class GeventedHTTPTransport(AsyncTransport, HTTPTransport):
scheme = ['gevent+http', 'gevent+https']
def __init__(self, parsed_url, maximum_outstanding_requests=100, *args, **kwargs):
if not has_gevent:
raise ImportError('GeventedHTTPTransport requires gevent.')
self._lock = Semaphore(maximum_outstanding_requests)
super(GeventedHTTPTransport, self).__init__(parsed_url, *args, **kwargs)
def async_send(self, data, headers, success_cb, failure_cb):
"""
Spawn an async request to a remote webserver.
"""
# this can be optimized by making a custom self.send that does not
# read the response since we don't use it.
self._lock.acquire()
return gevent.spawn(
super(GeventedHTTPTransport, self).send, data, headers
).link(lambda x: self._done(x, success_cb, failure_cb))
def _done(self, greenlet, success_cb, failure_cb, *args):
self._lock.release()
if greenlet.successful():
success_cb()
else:
failure_cb(greenlet.exception)
class GeventedHTTPTransport(AsyncTransport, HTTPTransport):
scheme = ["gevent+http", "gevent+https"]
def __init__(self, parsed_url, maximum_outstanding_requests=100):
if not has_gevent:
raise ImportError("GeventedHTTPTransport requires gevent.")
self._lock = Semaphore(maximum_outstanding_requests)
super(GeventedHTTPTransport, self).__init__(parsed_url)
# remove the gevent+ from the protocol, as it is not a real protocol
self._url = self._url.split("+", 1)[-1]
def async_send(self, data, headers, success_cb, failure_cb):
"""
Spawn an async request to a remote webserver.
"""
# this can be optimized by making a custom self.send that does not
# read the response since we don't use it.
self._lock.acquire()
return gevent.spawn(super(GeventedHTTPTransport, self).send, data, headers).link(
lambda x: self._done(x, success_cb, failure_cb)
)
def _done(self, greenlet, success_cb, failure_cb, *args):
self._lock.release()
if greenlet.successful():
success_cb()
else:
failure_cb(greenlet.value)
class Pool(Group):
def __init__(self, size=None, greenlet_class=None):
if size is not None and size < 0:
raise ValueError("size must not be negative: %r" % (size,))
Group.__init__(self)
self.size = size
if greenlet_class is not None:
self.greenlet_class = greenlet_class
if size is None:
self._semaphore = DummySemaphore()
else:
self._semaphore = Semaphore(size)
def wait_available(self):
self._semaphore.wait()
def full(self):
return self.free_count() <= 0
def free_count(self):
if self.size is None:
return 1
return max(0, self.size - len(self))
def add(self, greenlet):
self._semaphore.acquire()
try:
Group.add(self, greenlet)
except:
self._semaphore.release()
raise
def _discard(self, greenlet):
Group._discard(self, greenlet)
self._semaphore.release()
class GeventedHTTPTransport(HTTPTransport):
scheme = ['gevent+http', 'gevent+https']
def __init__(self, parsed_url, maximum_outstanding_requests=100):
if not has_gevent:
raise ImportError('GeventedHTTPTransport requires gevent.')
self._lock = Semaphore(maximum_outstanding_requests)
super(GeventedHTTPTransport, self).__init__(parsed_url)
# remove the gevent+ from the protocol, as it is not a real protocol
self._url = self._url.split('+', 1)[-1]
def send(self, data, headers):
"""
Spawn an async request to a remote webserver.
"""
# this can be optimized by making a custom self.send that does not
# read the response since we don't use it.
self._lock.acquire()
return gevent.spawn(
super(GeventedHTTPTransport, self).send, data,
headers).link(self._done, self)
def _done(self, *args):
self._lock.release()
class _RPCConnection(object):
def __init__(self, sock, use_greenlets):
self._sock = sock
self.use_greenlets = use_greenlets
if use_greenlets:
from gevent.lock import Semaphore
self._send_lock = Semaphore()
def recv(self, buf_size):
return self._sock.recv(buf_size)
def send(self, msg):
if self.use_greenlets:
self._send_lock.acquire()
try:
self._sock.sendall(msg)
finally:
self._send_lock.release()
else:
try:
self._sock.sendall(msg)
finally:
pass
def __del__(self):
try:
self._sock.close()
except:
pass
class GeventedHTTPTransport(HTTPTransport):
scheme = ['gevent+http', 'gevent+https']
def __init__(self, parsed_url, maximum_outstanding_requests=100):
if not has_gevent:
raise ImportError('GeventedHTTPTransport requires gevent.')
self._lock = Semaphore(maximum_outstanding_requests)
super(GeventedHTTPTransport, self).__init__(parsed_url)
# remove the gevent+ from the protocol, as it is not a real protocol
self._url = self._url.split('+', 1)[-1]
def send(self, data, headers):
"""
Spawn an async request to a remote webserver.
"""
# this can be optimized by making a custom self.send that does not
# read the response since we don't use it.
self._lock.acquire()
return gevent.spawn(super(GeventedHTTPTransport, self).send, data, headers).link(self._done, self)
def _done(self, *args):
self._lock.release()
class GeventedHTTPTransport(AsyncTransport, HTTPTransport):
scheme = ['gevent+http', 'gevent+https']
def __init__(self, maximum_outstanding_requests=100, *args, **kwargs):
if not has_gevent:
raise ImportError('GeventedHTTPTransport requires gevent.')
self._lock = Semaphore(maximum_outstanding_requests)
super().__init__(*args, **kwargs)
def async_send(self, url, data, headers, success_cb, failure_cb):
"""
Spawn an async request to a remote webserver.
"""
# this can be optimized by making a custom self.send that does not
# read the response since we don't use it.
self._lock.acquire()
return gevent.spawn(
super().send, url, data, headers
).link(lambda x: self._done(x, success_cb, failure_cb))
def _done(self, greenlet, success_cb, failure_cb, *args):
self._lock.release()
if greenlet.successful():
success_cb()
else:
failure_cb(greenlet.exception)
class Pool(Group):
def __init__(self, size=None, greenlet_class=None):
if size is not None and size < 0:
raise ValueError('size must not be negative: %r' % (size, ))
Group.__init__(self)
self.size = size
if greenlet_class is not None:
self.greenlet_class = greenlet_class
if size is None:
self._semaphore = DummySemaphore()
else:
self._semaphore = Semaphore(size)
def wait_available(self):
self._semaphore.wait()
def full(self):
return self.free_count() <= 0
def free_count(self):
if self.size is None:
return 1
return max(0, self.size - len(self))
def add(self, greenlet):
self._semaphore.acquire()
try:
Group.add(self, greenlet)
except:
self._semaphore.release()
raise
def _discard(self, greenlet):
Group._discard(self, greenlet)
self._semaphore.release()
def test_release_twice(self):
s = Semaphore()
result = []
s.rawlink(lambda s: result.append('a'))
s.release()
s.rawlink(lambda s: result.append('b'))
s.release()
gevent.sleep(0.001)
self.assertEqual(result, ['a', 'b'])
def test_release_twice(self):
s = Semaphore()
result = []
s.rawlink(lambda s: result.append('a'))
s.release()
s.rawlink(lambda s: result.append('b'))
s.release()
gevent.sleep(0.001)
self.assertEqual(result, ['a', 'b'])
def test_release_twice(self):
s = Semaphore()
result = []
s.rawlink(lambda s: result.append('a'))
s.release()
s.rawlink(lambda s: result.append('b'))
s.release()
gevent.sleep(0.001)
# The order, though, is not guaranteed.
self.assertEqual(sorted(result), ['a', 'b'])
def test_release_twice(self):
s = Semaphore()
result = []
s.rawlink(lambda s: result.append('a'))
s.release()
s.rawlink(lambda s: result.append('b'))
s.release()
gevent.sleep(0.001)
# The order, though, is not guaranteed.
self.assertEqual(sorted(result), ['a', 'b'])
class ModLogPump(object):
def __init__(self, channel, max_actions, action_time):
self.channel = channel
self.action_time = action_time
self._have = Event()
self._buffer = []
self._lock = Semaphore(max_actions)
self._emitter = gevent.spawn(self._emit_loop)
def _get_next_message(self):
data = ''
while self._buffer:
payload = self._buffer.pop(0)
if len(data) + len(payload) > 2000:
break
data += '\n'
data += payload
return data
def _emit_loop(self):
while True:
self._have.wait()
try:
self._emit()
except APIException as e:
# If send message is disabled, backoff (we'll drop events but
# thats ok)
if e.code == 40004:
gevent.sleep(5)
if not len(self._buffer):
self._have.clear()
def _emit(self):
self._lock.acquire()
msg = self._get_next_message()
if not msg:
self._lock.release()
return
self.channel.send_message(msg)
gevent.spawn(self._emit_unlock)
def _emit_unlock(self):
gevent.sleep(self.action_time)
self._lock.release()
def add_message(self, payload):
self._buffer.append(payload)
self._have.set()
class FifoQueue:
def __init__(self):
self._queue = deque()
self._semaphore = Semaphore(0)
def __len__(self):
return len(self._queue)
def push(self, request):
self._queue.append(request)
self._semaphore.release()
def pop(self):
self._semaphore.acquire()
return self._queue.popleft()
class GeventConnectionPool(psycopg2.pool.AbstractConnectionPool):
def __init__(self, minconn, maxconn, *args, **kwargs):
self.semaphore = Semaphore(maxconn)
psycopg2.pool.AbstractConnectionPool.__init__(self, minconn, maxconn, *args, **kwargs)
def getconn(self, *args, **kwargs):
self.semaphore.acquire()
return self._getconn(*args, **kwargs)
def putconn(self, *args, **kwargs):
self._putconn(*args, **kwargs)
self.semaphore.release()
# Not sure what to do about this one...
closeall = psycopg2.pool.AbstractConnectionPool._closeall
def test_renamed_label_refresh(db, default_account, thread, message, imapuid,
folder, mock_imapclient, monkeypatch):
# Check that imapuids see their labels refreshed after running
# the LabelRenameHandler.
msg_uid = imapuid.msg_uid
uid_dict = {msg_uid: GmailFlags((), ('stale label', ), ('23', ))}
update_metadata(default_account.id, folder.id, folder.canonical_name,
uid_dict, db.session)
new_flags = {
msg_uid: {
'FLAGS': ('\\Seen', ),
'X-GM-LABELS': ('new label', ),
'MODSEQ': ('23', )
}
}
mock_imapclient._data['[Gmail]/All mail'] = new_flags
mock_imapclient.add_folder_data(folder.name, new_flags)
monkeypatch.setattr(MockIMAPClient, 'search', lambda x, y: [msg_uid])
semaphore = Semaphore(value=1)
rename_handler = LabelRenameHandler(default_account.id,
default_account.namespace.id,
'new label', semaphore)
# Acquire the semaphore to check that LabelRenameHandlers block if
# the semaphore is in-use.
semaphore.acquire()
rename_handler.start()
# Wait 10 secs and check that the data hasn't changed.
gevent.sleep(10)
labels = list(imapuid.labels)
assert len(labels) == 1
assert labels[0].name == 'stale label'
semaphore.release()
rename_handler.join()
db.session.refresh(imapuid)
# Now check that the label got updated.
labels = list(imapuid.labels)
assert len(labels) == 1
assert labels[0].name == 'new label'
class JobSharedLock(object):
"""
Shared lock for jobs.
Each job method can specify a lock which will be shared
among all calls for that job and only one job can run at a time
for this lock.
"""
def __init__(self, queue, name):
self.queue = queue
self.name = name
self.jobs = []
self.semaphore = Semaphore()
def add_job(self, job):
self.jobs.append(job)
def get_jobs(self):
return self.jobs
def remove_job(self, job):
self.jobs.remove(job)
def locked(self):
return self.semaphore.locked()
def acquire(self):
return self.semaphore.acquire()
def release(self):
return self.semaphore.release()
class GeventConnectionPool(psycopg2.pool.AbstractConnectionPool):
def __init__(self, minconn, maxconn, *args, **kwargs):
self.semaphore = Semaphore(maxconn)
psycopg2.pool.AbstractConnectionPool.__init__(self, minconn, maxconn,
*args, **kwargs)
def getconn(self, *args, **kwargs):
self.semaphore.acquire()
return self._getconn(*args, **kwargs)
def putconn(self, *args, **kwargs):
self._putconn(*args, **kwargs)
self.semaphore.release()
# Not sure what to do about this one...
closeall = psycopg2.pool.AbstractConnectionPool._closeall
class PriorityQueue:
def __init__(self):
self._queue = []
self._semaphore = Semaphore(0)
def __len__(self):
return len(self._queue)
def push(self, request):
heappush(self._queue, _PriorityQueueItem(request))
self._semaphore.release()
def pop(self):
self._semaphore.acquire()
item = heappop(self._queue)
return item.request
class JobSharedLock(object):
"""
Shared lock for jobs.
Each job method can specify a lock which will be shared
among all calls for that job and only one job can run at a time
for this lock.
"""
def __init__(self, queue, name):
self.queue = queue
self.name = name
self.jobs = []
self.semaphore = Semaphore()
def add_job(self, job):
self.jobs.append(job)
def get_jobs(self):
return self.jobs
def remove_job(self, job):
self.jobs.remove(job)
def locked(self):
return self.semaphore.locked()
def acquire(self):
return self.semaphore.acquire()
def release(self):
return self.semaphore.release()
class Puppet(object):
def __init__(self):
self.results = {}
self.status = ("Idle", None)
self.worker = None
self.operator = None
self.lock = Semaphore()
def receive_info(self):
while True:
(info_id, info) = self.task_queue.get()
if info_id > 0:
self.results[info_id].set(info)
elif info_id == -1:
self.status = info
def submit_task(self, tid, task_info):
self.results[tid] = AsyncResult()
self.lock.acquire()
self.task_queue.put((tid, task_info))
self.lock.release()
def fetch_result(self, tid):
res = self.results[tid].get()
del self.results[tid]
return res
def hire_worker(self):
tq_worker, self.task_queue = gipc.pipe(duplex=True)
self.worker = gipc.start_process(target=process_worker,
args=(tq_worker, ))
self.operator = gevent.spawn(self.receive_info)
def fire_worker(self):
self.worker.terminate()
self.operator.kill()
def get_attr(self, attr):
if attr == "cpu":
return psutil.cpu_percent(interval=None)
elif attr == "memory":
return psutil.virtual_memory().percent
else:
return getattr(self, attr, "No such attribute")
def current_tasks(self):
return self.results.keys()
def test_renamed_label_refresh(db, default_account, thread, message, imapuid,
folder, mock_imapclient, monkeypatch):
# Check that imapuids see their labels refreshed after running
# the LabelRenameHandler.
msg_uid = imapuid.msg_uid
uid_dict = {msg_uid: GmailFlags((), ("stale label", ), ("23", ))}
update_metadata(default_account.id, folder.id, folder.canonical_name,
uid_dict, db.session)
new_flags = {
msg_uid: {
b"FLAGS": (b"\\Seen", ),
b"X-GM-LABELS": (b"new label", ),
b"MODSEQ": (23, ),
}
}
mock_imapclient._data["[Gmail]/All mail"] = new_flags
mock_imapclient.add_folder_data(folder.name, new_flags)
monkeypatch.setattr(MockIMAPClient, "search", lambda x, y: [msg_uid])
semaphore = Semaphore(value=1)
rename_handler = LabelRenameHandler(default_account.id,
default_account.namespace.id,
"new label", semaphore)
# Acquire the semaphore to check that LabelRenameHandlers block if
# the semaphore is in-use.
semaphore.acquire()
rename_handler.start()
gevent.sleep(0) # yield to the handler
labels = list(imapuid.labels)
assert len(labels) == 1
assert labels[0].name == "stale label"
semaphore.release()
rename_handler.join()
db.session.refresh(imapuid)
# Now check that the label got updated.
labels = list(imapuid.labels)
assert len(labels) == 1
assert labels[0].name == "new label"
class Puppet(object):
def __init__(self):
self.results = {}
self.status = ("Idle", None)
self.worker = None
self.operator = None
self.lock = Semaphore()
def receive_info(self):
while True:
(info_id, info) = self.task_queue.get()
if info_id > 0:
self.results[info_id].set(info)
elif info_id == -1:
self.status = info
def submit_task(self, tid, task_info):
self.results[tid] = AsyncResult()
self.lock.acquire()
self.task_queue.put((tid, task_info))
self.lock.release()
def fetch_result(self, tid):
res = self.results[tid].get()
del self.results[tid]
return res
def hire_worker(self):
tq_worker, self.task_queue = gipc.pipe(duplex=True)
self.worker = gipc.start_process(target=process_worker, args=(tq_worker,))
self.operator = gevent.spawn(self.receive_info)
def fire_worker(self):
self.worker.terminate()
self.operator.kill()
def get_attr(self, attr):
if attr == "cpu":
return psutil.cpu_percent(interval=None)
elif attr == "memory":
return psutil.virtual_memory().percent
else:
return getattr(self, attr, "No such attribute")
def current_tasks(self):
return self.results.keys()
def test_renamed_label_refresh(db, default_account, thread, message,
imapuid, folder, mock_imapclient, monkeypatch):
# Check that imapuids see their labels refreshed after running
# the LabelRenameHandler.
msg_uid = imapuid.msg_uid
uid_dict = {msg_uid: GmailFlags((), ('stale label',), ('23',))}
update_metadata(default_account.id, folder.id, folder.canonical_name,
uid_dict, db.session)
new_flags = {msg_uid: {'FLAGS': ('\\Seen',), 'X-GM-LABELS': ('new label',),
'MODSEQ': ('23',)}}
mock_imapclient._data['[Gmail]/All mail'] = new_flags
mock_imapclient.add_folder_data(folder.name, new_flags)
monkeypatch.setattr(MockIMAPClient, 'search',
lambda x, y: [msg_uid])
semaphore = Semaphore(value=1)
rename_handler = LabelRenameHandler(default_account.id,
default_account.namespace.id,
'new label', semaphore)
# Acquire the semaphore to check that LabelRenameHandlers block if
# the semaphore is in-use.
semaphore.acquire()
rename_handler.start()
# Wait 10 secs and check that the data hasn't changed.
gevent.sleep(10)
labels = list(imapuid.labels)
assert len(labels) == 1
assert labels[0].name == 'stale label'
semaphore.release()
rename_handler.join()
db.session.refresh(imapuid)
# Now check that the label got updated.
labels = list(imapuid.labels)
assert len(labels) == 1
assert labels[0].name == 'new label'
class GeventedHTTPTransport(AsyncTransport, HTTPTransport):
scheme = ['gevent+http', 'gevent+https']
def __init__(self, parsed_url, maximum_outstanding_requests = 100, *args, **kwargs):
if not has_gevent:
raise ImportError('GeventedHTTPTransport requires gevent.')
self._lock = Semaphore(maximum_outstanding_requests)
super(GeventedHTTPTransport, self).__init__(parsed_url, *args, **kwargs)
def async_send(self, data, headers, success_cb, failure_cb):
self._lock.acquire()
return gevent.spawn(super(GeventedHTTPTransport, self).send, data, headers).link(lambda x: self._done(x, success_cb, failure_cb))
def _done(self, greenlet, success_cb, failure_cb, *args):
self._lock.release()
if greenlet.successful():
success_cb()
else:
failure_cb(greenlet.exception)
class DatagramServer(BaseServer):
"""A UDP server"""
reuse_addr = DEFAULT_REUSE_ADDR
def __init__(self, *args, **kwargs):
# The raw (non-gevent) socket, if possible
self._socket = None
BaseServer.__init__(self, *args, **kwargs)
from gevent.lock import Semaphore
self._writelock = Semaphore()
def init_socket(self):
if not hasattr(self, 'socket'):
# FIXME: clean up the socket lifetime
# pylint:disable=attribute-defined-outside-init
self.socket = self.get_listener(self.address, self.family)
self.address = self.socket.getsockname()
self._socket = self.socket
try:
self._socket = self._socket._sock
except AttributeError:
pass
@classmethod
def get_listener(cls, address, family=None):
return _udp_socket(address, reuse_addr=cls.reuse_addr, family=family)
def do_read(self):
try:
data, address = self._socket.recvfrom(8192)
except _socket.error as err:
if err.args[0] == EWOULDBLOCK:
return
raise
return data, address
def sendto(self, *args):
self._writelock.acquire()
try:
self.socket.sendto(*args)
finally:
self._writelock.release()
class DatagramServer(BaseServer):
"""A UDP server"""
reuse_addr = DEFAULT_REUSE_ADDR
def __init__(self, *args, **kwargs):
# The raw (non-gevent) socket, if possible
self._socket = None
BaseServer.__init__(self, *args, **kwargs)
from gevent.lock import Semaphore
self._writelock = Semaphore()
def init_socket(self):
if not hasattr(self, 'socket'):
# FIXME: clean up the socket lifetime
# pylint:disable=attribute-defined-outside-init
self.socket = self.get_listener(self.address, self.family)
self.address = self.socket.getsockname()
self._socket = self.socket
try:
self._socket = self._socket._sock
except AttributeError:
pass
@classmethod
def get_listener(cls, address, family=None):
return _udp_socket(address, reuse_addr=cls.reuse_addr, family=family)
def do_read(self):
try:
data, address = self._socket.recvfrom(8192)
except SocketError as err:
if err.args[0] == EWOULDBLOCK:
return
raise
return data, address
def sendto(self, *args):
self._writelock.acquire()
try:
self.socket.sendto(*args)
finally:
self._writelock.release()
class ProxyGuard(object):
"""
Helper class that can be used to protect an proxy object that represents a remote resource.
