],
u'key1': 9000111222333444555666777888999000L
}, 'key1,9000111222333444555666777888999000\nkey2,10,20,30'),
# non-ascii values
({
'target': 'zażółć',
u'client': [u'jaźń', u'gęślą']
}, 'client,gęślą,jaźń\ntarget,zażółć'),
({
u'target': u'zażółć',
'client': ['jaźń', 'gęślą']
}, 'client,gęślą,jaźń\ntarget,zażółć'),
# subdicts
({
'dip': u'3.3.3.3',
u'address': [{
'ip': '255.255.255.0'
}, {
'ip': '127.0.0.1'
}]
},
"address,{'ip': '127.0.0.1'},{'ip': '255.255.255.0'}\ndip,3.3.3.3"
),
# non-ascii subdict keys/values
({
u'key2': [{
'ką2': 'vą2'
}, {
'ką1': 'vą1'
}],
'key1': {
'ką': 'vą'
}
}, "key1,{'k\\xc4\\x85': 'v\\xc4\\x85'}\n" +
"key2,{'k\\xc4\\x851': 'v\\xc4\\x851'},{'k\\xc4\\x852': 'v\\xc4\\x852'}"
),
# proper encoding of unicode keys/values + proper sorting of whole subdicts
({
'key1': {
u'ką': u'vą'
},
'key2': [{
u'ką2': 'vą2'
}, {
'ką1': u'vą1'
}]
}, "key1,{'k\\xc4\\x85': 'v\\xc4\\x85'}\n" +
"key2,{'k\\xc4\\x851': 'v\\xc4\\x851'},{'k\\xc4\\x852': 'v\\xc4\\x852'}"
),
# ...as well as proper int/long normalization/representation
({
'key1': {
u'k': 2L
},
'key2': [{
'k2': 2L
}, {
u'k1': 1
}]
}, "key1,{'k': 2}\nkey2,{'k1': 1},{'k2': 2}"),
({
u'key2': [{
'k2': 2
}, {
'k1': 1
}],
'key1': {
'k': 3
}
}, "key1,{'k': 3}\nkey2,{'k1': 1},{'k2': 2}"),
({
u'key2': [{
'k2': 2L
}, {
'k1': 1L
}],
'key1': {
'k': 9000111222333444555666777888999000L
}
},
"key1,{'k': 9000111222333444555666777888999000}\nkey2,{'k1': 1},{'k2': 2}"
),
# proper sorting of multiple items in subdicts
({
'key1': {
'c': 2,
u'a': 3L,
u'b': 1L
},
'key2': [{
'c': 2,
u'a': 3L,
u'b': 1L
}, {
'd': 3,
u'a': 2L,
u'b': 1L
}]
}, "key1,{'a': 3, 'b': 1, 'c': 2}\n" +
"key2,{'a': 2, 'b': 1, 'd': 3},{'a': 3, 'b': 1, 'c': 2}"),
]
class _RecordDict(RecordDict):
adjust_key1 = adjust_key2 = None
optional_keys = RecordDict.optional_keys | {'key1', 'key2'}
parser = BaseParser.__new__(BaseParser)
for input_dict, expected_base in inputs_and_resultant_hash_bases:
record_dict = _RecordDict(input_dict)
expected_result = hashlib.md5(expected_base).hexdigest()
result = parser.get_output_message_id(record_dict)
self.assertIsInstance(result, str)
self.assertEqual(result, expected_result)
def test__get_output_message_id__errors(self):
inputs_and_exc_classes = [
# bad subdict key type
(
{
'key1': {
32: 2
}
},
TypeError,
),
(
{
'key1': [{
32: 2
}]
},
TypeError,
),
# bad subdict value type
(
{
'key1': {
'k': 2.3
}
},
TypeError,
),
(
{
'key1': [{
'k': 2.3
}]
},
TypeError,
),
(
{
'key1': {
'k': {
'k': 2
}
}
}, # nesting is illegal
TypeError,
),
(
{
'key1': [{
'k': {
'k': 2
}
}]
}, # nesting is illegal
TypeError,
),
# bad value type
(
{
'key1': 2.3
},
TypeError,
),
(
{
'key1': [2.3]
},
TypeError,
),
(
{
'key1': [[2]]
}, # nesting is illegal
TypeError,
),
]
class _RecordDict(RecordDict):
adjust_key1 = adjust_key2 = None
optional_keys = RecordDict.optional_keys | {'key1', 'key2'}
parser = BaseParser.__new__(BaseParser)
for input_dict, exc_class in inputs_and_exc_classes:
record_dict = _RecordDict(input_dict)
with self.assertRaises(exc_class):
parser.get_output_message_id(record_dict)
def test__postprocess_parsed__without__do_not_resolve_fqdn_to_ip(self):
data = {}
parsed = RecordDict()
self.meth.postprocess_parsed(data, parsed, 1, item_no=1)
self.assertEqual(parsed, {})
def test__postprocess_parsed__with__do_not_resolve_fqdn_to_ip__False(self):
data = {'_do_not_resolve_fqdn_to_ip': False}
parsed = RecordDict()
self.meth.postprocess_parsed(data, parsed, 1, item_no=1)
self.assertEqual(parsed, {})
def test__postprocess_parsed__with__do_not_resolve_fqdn_to_ip__True(self):
data = {'_do_not_resolve_fqdn_to_ip': True}
parsed = RecordDict()
self.meth.postprocess_parsed(data, parsed, 1, item_no=1)
self.assertEqual(parsed, {'_do_not_resolve_fqdn_to_ip': True})
def prepare_config_full(cls, pyramid_configurator):
# type: (Configurator) -> Config
config_constructor_kwargs = cls.prepare_config_constructor_kwargs(pyramid_configurator)
return Config(cls.config_spec, **config_constructor_kwargs)