It is possible to specify the maximum number of parallel calls of any method.
This is helpful in the case that the resource limits the number of parallel connections.
"""
def __init__(self,
proxy: Any,
max_parallel_access=1,
attr_name: Optional[str] = None,
semaphore: Optional[Semaphore] = None) -> None:
self._attr_name = attr_name
self._proxy = proxy
self._max_parallel_access = max_parallel_access
if semaphore is not None:
self._semaphore = semaphore
else:
self._semaphore = Semaphore(value=max_parallel_access)
def __getattr__(self, name):
return ProxyGuard(self._proxy,
max_parallel_access=self._max_parallel_access,
attr_name=name,
semaphore=self._semaphore)
def __call__(self, *args, **kwargs) -> Any:
self._semaphore.acquire(blocking=True)
ex = None
result = None
try:
result = getattr(self._proxy, self._attr_name)(*args, **kwargs)
except BaseException as base_exception:
ex = base_exception
finally:
self._semaphore.release()
if ex is not None:
raise ex
return result
class DatagramServer(BaseServer):
"""A UDP server"""
reuse_addr = DEFAULT_REUSE_ADDR
def __init__(self, *args, **kwargs):
BaseServer.__init__(self, *args, **kwargs)
from gevent.lock import Semaphore
self._writelock = Semaphore()
def init_socket(self):
if not hasattr(self, 'socket'):
self.socket = self.get_listener(self.address, self.family)
self.address = self.socket.getsockname()
self._socket = self.socket
try:
if sys.version_info[0] == 2:
self._socket = self._socket._sock
else:
self._socket = super(socket, self._socket)
except AttributeError:
pass
@classmethod
def get_listener(self, address, family=None):
return _udp_socket(address, reuse_addr=self.reuse_addr, family=family)
def do_read(self):
try:
data, address = self._socket.recvfrom(8192)
except _socket.error as err:
if err.args[0] == EWOULDBLOCK:
return
raise
return data, address
def sendto(self, *args):
self._writelock.acquire()
try:
self.socket.sendto(*args)
finally:
self._writelock.release()
class Hub:
def __init__(self, num):
self.need = 0
self.block_queue = Queue(maxsize=1)
self.total_player = []
for n in range(num):
self.total_player.append(n)
self.semaphore = Semaphore()
def _get_player(self, num):
if len(self.total_player) < num:
# block here
self.need = num
print "_get_player set need=%s" % self.need
self.block_queue.get(block=True)
assert len(self.total_player) >= num, (
"_get_player error total_player=%s num=%s" %
(len(self.total_player), num))
size = len(self.total_player)
player_list = self.total_player[size - num:]
self.total_player = self.total_player[:size - num]
return player_list
def acquire_player(self, num):
# lock
self.semaphore.acquire()
print "acquire_player num=%s" % num
player_list = self._get_player(num)
# unlock
self.semaphore.release()
return player_list
def release_player(self, player_list):
self.total_player.extend(player_list)
print "release_player need=%s len(self.total_player)=%s" % (
self.need, len(self.total_player))
if self.need == 0 or self.need > len(self.total_player):
return
self.need = 0
self.block_queue.put(1, block=True)
class TaskRunner(object):
def __init__(self, name, task):
self.name = name
self.task = task
self.lock = Semaphore(task.max_concurrent)
def process(self, job):
log.info('[%s] Running job %s...', job['id'], self.name)
start = time.time()
try:
self.task(*job['args'], **job['kwargs'])
if self.task.buffer_time:
time.sleep(self.task.buffer_time)
except:
log.exception('[%s] Failed in %ss', job['id'], time.time() - start)
log.info('[%s] Completed in %ss', job['id'], time.time() - start)
def run(self, job):
lock = None
if self.task.global_lock:
lock = rdb.lock('{}:{}'.format(
self.task.name,
self.task.global_lock(
*job['args'],
**job['kwargs']
)
))
lock.acquire()
if self.task.max_concurrent:
self.lock.acquire()
self.process(job)
if lock:
lock.release()
if self.task.max_concurrent:
self.lock.release()
class DatagramServer(BaseServer):
"""A UDP server"""
reuse_addr = DEFAULT_REUSE_ADDR
def __init__(self, *args, **kwargs):
BaseServer.__init__(self, *args, **kwargs)
from gevent.lock import Semaphore
self._writelock = Semaphore()
def init_socket(self):
if not hasattr(self, 'socket'):
self.socket = self.get_listener(self.address, self.family)
self.address = self.socket.getsockname()
self._socket = self.socket
try:
self._socket = self._socket._sock
except AttributeError:
pass
@classmethod
def get_listener(self, address, family=None):
return _udp_socket(address, reuse_addr=self.reuse_addr, family=family)
def do_read(self):
try:
data, address = self._socket.recvfrom(8192)
except _socket.error:
err = sys.exc_info()[1]
if err.args[0] == EWOULDBLOCK:
return
raise
return data, address
def sendto(self, *args):
self._writelock.acquire()
try:
self.socket.sendto(*args)
finally:
self._writelock.release()
class LimitEngine(EngineBase):
def __init__(self, provider, n, **kwargs):
super(LimitEngine, self).__init__(provider, **kwargs)
self.conf = self.provider.configurations()[0]
self.n = n
self.services = []
self.num_unscheduled = -1
self.le_lock = Semaphore()
def before_eval(self):
self.num_unscheduled = len(self.dag)
def after_eval(self):
for s in self.services:
self.provider.stop_service(s)
self.services = []
def which_service(self, task):
self.num_unscheduled -= 1
for s in self.services:
if len(s.tasks) == 0:
return s
if len(self.services) < self.n:
s = self.provider.start_service(len(self.services) + 1, self.conf)
self.services.append(s)
else:
s = min(self.services, key=lambda x: len(x.tasks))
return s
def after_task(self, task, service):
self.le_lock.acquire()
if self.num_unscheduled == 0 and len(service.tasks) == 0:
self.provider.stop_service(service)
for s in self.services:
if len(s.tasks) == 0:
self.provider.stop_service(s)
self.le_lock.release()
def current_services(self):
return self.services
class BlockingDeque(deque):
def __init__(self, *args, **kwargs):
super(BlockingDeque, self).__init__(*args, **kwargs)
self.sema = Semaphore(len(self))
def append(self, *args, **kwargs):
ret = super(BlockingDeque, self).append(*args, **kwargs)
self.sema.release()
return ret
def appendleft(self, *args, **kwargs):
ret = super(BlockingDeque, self).appendleft(*args, **kwargs)
self.sema.release()
return ret
def clear(self, *args, **kwargs):
ret = super(BlockingDeque, self).clear(*args, **kwargs)
while not self.sema.locked():
self.sema.acquire(blocking=False)
return ret
def extend(self, *args, **kwargs):
pre_n = len(self)
ret = super(BlockingDeque, self).extend(*args, **kwargs)
post_n = len(self)
for i in range(pre_n, post_n):
self.sema.release()
return ret
def extendleft(self, *args, **kwargs):
pre_n = len(self)
ret = super(BlockingDeque, self).extendleft(*args, **kwargs)
post_n = len(self)
for i in range(pre_n, post_n):
self.sema.release()
return ret
def pop(self, *args, **kwargs):
self.sema.acquire()
return super(BlockingDeque, self).pop(*args, **kwargs)
def popleft(self, *args, **kwargs):
self.sema.acquire()
return super(BlockingDeque, self).popleft(*args, **kwargs)
def remove(self, *args, **kwargs):
ret = super(BlockingDeque, self).remove(*args, **kwargs)
self.sema.acquire()
return ret
class GeventedHTTPTransport(AsyncTransport, HTTPTransport):
scheme = ['gevent+http', 'gevent+https']
def __init__(self, parsed_url, maximum_outstanding_requests=100,
oneway=False):
if not has_gevent:
raise ImportError('GeventedHTTPTransport requires gevent.')
self._lock = Semaphore(maximum_outstanding_requests)
self._oneway = oneway == 'true'
super(GeventedHTTPTransport, self).__init__(parsed_url)
# remove the gevent+ from the protocol, as it is not a real protocol
self._url = self._url.split('+', 1)[-1]
def async_send(self, data, headers, success_cb, failure_cb):
"""
Spawn an async request to a remote webserver.
"""
if not self._oneway:
self._lock.acquire()
return gevent.spawn(
super(GeventedHTTPTransport, self).send, data, headers
).link(lambda x: self._done(x, success_cb, failure_cb))
else:
req = urllib2.Request(self._url, headers=headers)
return urlopen(url=req,
data=data,
timeout=self.timeout,
verify_ssl=self.verify_ssl,
ca_certs=self.ca_certs)
def _done(self, greenlet, success_cb, failure_cb, *args):
self._lock.release()
if greenlet.successful():
success_cb()
else:
failure_cb(greenlet.exception)
class BlockingDeque(deque):
def __init__(self, *args, **kwargs):
super(BlockingDeque, self).__init__(*args, **kwargs)
self.sema = Semaphore(len(self))
def append(self, *args, **kwargs):
ret = super(BlockingDeque, self).append(*args, **kwargs)
self.sema.release()
return ret
def appendleft(self, *args, **kwargs):
ret = super(BlockingDeque, self).appendleft(*args, **kwargs)
self.sema.release()
return ret
def clear(self, *args, **kwargs):
ret = super(BlockingDeque, self).clear(*args, **kwargs)
while not self.sema.locked():
self.sema.acquire(blocking=False)
return ret
def extend(self, *args, **kwargs):
pre_n = len(self)
ret = super(BlockingDeque, self).extend(*args, **kwargs)
post_n = len(self)
for i in range(pre_n, post_n):
self.sema.release()
return ret
def extendleft(self, *args, **kwargs):
pre_n = len(self)
ret = super(BlockingDeque, self).extendleft(*args, **kwargs)
post_n = len(self)
for i in range(pre_n, post_n):
self.sema.release()
return ret
def pop(self, *args, **kwargs):
self.sema.acquire()
return super(BlockingDeque, self).pop(*args, **kwargs)
def popleft(self, *args, **kwargs):
self.sema.acquire()
return super(BlockingDeque, self).popleft(*args, **kwargs)
def remove(self, *args, **kwargs):
ret = super(BlockingDeque, self).remove(*args, **kwargs)
self.sema.acquire()
return ret
def test_fair_or_hangs(self):
# If the lock isn't fair, this hangs, spinning between
# the last two greenlets.
# See https://github.com/gevent/gevent/issues/1487
sem = Semaphore()
should_quit = []
keep_going1 = FirstG.spawn(acquire_then_spawn, sem, should_quit)
keep_going2 = FirstG.spawn(acquire_then_spawn, sem, should_quit)
exiting = LastG.spawn(acquire_then_exit, sem, should_quit)
with self.assertRaises(gevent.exceptions.LoopExit):
gevent.joinall([keep_going1, keep_going2, exiting])
self.assertTrue(exiting.dead, exiting)
self.assertTrue(keep_going2.dead, keep_going2)
self.assertFalse(keep_going1.dead, keep_going1)
sem.release()
keep_going1.kill()
keep_going2.kill()
exiting.kill()
gevent.idle()
class RollingNumber(object):
_last_idx = 0
@property
def now(self):
return int(time.time() * 1000)
def __init__(self, window=10 * 1000, interval=1000):
self.window = window
self.interval = interval
self.bucket_num = window / self.interval
self._new_bucket_lock = Semaphore()
self.reset(self.now)
def reset(self, start=None):
self._buckets = deque([0], maxlen=self.bucket_num)
self._start = start or self.now
def get_current_bucket_index(self):
if not self._new_bucket_lock.acquire(0):
return self._last_idx
now = self.now
elapsed = now - self._start
if elapsed > self.interval:
if elapsed > self.window:
self.reset(now)
self._new_bucket_lock.release()
return self._last_idx
t = elapsed - self.interval
while t > 0:
self._buckets.appendleft(0)
t1, t = t, t - self.interval
if t < 0:
self._start = now - t1
self._new_bucket_lock.release()
return self._last_idx
self._new_bucket_lock.release()
return self._last_idx
@property
def current_bucket_count(self):
idx = self.get_current_bucket_index()
return self._buckets[idx]
def inc(self):
i = self.get_current_bucket_index()
self._buckets[i] += 1
@property
def sum(self):
return sum(self._buckets)
class AnalyticsDiscovery(gevent.Greenlet):
def _sandesh_connection_info_update(self, status, message):
new_conn_state = getattr(ConnectionStatus, status)
ConnectionState.update(conn_type = ConnectionType.ZOOKEEPER,
name = self._svc_name, status = new_conn_state,
message = message,
server_addrs = self._zk_server.split(','))
if (self._conn_state and self._conn_state != ConnectionStatus.DOWN and
new_conn_state == ConnectionStatus.DOWN):
msg = 'Connection to Zookeeper down: %s' %(message)
self._logger.error(msg)
if (self._conn_state and self._conn_state != new_conn_state and
new_conn_state == ConnectionStatus.UP):
msg = 'Connection to Zookeeper ESTABLISHED'
self._logger.error(msg)
self._conn_state = new_conn_state
# end _sandesh_connection_info_update
def _zk_listen(self, state):
self._logger.error("Analytics Discovery listen %s" % str(state))
if state == KazooState.CONNECTED:
self._sandesh_connection_info_update(status='UP', message='')
self._logger.error("Analytics Discovery to publish %s" % str(self._pubinfo))
self._reconnect = True
elif state == KazooState.LOST:
self._logger.error("Analytics Discovery connection LOST")
# Lost the session with ZooKeeper Server
# Best of option we have is to exit the process and restart all
# over again
self._sandesh_connection_info_update(status='DOWN',
message='Connection to Zookeeper lost')
os._exit(2)
elif state == KazooState.SUSPENDED:
self._logger.error("Analytics Discovery connection SUSPENDED")
# Update connection info
self._sandesh_connection_info_update(status='INIT',
message = 'Connection to zookeeper lost. Retrying')
def _zk_datawatch(self, watcher, child, data, stat, event="unknown"):
self._logger.error(\
"Analytics Discovery %s ChildData : child %s, data %s, event %s" % \
(watcher, child, data, event))
if data:
data_dict = json.loads(data)
self._wchildren[watcher][child] = OrderedDict(sorted(data_dict.items()))
else:
if child in self._wchildren[watcher]:
del self._wchildren[watcher][child]
if self._watchers[watcher]:
self._pendingcb.add(watcher)
def _zk_watcher(self, watcher, children):
self._logger.error("Analytics Discovery Children %s" % children)
self._reconnect = True
def __init__(self, logger, zkservers, svc_name, inst,
watchers={}, zpostfix="", freq=10):
gevent.Greenlet.__init__(self)
self._svc_name = svc_name
self._inst = inst
self._zk_server = zkservers
# initialize logging and other stuff
if logger is None:
logging.basicConfig()
self._logger = logging
else:
self._logger = logger
self._conn_state = None
self._sandesh_connection_info_update(status='INIT', message='')
self._zkservers = zkservers
self._zk = None
self._pubinfo = None
self._publock = Semaphore()
self._watchers = watchers
self._wchildren = {}
self._pendingcb = set()
self._zpostfix = zpostfix
self._basepath = "/analytics-discovery-" + self._zpostfix
self._reconnect = None
self._freq = freq
def publish(self, pubinfo):
# This function can be called concurrently by the main AlarmDiscovery
# processing loop as well as by clients.
# It is NOT re-entrant
self._publock.acquire()
self._pubinfo = pubinfo
if self._conn_state == ConnectionStatus.UP:
try:
self._logger.error("ensure %s" % (self._basepath + "/" + self._svc_name))
self._logger.error("zk state %s (%s)" % (self._zk.state, self._zk.client_state))
self._zk.ensure_path(self._basepath + "/" + self._svc_name)
self._logger.error("check for %s/%s/%s" % \
(self._basepath, self._svc_name, self._inst))
if pubinfo is not None:
if self._zk.exists("%s/%s/%s" % \
(self._basepath, self._svc_name, self._inst)):
self._zk.set("%s/%s/%s" % \
(self._basepath, self._svc_name, self._inst),
self._pubinfo)
else:
self._zk.create("%s/%s/%s" % \
(self._basepath, self._svc_name, self._inst),
self._pubinfo, ephemeral=True)
else:
if self._zk.exists("%s/%s/%s" % \
(self._basepath, self._svc_name, self._inst)):
self._logger.error("withdrawing published info!")
self._zk.delete("%s/%s/%s" % \
(self._basepath, self._svc_name, self._inst))
except Exception as ex:
template = "Exception {0} in AnalyticsDiscovery publish. Args:\n{1!r}"
messag = template.format(type(ex).__name__, ex.args)
self._logger.error("%s : traceback %s for %s info %s" % \
(messag, traceback.format_exc(), self._svc_name, str(self._pubinfo)))
self._sandesh_connection_info_update(status='DOWN', message='')
self._reconnect = True
else:
self._logger.error("Analytics Discovery cannot publish while down")
self._publock.release()
def _run(self):
while True:
self._logger.error("Analytics Discovery zk start")
self._zk = KazooClient(hosts=self._zkservers)
self._zk.add_listener(self._zk_listen)
try:
self._zk.start()
while self._conn_state != ConnectionStatus.UP:
gevent.sleep(1)
break
except Exception as e:
# Update connection info
self._sandesh_connection_info_update(status='DOWN',
message=str(e))
self._zk.remove_listener(self._zk_listen)
try:
self._zk.stop()
self._zk.close()
except Exception as ex:
template = "Exception {0} in AnalyticsDiscovery zk stop/close. Args:\n{1!r}"
messag = template.format(type(ex).__name__, ex.args)
self._logger.error("%s : traceback %s for %s" % \
(messag, traceback.format_exc(), self._svc_name))
finally:
self._zk = None
gevent.sleep(1)
try:
# Update connection info
self._sandesh_connection_info_update(status='UP', message='')
self._reconnect = False
# Done connecting to ZooKeeper
for wk in self._watchers.keys():
self._zk.ensure_path(self._basepath + "/" + wk)
self._wchildren[wk] = {}
self._zk.ChildrenWatch(self._basepath + "/" + wk,
partial(self._zk_watcher, wk))
# Trigger the initial publish
self._reconnect = True
while True:
try:
if not self._reconnect:
pending_list = list(self._pendingcb)
self._pendingcb = set()
for wk in pending_list:
if self._watchers[wk]:
self._watchers[wk](\
sorted(self._wchildren[wk].values()))
# If a reconnect happens during processing, don't lose it
while self._reconnect:
self._logger.error("Analytics Discovery %s reconnect" \
% self._svc_name)
self._reconnect = False
self._pendingcb = set()
self.publish(self._pubinfo)
for wk in self._watchers.keys():
self._zk.ensure_path(self._basepath + "/" + wk)
children = self._zk.get_children(self._basepath + "/" + wk)
old_children = set(self._wchildren[wk].keys())
new_children = set(children)
# Remove contents for the children who are gone
# (DO NOT remove the watch)
for elem in old_children - new_children:
del self._wchildren[wk][elem]
# Overwrite existing children, or create new ones
for elem in new_children:
# Create a watch for new children
if elem not in self._wchildren[wk]:
self._zk.DataWatch(self._basepath + "/" + \
wk + "/" + elem,
partial(self._zk_datawatch, wk, elem))
data_str, _ = self._zk.get(\
self._basepath + "/" + wk + "/" + elem)
data_dict = json.loads(data_str)
self._wchildren[wk][elem] = \
OrderedDict(sorted(data_dict.items()))
self._logger.error(\
"Analytics Discovery %s ChildData : child %s, data %s, event %s" % \
(wk, elem, self._wchildren[wk][elem], "GET"))
if self._watchers[wk]:
self._watchers[wk](sorted(self._wchildren[wk].values()))
gevent.sleep(self._freq)
except gevent.GreenletExit:
self._logger.error("Exiting AnalyticsDiscovery for %s" % \
self._svc_name)
self._zk.remove_listener(self._zk_listen)
gevent.sleep(1)
try:
self._zk.stop()
except:
self._logger.error("Stopping kazooclient failed")
else:
self._logger.error("Stopping kazooclient successful")
try:
self._zk.close()
except:
self._logger.error("Closing kazooclient failed")
else:
self._logger.error("Closing kazooclient successful")
break
except Exception as ex:
template = "Exception {0} in AnalyticsDiscovery reconnect. Args:\n{1!r}"
messag = template.format(type(ex).__name__, ex.args)
self._logger.error("%s : traceback %s for %s info %s" % \
(messag, traceback.format_exc(), self._svc_name, str(self._pubinfo)))
self._reconnect = True
except Exception as ex:
template = "Exception {0} in AnalyticsDiscovery run. Args:\n{1!r}"
messag = template.format(type(ex).__name__, ex.args)
self._logger.error("%s : traceback %s for %s info %s" % \
(messag, traceback.format_exc(), self._svc_name, str(self._pubinfo)))
raise SystemExit
class ThreadPool(object):
def __init__(self, maxsize, hub=None):
if hub is None:
hub = get_hub()
self.hub = hub
self._maxsize = 0
self.manager = None
self.pid = os.getpid()
self.fork_watcher = hub.loop.fork(ref=False)
self._init(maxsize)
def _set_maxsize(self, maxsize):
if not isinstance(maxsize, integer_types):
raise TypeError('maxsize must be integer: %r' % (maxsize, ))
if maxsize < 0:
raise ValueError('maxsize must not be negative: %r' % (maxsize, ))
difference = maxsize - self._maxsize
self._semaphore.counter += difference
self._maxsize = maxsize
self.adjust()
# make sure all currently blocking spawn() start unlocking if maxsize increased
self._semaphore._start_notify()
def _get_maxsize(self):
return self._maxsize
maxsize = property(_get_maxsize, _set_maxsize)
def __repr__(self):
return '<%s at 0x%x %s/%s/%s>' % (self.__class__.__name__, id(self), len(self), self.size, self.maxsize)
def __len__(self):
# XXX just do unfinished_tasks property
return self.task_queue.unfinished_tasks
def _get_size(self):
return self._size
def _set_size(self, size):
if size < 0:
raise ValueError('Size of the pool cannot be negative: %r' % (size, ))
if size > self._maxsize:
raise ValueError('Size of the pool cannot be bigger than maxsize: %r > %r' % (size, self._maxsize))
if self.manager:
self.manager.kill()
while self._size < size:
self._add_thread()
delay = 0.0001
while self._size > size:
while self._size - size > self.task_queue.unfinished_tasks:
self.task_queue.put(None)
if getcurrent() is self.hub:
break
sleep(delay)
delay = min(delay * 2, .05)
if self._size:
self.fork_watcher.start(self._on_fork)
else:
self.fork_watcher.stop()
size = property(_get_size, _set_size)
def _init(self, maxsize):
self._size = 0
self._semaphore = Semaphore(1)
self._lock = Lock()
self.task_queue = Queue()
self._set_maxsize(maxsize)
def _on_fork(self):
# fork() only leaves one thread; also screws up locks;
# let's re-create locks and threads
pid = os.getpid()
if pid != self.pid:
self.pid = pid
# Do not mix fork() and threads; since fork() only copies one thread
# all objects referenced by other threads has refcount that will never
# go down to 0.
self._init(self._maxsize)
def join(self):
delay = 0.0005
while self.task_queue.unfinished_tasks > 0:
sleep(delay)
delay = min(delay * 2, .05)
def kill(self):
self.size = 0
def _adjust_step(self):
# if there is a possibility & necessity for adding a thread, do it
while self._size < self._maxsize and self.task_queue.unfinished_tasks > self._size:
self._add_thread()
# while the number of threads is more than maxsize, kill one
# we do not check what's already in task_queue - it could be all Nones
while self._size - self._maxsize > self.task_queue.unfinished_tasks:
self.task_queue.put(None)
if self._size:
self.fork_watcher.start(self._on_fork)
else:
self.fork_watcher.stop()
def _adjust_wait(self):
delay = 0.0001
while True:
self._adjust_step()
if self._size <= self._maxsize:
return
sleep(delay)
delay = min(delay * 2, .05)
def adjust(self):
self._adjust_step()
if not self.manager and self._size > self._maxsize:
# might need to feed more Nones into the pool
self.manager = Greenlet.spawn(self._adjust_wait)
def _add_thread(self):
with self._lock:
self._size += 1
try:
start_new_thread(self._worker, ())
except:
with self._lock:
self._size -= 1
raise
def spawn(self, func, *args, **kwargs):
while True:
semaphore = self._semaphore
semaphore.acquire()
if semaphore is self._semaphore:
break
try:
task_queue = self.task_queue
result = AsyncResult()
thread_result = ThreadResult(result, hub=self.hub)
task_queue.put((func, args, kwargs, thread_result))
self.adjust()
# rawlink() must be the last call
result.rawlink(lambda *args: self._semaphore.release())
# XXX this _semaphore.release() is competing for order with get()
# XXX this is not good, just make ThreadResult release the semaphore before doing anything else
except:
semaphore.release()
raise
return result
def _decrease_size(self):
if sys is None:
return
_lock = getattr(self, '_lock', None)
if _lock is not None:
with _lock:
self._size -= 1
def _worker(self):
need_decrease = True
try:
while True:
task_queue = self.task_queue
task = task_queue.get()
try:
if task is None:
need_decrease = False
self._decrease_size()
# we want first to decrease size, then decrease unfinished_tasks
# otherwise, _adjust might think there's one more idle thread that
# needs to be killed
return
func, args, kwargs, thread_result = task
try:
value = func(*args, **kwargs)
except:
exc_info = getattr(sys, 'exc_info', None)
if exc_info is None:
return
thread_result.handle_error((self, func), exc_info())
else:
if sys is None:
return
thread_result.set(value)
del value
finally:
del func, args, kwargs, thread_result, task
finally:
if sys is None:
return
task_queue.task_done()
finally:
if need_decrease:
self._decrease_size()
# XXX apply() should re-raise error by default
# XXX because that's what builtin apply does
# XXX check gevent.pool.Pool.apply and multiprocessing.Pool.apply
def apply_e(self, expected_errors, function, args=None, kwargs=None):
if args is None:
args = ()
if kwargs is None:
kwargs = {}
success, result = self.spawn(wrap_errors, expected_errors, function, args, kwargs).get()
if success:
return result
result = [result]
raise result.pop()
def apply(self, func, args=None, kwds=None):
"""Equivalent of the apply() builtin function. It blocks till the result is ready."""
if args is None:
args = ()
if kwds is None:
kwds = {}
return self.spawn(func, *args, **kwds).get()
def apply_cb(self, func, args=None, kwds=None, callback=None):
result = self.apply(func, args, kwds)
if callback is not None:
callback(result)
return result
def apply_async(self, func, args=None, kwds=None, callback=None):
"""A variant of the apply() method which returns a Greenlet object.
If callback is specified then it should be a callable which accepts a single argument. When the result becomes ready
callback is applied to it (unless the call failed)."""
if args is None:
args = ()
if kwds is None:
kwds = {}
return Greenlet.spawn(self.apply_cb, func, args, kwds, callback)
def map(self, func, iterable):
return list(self.imap(func, iterable))
def map_cb(self, func, iterable, callback=None):
result = self.map(func, iterable)
if callback is not None:
callback(result)
return result
def map_async(self, func, iterable, callback=None):
"""
A variant of the map() method which returns a Greenlet object.
If callback is specified then it should be a callable which accepts a
single argument.
"""
return Greenlet.spawn(self.map_cb, func, iterable, callback)
def imap(self, func, iterable):
"""An equivalent of itertools.imap()"""
return IMap.spawn(func, iterable, spawn=self.spawn)
def imap_unordered(self, func, iterable):
"""The same as imap() except that the ordering of the results from the
returned iterator should be considered in arbitrary order."""
return IMapUnordered.spawn(func, iterable, spawn=self.spawn)
class Pool(Group):
def __init__(self, size=None, greenlet_class=None):
"""
Create a new pool.
A pool is like a group, but the maximum number of members
is governed by the *size* parameter.
:keyword int size: If given, this non-negative integer is the
maximum count of active greenlets that will be allowed in
this pool. A few values have special significance:
* ``None`` (the default) places no limit on the number of
greenlets. This is useful when you need to track, but not limit,
greenlets, as with :class:`gevent.pywsgi.WSGIServer`. A :class:`Group`
may be a more efficient way to achieve the same effect.
* ``0`` creates a pool that can never have any active greenlets. Attempting
to spawn in this pool will block forever. This is only useful
if an application uses :meth:`wait_available` with a timeout and checks
:meth:`free_count` before attempting to spawn.
"""
if size is not None and size < 0:
raise ValueError('size must not be negative: %r' % (size, ))
Group.__init__(self)
self.size = size
if greenlet_class is not None:
self.greenlet_class = greenlet_class
if size is None:
self._semaphore = DummySemaphore()
else:
self._semaphore = Semaphore(size)
def wait_available(self, timeout=None):
"""
Wait until it's possible to spawn a greenlet in this pool.
:param float timeout: If given, only wait the specified number
of seconds.
.. warning:: If the pool was initialized with a size of 0, this
method will block forever unless a timeout is given.
:return: A number indicating how many new greenlets can be put into
the pool without blocking.
.. versionchanged:: 1.1a3
Added the ``timeout`` parameter.
"""
return self._semaphore.wait(timeout=timeout)
def full(self):
"""
Return a boolean indicating whether this pool has any room for
members. (True if it does, False if it doesn't.)
"""
return self.free_count() <= 0
def free_count(self):
"""
Return a number indicating *approximately* how many more members
can be added to this pool.
"""
if self.size is None:
return 1
return max(0, self.size - len(self))
def add(self, greenlet):
"""
Begin tracking the given greenlet, blocking until space is available.
.. seealso:: :meth:`Group.add`
"""
self._semaphore.acquire()
try:
Group.add(self, greenlet)
except:
self._semaphore.release()
raise
def _discard(self, greenlet):
Group._discard(self, greenlet)
self._semaphore.release()
class VncKombuClientBase(object):
def _update_sandesh_status(self, status, msg=''):
ConnectionState.update(conn_type=ConnectionType.DATABASE,
name='RabbitMQ', status=status, message=msg,
server_addrs=["%s:%s" % (self._rabbit_ip, self._rabbit_port)])
# end _update_sandesh_status
def publish(self, message):
self._publish_queue.put(message)
# end publish
def __init__(self, rabbit_ip, rabbit_port, rabbit_user, rabbit_password,
rabbit_vhost, rabbit_ha_mode, q_name, subscribe_cb, logger):
self._rabbit_ip = rabbit_ip
self._rabbit_port = rabbit_port
self._rabbit_user = rabbit_user
self._rabbit_password = rabbit_password
self._rabbit_vhost = rabbit_vhost
self._subscribe_cb = subscribe_cb
self._logger = logger
self._publish_queue = Queue()
self._conn_lock = Semaphore()
self.obj_upd_exchange = kombu.Exchange('vnc_config.object-update', 'fanout',
durable=False)
def num_pending_messages(self):
return self._publish_queue.qsize()
# end num_pending_messages
def prepare_to_consume(self):
# override this method
return
def _reconnect(self, delete_old_q=False):
if self._conn_lock.locked():
# either connection-monitor or publisher should have taken
# the lock. The one who acquired the lock would re-establish
# the connection and releases the lock, so the other one can
# just wait on the lock, till it gets released
self._conn_lock.wait()
return
self._conn_lock.acquire()
msg = "RabbitMQ connection down"
self._logger(msg, level=SandeshLevel.SYS_ERR)
self._update_sandesh_status(ConnectionStatus.DOWN)
self._conn_state = ConnectionStatus.DOWN
self._conn.close()
self._conn.ensure_connection()
self._conn.connect()
self._update_sandesh_status(ConnectionStatus.UP)
self._conn_state = ConnectionStatus.UP
msg = 'RabbitMQ connection ESTABLISHED %s' % repr(self._conn)
self._logger(msg, level=SandeshLevel.SYS_NOTICE)
self._channel = self._conn.channel()
if delete_old_q:
# delete the old queue in first-connect context
# as db-resync would have caught up with history.
try:
bound_q = self._update_queue_obj(self._channel)
bound_q.delete()
except Exception as e:
msg = 'Unable to delete the old ampq queue: %s' %(str(e))
self._logger(msg, level=SandeshLevel.SYS_ERR)
self._consumer = kombu.Consumer(self._channel,
queues=self._update_queue_obj,
callbacks=[self._subscribe])
self._producer = kombu.Producer(self._channel, exchange=self.obj_upd_exchange)
self._conn_lock.release()
# end _reconnect
def _connection_watch(self):
self.prepare_to_consume()
while True:
try:
self._consumer.consume()
self._conn.drain_events()
except self._conn.connection_errors + self._conn.channel_errors as e:
self._reconnect()
# end _connection_watch
def _publisher(self):
message = None
while True:
try:
if not message:
# earlier was sent fine, dequeue one more
message = self._publish_queue.get()
while True:
try:
self._producer.publish(message)
message = None
break
except self._conn.connection_errors + self._conn.channel_errors as e:
self._reconnect()
except Exception as e:
log_str = "Unknown exception in _publisher greenlet" + str(e)
self._logger(log_str, level=SandeshLevel.SYS_ERR)
# end _publisher
def _subscribe(self, body, message):
try:
self._subscribe_cb(body)
finally:
message.ack()
def _start(self):
self._reconnect(delete_old_q=True)
self._publisher_greenlet = gevent.spawn(self._publisher)
self._connection_monitor_greenlet = gevent.spawn(self._connection_watch)
def shutdown(self):
self._publisher_greenlet.kill()
self._connection_monitor_greenlet.kill()
self._producer.close()
self._consumer.close()
self._conn.close()
class Rotkehlchen():
def __init__(self, args: argparse.Namespace) -> None:
"""Initialize the Rotkehlchen object
May Raise:
- SystemPermissionError if the given data directory's permissions
are not correct.
"""
self.lock = Semaphore()
self.lock.acquire()
# Can also be None after unlock if premium credentials did not
# authenticate or premium server temporarily offline
self.premium: Optional[Premium] = None
self.user_is_logged_in: bool = False
configure_logging(args)
self.sleep_secs = args.sleep_secs
if args.data_dir is None:
self.data_dir = default_data_directory()
else:
self.data_dir = Path(args.data_dir)
if not os.access(self.data_dir, os.W_OK | os.R_OK):
raise SystemPermissionError(
f'The given data directory {self.data_dir} is not readable or writable',
)
self.main_loop_spawned = False
self.args = args
self.api_task_greenlets: List[gevent.Greenlet] = []
self.msg_aggregator = MessagesAggregator()
self.greenlet_manager = GreenletManager(
msg_aggregator=self.msg_aggregator)
self.exchange_manager = ExchangeManager(
msg_aggregator=self.msg_aggregator)
# Initialize the AssetResolver singleton
AssetResolver(data_directory=self.data_dir)
self.data = DataHandler(self.data_dir, self.msg_aggregator)
self.cryptocompare = Cryptocompare(data_directory=self.data_dir,
database=None)
self.coingecko = Coingecko(data_directory=self.data_dir)
self.icon_manager = IconManager(data_dir=self.data_dir,
coingecko=self.coingecko)
self.greenlet_manager.spawn_and_track(
after_seconds=None,
task_name='periodically_query_icons_until_all_cached',
exception_is_error=False,
method=self.icon_manager.periodically_query_icons_until_all_cached,
batch_size=ICONS_BATCH_SIZE,
sleep_time_secs=ICONS_QUERY_SLEEP,
)
# Initialize the Inquirer singleton
Inquirer(
data_dir=self.data_dir,
cryptocompare=self.cryptocompare,
coingecko=self.coingecko,
)
# Keeps how many trades we have found per location. Used for free user limiting
self.actions_per_location: Dict[str, Dict[Location, int]] = {
'trade': defaultdict(int),
'asset_movement': defaultdict(int),
}
self.lock.release()
self.task_manager: Optional[TaskManager] = None
self.shutdown_event = gevent.event.Event()
def reset_after_failed_account_creation_or_login(self) -> None:
"""If the account creation or login failed make sure that the Rotki instance is clear
Tricky instances are when after either failed premium credentials or user refusal
to sync premium databases we relogged in.
"""
self.cryptocompare.db = None
def unlock_user(
self,
user: str,
password: str,
create_new: bool,
sync_approval: Literal['yes', 'no', 'unknown'],
premium_credentials: Optional[PremiumCredentials],
initial_settings: Optional[ModifiableDBSettings] = None,
) -> None:
"""Unlocks an existing user or creates a new one if `create_new` is True
May raise:
- PremiumAuthenticationError if the password can't unlock the database.
- AuthenticationError if premium_credentials are given and are invalid
or can't authenticate with the server
- DBUpgradeError if the rotki DB version is newer than the software or
there is a DB upgrade and there is an error.
- SystemPermissionError if the directory or DB file can not be accessed
"""
log.info(
'Unlocking user',
user=user,
create_new=create_new,
sync_approval=sync_approval,
initial_settings=initial_settings,
)
# unlock or create the DB
self.password = password
self.user_directory = self.data.unlock(user, password, create_new,
initial_settings)
self.data_importer = DataImporter(db=self.data.db)
self.last_data_upload_ts = self.data.db.get_last_data_upload_ts()
self.premium_sync_manager = PremiumSyncManager(data=self.data,
password=password)
# set the DB in the external services instances that need it
self.cryptocompare.set_database(self.data.db)
# Anything that was set above here has to be cleaned in case of failure in the next step
# by reset_after_failed_account_creation_or_login()
try:
self.premium = self.premium_sync_manager.try_premium_at_start(
given_premium_credentials=premium_credentials,
username=user,
create_new=create_new,
sync_approval=sync_approval,
)
except PremiumAuthenticationError:
# Reraise it only if this is during the creation of a new account where
# the premium credentials were given by the user
if create_new:
raise
self.msg_aggregator.add_warning(
'Could not authenticate the Rotki premium API keys found in the DB.'
' Has your subscription expired?', )
# else let's just continue. User signed in succesfully, but he just
# has unauthenticable/invalid premium credentials remaining in his DB
settings = self.get_settings()
self.greenlet_manager.spawn_and_track(
after_seconds=None,
task_name='submit_usage_analytics',
exception_is_error=False,
method=maybe_submit_usage_analytics,
should_submit=settings.submit_usage_analytics,
)
self.etherscan = Etherscan(database=self.data.db,
msg_aggregator=self.msg_aggregator)
self.beaconchain = BeaconChain(database=self.data.db,
msg_aggregator=self.msg_aggregator)
eth_rpc_endpoint = settings.eth_rpc_endpoint
# Initialize the price historian singleton
PriceHistorian(
data_directory=self.data_dir,
cryptocompare=self.cryptocompare,
coingecko=self.coingecko,
)
PriceHistorian().set_oracles_order(settings.historical_price_oracles)
self.accountant = Accountant(
db=self.data.db,
user_directory=self.user_directory,
msg_aggregator=self.msg_aggregator,
create_csv=True,
)
# Initialize the rotkehlchen logger
LoggingSettings(anonymized_logs=settings.anonymized_logs)
exchange_credentials = self.data.db.get_exchange_credentials()
self.exchange_manager.initialize_exchanges(
exchange_credentials=exchange_credentials,
database=self.data.db,
)
# Initialize blockchain querying modules
ethereum_manager = EthereumManager(
ethrpc_endpoint=eth_rpc_endpoint,
etherscan=self.etherscan,
database=self.data.db,
msg_aggregator=self.msg_aggregator,
greenlet_manager=self.greenlet_manager,
connect_at_start=ETHEREUM_NODES_TO_CONNECT_AT_START,
)
kusama_manager = SubstrateManager(
chain=SubstrateChain.KUSAMA,
msg_aggregator=self.msg_aggregator,
greenlet_manager=self.greenlet_manager,
connect_at_start=KUSAMA_NODES_TO_CONNECT_AT_START,
connect_on_startup=self._connect_ksm_manager_on_startup(),
own_rpc_endpoint=settings.ksm_rpc_endpoint,
)
Inquirer().inject_ethereum(ethereum_manager)
Inquirer().set_oracles_order(settings.current_price_oracles)
self.chain_manager = ChainManager(
blockchain_accounts=self.data.db.get_blockchain_accounts(),
ethereum_manager=ethereum_manager,
kusama_manager=kusama_manager,
msg_aggregator=self.msg_aggregator,
database=self.data.db,
greenlet_manager=self.greenlet_manager,
premium=self.premium,
eth_modules=settings.active_modules,
data_directory=self.data_dir,
beaconchain=self.beaconchain,
btc_derivation_gap_limit=settings.btc_derivation_gap_limit,
)
self.events_historian = EventsHistorian(
user_directory=self.user_directory,
db=self.data.db,
msg_aggregator=self.msg_aggregator,
exchange_manager=self.exchange_manager,
chain_manager=self.chain_manager,
)
self.task_manager = TaskManager(
max_tasks_num=DEFAULT_MAX_TASKS_NUM,
greenlet_manager=self.greenlet_manager,
api_task_greenlets=self.api_task_greenlets,
database=self.data.db,
cryptocompare=self.cryptocompare,
premium_sync_manager=self.premium_sync_manager,
chain_manager=self.chain_manager,
exchange_manager=self.exchange_manager,
)
self.user_is_logged_in = True
log.debug('User unlocking complete')
def logout(self) -> None:
if not self.user_is_logged_in:
return
user = self.data.username
log.info(
'Logging out user',
user=user,
)
self.greenlet_manager.clear()
del self.chain_manager
self.exchange_manager.delete_all_exchanges()
# Reset rotkehlchen logger to default
LoggingSettings(anonymized_logs=DEFAULT_ANONYMIZED_LOGS)
del self.accountant
del self.events_historian
del self.data_importer
if self.premium is not None:
del self.premium
self.data.logout()
self.password = ''
self.cryptocompare.unset_database()
# Make sure no messages leak to other user sessions
self.msg_aggregator.consume_errors()
self.msg_aggregator.consume_warnings()
self.task_manager = None
self.user_is_logged_in = False
log.info(
'User successfully logged out',
user=user,
)
def set_premium_credentials(self, credentials: PremiumCredentials) -> None:
"""
Sets the premium credentials for Rotki
Raises PremiumAuthenticationError if the given key is rejected by the Rotkehlchen server
"""
log.info('Setting new premium credentials')
if self.premium is not None:
self.premium.set_credentials(credentials)
else:
self.premium = premium_create_and_verify(credentials)
self.data.db.set_rotkehlchen_premium(credentials)
def delete_premium_credentials(self) -> Tuple[bool, str]:
"""Deletes the premium credentials for Rotki"""
msg = ''
success = self.data.db.del_rotkehlchen_premium()
if success is False:
msg = 'The database was unable to delete the Premium keys for the logged-in user'
self.deactivate_premium_status()
return success, msg
def deactivate_premium_status(self) -> None:
"""Deactivate premium in the current session"""
self.premium = None
self.premium_sync_manager.premium = None
self.chain_manager.deactivate_premium_status()
def start(self) -> gevent.Greenlet:
assert not self.main_loop_spawned, 'Tried to spawn the main loop twice'
greenlet = gevent.spawn(self.main_loop)
self.main_loop_spawned = True
return greenlet
def main_loop(self) -> None:
"""Rotki main loop that fires often and runs the task manager's scheduler"""
while self.shutdown_event.wait(
timeout=MAIN_LOOP_SECS_DELAY) is not True:
if self.task_manager is not None:
self.task_manager.schedule()
def get_blockchain_account_data(
self,
blockchain: SupportedBlockchain,
) -> Union[List[BlockchainAccountData], Dict[str, Any]]:
account_data = self.data.db.get_blockchain_account_data(blockchain)
if blockchain != SupportedBlockchain.BITCOIN:
return account_data
xpub_data = self.data.db.get_bitcoin_xpub_data()
addresses_to_account_data = {x.address: x for x in account_data}
address_to_xpub_mappings = self.data.db.get_addresses_to_xpub_mapping(
list(addresses_to_account_data.keys()), # type: ignore
)
xpub_mappings: Dict['XpubData', List[BlockchainAccountData]] = {}
for address, xpub_entry in address_to_xpub_mappings.items():
if xpub_entry not in xpub_mappings:
xpub_mappings[xpub_entry] = []
xpub_mappings[xpub_entry].append(
addresses_to_account_data[address])
data: Dict[str, Any] = {'standalone': [], 'xpubs': []}
# Add xpub data
for xpub_entry in xpub_data:
data_entry = xpub_entry.serialize()
addresses = xpub_mappings.get(xpub_entry, None)
data_entry['addresses'] = addresses if addresses and len(
addresses) != 0 else None
data['xpubs'].append(data_entry)
# Add standalone addresses
for account in account_data:
if account.address not in address_to_xpub_mappings:
data['standalone'].append(account)
return data
def add_blockchain_accounts(
self,
blockchain: SupportedBlockchain,
account_data: List[BlockchainAccountData],
) -> BlockchainBalancesUpdate:
"""Adds new blockchain accounts
Adds the accounts to the blockchain instance and queries them to get the
updated balances. Also adds them in the DB
May raise:
- EthSyncError from modify_blockchain_account
- InputError if the given accounts list is empty.
- TagConstraintError if any of the given account data contain unknown tags.
- RemoteError if an external service such as Etherscan is queried and
there is a problem with its query.
"""
self.data.db.ensure_tags_exist(
given_data=account_data,
action='adding',
data_type='blockchain accounts',
)
address_type = blockchain.get_address_type()
updated_balances = self.chain_manager.add_blockchain_accounts(
blockchain=blockchain,
accounts=[address_type(entry.address) for entry in account_data],
)
self.data.db.add_blockchain_accounts(
blockchain=blockchain,
account_data=account_data,
)
return updated_balances
def edit_blockchain_accounts(
self,
blockchain: SupportedBlockchain,
account_data: List[BlockchainAccountData],
) -> None:
"""Edits blockchain accounts
Edits blockchain account data for the given accounts
May raise:
- InputError if the given accounts list is empty or if
any of the accounts to edit do not exist.
- TagConstraintError if any of the given account data contain unknown tags.
"""
# First check for validity of account data addresses
if len(account_data) == 0:
raise InputError(
'Empty list of blockchain account data to edit was given')
accounts = [x.address for x in account_data]
unknown_accounts = set(accounts).difference(
self.chain_manager.accounts.get(blockchain))
if len(unknown_accounts) != 0:
raise InputError(
f'Tried to edit unknown {blockchain.value} '
f'accounts {",".join(unknown_accounts)}', )
self.data.db.ensure_tags_exist(
given_data=account_data,
action='editing',
data_type='blockchain accounts',
)
# Finally edit the accounts
self.data.db.edit_blockchain_accounts(
blockchain=blockchain,
account_data=account_data,
)
def remove_blockchain_accounts(
self,
blockchain: SupportedBlockchain,
accounts: ListOfBlockchainAddresses,
) -> BlockchainBalancesUpdate:
"""Removes blockchain accounts
Removes the accounts from the blockchain instance and queries them to get
the updated balances. Also removes them from the DB
May raise:
- RemoteError if an external service such as Etherscan is queried and
there is a problem with its query.
- InputError if a non-existing account was given to remove
"""
balances_update = self.chain_manager.remove_blockchain_accounts(
blockchain=blockchain,
accounts=accounts,
)
self.data.db.remove_blockchain_accounts(blockchain, accounts)
return balances_update
def get_history_query_status(self) -> Dict[str, str]:
if self.events_historian.progress < FVal('100'):
processing_state = self.events_historian.processing_state_name
progress = self.events_historian.progress / 2
elif self.accountant.first_processed_timestamp == -1:
processing_state = 'Processing all retrieved historical events'
progress = FVal(50)
else:
processing_state = 'Processing all retrieved historical events'
# start_ts is min of the query start or the first action timestamp since action
# processing can start well before query start to calculate cost basis
start_ts = min(
self.accountant.events.query_start_ts,
self.accountant.first_processed_timestamp,
)
diff = self.accountant.events.query_end_ts - start_ts
progress = 50 + 100 * (
FVal(self.accountant.currently_processing_timestamp - start_ts)
/ FVal(diff) / 2)
return {
'processing_state': str(processing_state),
'total_progress': str(progress)
}
def process_history(
self,
start_ts: Timestamp,
end_ts: Timestamp,
) -> Tuple[Dict[str, Any], str]:
(
error_or_empty,
history,
loan_history,
asset_movements,
eth_transactions,
defi_events,
ledger_actions,
) = self.events_historian.get_history(
start_ts=start_ts,
end_ts=end_ts,
has_premium=self.premium is not None,
)
result = self.accountant.process_history(
start_ts=start_ts,
end_ts=end_ts,
trade_history=history,
loan_history=loan_history,
asset_movements=asset_movements,
eth_transactions=eth_transactions,
defi_events=defi_events,
ledger_actions=ledger_actions,
)
return result, error_or_empty
@overload
def _apply_actions_limit(
self,
location: Location,
action_type: Literal['trade'],
location_actions: TRADES_LIST,
all_actions: TRADES_LIST,
) -> TRADES_LIST:
...
@overload
def _apply_actions_limit(
self,
location: Location,
action_type: Literal['asset_movement'],
location_actions: List[AssetMovement],
all_actions: List[AssetMovement],
) -> List[AssetMovement]:
...
def _apply_actions_limit(
self,
location: Location,
action_type: Literal['trade', 'asset_movement'],
location_actions: Union[TRADES_LIST, List[AssetMovement]],
all_actions: Union[TRADES_LIST, List[AssetMovement]],
) -> Union[TRADES_LIST, List[AssetMovement]]:
"""Take as many actions from location actions and add them to all actions as the limit permits
Returns the modified (or not) all_actions
"""
# If we are already at or above the limit return current actions disregarding this location
actions_mapping = self.actions_per_location[action_type]
current_num_actions = sum(x for _, x in actions_mapping.items())
limit = LIMITS_MAPPING[action_type]
if current_num_actions >= limit:
return all_actions
# Find out how many more actions can we return, and depending on that get
# the number of actions from the location actions and add them to the total
remaining_num_actions = limit - current_num_actions
if remaining_num_actions < 0:
remaining_num_actions = 0
num_actions_to_take = min(len(location_actions), remaining_num_actions)
actions_mapping[location] = num_actions_to_take
all_actions.extend(
location_actions[0:num_actions_to_take]) # type: ignore
return all_actions
def query_trades(
self,
from_ts: Timestamp,
to_ts: Timestamp,
location: Optional[Location],
) -> TRADES_LIST:
"""Queries trades for the given location and time range.
If no location is given then all external, all exchange and DEX trades are queried.
DEX Trades are queried only if the user has premium
If the user does not have premium then a trade limit is applied.
May raise:
- RemoteError: If there are problems connecting to any of the remote exchanges
"""
trades: TRADES_LIST
if location is not None:
trades = self.query_location_trades(from_ts, to_ts, location)
else:
trades = self.query_location_trades(from_ts, to_ts,
Location.EXTERNAL)
# crypto.com is not an API key supported exchange but user can import from CSV
trades.extend(
self.query_location_trades(from_ts, to_ts, Location.CRYPTOCOM))
for name, exchange in self.exchange_manager.connected_exchanges.items(
):
exchange_trades = exchange.query_trade_history(
start_ts=from_ts, end_ts=to_ts)
if self.premium is None:
trades = self._apply_actions_limit(
location=deserialize_location(name),
action_type='trade',
location_actions=exchange_trades,
all_actions=trades,
)
else:
trades.extend(exchange_trades)
# for all trades we also need uniswap trades
if self.premium is not None:
uniswap = self.chain_manager.uniswap
if uniswap is not None:
trades.extend(
uniswap.get_trades(
addresses=self.chain_manager.
queried_addresses_for_module('uniswap'),
from_timestamp=from_ts,
to_timestamp=to_ts,
), )
# return trades with most recent first
trades.sort(key=lambda x: x.timestamp, reverse=True)
return trades
def query_location_trades(
self,
from_ts: Timestamp,
to_ts: Timestamp,
location: Location,
) -> TRADES_LIST:
# clear the trades queried for this location
self.actions_per_location['trade'][location] = 0
location_trades: TRADES_LIST
if location in (Location.EXTERNAL, Location.CRYPTOCOM):
location_trades = self.data.db.get_trades( # type: ignore # list invariance
from_ts=from_ts,
to_ts=to_ts,
location=location,
)
elif location == Location.UNISWAP:
if self.premium is not None:
uniswap = self.chain_manager.uniswap
if uniswap is not None:
location_trades = uniswap.get_trades( # type: ignore # list invariance
addresses=self.chain_manager.
queried_addresses_for_module('uniswap'),
from_timestamp=from_ts,
to_timestamp=to_ts,
)
else:
# should only be an exchange
exchange = self.exchange_manager.get(str(location))
if not exchange:
logger.warning(
f'Tried to query trades from {location} which is either not an '
f'exchange or not an exchange the user has connected to', )
return []
location_trades = exchange.query_trade_history(start_ts=from_ts,
end_ts=to_ts)
trades: TRADES_LIST = []
if self.premium is None:
trades = self._apply_actions_limit(
location=location,
action_type='trade',
location_actions=location_trades,
all_actions=trades,
)
else:
trades = location_trades
return trades
def query_balances(
self,
requested_save_data: bool = False,
timestamp: Timestamp = None,
ignore_cache: bool = False,
) -> Dict[str, Any]:
"""Query all balances rotkehlchen can see.
If requested_save_data is True then the data are always saved in the DB,
if it is False then data are saved if self.data.should_save_balances()
is True.
If timestamp is None then the current timestamp is used.
If a timestamp is given then that is the time that the balances are going
to be saved in the DB
If ignore_cache is True then all underlying calls that have a cache ignore it
Returns a dictionary with the queried balances.
"""
log.info('query_balances called',
requested_save_data=requested_save_data)
balances: Dict[str, Dict[Asset, Balance]] = {}
problem_free = True
for _, exchange in self.exchange_manager.connected_exchanges.items():
exchange_balances, _ = exchange.query_balances(
ignore_cache=ignore_cache)
# If we got an error, disregard that exchange but make sure we don't save data
if not isinstance(exchange_balances, dict):
problem_free = False
else:
balances[exchange.name] = exchange_balances
liabilities: Dict[Asset, Balance]
try:
blockchain_result = self.chain_manager.query_balances(
blockchain=None,
force_token_detection=ignore_cache,
ignore_cache=ignore_cache,
)
balances[str(
Location.BLOCKCHAIN)] = blockchain_result.totals.assets
liabilities = blockchain_result.totals.liabilities
except (RemoteError, EthSyncError) as e:
problem_free = False
liabilities = {}
log.error(f'Querying blockchain balances failed due to: {str(e)}')
balances = account_for_manually_tracked_balances(db=self.data.db,
balances=balances)
# Calculate usd totals
assets_total_balance: DefaultDict[Asset,
Balance] = defaultdict(Balance)
total_usd_per_location: Dict[str, FVal] = {}
for location, asset_balance in balances.items():
total_usd_per_location[location] = ZERO
for asset, balance in asset_balance.items():
assets_total_balance[asset] += balance
total_usd_per_location[location] += balance.usd_value
net_usd = sum((balance.usd_value
for _, balance in assets_total_balance.items()), ZERO)
liabilities_total_usd = sum(
(liability.usd_value for _, liability in liabilities.items()),
ZERO) # noqa: E501
net_usd -= liabilities_total_usd
# Calculate location stats
location_stats: Dict[str, Any] = {}
for location, total_usd in total_usd_per_location.items():
if location == str(Location.BLOCKCHAIN):
total_usd -= liabilities_total_usd
percentage = (total_usd /
net_usd).to_percentage() if net_usd != ZERO else '0%'
location_stats[location] = {
'usd_value': total_usd,
'percentage_of_net_value': percentage,
}
# Calculate 'percentage_of_net_value' per asset
assets_total_balance_as_dict: Dict[Asset, Dict[str, Any]] = {
asset: balance.to_dict()
for asset, balance in assets_total_balance.items()
}
liabilities_as_dict: Dict[Asset, Dict[str, Any]] = {
asset: balance.to_dict()
for asset, balance in liabilities.items()
}
for asset, balance_dict in assets_total_balance_as_dict.items():
percentage = (balance_dict['usd_value'] / net_usd).to_percentage(
) if net_usd != ZERO else '0%' # noqa: E501
assets_total_balance_as_dict[asset][
'percentage_of_net_value'] = percentage
for asset, balance_dict in liabilities_as_dict.items():
percentage = (balance_dict['usd_value'] / net_usd).to_percentage(
) if net_usd != ZERO else '0%' # noqa: E501
liabilities_as_dict[asset]['percentage_of_net_value'] = percentage
# Compose balances response
result_dict = {
'assets': assets_total_balance_as_dict,
'liabilities': liabilities_as_dict,
'location': location_stats,
'net_usd': net_usd,
}
allowed_to_save = requested_save_data or self.data.should_save_balances(
)
if problem_free and allowed_to_save:
if not timestamp:
timestamp = Timestamp(int(time.time()))
self.data.db.save_balances_data(data=result_dict,
timestamp=timestamp)
log.debug('query_balances data saved')
else:
log.debug(
'query_balances data not saved',
allowed_to_save=allowed_to_save,
problem_free=problem_free,
)
return result_dict
def _query_exchange_asset_movements(
self,
from_ts: Timestamp,
to_ts: Timestamp,
all_movements: List[AssetMovement],
exchange: Union[ExchangeInterface, Location],
) -> List[AssetMovement]:
if isinstance(exchange, ExchangeInterface):
location = deserialize_location(exchange.name)
# clear the asset movements queried for this exchange
self.actions_per_location['asset_movement'][location] = 0
location_movements = exchange.query_deposits_withdrawals(
start_ts=from_ts,
end_ts=to_ts,
)
else:
assert isinstance(exchange,
Location), 'only a location should make it here'
assert exchange == Location.CRYPTOCOM, 'only cryptocom should make it here'
location = exchange
# cryptocom has no exchange integration but we may have DB entries
self.actions_per_location['asset_movement'][location] = 0
location_movements = self.data.db.get_asset_movements(
from_ts=from_ts,
to_ts=to_ts,
location=location,
)
movements: List[AssetMovement] = []
if self.premium is None:
movements = self._apply_actions_limit(
location=location,
action_type='asset_movement',
location_actions=location_movements,
all_actions=all_movements,
)
else:
all_movements.extend(location_movements)
movements = all_movements
return movements
def query_asset_movements(
self,
from_ts: Timestamp,
to_ts: Timestamp,
location: Optional[Location],
) -> List[AssetMovement]:
"""Queries AssetMovements for the given location and time range.
If no location is given then all exchange asset movements are queried.
If the user does not have premium then a limit is applied.
May raise:
- RemoteError: If there are problems connecting to any of the remote exchanges
"""
movements: List[AssetMovement] = []
if location is not None:
if location == Location.CRYPTOCOM:
movements = self._query_exchange_asset_movements(
from_ts=from_ts,
to_ts=to_ts,
all_movements=movements,
exchange=Location.CRYPTOCOM,
)
else:
exchange = self.exchange_manager.get(str(location))
if not exchange:
logger.warning(
f'Tried to query deposits/withdrawals from {location} which is either '
f'not at exchange or not an exchange the user has connected to',
)
return []
movements = self._query_exchange_asset_movements(
from_ts=from_ts,
to_ts=to_ts,
all_movements=movements,
exchange=exchange,
)
else:
# cryptocom has no exchange integration but we may have DB entries due to csv import
movements = self._query_exchange_asset_movements(
from_ts=from_ts,
to_ts=to_ts,
all_movements=movements,
exchange=Location.CRYPTOCOM,
)
for _, exchange in self.exchange_manager.connected_exchanges.items(
):
self._query_exchange_asset_movements(
from_ts=from_ts,
to_ts=to_ts,
all_movements=movements,
exchange=exchange,
)
# return movements with most recent first
movements.sort(key=lambda x: x.timestamp, reverse=True)
return movements
def set_settings(self, settings: ModifiableDBSettings) -> Tuple[bool, str]:
"""Tries to set new settings. Returns True in success or False with message if error"""
with self.lock:
if settings.eth_rpc_endpoint is not None:
result, msg = self.chain_manager.set_eth_rpc_endpoint(
settings.eth_rpc_endpoint)
if not result:
return False, msg
if settings.ksm_rpc_endpoint is not None:
result, msg = self.chain_manager.set_ksm_rpc_endpoint(
settings.ksm_rpc_endpoint)
if not result:
return False, msg
if settings.kraken_account_type is not None:
kraken = self.exchange_manager.get('kraken')
if kraken:
kraken.set_account_type(
settings.kraken_account_type) # type: ignore
if settings.btc_derivation_gap_limit is not None:
self.chain_manager.btc_derivation_gap_limit = settings.btc_derivation_gap_limit
if settings.current_price_oracles is not None:
Inquirer().set_oracles_order(settings.current_price_oracles)
if settings.historical_price_oracles is not None:
PriceHistorian().set_oracles_order(
settings.historical_price_oracles)
self.data.db.set_settings(settings)
return True, ''
def get_settings(self) -> DBSettings:
"""Returns the db settings with a check whether premium is active or not"""
db_settings = self.data.db.get_settings(
have_premium=self.premium is not None)
return db_settings
def setup_exchange(
self,
name: str,
api_key: ApiKey,
api_secret: ApiSecret,
passphrase: Optional[str] = None,
) -> Tuple[bool, str]:
"""
Setup a new exchange with an api key and an api secret and optionally a passphrase
By default the api keys are always validated unless validate is False.
"""
is_success, msg = self.exchange_manager.setup_exchange(
name=name,
api_key=api_key,
api_secret=api_secret,
database=self.data.db,
passphrase=passphrase,
)
if is_success:
# Success, save the result in the DB
self.data.db.add_exchange(name,
api_key,
api_secret,
passphrase=passphrase)
return is_success, msg
def remove_exchange(self, name: str) -> Tuple[bool, str]:
if not self.exchange_manager.has_exchange(name):
return False, 'Exchange {} is not registered'.format(name)
self.exchange_manager.delete_exchange(name)
# Success, remove it also from the DB
self.data.db.remove_exchange(name)
self.data.db.delete_used_query_range_for_exchange(name)
return True, ''
def query_periodic_data(self) -> Dict[str, Union[bool, Timestamp]]:
"""Query for frequently changing data"""
result: Dict[str, Union[bool, Timestamp]] = {}
if self.user_is_logged_in:
result[
'last_balance_save'] = self.data.db.get_last_balance_save_time(
)
result[
'eth_node_connection'] = self.chain_manager.ethereum.web3_mapping.get(
NodeName.OWN, None) is not None # noqa : E501
result['last_data_upload_ts'] = Timestamp(
self.premium_sync_manager.last_data_upload_ts) # noqa : E501
return result
def shutdown(self) -> None:
self.logout()
self.shutdown_event.set()
def _connect_ksm_manager_on_startup(self) -> bool:
return bool(self.data.db.get_blockchain_accounts().ksm)
def create_oracle_cache(
self,
oracle: HistoricalPriceOracle,
from_asset: Asset,
to_asset: Asset,
purge_old: bool,
) -> None:
"""Creates the cache of the given asset pair from the start of time
until now for the given oracle.
if purge_old is true then any old cache in memory and in a file is purged
May raise:
- RemoteError if there is a problem reaching the oracle
- UnsupportedAsset if any of the two assets is not supported by the oracle
"""
if oracle != HistoricalPriceOracle.CRYPTOCOMPARE:
return # only for cryptocompare for now
self.cryptocompare.create_cache(from_asset, to_asset, purge_old)
def delete_oracle_cache(
self,
oracle: HistoricalPriceOracle,
from_asset: Asset,
to_asset: Asset,
) -> None:
if oracle != HistoricalPriceOracle.CRYPTOCOMPARE:
return # only for cryptocompare for now
self.cryptocompare.delete_cache(from_asset, to_asset)
def get_oracle_cache(
self, oracle: HistoricalPriceOracle) -> List[Dict[str, Any]]:
if oracle != HistoricalPriceOracle.CRYPTOCOMPARE:
return [] # only for cryptocompare for now
return self.cryptocompare.get_all_cache_data()
class Spider:
def __init__(self, student_shortcut, job_manager, db_manager):
self.shortcut = student_shortcut
self.job_manager = job_manager
self.db_manager = db_manager
self.term_course_mutex = Semaphore()
self.term_course_flags = set()
# 避免重复写数据库
# 学期, 专业, 计划, 班级学生关系记录 不会重复抓取
# 课程记录使用 课程代码 区分
# 课程学期教学班使用 学期代码 + 课程代码 区分
# 学生记录使用 学号 区分
# 三者不会有相同情况, 只用一个集合即可
# 在这里没有使用锁, 因此这个标记不一定会命中
# 使用锁会增加很多等待时间, 还不如直接更新数据库
# 并没有提升多少性能
# self._flags = {
# 'course': ('课程代码', Semaphore(), set()),
# 'student': ('学号', Semaphore(), set()),
# }
# 统计信息: 内存标志命中次数
# self._flags_hit_count = {
# 'course': 0,
# 'student': 0,
# }
@elapse_dec
def crawl(self, dfs_mode=False):
self.job_manager.jobs = (self.iter_term_and_major,
self.iter_term_and_course,
self.iter_teaching_class, self.sync_students)
logger.info('Crawl start!'.center(72, '='))
self.job_manager.start(dfs_mode)
logger.info(
'Jobs are all dispatched. Waiting for database requests handling.')
self.db_manager.join()
logger.info('Crawl finished!'.center(72, '='))
report_db(self.db_manager.db)
# 以下是任务
def iter_term_and_major(self):
# @structure {'专业': [{'专业代码': str, '专业名称': str}], '学期': [{'学期代码': str, '学期名称': str}]}
code = self.shortcut.get_code()
terms = code['学期']
majors = code['专业']
for term in terms:
term_code = term['学期代码']
self.db_manager.request(
'term', InsertOnNotExist({'学期代码': term_code}, term))
yield term_code, None
max_term_number = int(terms[-1]['学期代码'])
# 一些专业被删掉了, 因此很多记录都没了= =
for major in majors:
major_code = major['专业代码']
self.db_manager.request(
'major', InsertOnNotExist({'专业代码': major_code}, major))
for i in term_range(major['专业名称'], max_term_number):
term_code = '%03d' % i
yield term_code, major_code
def iter_term_and_course(self, term_code, major_code=None):
if major_code:
courses = self.shortcut.get_teaching_plan(xqdm=term_code,
zydm=major_code)
for course in courses:
course_code = course['课程代码']
self.db_manager.request(
'course', InsertOnNotExist({'课程代码': course_code}, course))
plan_doc = {
'课程代码': course_code,
'学期代码': term_code,
'专业代码': major_code
}
self.db_manager.request('plan',
InsertOnNotExist(plan_doc, plan_doc))
yield term_code, course_code
else:
courses = self.shortcut.get_teaching_plan(xqdm=term_code, kclx='x')
for course in courses:
course_code = course['课程代码']
self.db_manager.request(
'course', InsertOnNotExist({'课程代码': course_code}, course))
yield term_code, course_code
def iter_teaching_class(self,
term_code,
course_code=None,
course_name=None):
if course_code is None:
is_new = True
else:
key = term_code + course_code
self.term_course_mutex.acquire()
is_new = key not in self.term_course_flags
if is_new:
self.term_course_flags.add(key)
# 在 if 内释放锁会导致出现重复键时锁无法释放
self.term_course_mutex.release()
if is_new:
# @structure [{'任课教师': str, '课程名称': str, '教学班号': str, 'c': str, '班级容量': int}]
classes = self.shortcut.search_course(xqdm=term_code,
kcdm=course_code,
kcmc=course_name)
for teaching_class in classes:
course_code = teaching_class['课程代码']
class_code = teaching_class['教学班号']
# @structure {'校区': str,'开课单位': str,'考核类型': str,'课程类型': str,'课程名称': str,'教学班号': str,
# '起止周': str, '时间地点': str,'学分': float,'性别限制': str,'优选范围': str,'禁选范围': str,'选中人数': int,'备 注': str}
class_info = self.db_manager.db['class'].find_one({
'学期代码':
term_code,
'课程代码':
course_code,
'教学班号':
class_code
})
if not class_info:
class_info = self.shortcut.get_class_info(xqdm=term_code,
kcdm=course_code,
jxbh=class_code)
class_info.update(teaching_class)
# 接口没有学期代码参数
class_info['学期代码'] = term_code
self.db_manager.request('class', InsertOne(class_info))
yield term_code, course_code, class_code
def sync_students(self, term_code, course_code, class_code):
# @structure {'学期': str, '班级名称': str, '学生': [{'姓名': str, '学号': int}]}
students = self.shortcut.get_class_students(xqdm=term_code,
kcdm=course_code,
jxbh=class_code)
# 可能没有结果
if students:
students = students['学生']
for student in students:
student_code = student['学号']
student_name = student['姓名']
student['性别'] = '女' if student_name.endswith('*') else '男'
student['姓名'] = student_name.rstrip('*')
self.db_manager.request(
'student', InsertOnNotExist({'学号': student_code}, student))
class_student_doc = {
'学期代码': term_code,
'课程代码': course_code,
'教学班号': class_code,
'学号': student_code
}
self.db_manager.request(
'class_student',
InsertOnNotExist(class_student_doc, class_student_doc))
class TransportChannel(object):
def __init__(self, uplink=None, downlink=None):
self.log = logging.getLogger("{module}.{name}".format(
module=self.__class__.__module__, name=self.__class__.__name__))
self.downlink = downlink
self.uplink = uplink
self.recv_callback = None
self.context = zmq.Context()
self.poller = zmq.Poller()
self.ul_socket = self.context.socket(zmq.SUB) # one SUB socket for uplink communication over topics
if sys.version_info.major >= 3:
self.ul_socket.setsockopt_string(zmq.SUBSCRIBE, "NEW_NODE")
self.ul_socket.setsockopt_string(zmq.SUBSCRIBE, "NODE_EXIT")
else:
self.ul_socket.setsockopt(zmq.SUBSCRIBE, "NEW_NODE")
self.ul_socket.setsockopt(zmq.SUBSCRIBE, "NODE_EXIT")
self.downlinkSocketLock = Semaphore(value=1)
self.dl_socket = self.context.socket(zmq.PUB) # one PUB socket for downlink communication over topics
#register UL socket in poller
self.poller.register(self.ul_socket, zmq.POLLIN)
self.importedPbClasses = {}
def set_downlink(self, downlink):
self.log.debug("Set Downlink: {}".format(downlink))
self.downlink = downlink
def set_uplink(self, uplink):
self.log.debug("Set Uplink: {}".format(uplink))
self.uplink = uplink
def subscribe_to(self, topic):
self.log.debug("Controller subscribes to topic: {}".format(topic))
if sys.version_info.major >= 3:
self.ul_socket.setsockopt_string(zmq.SUBSCRIBE, topic)
else:
self.ul_socket.setsockopt(zmq.SUBSCRIBE, topic)
def set_recv_callback(self, callback):
self.recv_callback = callback
def send_downlink_msg(self, msgContainer):
msgContainer[0] = msgContainer[0].encode('utf-8')
cmdDesc = msgContainer[1]
msg = msgContainer[2]
if cmdDesc.serialization_type == msgs.CmdDesc.PICKLE:
try:
msg = pickle.dumps(msg)
except:
msg = dill.dumps(msg)
elif cmdDesc.serialization_type == msgs.CmdDesc.PROTOBUF:
msg = msg.SerializeToString()
msgContainer[1] = cmdDesc.SerializeToString()
msgContainer[2] = msg
self.downlinkSocketLock.acquire()
try:
self.dl_socket.send_multipart(msgContainer)
finally:
self.downlinkSocketLock.release()
def deserialize_protobuff(self, message, typ):
class_ = self.importedPbClasses.get(typ, None)
if class_ is None:
module_, class_ = typ.rsplit('.', 1)
class_ = getattr(import_module(module_), class_)
self.importedPbClasses[typ] = class_
rv = class_()
rv.ParseFromString(message)
return rv
def start_receiving(self):
socks = dict(self.poller.poll())
if self.ul_socket in socks and socks[self.ul_socket] == zmq.POLLIN:
try:
msgContainer = self.ul_socket.recv_multipart(zmq.NOBLOCK)
except zmq.ZMQError:
raise zmq.ZMQError
assert len(msgContainer) == 3, msgContainer
dest = msgContainer[0]
cmdDesc = msgs.CmdDesc()
cmdDesc.ParseFromString(msgContainer[1])
msg = msgContainer[2]
if cmdDesc.serialization_type == msgs.CmdDesc.PICKLE:
try:
msg = pickle.loads(msg)
except:
msg = dill.loads(msg)
elif cmdDesc.serialization_type == msgs.CmdDesc.PROTOBUF:
if cmdDesc.pb_full_name:
msg = self.deserialize_protobuff(msg, cmdDesc.pb_full_name)
msgContainer[0] = dest.decode('utf-8')
msgContainer[1] = cmdDesc
msgContainer[2] = msg
self.recv_callback(msgContainer)
def start(self):
self.log.debug("Controller on DL-{}, UP-{}".format(self.downlink, self.uplink))
self.dl_socket.bind(self.downlink)
self.ul_socket.bind(self.uplink)
def stop(self):
self.ul_socket.setsockopt(zmq.LINGER, 0)
self.dl_socket.setsockopt(zmq.LINGER, 0)
self.ul_socket.close()
self.dl_socket.close()
self.context.term()
class Stream(object):
"""
参考tornado的iostream
"""
read_lock = None
write_lock = None
reading = False
writing = False
def __init__(self, sock, max_buffer_size=-1,
read_chunk_size=None, use_gevent=False, lock_mode=LOCK_MODE_RDWR):
self.sock = sock
self.max_buffer_size = max_buffer_size
self.read_chunk_size = read_chunk_size or READ_CHUNK_SIZE
self._read_buffer = collections.deque()
self._read_buffer_size = 0
self._read_delimiter = None
self._read_regex = None
self._read_bytes = None
self._read_until_close = False
self._read_checker = None
if use_gevent:
from gevent.lock import Semaphore as Lock
else:
from threading import Lock
if lock_mode & LOCK_MODE_READ:
self.read_lock = Lock()
if lock_mode & LOCK_MODE_WRITE:
self.write_lock = Lock()
def close(self, exc_info=False):
if self.closed():
# 如果已经关闭过，就直接返回了
return
self.close_fd()
def shutdown(self, how=2):
"""
gevent的close只是把sock替换为另一个类的实例。
这个实例的任何方法都会报错，但只有当真正调用recv、write或者有recv or send事件的时候，才会调用到这些函数，才可能检测到。
而我们在endpoint对应的函数里spawn_later一个新greenlet而不做join的话，connection的while循环此时已经开始read了。
之所以不把这个函数实现到connection，是因为shutdown更类似于触发一个close的事件
用shutdown可以直接触发. how: 0: SHUT_RD, 1: SHUT_WR, else: all
shutdown后还是会触发close事件以及相关的回调函数，不必担心
"""
if self.closed():
return
self.shutdown_fd(how)
@lock_read
def read_until_regex(self, regex):
"""Run when we read the given regex pattern.
"""
self._read_regex = re.compile(regex)
while 1:
ret, data = self._try_inline_read()
if ret <= 0:
return data
@lock_read
def read_until(self, delimiter):
"""
为了兼容调用的方法
"""
self._read_delimiter = delimiter
while 1:
ret, data = self._try_inline_read()
if ret <= 0:
return data
@lock_read
def read_bytes(self, num_bytes):
"""Run when we read the given number of bytes.
"""
assert isinstance(num_bytes, numbers.Integral)
self._read_bytes = num_bytes
while 1:
ret, data = self._try_inline_read()
if ret <= 0:
return data
@lock_read
def read_until_close(self):
"""Reads all data from the socket until it is closed.
"""
if self.closed():
return self._consume(self._read_buffer_size)
self._read_until_close = True
while 1:
ret, data = self._try_inline_read()
if ret <= 0:
return data
@lock_read
def read_with_checker(self, checker):
"""
checker(buf):
0 继续接收
>0 使用的长度
<0 异常
"""
self._read_checker = checker
while 1:
ret, data = self._try_inline_read()
if ret <= 0:
return data
@lock_write
def write(self, data):
"""
写数据
"""
if self.closed():
return False
while data:
num_bytes = self.write_to_fd(data)
if num_bytes is None:
return False
data = data[num_bytes:]
return True
def closed(self):
return not self.sock
def acquire_read_lock(self):
self.reading = True
if self.read_lock:
self.read_lock.acquire()
def release_read_lock(self):
self.reading = False
if self.read_lock:
self.read_lock.release()
def acquire_write_lock(self):
self.writing = True
if self.write_lock:
self.write_lock.acquire()
def release_write_lock(self):
self.writing = False
if self.write_lock:
self.write_lock.release()
def read_from_fd(self):
"""
从fd里读取数据。
超时不捕获异常，由外面捕获。
中断也不应该关闭链接，比如kill -USR1 会抛出中断异常
其他错误则直接关闭连接
:return:
"""
try:
chunk = self.sock.recv(self.read_chunk_size)
except socket.timeout, e:
# 服务器是不会recv超时的
raise e
except socket.error, e:
if e.errno == errno.EINTR:
# 中断，返回空字符串，但不断掉连接
return ''
else:
# Connection reset by peer 的原因说明:
# 网上说是对端非正常关闭连接，比如对端程序异常退出之类
# 我重现的方法是: C向S发送数据，如果S有回应，而C没有读取，C就调用close或者被析构的话
logger.error('exc occur.', exc_info=True)
self.close()
return None
class Match(Actor):
'''**Pattern matching on a key/value document stream.**
This module routes messages to a queue associated to the matching rule
set. The event['data'] payload has to be of <type 'dict'>. Typically,
the source data is JSON converted to a Python dictionary.
The match rules can be either stored on disk or directly defined into the
bootstrap file.
A match rule is written in YAML syntax and consists out of 2 parts:
- condition:
A list of dictionaries holding with the individual conditions which
ALL have to match for the complete rule to match.
::
re: Regex matching
!re: Negative regex matching
>: Bigger than
>=: Bigger or equal than
<: Smaller than
<=: Smaller or equal than
=: Equal than
in: Evaluate list membership
!in: Evaluate negative list membership
- queue:
The queue section contains a list of dictionaries/maps each containing
1 key with another dictionary/map as a value. These key/value pairs
are added to the *header section* of the event and stored under the
queue name key.
If you are not interested in adding any information to the header you
can leave the dictionary empty. So this would be valid:
All rules will be evaluated sequentially in no particular order. When a
rule matches, evaluation the other rules will continue untill all rules
are processed.
Examples
~~~~~~~~
This example would route the events - with field "greeting" containing
the value "hello" - to the outbox queue without adding any information
to the header of the event itself.
::
condition:
- greeting: re:^hello$
queue:
- outbox:
This example combines multiple conditions and stores 4 variables under
event["header"][self.name] while submitting the event to the modules'
**email** queue.
::
condition:
- check_command: re:check:host.alive
- hostproblemid: re:\d*
- hostgroupnames: in:tag:development
queue:
- email:
from: [email protected]
to:
- [email protected]
subject: UMI - Host {{ hostname }} is {{ hoststate }}.
template: host_email_alert
Parameters:
- name(str)
| The name of the module.
- size(int)
| The default max length of each queue.
- frequency(int)
| The frequency in seconds to generate metrics.
- location(str)("")
| The directory containing rules.
| If empty, no rules are read from disk.
- rules(dict)({})
| A dict of rules in the above described format.
| For example:
| {"omg": {"condition": [{"greeting": "re:^hello$"}], "queue": [{"outbox": {"one": 1}}]}}
Queues:
- inbox
| Incoming events
- <queue_name>
| The queue which matches a rule
- nomatch
| The queue receiving event without matches
'''
def __init__(self, actor_config, location="", rules={}):
Actor.__init__(self, actor_config)
self.pool.createQueue("inbox")
self.pool.createQueue("nomatch")
self.registerConsumer(self.consume, "inbox")
self.__active_rules = {}
self.match = MatchRules()
self.rule_lock = Semaphore()
def preHook(self):
if self.kwargs.location != "":
self.createDir()
self.logging.info("Rules directoy '%s' defined." % (self.kwargs.location))
self.sendToBackground(self.monitorRuleDirectory)
else:
self.__active_rules.update(self.uplook.dump()["rules"])
self.logging.info("No rules directory defined, not reading rules from disk.")
def createDir(self):
if os.path.exists(self.kwargs.location):
if not os.path.isdir(self.kwargs.location):
raise Exception("%s exists but is not a directory" % (self.kwargs.location))
else:
self.logging.info("Directory %s exists so I'm using it." % (self.kwargs.location))
else:
self.logging.info("Directory %s does not exist so I'm creating it." % (self.kwargs.location))
os.makedirs(self.kwargs.location)
def monitorRuleDirectory(self):
'''
Loads new rules when changes happen.
'''
self.logging.info("Monitoring directory '%s' for changes" % (self.kwargs.location))
self.readRules = ReadRulesDisk(self.logging, self.kwargs.location)
new_rules = self.readRules.readDirectory()
new_rules.update(self.kwargs.rules)
self.__active_rules = new_rules
self.logging.info("Read %s rules from disk and %s defined in config." % (len(new_rules), len(self.kwargs.rules)))
while self.loop():
try:
new_rules = self.readRules.waitForChanges()
new_rules.update(self.kwargs.rules)
if cmp(self.__active_rules, new_rules) != 0:
self.rule_lock.acquire()
self.__active_rules = new_rules
self.rule_lock.release()
self.logging.info("Read %s rules from disk and %s defined in config." % (len(new_rules), len(self.kwargs.rules)))
except Exception as err:
self.logging.warning("Problem reading rules directory. Reason: %s" % (err))
sleep(1)
def consume(self, event):
'''Submits matching documents to the defined queue along with
the defined header.'''
if isinstance(event.data, dict):
self.rule_lock.acquire()
for rule in self.__active_rules:
e = deepcopy(event)
if self.evaluateCondition(self.__active_rules[rule]["condition"], e.data):
e.setHeaderValue("rule", rule)
for queue in self.__active_rules[rule]["queue"]:
event_copy = deepcopy(e)
for name in queue:
if queue[name] is not None:
for key, value in queue[name].iteritems():
event_copy.setHeaderValue(key, value)
# event_copy["header"][self.name].update(queue[name])
self.submit(event_copy, self.pool.getQueue(name))
else:
e.setHeaderValue("rule", rule)
self.submit(e, self.pool.queue.nomatch)
# self.logging.debug("No match for rule %s." % (rule))
self.rule_lock.release()
else:
raise Exception("Incoming data is not of type dict, dropped.")
def evaluateCondition(self, conditions, fields):
for condition in conditions:
for field in condition:
if field in fields:
if not self.match.do(condition[field], fields[field]):
# self.logging.debug("field %s with condition %s DOES NOT MATCH value %s" % (field, condition[field], fields[field]))
return False
else:
pass
# self.logging.debug("field %s with condition %s MATCHES value %s" % (field, condition[field], fields[field]))
else:
return False
return True
class ConnectionInstance(object):
def __init__(self, parent, io_loop=None):
self._parent = parent
self._closing = False
self._user_queries = { }
self._cursor_cache = { }
self._write_mutex = Semaphore()
self._socket = None
def connect(self, timeout):
with gevent.Timeout(timeout, RqlTimeoutError(self._parent.host, self._parent.port)) as timeout:
self._socket = SocketWrapper(self)
# Start a parallel coroutine to perform reads
gevent.spawn(self._reader)
return self._parent
def is_open(self):
return self._socket is not None and self._socket.is_open()
def close(self, noreply_wait, token, exception=None):
self._closing = True
if exception is not None:
err_message = "Connection is closed (%s)." % str(exception)
else:
err_message = "Connection is closed."
# Cursors may remove themselves when errored, so copy a list of them
for cursor in list(self._cursor_cache.values()):
cursor._error(err_message)
for query, async_res in iter(self._user_queries.values()):
async_res.set_exception(RqlDriverError(err_message))
self._user_queries = { }
self._cursor_cache = { }
if noreply_wait:
noreply = net.Query(pQuery.NOREPLY_WAIT, token, None, None)
self.run_query(noreply, False)
try:
self._socket.close()
except:
pass
# TODO: make connection recoverable if interrupted by a user's gevent.Timeout?
def run_query(self, query, noreply):
self._write_mutex.acquire()
try:
self._socket.sendall(query.serialize(self._parent._get_json_encoder(query)))
finally:
self._write_mutex.release()
if noreply:
return None
async_res = AsyncResult()
self._user_queries[query.token] = (query, async_res)
return async_res.get()
# The _reader coroutine runs in its own coroutine in parallel, reading responses
# off of the socket and forwarding them to the appropriate AsyncResult or Cursor.
# This is shut down as a consequence of closing the stream, or an error in the
# socket/protocol from the server. Unexpected errors in this coroutine will
# close the ConnectionInstance and be passed to any open AsyncResult or Cursors.
def _reader(self):
try:
while True:
buf = self._socket.recvall(12)
(token, length,) = struct.unpack("<qL", buf)
buf = self._socket.recvall(length)
cursor = self._cursor_cache.get(token)
if cursor is not None:
cursor._extend(buf)
elif token in self._user_queries:
# Do not pop the query from the dict until later, so
# we don't lose track of it in case of an exception
query, async_res = self._user_queries[token]
res = net.Response(token, buf, self._parent._get_json_decoder(query))
if res.type == pResponse.SUCCESS_ATOM:
async_res.set(net.maybe_profile(res.data[0], res))
elif res.type in (pResponse.SUCCESS_SEQUENCE,
pResponse.SUCCESS_PARTIAL):
cursor = GeventCursor(self, query, res)
async_res.set(net.maybe_profile(cursor, res))
elif res.type == pResponse.WAIT_COMPLETE:
async_res.set(None)
else:
async_res.set_exception(res.make_error(query))
del self._user_queries[token]
elif not self._closing:
raise RqlDriverError("Unexpected response received.")
except Exception as ex:
if not self._closing:
self.close(False, None, ex)
class PostgresConnectionPool(object):
"""A pool of psycopg2 connections shared between multiple greenlets."""
@classmethod
def for_url(cls, url, initial=1, limit=20):
"""Construct a connection pool instance given a URL."""
from . import ConnectionFailed
params = parse_database_url(url)
try:
db = cls(initial=initial, limit=limit, **params)
except ConnectionFailed as e:
raise ConnectionFailed(
'Failed to connect to %s' % make_safe_url(params), *e.args
)
db.url = url # record the URL for debugging
return db
@classmethod
def for_name(cls, name, initial=1, limit=20):
"""Construct a connection pool instance from the named setting."""
from ..config import settings
url = getattr(settings, name)
return cls.for_url(url, initial=initial, limit=limit)
def __init__(self, initial=1, limit=20, **settings):
"""Construct a pool of connections.
settings are passed straight to psycopg2.connect. initial connections
are opened immediately. More connections may be opened if they are
required, but at most limit connections may be open at any one time.
"""
self.settings = settings
self.sem = Semaphore(limit)
self.size = 0
self.pool = []
for i in xrange(initial):
self.pool.append(self._connect())
def _connect(self):
"""Connect to PostgreSQL using the stored connection settings."""
from . import ConnectionFailed, OperationalError
try:
pg = psycopg2.connect(**self.settings)
except OperationalError as e:
url = make_safe_url(self.settings)
raise ConnectionFailed(
'Failed to connect using %s' % url, *e.args
)
self.size += 1
print "PostgreSQL connection pool size:", self.size
return pg
@contextmanager
def cursor(self):
"""Obtain a cursor from the pool. We reserve the connection exclusively
for this cursor, as cursors are apparently not safe to be used in multiple
greenlets.
"""
with self.connection() as conn:
yield conn.cursor()
@contextmanager
def connection(self):
"""Obtain a connection from the pool, as a context manager, so that the
connection will eventually be returned to the pool.
>>> pool = PostgresConnectionPool(**settings)
>>> with pool.connection() as conn:
... c = conn.cursor()
... c.execute(...)
... conn.commit()
"""
self.sem.acquire()
try:
conn = self.pool.pop(0)
except IndexError:
conn = self._connect()
try:
yield conn
except psycopg2.OperationalError:
# Connection errors should result in the connection being removed
# from the pool.
#
# Unfortunately OperationalError could possibly mean other things and
# we don't know enough to determine which
try:
conn.close()
finally:
self.size -= 1
conn = None
raise
except:
conn.rollback()
raise
else:
conn.commit()
finally:
if conn is not None:
conn.reset()
self.pool.append(conn)
self.sem.release()
class Rotkehlchen(object):
def __init__(self, args):
self.lock = Semaphore()
self.lock.acquire()
self.results_cache: typing.ResultCache = dict()
self.connected_exchanges = []
logfilename = None
if args.logtarget == 'file':
logfilename = args.logfile
if args.loglevel == 'debug':
loglevel = logging.DEBUG
elif args.loglevel == 'info':
loglevel = logging.INFO
elif args.loglevel == 'warn':
loglevel = logging.WARN
elif args.loglevel == 'error':
loglevel = logging.ERROR
elif args.loglevel == 'critical':
loglevel = logging.CRITICAL
else:
raise ValueError('Should never get here. Illegal log value')
logging.basicConfig(
filename=logfilename,
filemode='w',
level=loglevel,
format='%(asctime)s -- %(levelname)s:%(name)s:%(message)s',
datefmt='%d/%m/%Y %H:%M:%S %Z',
)
if not args.logfromothermodules:
logging.getLogger('zerorpc').setLevel(logging.CRITICAL)
logging.getLogger('zerorpc.channel').setLevel(logging.CRITICAL)
logging.getLogger('urllib3').setLevel(logging.CRITICAL)
logging.getLogger('urllib3.connectionpool').setLevel(
logging.CRITICAL)
self.sleep_secs = args.sleep_secs
self.data_dir = args.data_dir
self.args = args
self.last_data_upload_ts = 0
self.poloniex = None
self.kraken = None
self.bittrex = None
self.bitmex = None
self.binance = None
self.data = DataHandler(self.data_dir)
self.lock.release()
self.shutdown_event = gevent.event.Event()
def initialize_exchanges(self, secret_data):
# initialize exchanges for which we have keys and are not already initialized
if self.kraken is None and 'kraken' in secret_data:
self.kraken = Kraken(
str.encode(secret_data['kraken']['api_key']),
str.encode(secret_data['kraken']['api_secret']), self.data_dir)
self.connected_exchanges.append('kraken')
self.trades_historian.set_exchange('kraken', self.kraken)
if self.poloniex is None and 'poloniex' in secret_data:
self.poloniex = Poloniex(
str.encode(secret_data['poloniex']['api_key']),
str.encode(secret_data['poloniex']['api_secret']),
self.inquirer, self.data_dir)
self.connected_exchanges.append('poloniex')
self.trades_historian.set_exchange('poloniex', self.poloniex)
if self.bittrex is None and 'bittrex' in secret_data:
self.bittrex = Bittrex(
str.encode(secret_data['bittrex']['api_key']),
str.encode(secret_data['bittrex']['api_secret']),
self.inquirer, self.data_dir)
self.connected_exchanges.append('bittrex')
self.trades_historian.set_exchange('bittrex', self.bittrex)
if self.binance is None and 'binance' in secret_data:
self.binance = Binance(
str.encode(secret_data['binance']['api_key']),
str.encode(secret_data['binance']['api_secret']),
self.inquirer, self.data_dir)
self.connected_exchanges.append('binance')
self.trades_historian.set_exchange('binance', self.binance)
if self.bitmex is None and 'bitmex' in secret_data:
self.bitmex = Bitmex(
str.encode(secret_data['bitmex']['api_key']),
str.encode(secret_data['bitmex']['api_secret']), self.inquirer,
self.data_dir)
self.connected_exchanges.append('bitmex')
self.trades_historian.set_exchange('bitmex', self.bitmex)
def try_premium_at_start(self, api_key, api_secret, create_new,
sync_approval, user_dir):
"""Check if new user provided api pair or we already got one in the DB"""
if api_key != '':
self.premium, valid, empty_or_error = premium_create_and_verify(
api_key, api_secret)
if not valid:
log.error('Given API key is invalid')
# At this point we are at a new user trying to create an account with
# premium API keys and we failed. But a directory was created. Remove it.
shutil.rmtree(user_dir)
raise AuthenticationError(
'Could not verify keys for the new account. '
'{}'.format(empty_or_error))
else:
# If we got premium initialize it and try to sync with the server
premium_credentials = self.data.db.get_rotkehlchen_premium()
if premium_credentials:
api_key = premium_credentials[0]
api_secret = premium_credentials[1]
self.premium, valid, empty_or_error = premium_create_and_verify(
api_key, api_secret)
if not valid:
log.error(
'The API keys found in the Database are not valid. Perhaps '
'they expired?')
del self.premium
return
else:
# no premium credentials in the DB
return
if self.can_sync_data_from_server():
if sync_approval == 'unknown' and not create_new:
log.info('DB data at server newer than local')
raise PermissionError(
'Rotkehlchen Server has newer version of your DB data. '
'Should we replace local data with the server\'s?')
elif sync_approval == 'yes' or sync_approval == 'unknown' and create_new:
log.info('User approved data sync from server')
if self.sync_data_from_server():
if create_new:
# if we successfully synced data from the server and this is
# a new account, make sure the api keys are properly stored
# in the DB
self.data.db.set_rotkehlchen_premium(
api_key, api_secret)
else:
log.debug('Could sync data from server but user refused')
def unlock_user(self, user, password, create_new, sync_approval, api_key,
api_secret):
log.info(
'Unlocking user',
user=user,
create_new=create_new,
sync_approval=sync_approval,
)
# unlock or create the DB
self.password = password
user_dir = self.data.unlock(user, password, create_new)
self.try_premium_at_start(api_key, api_secret, create_new,
sync_approval, user_dir)
secret_data = self.data.db.get_exchange_secrets()
settings = self.data.db.get_settings()
historical_data_start = settings['historical_data_start']
eth_rpc_port = settings['eth_rpc_port']
self.trades_historian = TradesHistorian(
self.data_dir,
self.data.db,
self.data.get_eth_accounts(),
historical_data_start,
)
self.price_historian = PriceHistorian(
self.data_dir,
historical_data_start,
)
db_settings = self.data.db.get_settings()
self.accountant = Accountant(
price_historian=self.price_historian,
profit_currency=self.data.main_currency(),
user_directory=user_dir,
create_csv=True,
ignored_assets=self.data.db.get_ignored_assets(),
include_crypto2crypto=db_settings['include_crypto2crypto'],
taxfree_after_period=db_settings['taxfree_after_period'],
)
# Initialize the rotkehlchen logger
LoggingSettings(anonymized_logs=db_settings['anonymized_logs'])
self.inquirer = Inquirer(kraken=self.kraken)
self.initialize_exchanges(secret_data)
ethchain = Ethchain(eth_rpc_port)
self.blockchain = Blockchain(
blockchain_accounts=self.data.db.get_blockchain_accounts(),
all_eth_tokens=self.data.eth_tokens,
owned_eth_tokens=self.data.db.get_owned_tokens(),
inquirer=self.inquirer,
ethchain=ethchain,
)
def set_premium_credentials(self, api_key, api_secret):
log.info('Setting new premium credentials')
if hasattr(self, 'premium'):
valid, empty_or_error = self.premium.set_credentials(
api_key, api_secret)
else:
self.premium, valid, empty_or_error = premium_create_and_verify(
api_key, api_secret)
if valid:
self.data.set_premium_credentials(api_key, api_secret)
return True, ''
log.error('Setting new premium credentials failed',
error=empty_or_error)
return False, empty_or_error
def maybe_upload_data_to_server(self):
# upload only if unlocked user has premium
if not hasattr(self, 'premium'):
return
# upload only once per hour
diff = ts_now() - self.last_data_upload_ts
if diff > 3600:
self.upload_data_to_server()
def upload_data_to_server(self):
log.debug('upload to server -- start')
data, our_hash = self.data.compress_and_encrypt_db(self.password)
success, result_or_error = self.premium.query_last_data_metadata()
if not success:
log.debug(
'upload to server -- query last metadata failed',
error=result_or_error,
)
return
log.debug(
'CAN_PUSH',
ours=our_hash,
theirs=result_or_error['data_hash'],
)
if our_hash == result_or_error['data_hash']:
log.debug('upload to server -- same hash')
# same hash -- no need to upload anything
return
our_last_write_ts = self.data.db.get_last_write_ts()
if our_last_write_ts <= result_or_error['last_modify_ts']:
# Server's DB was modified after our local DB
log.debug("CAN_PUSH -> 3")
log.debug('upload to server -- remote db more recent than local')
return
success, result_or_error = self.premium.upload_data(
data, our_hash, our_last_write_ts, 'zlib')
if not success:
log.debug('upload to server -- upload error',
error=result_or_error)
return
self.last_data_upload_ts = ts_now()
log.debug('upload to server -- success')
def can_sync_data_from_server(self):
log.debug('sync data from server -- start')
data, our_hash = self.data.compress_and_encrypt_db(self.password)
success, result_or_error = self.premium.query_last_data_metadata()
if not success:
log.debug('sync data from server failed', error=result_or_error)
return False
log.debug(
'CAN_PULL',
ours=our_hash,
theirs=result_or_error['data_hash'],
)
if our_hash == result_or_error['data_hash']:
log.debug('sync from server -- same hash')
# same hash -- no need to get anything
return False
our_last_write_ts = self.data.db.get_last_write_ts()
if our_last_write_ts >= result_or_error['last_modify_ts']:
# Local DB is newer than Server DB
log.debug('sync from server -- local DB more recent than remote')
return False
return True
def sync_data_from_server(self):
success, error_or_result = self.premium.pull_data()
if not success:
log.debug('sync from server -- pulling failed.',
error=error_or_result)
return False
self.data.decompress_and_decrypt_db(self.password,
error_or_result['data'])
return True
def start(self):
return gevent.spawn(self.main_loop)
def main_loop(self):
while True and not self.shutdown_event.is_set():
log.debug('Main loop start')
if self.poloniex is not None:
self.poloniex.main_logic()
if self.kraken is not None:
self.kraken.main_logic()
self.maybe_upload_data_to_server()
log.debug('Main loop end')
gevent.sleep(MAIN_LOOP_SECS_DELAY)
def add_blockchain_account(self, blockchain, account):
try:
new_data = self.blockchain.add_blockchain_account(
blockchain, account)
except (InputError, EthSyncError) as e:
return simple_result(False, str(e))
self.data.add_blockchain_account(blockchain, account)
return accounts_result(new_data['per_account'], new_data['totals'])
def remove_blockchain_account(self, blockchain, account):
try:
new_data = self.blockchain.remove_blockchain_account(
blockchain, account)
except (InputError, EthSyncError) as e:
return simple_result(False, str(e))
self.data.remove_blockchain_account(blockchain, account)
return accounts_result(new_data['per_account'], new_data['totals'])
def add_owned_eth_tokens(self, tokens):
try:
new_data = self.blockchain.track_new_tokens(tokens)
except (InputError, EthSyncError) as e:
return simple_result(False, str(e))
self.data.write_owned_eth_tokens(self.blockchain.owned_eth_tokens)
return accounts_result(new_data['per_account'], new_data['totals'])
def remove_owned_eth_tokens(self, tokens):
try:
new_data = self.blockchain.remove_eth_tokens(tokens)
except InputError as e:
return simple_result(False, str(e))
self.data.write_owned_eth_tokens(self.blockchain.owned_eth_tokens)
return accounts_result(new_data['per_account'], new_data['totals'])
def process_history(self, start_ts, end_ts):
(
error_or_empty, history, margin_history, loan_history,
asset_movements, eth_transactions
) = self.trades_historian.get_history(
start_ts=
0, # For entire history processing we need to have full history available
end_ts=ts_now(),
end_at_least_ts=end_ts)
result = self.accountant.process_history(start_ts, end_ts, history,
margin_history, loan_history,
asset_movements,
eth_transactions)
return result, error_or_empty
def query_fiat_balances(self):
log.info('query_fiat_balances called')
result = {}
balances = self.data.get_fiat_balances()
for currency, amount in balances.items():
amount = FVal(amount)
usd_rate = query_fiat_pair(currency, 'USD')
result[currency] = {
'amount': amount,
'usd_value': amount * usd_rate
}
return result
def query_balances(self, requested_save_data=False):
log.info('query_balances called',
requested_save_data=requested_save_data)
balances = {}
problem_free = True
for exchange in self.connected_exchanges:
exchange_balances, msg = getattr(self, exchange).query_balances()
# If we got an error, disregard that exchange but make sure we don't save data
if not exchange_balances:
problem_free = False
else:
balances[exchange] = exchange_balances
result, error_or_empty = self.blockchain.query_balances()
if error_or_empty == '':
balances['blockchain'] = result['totals']
else:
problem_free = False
result = self.query_fiat_balances()
if result != {}:
balances['banks'] = result
combined = combine_stat_dicts([v for k, v in balances.items()])
total_usd_per_location = [(k, dict_get_sumof(v, 'usd_value'))
for k, v in balances.items()]
# calculate net usd value
net_usd = FVal(0)
for k, v in combined.items():
net_usd += FVal(v['usd_value'])
stats = {'location': {}, 'net_usd': net_usd}
for entry in total_usd_per_location:
name = entry[0]
total = entry[1]
if net_usd != FVal(0):
percentage = (total / net_usd).to_percentage()
else:
percentage = '0%'
stats['location'][name] = {
'usd_value': total,
'percentage_of_net_value': percentage,
}
for k, v in combined.items():
if net_usd != FVal(0):
percentage = (v['usd_value'] / net_usd).to_percentage()
else:
percentage = '0%'
combined[k]['percentage_of_net_value'] = percentage
result_dict = merge_dicts(combined, stats)
allowed_to_save = requested_save_data or self.data.should_save_balances(
)
if problem_free and allowed_to_save:
self.data.save_balances_data(result_dict)
# After adding it to the saved file we can overlay additional data that
# is not required to be saved in the history file
try:
details = self.data.accountant.details
for asset, (tax_free_amount, average_buy_value) in details.items():
if asset not in result_dict:
continue
result_dict[asset]['tax_free_amount'] = tax_free_amount
result_dict[asset]['average_buy_value'] = average_buy_value
current_price = result_dict[asset]['usd_value'] / result_dict[
asset]['amount']
if average_buy_value != FVal(0):
result_dict[asset]['percent_change'] = (
((current_price - average_buy_value) /
average_buy_value) * 100)
else:
result_dict[asset]['percent_change'] = 'INF'
except AttributeError:
pass
return result_dict
def set_main_currency(self, currency):
with self.lock:
self.data.set_main_currency(currency, self.accountant)
if currency != 'USD':
self.usd_to_main_currency_rate = query_fiat_pair(
'USD', currency)
def set_settings(self, settings):
log.info('Add new settings')
message = ''
with self.lock:
if 'eth_rpc_port' in settings:
result, msg = self.blockchain.set_eth_rpc_port(
settings['eth_rpc_port'])
if not result:
# Don't save it in the DB
del settings['eth_rpc_port']
message += "\nEthereum RPC port not set: " + msg
if 'main_currency' in settings:
main_currency = settings['main_currency']
if main_currency != 'USD':
self.usd_to_main_currency_rate = query_fiat_pair(
'USD', main_currency)
res, msg = self.accountant.customize(settings)
if not res:
message += '\n' + msg
return False, message
_, msg, = self.data.set_settings(settings, self.accountant)
if msg != '':
message += '\n' + msg
# Always return success here but with a message
return True, message
def usd_to_main_currency(self, amount):
main_currency = self.data.main_currency()
if main_currency != 'USD' and not hasattr(self,
'usd_to_main_currency_rate'):
self.usd_to_main_currency_rate = query_fiat_pair(
'USD', main_currency)
return self.usd_to_main_currency_rate * amount
def setup_exchange(self, name, api_key, api_secret):
log.info('setup_exchange', name=name)
if name not in SUPPORTED_EXCHANGES:
return False, 'Attempted to register unsupported exchange {}'.format(
name)
if getattr(self, name) is not None:
return False, 'Exchange {} is already registered'.format(name)
secret_data = {}
secret_data[name] = {
'api_key': api_key,
'api_secret': api_secret,
}
self.initialize_exchanges(secret_data)
exchange = getattr(self, name)
result, message = exchange.validate_api_key()
if not result:
log.error(
'Failed to validate API key for exchange',
name=name,
error=message,
)
self.delete_exchange_data(name)
return False, message
# Success, save the result in the DB
self.data.db.add_exchange(name, api_key, api_secret)
return True, ''
def delete_exchange_data(self, name):
self.connected_exchanges.remove(name)
self.trades_historian.set_exchange(name, None)
delattr(self, name)
setattr(self, name, None)
def remove_exchange(self, name):
if getattr(self, name) is None:
return False, 'Exchange {} is not registered'.format(name)
self.delete_exchange_data(name)
# Success, remove it also from the DB
self.data.db.remove_exchange(name)
return True, ''
def shutdown(self):
log.info("Shutting Down")
self.shutdown_event.set()
class Queue(Greenlet):
"""Manages the queue of |Envelope| objects waiting for delivery. This is
not a standard FIFO queue, a message's place in the queue depends entirely
on the timestamp of its next delivery attempt.
:param store: Object implementing :class:`QueueStorage`.
:param relay: |Relay| object used to attempt message deliveries. If this
is not given, no deliveries will be attempted on received
messages.
:param backoff: Function that, given an |Envelope| and number of delivery
attempts, will return the number of seconds before the next
attempt. If it returns ``None``, the message will be
permanently failed. The default backoff function simply
returns ``None`` and messages are never retried.
:param bounce_factory: Function that produces a |Bounce| object given the
same parameters as the |Bounce| constructor. If the
function returns ``None``, no bounce is delivered.
By default, a new |Bounce| is created in every case.
:param bounce_queue: |Queue| object that will be used for delivering bounce
messages. The default is ``self``.
:param store_pool: Number of simultaneous operations performable against
the ``store`` object. Default is unlimited.
:param relay_pool: Number of simultaneous operations performable against
the ``relay`` object. Default is unlimited.
"""
def __init__(self, store, relay=None, backoff=None, bounce_factory=None,
bounce_queue=None, store_pool=None, relay_pool=None):
super(Queue, self).__init__()
self.store = store
self.relay = relay
self.backoff = backoff or self._default_backoff
self.bounce_factory = bounce_factory or Bounce
self.bounce_queue = bounce_queue or self
self.wake = Event()
self.queued = []
self.active_ids = set()
self.queued_ids = set()
self.queued_lock = Semaphore(1)
self.queue_policies = []
self._use_pool('store_pool', store_pool)
self._use_pool('relay_pool', relay_pool)
def add_policy(self, policy):
"""Adds a |QueuePolicy| to be executed before messages are persisted
to storage.
:param policy: |QueuePolicy| object to execute.
"""
if isinstance(policy, QueuePolicy):
self.queue_policies.append(policy)
else:
raise TypeError('Argument not a QueuePolicy.')
@staticmethod
def _default_backoff(envelope, attempts):
pass
def _run_policies(self, envelope):
results = [envelope]
def recurse(current, i):
try:
policy = self.queue_policies[i]
except IndexError:
return
ret = policy.apply(current)
if ret:
results.remove(current)
results.extend(ret)
for env in ret:
recurse(env, i+1)
else:
recurse(current, i+1)
recurse(envelope, 0)
return results
def _use_pool(self, attr, pool):
if pool is None:
pass
elif isinstance(pool, Pool):
setattr(self, attr, pool)
else:
setattr(self, attr, Pool(pool))
def _pool_run(self, which, func, *args, **kwargs):
pool = getattr(self, which+'_pool', None)
if pool:
ret = pool.spawn(func, *args, **kwargs)
return ret.get()
else:
return func(*args, **kwargs)
def _pool_imap(self, which, func, *iterables):
pool = getattr(self, which+'_pool', gevent)
threads = imap(pool.spawn, repeat(func), *iterables)
ret = []
for thread in threads:
thread.join()
ret.append(thread.exception or thread.value)
return ret
def _pool_spawn(self, which, func, *args, **kwargs):
pool = getattr(self, which+'_pool', gevent)
return pool.spawn(func, *args, **kwargs)
def _add_queued(self, entry):
timestamp, id = entry
if id not in self.queued_ids | self.active_ids:
bisect.insort(self.queued, entry)
self.queued_ids.add(id)
self.wake.set()
def enqueue(self, envelope):
"""Drops a new message in the queue for delivery. The first delivery
attempt is made immediately (depending on relay pool availability).
This method is not typically called directly, |Edge| objects use it
when they receive new messages.
:param envelope: |Envelope| object to enqueue.
:returns: Zipped list of envelopes and their respective queue IDs (or
thrown :exc:`QueueError` objects).
"""
now = time.time()
envelopes = self._run_policies(envelope)
ids = self._pool_imap('store', self.store.write, envelopes,
repeat(now))
results = list(zip(envelopes, ids))
for env, id in results:
if not isinstance(id, BaseException):
if self.relay and id not in self.active_ids:
self.active_ids.add(id)
self._pool_spawn('relay', self._attempt, id, env, 0)
elif not isinstance(id, QueueError):
raise id # Re-raise exceptions that are not QueueError.
return results
def _load_all(self):
for entry in self.store.load():
self._add_queued(entry)
def _remove(self, id):
self._pool_spawn('store', self.store.remove, id)
self.queued_ids.discard(id)
self.active_ids.discard(id)
def _bounce(self, envelope, reply):
bounce = self.bounce_factory(envelope, reply)
if bounce:
return self.bounce_queue.enqueue(bounce)
def _perm_fail(self, id, envelope, reply):
if id is not None:
self._remove(id)
if envelope.sender: # Can't bounce to null-sender.
self._pool_spawn('bounce', self._bounce, envelope, reply)
def _split_by_reply(self, envelope, replies):
if isinstance(replies, Reply):
return [(replies, envelope)]
groups = []
for i, rcpt in enumerate(envelope.recipients):
for reply, group_env in groups:
if replies[i] == reply:
group_env.recipients.append(rcpt)
break
else:
group_env = envelope.copy([rcpt])
groups.append((replies[i], group_env))
return groups
def _retry_later(self, id, envelope, replies):
attempts = self.store.increment_attempts(id)
wait = self.backoff(envelope, attempts)
if wait is None:
for reply, group_env in self._split_by_reply(envelope, replies):
reply.message += ' (Too many retries)'
self._perm_fail(None, group_env, reply)
self._remove(id)
return False
else:
when = time.time() + wait
self.store.set_timestamp(id, when)
self.active_ids.discard(id)
self._add_queued((when, id))
return True
def _attempt(self, id, envelope, attempts):
try:
results = self.relay._attempt(envelope, attempts)
except TransientRelayError as e:
self._pool_spawn('store', self._retry_later, id, envelope, e.reply)
except PermanentRelayError as e:
self._perm_fail(id, envelope, e.reply)
except Exception as e:
logging.log_exception(__name__)
reply = Reply('450', '4.0.0 Unhandled delivery error: '+str(e))
self._pool_spawn('store', self._retry_later, id, envelope, reply)
raise
else:
if isinstance(results, collections.Mapping):
self._handle_partial_relay(id, envelope, attempts, results)
elif isinstance(results, collections.Sequence):
results = dict(zip(envelope.recipients, results))
self._handle_partial_relay(id, envelope, attempts, results)
else:
self._remove(id)
def _handle_partial_relay(self, id, envelope, attempts, results):
delivered = set()
tempfails = []
permfails = []
for rcpt, rcpt_res in results.items():
if rcpt_res is None or isinstance(rcpt_res, Reply):
delivered.add(envelope.recipients.index(rcpt))
elif isinstance(rcpt_res, PermanentRelayError):
delivered.add(envelope.recipients.index(rcpt))
permfails.append((rcpt, rcpt_res.reply))
elif isinstance(rcpt_res, TransientRelayError):
tempfails.append((rcpt, rcpt_res.reply))
if permfails:
rcpts, replies = zip(*permfails)
fail_env = envelope.copy(rcpts)
for reply, group_env in self._split_by_reply(fail_env, replies):
self._perm_fail(None, group_env, reply)
if tempfails:
rcpts, replies = zip(*tempfails)
fail_env = envelope.copy(rcpts)
if not self._retry_later(id, fail_env, replies):
return
else:
self.store.remove(id)
return
self.store.set_recipients_delivered(id, delivered)
def _dequeue(self, id):
try:
envelope, attempts = self.store.get(id)
except KeyError:
return
if id not in self.active_ids:
self.active_ids.add(id)
self._pool_spawn('relay', self._attempt, id, envelope, attempts)
def _check_ready(self, now):
last_i = 0
for i, entry in enumerate(self.queued):
timestamp, entry_id = entry
if now >= timestamp:
self._pool_spawn('store', self._dequeue, entry_id)
last_i = i+1
else:
break
if last_i > 0:
self.queued = self.queued[last_i:]
self.queued_ids = set([id for _, id in self.queued])
def _wait_store(self):
while True:
try:
for entry in self.store.wait():
self._add_queued(entry)
except NotImplementedError:
return
def _wait_ready(self, now):
try:
first = self.queued[0]
except IndexError:
self.wake.wait()
self.wake.clear()
return
first_timestamp = first[0]
if first_timestamp > now:
self.wake.wait(first_timestamp-now)
self.wake.clear()
def flush(self):
"""Attempts to immediately flush all messages waiting in the queue,
regardless of their retry timers.
.. warning::
This can be a very expensive operation, use with care.
"""
self.wake.set()
self.wake.clear()
self.queued_lock.acquire()
try:
for entry in self.queued:
self._pool_spawn('store', self._dequeue, entry[1])
self.queued = []
finally:
self.queued_lock.release()
def kill(self):
"""This method is used by |Queue| and |Queue|-like objects to properly
end any associated services (such as running :class:`~gevent.Greenlet`
threads) and close resources.
"""
super(Queue, self).kill()
def _run(self):
if not self.relay:
return
self._pool_spawn('store', self._load_all)
self._pool_spawn('store', self._wait_store)
while True:
self.queued_lock.acquire()
try:
now = time.time()
self._check_ready(now)
self._wait_ready(now)
finally:
self.queued_lock.release()
class TarantoolInspector(StreamServer):
"""
Tarantool inspector daemon. Usage:
inspector = TarantoolInspector('localhost', 8080)
inspector.start()
# run some tests
inspector.stop()
"""
def __init__(self, host, port):
# When specific port range was acquired for current worker, don't allow
# OS set port for us that isn't from specified range.
if port == 0:
port = find_port()
super(TarantoolInspector, self).__init__((host, port))
self.parser = None
def start(self):
super(TarantoolInspector, self).start()
os.environ['INSPECTOR'] = str(self.server_port)
def stop(self):
del os.environ['INSPECTOR']
def set_parser(self, parser):
self.parser = parser
self.sem = Semaphore()
@staticmethod
def readline(socket, delimiter='\n', size=4096):
result = ''
data = True
while data:
try:
data = socket.recv(size)
except IOError:
# catch instance halt connection refused errors
data = ''
result += data
while result.find(delimiter) != -1:
line, result = result.split(delimiter, 1)
yield line
return
def handle(self, socket, addr):
if self.parser is None:
raise AttributeError('Parser is not defined')
self.sem.acquire()
for line in self.readline(socket):
try:
result = self.parser.parse_preprocessor(line)
except (KeyboardInterrupt, TarantoolStartError):
# propagate to the main greenlet
raise
except Exception as e:
self.parser.kill_current_test()
color_stdout('\nTarantoolInpector.handle() received the following error:\n' +
traceback.format_exc() + '\n', schema='error')
result = { "error": repr(e) }
if result == None:
result = True
result = yaml.dump(result)
if not result.endswith('...\n'):
result = result + '...\n'
socket.sendall(result)
self.sem.release()
def cleanup_nondefault(self):
if self.parser:
self.parser.cleanup_nondefault()
class ServiceBase(object):
def __init__(self, s_id, conf):
self.s_id = s_id
self.conf = conf
self.start_time = None
self.finish_time = None
self.puppet = None
self.lock = Semaphore()
self.queue = set()
self._status = "Unknown"
self.started = False
def start(self):
self.lock.acquire()
if self.started:
self.lock.release()
return
# print "Starting service", self.s_id
self.start_time = time.time()
self._status = "Booting"
self.real_start()
self.started = True
self.lock.release()
def terminate(self):
if not self.puppet:
return
self.real_terminate()
self._status = "Terminated"
self.puppet = None
self.finish_time = time.time()
def real_start(self):
raise NotImplementedError
def real_terminate(self):
raise NotImplementedError
def record_task(self, task):
self.queue.add(task.tid)
def run(self, tid, task, *argv, **kwargs):
task_info = Husky.dumps((task, argv, kwargs))
self.puppet.submit_task(tid, task_info)
res = self.puppet.fetch_result(tid)
self.queue.discard(tid)
return Husky.loads(res)
def __getattr__(self, item):
try:
if item == "tasks":
return list(self.queue)
elif item == "status":
if self.puppet:
return self.puppet.get_attr("status")
else:
return self._status
elif item in ["cpu", "memory"]:
if self.puppet:
return self.puppet.get_attr(item)
else:
return None
except:
return "Unknown"
def __repr__(self):
return "%s-%d" % (self.conf, self.s_id)
class Rotkehlchen():
def __init__(self, args):
self.lock = Semaphore()
self.lock.acquire()
self.results_cache: ResultCache = dict()
self.premium = None
self.connected_exchanges = []
self.user_is_logged_in = False
logfilename = None
if args.logtarget == 'file':
logfilename = args.logfile
if args.loglevel == 'debug':
loglevel = logging.DEBUG
elif args.loglevel == 'info':
loglevel = logging.INFO
elif args.loglevel == 'warn':
loglevel = logging.WARN
elif args.loglevel == 'error':
loglevel = logging.ERROR
elif args.loglevel == 'critical':
loglevel = logging.CRITICAL
else:
raise ValueError('Should never get here. Illegal log value')
logging.basicConfig(
filename=logfilename,
filemode='w',
level=loglevel,
format='%(asctime)s -- %(levelname)s:%(name)s:%(message)s',
datefmt='%d/%m/%Y %H:%M:%S %Z',
)
if not args.logfromothermodules:
logging.getLogger('zerorpc').setLevel(logging.CRITICAL)
logging.getLogger('zerorpc.channel').setLevel(logging.CRITICAL)
logging.getLogger('urllib3').setLevel(logging.CRITICAL)
logging.getLogger('urllib3.connectionpool').setLevel(
logging.CRITICAL)
self.sleep_secs = args.sleep_secs
self.data_dir = args.data_dir
self.args = args
self.poloniex = None
self.kraken = None
self.bittrex = None
self.bitmex = None
self.binance = None
self.msg_aggregator = MessagesAggregator()
self.data = DataHandler(self.data_dir, self.msg_aggregator)
# Initialize the Inquirer singleton
Inquirer(data_dir=self.data_dir)
self.lock.release()
self.shutdown_event = gevent.event.Event()
def initialize_exchanges(
self, exchange_credentials: Dict[str, ApiCredentials]) -> None:
# initialize exchanges for which we have keys and are not already initialized
if self.kraken is None and 'kraken' in exchange_credentials:
self.kraken = Kraken(
api_key=exchange_credentials['kraken'].api_key,
secret=exchange_credentials['kraken'].api_secret,
user_directory=self.user_directory,
msg_aggregator=self.msg_aggregator,
usd_eur_price=Inquirer().query_fiat_pair(S_EUR, S_USD),
)
self.connected_exchanges.append('kraken')
self.trades_historian.set_exchange('kraken', self.kraken)
if self.poloniex is None and 'poloniex' in exchange_credentials:
self.poloniex = Poloniex(
api_key=exchange_credentials['poloniex'].api_key,
secret=exchange_credentials['poloniex'].api_secret,
user_directory=self.user_directory,
msg_aggregator=self.msg_aggregator,
)
self.connected_exchanges.append('poloniex')
self.trades_historian.set_exchange('poloniex', self.poloniex)
if self.bittrex is None and 'bittrex' in exchange_credentials:
self.bittrex = Bittrex(
api_key=exchange_credentials['bittrex'].api_key,
secret=exchange_credentials['bittrex'].api_secret,
user_directory=self.user_directory,
msg_aggregator=self.msg_aggregator,
)
self.connected_exchanges.append('bittrex')
self.trades_historian.set_exchange('bittrex', self.bittrex)
if self.binance is None and 'binance' in exchange_credentials:
self.binance = Binance(
api_key=exchange_credentials['binance'].api_key,
secret=exchange_credentials['binance'].api_secret,
data_dir=self.user_directory,
msg_aggregator=self.msg_aggregator,
)
self.connected_exchanges.append('binance')
self.trades_historian.set_exchange('binance', self.binance)
if self.bitmex is None and 'bitmex' in exchange_credentials:
self.bitmex = Bitmex(
api_key=exchange_credentials['bitmex'].api_key,
secret=exchange_credentials['bitmex'].api_secret,
user_directory=self.user_directory,
)
self.connected_exchanges.append('bitmex')
self.trades_historian.set_exchange('bitmex', self.bitmex)
def remove_all_exchanges(self):
if self.kraken is not None:
self.delete_exchange_data('kraken')
if self.poloniex is not None:
self.delete_exchange_data('poloniex')
if self.bittrex is not None:
self.delete_exchange_data('bittrex')
if self.binance is not None:
self.delete_exchange_data('binance')
if self.bitmex is not None:
self.delete_exchange_data('bitmex')
def unlock_user(
self,
user: str,
password: str,
create_new: bool,
sync_approval: bool,
api_key: ApiKey,
api_secret: ApiSecret,
) -> None:
"""Unlocks an existing user or creates a new one if `create_new` is True"""
log.info(
'Unlocking user',
user=user,
create_new=create_new,
sync_approval=sync_approval,
)
# unlock or create the DB
self.password = password
self.user_directory = self.data.unlock(user, password, create_new)
self.last_data_upload_ts = self.data.db.get_last_data_upload_ts()
self.premium_sync_manager = PremiumSyncManager(data=self.data,
password=password)
try:
self.premium = self.premium_sync_manager.try_premium_at_start(
api_key=api_key,
api_secret=api_secret,
username=user,
create_new=create_new,
sync_approval=sync_approval,
)
except AuthenticationError:
# It means that our credentials were not accepted by the server
# or some other error happened
pass
exchange_credentials = self.data.db.get_exchange_credentials()
settings = self.data.db.get_settings()
historical_data_start = settings['historical_data_start']
eth_rpc_endpoint = settings['eth_rpc_endpoint']
self.trades_historian = TradesHistorian(
user_directory=self.user_directory,
db=self.data.db,
eth_accounts=self.data.get_eth_accounts(),
msg_aggregator=self.msg_aggregator,
)
# Initialize the price historian singleton
PriceHistorian(
data_directory=self.data_dir,
history_date_start=historical_data_start,
cryptocompare=Cryptocompare(data_directory=self.data_dir),
)
db_settings = self.data.db.get_settings()
self.accountant = Accountant(
profit_currency=self.data.main_currency(),
user_directory=self.user_directory,
msg_aggregator=self.msg_aggregator,
create_csv=True,
ignored_assets=self.data.db.get_ignored_assets(),
include_crypto2crypto=db_settings['include_crypto2crypto'],
taxfree_after_period=db_settings['taxfree_after_period'],
include_gas_costs=db_settings['include_gas_costs'],
)
# Initialize the rotkehlchen logger
LoggingSettings(anonymized_logs=db_settings['anonymized_logs'])
self.initialize_exchanges(exchange_credentials)
ethchain = Ethchain(eth_rpc_endpoint)
self.blockchain = Blockchain(
blockchain_accounts=self.data.db.get_blockchain_accounts(),
owned_eth_tokens=self.data.db.get_owned_tokens(),
ethchain=ethchain,
msg_aggregator=self.msg_aggregator,
)
self.user_is_logged_in = True
def logout(self) -> None:
if not self.user_is_logged_in:
return
user = self.data.username
log.info(
'Logging out user',
user=user,
)
del self.blockchain
self.remove_all_exchanges()
# Reset rotkehlchen logger to default
LoggingSettings(anonymized_logs=DEFAULT_ANONYMIZED_LOGS)
del self.accountant
del self.trades_historian
if self.premium is not None:
del self.premium
self.data.logout()
self.password = ''
self.user_is_logged_in = False
log.info(
'User successfully logged out',
user=user,
)
def set_premium_credentials(self, api_key: ApiKey,
api_secret: ApiSecret) -> None:
"""
Raises IncorrectApiKeyFormat if the given key is not in a proper format
Raises AuthenticationError if the given key is rejected by the Rotkehlchen server
"""
log.info('Setting new premium credentials')
if self.premium is not None:
self.premium.set_credentials(api_key, api_secret)
else:
self.premium = premium_create_and_verify(api_key, api_secret)
self.data.set_premium_credentials(api_key, api_secret)
def start(self):
return gevent.spawn(self.main_loop)
def main_loop(self):
while self.shutdown_event.wait(MAIN_LOOP_SECS_DELAY) is not True:
log.debug('Main loop start')
if self.poloniex is not None:
self.poloniex.main_logic()
if self.kraken is not None:
self.kraken.main_logic()
self.premium_sync_manager.maybe_upload_data_to_server()
log.debug('Main loop end')
def add_blockchain_account(
self,
blockchain: SupportedBlockchain,
account: BlockchainAddress,
) -> Dict:
try:
new_data = self.blockchain.add_blockchain_account(
blockchain, account)
except (InputError, EthSyncError) as e:
return simple_result(False, str(e))
self.data.add_blockchain_account(blockchain, account)
return accounts_result(new_data['per_account'], new_data['totals'])
def remove_blockchain_account(
self,
blockchain: SupportedBlockchain,
account: BlockchainAddress,
):
try:
new_data = self.blockchain.remove_blockchain_account(
blockchain, account)
except (InputError, EthSyncError) as e:
return simple_result(False, str(e))
self.data.remove_blockchain_account(blockchain, account)
return accounts_result(new_data['per_account'], new_data['totals'])
def add_owned_eth_tokens(self, tokens: List[str]):
ethereum_tokens = [
EthereumToken(identifier=identifier) for identifier in tokens
]
try:
new_data = self.blockchain.track_new_tokens(ethereum_tokens)
except (InputError, EthSyncError) as e:
return simple_result(False, str(e))
self.data.write_owned_eth_tokens(self.blockchain.owned_eth_tokens)
return accounts_result(new_data['per_account'], new_data['totals'])
def remove_owned_eth_tokens(self, tokens: List[str]):
ethereum_tokens = [
EthereumToken(identifier=identifier) for identifier in tokens
]
try:
new_data = self.blockchain.remove_eth_tokens(ethereum_tokens)
except InputError as e:
return simple_result(False, str(e))
self.data.write_owned_eth_tokens(self.blockchain.owned_eth_tokens)
return accounts_result(new_data['per_account'], new_data['totals'])
def process_history(self, start_ts, end_ts):
(
error_or_empty,
history,
margin_history,
loan_history,
asset_movements,
eth_transactions,
) = self.trades_historian.get_history(
start_ts=
0, # For entire history processing we need to have full history available
end_ts=ts_now(),
end_at_least_ts=end_ts,
)
result = self.accountant.process_history(
start_ts,
end_ts,
history,
margin_history,
loan_history,
asset_movements,
eth_transactions,
)
return result, error_or_empty
def query_fiat_balances(self):
log.info('query_fiat_balances called')
result = {}
balances = self.data.get_fiat_balances()
for currency, amount in balances.items():
amount = FVal(amount)
usd_rate = Inquirer().query_fiat_pair(currency, 'USD')
result[currency] = {
'amount': amount,
'usd_value': amount * usd_rate,
}
return result
def query_balances(
self,
requested_save_data: bool = False,
timestamp: Timestamp = None,
) -> Dict[str, Any]:
"""Query all balances rotkehlchen can see.
If requested_save_data is True then the data are saved in the DB.
If timestamp is None then the current timestamp is used.
If a timestamp is given then that is the time that the balances are going
to be saved in the DB
Returns a dictionary with the queried balances.
"""
log.info('query_balances called',
requested_save_data=requested_save_data)
balances = {}
problem_free = True
for exchange in self.connected_exchanges:
exchange_balances, _ = getattr(self, exchange).query_balances()
# If we got an error, disregard that exchange but make sure we don't save data
if not isinstance(exchange_balances, dict):
problem_free = False
else:
balances[exchange] = exchange_balances
result, error_or_empty = self.blockchain.query_balances()
if error_or_empty == '':
balances['blockchain'] = result['totals']
else:
problem_free = False
result = self.query_fiat_balances()
if result != {}:
balances['banks'] = result
combined = combine_stat_dicts([v for k, v in balances.items()])
total_usd_per_location = [(k, dict_get_sumof(v, 'usd_value'))
for k, v in balances.items()]
# calculate net usd value
net_usd = FVal(0)
for _, v in combined.items():
net_usd += FVal(v['usd_value'])
stats: Dict[str, Any] = {
'location': {},
'net_usd': net_usd,
}
for entry in total_usd_per_location:
name = entry[0]
total = entry[1]
if net_usd != FVal(0):
percentage = (total / net_usd).to_percentage()
else:
percentage = '0%'
stats['location'][name] = {
'usd_value': total,
'percentage_of_net_value': percentage,
}
for k, v in combined.items():
if net_usd != FVal(0):
percentage = (v['usd_value'] / net_usd).to_percentage()
else:
percentage = '0%'
combined[k]['percentage_of_net_value'] = percentage
result_dict = merge_dicts(combined, stats)
allowed_to_save = requested_save_data or self.data.should_save_balances(
)
if problem_free and allowed_to_save:
if not timestamp:
timestamp = Timestamp(int(time.time()))
self.data.save_balances_data(data=result_dict, timestamp=timestamp)
log.debug('query_balances data saved')
else:
log.debug(
'query_balances data not saved',
allowed_to_save=allowed_to_save,
problem_free=problem_free,
)
# After adding it to the saved file we can overlay additional data that
# is not required to be saved in the history file
try:
details = self.data.accountant.details
for asset, (tax_free_amount, average_buy_value) in details.items():
if asset not in result_dict:
continue
result_dict[asset]['tax_free_amount'] = tax_free_amount
result_dict[asset]['average_buy_value'] = average_buy_value
current_price = result_dict[asset]['usd_value'] / result_dict[
asset]['amount']
if average_buy_value != FVal(0):
result_dict[asset]['percent_change'] = (
((current_price - average_buy_value) /
average_buy_value) * 100)
else:
result_dict[asset]['percent_change'] = 'INF'
except AttributeError:
pass
return result_dict
def set_main_currency(self, currency):
with self.lock:
self.data.set_main_currency(currency, self.accountant)
if currency != S_USD:
self.usd_to_main_currency_rate = Inquirer().query_fiat_pair(
S_USD, currency)
def set_settings(self, settings):
log.info('Add new settings')
message = ''
with self.lock:
if 'eth_rpc_endpoint' in settings:
result, msg = self.blockchain.set_eth_rpc_endpoint(
settings['eth_rpc_endpoint'])
if not result:
# Don't save it in the DB
del settings['eth_rpc_endpoint']
message += "\nEthereum RPC endpoint not set: " + msg
if 'main_currency' in settings:
given_symbol = settings['main_currency']
try:
main_currency = Asset(given_symbol)
except UnknownAsset:
return False, f'Unknown fiat currency {given_symbol} provided'
if not main_currency.is_fiat():
msg = (
f'Provided symbol for main currency {given_symbol} is '
f'not a fiat currency')
return False, msg
if main_currency != A_USD:
self.usd_to_main_currency_rate = Inquirer(
).query_fiat_pair(
S_USD,
main_currency.identifier,
)
res, msg = self.accountant.customize(settings)
if not res:
message += '\n' + msg
return False, message
_, msg, = self.data.set_settings(settings, self.accountant)
if msg != '':
message += '\n' + msg
# Always return success here but with a message
return True, message
def setup_exchange(
self,
name: str,
api_key: str,
api_secret: str,
) -> Tuple[bool, str]:
"""
Setup a new exchange with an api key and an api secret
By default the api keys are always validated unless validate is False.
"""
log.info('setup_exchange', name=name)
if name not in SUPPORTED_EXCHANGES:
return False, 'Attempted to register unsupported exchange {}'.format(
name)
if getattr(self, name) is not None:
return False, 'Exchange {} is already registered'.format(name)
credentials_dict = {}
api_credentials = ApiCredentials.serialize(api_key=api_key,
api_secret=api_secret)
credentials_dict[name] = api_credentials
self.initialize_exchanges(credentials_dict)
exchange = getattr(self, name)
result, message = exchange.validate_api_key()
if not result:
log.error(
'Failed to validate API key for exchange',
name=name,
error=message,
)
self.delete_exchange_data(name)
return False, message
# Success, save the result in the DB
self.data.db.add_exchange(name, api_key, api_secret)
return True, ''
def delete_exchange_data(self, name):
self.connected_exchanges.remove(name)
self.trades_historian.set_exchange(name, None)
delattr(self, name)
setattr(self, name, None)
def remove_exchange(self, name):
if getattr(self, name) is None:
return False, 'Exchange {} is not registered'.format(name)
self.delete_exchange_data(name)
# Success, remove it also from the DB
self.data.db.remove_exchange(name)
return True, ''
def query_periodic_data(self) -> Dict[str, Union[bool, Timestamp]]:
"""Query for frequently changing data"""
result = {}
if self.user_is_logged_in:
result[
'last_balance_save'] = self.data.db.get_last_balance_save_time(
)
result['eth_node_connection'] = self.blockchain.ethchain.connected
result[
'history_process_start_ts'] = self.accountant.started_processing_timestamp
result[
'history_process_current_ts'] = self.accountant.currently_processing_timestamp
return result
def shutdown(self):
self.logout()
self.shutdown_event.set()
class ThreadPool(GroupMappingMixin):
def __init__(self, maxsize, hub=None):
if hub is None:
hub = get_hub()
self.hub = hub
self._maxsize = 0
self.manager = None
self.pid = os.getpid()
self.fork_watcher = hub.loop.fork(ref=False)
self._init(maxsize)
def _set_maxsize(self, maxsize):
if not isinstance(maxsize, integer_types):
raise TypeError('maxsize must be integer: %r' % (maxsize, ))
if maxsize < 0:
raise ValueError('maxsize must not be negative: %r' % (maxsize, ))
difference = maxsize - self._maxsize
self._semaphore.counter += difference
self._maxsize = maxsize
self.adjust()
# make sure all currently blocking spawn() start unlocking if maxsize increased
self._semaphore._start_notify()
def _get_maxsize(self):
return self._maxsize
maxsize = property(_get_maxsize, _set_maxsize)
def __repr__(self):
return '<%s at 0x%x %s/%s/%s>' % (self.__class__.__name__, id(self), len(self), self.size, self.maxsize)
def __len__(self):
# XXX just do unfinished_tasks property
return self.task_queue.unfinished_tasks
def _get_size(self):
return self._size
def _set_size(self, size):
if size < 0:
raise ValueError('Size of the pool cannot be negative: %r' % (size, ))
if size > self._maxsize:
raise ValueError('Size of the pool cannot be bigger than maxsize: %r > %r' % (size, self._maxsize))
if self.manager:
self.manager.kill()
while self._size < size:
self._add_thread()
delay = 0.0001
while self._size > size:
while self._size - size > self.task_queue.unfinished_tasks:
self.task_queue.put(None)
if getcurrent() is self.hub:
break
sleep(delay)
delay = min(delay * 2, .05)
if self._size:
self.fork_watcher.start(self._on_fork)
else:
self.fork_watcher.stop()
size = property(_get_size, _set_size)
def _init(self, maxsize):
self._size = 0
self._semaphore = Semaphore(1)
self._lock = Lock()
self.task_queue = Queue()
self._set_maxsize(maxsize)
def _on_fork(self):
# fork() only leaves one thread; also screws up locks;
# let's re-create locks and threads
pid = os.getpid()
if pid != self.pid:
self.pid = pid
# Do not mix fork() and threads; since fork() only copies one thread
# all objects referenced by other threads has refcount that will never
# go down to 0.
self._init(self._maxsize)
def join(self):
delay = 0.0005
while self.task_queue.unfinished_tasks > 0:
sleep(delay)
delay = min(delay * 2, .05)
def kill(self):
self.size = 0
def _adjust_step(self):
# if there is a possibility & necessity for adding a thread, do it
while self._size < self._maxsize and self.task_queue.unfinished_tasks > self._size:
self._add_thread()
# while the number of threads is more than maxsize, kill one
# we do not check what's already in task_queue - it could be all Nones
while self._size - self._maxsize > self.task_queue.unfinished_tasks:
self.task_queue.put(None)
if self._size:
self.fork_watcher.start(self._on_fork)
else:
self.fork_watcher.stop()
def _adjust_wait(self):
delay = 0.0001
while True:
self._adjust_step()
if self._size <= self._maxsize:
return
sleep(delay)
delay = min(delay * 2, .05)
def adjust(self):
self._adjust_step()
if not self.manager and self._size > self._maxsize:
# might need to feed more Nones into the pool
self.manager = Greenlet.spawn(self._adjust_wait)
def _add_thread(self):
with self._lock:
self._size += 1
try:
start_new_thread(self._worker, ())
except:
with self._lock:
self._size -= 1
raise
def spawn(self, func, *args, **kwargs):
while True:
semaphore = self._semaphore
semaphore.acquire()
if semaphore is self._semaphore:
break
try:
task_queue = self.task_queue
result = AsyncResult()
thread_result = ThreadResult(result, hub=self.hub)
task_queue.put((func, args, kwargs, thread_result))
self.adjust()
# rawlink() must be the last call
result.rawlink(lambda *args: self._semaphore.release())
# XXX this _semaphore.release() is competing for order with get()
# XXX this is not good, just make ThreadResult release the semaphore before doing anything else
except:
semaphore.release()
raise
return result
def _decrease_size(self):
if sys is None:
return
_lock = getattr(self, '_lock', None)
if _lock is not None:
with _lock:
self._size -= 1
def _worker(self):
need_decrease = True
try:
while True:
task_queue = self.task_queue
task = task_queue.get()
try:
if task is None:
need_decrease = False
self._decrease_size()
# we want first to decrease size, then decrease unfinished_tasks
# otherwise, _adjust might think there's one more idle thread that
# needs to be killed
return
func, args, kwargs, thread_result = task
try:
value = func(*args, **kwargs)
except:
exc_info = getattr(sys, 'exc_info', None)
if exc_info is None:
return
thread_result.handle_error((self, func), exc_info())
else:
if sys is None:
return
thread_result.set(value)
del value
finally:
del func, args, kwargs, thread_result, task
finally:
if sys is None:
return
task_queue.task_done()
finally:
if need_decrease:
self._decrease_size()
def apply_e(self, expected_errors, function, args=None, kwargs=None):
# Deprecated but never documented. In the past, before
# self.apply() allowed all errors to be raised to the caller,
# expected_errors allowed a caller to specify a set of errors
# they wanted to be raised, through the wrap_errors function.
# In practice, it always took the value Exception or
# BaseException.
return self.apply(function, args, kwargs)
def _apply_immediately(self):
# we always pass apply() off to the threadpool
return False
def _apply_async_cb_spawn(self, callback, result):
callback(result)
def _apply_async_use_greenlet(self):
# Always go to Greenlet because our self.spawn uses threads
return True
class Mime(object):
"""
Our mime response object.
"""
def __init__(self, magic_file=None):
"""
Create a new libmagic wrapper.
magic_file - Another magic file other then the default
"""
# The lock allows us to be thread-safe
self.lock = Semaphore(value=1)
# Tracks the errno/errstr set after each call
self.errno = 0
self.errstr = ''
# Load our magic file
self.magic_file = magic_file
# Initialize our flags
# our flags
self.flags = (MAGIC_MIME | MAGIC_MIME_ENCODING)
def from_content(self, content, uncompress=False, fullscan=False):
"""
detect the type based on the content specified.
content - the content to process
uncompress - Try to look inside compressed files.
fullscan - Scan entire file and extract as much details as possible
"""
if not HAS_LIBMAGIC:
return None
if not content:
return MimeResponse()
# our pointer to our libmagic object
ptr = None
# our flags
flags = self.flags
if fullscan:
flags |= MAGIC_CONTINUE
if uncompress:
flags |= MAGIC_COMPRESS
try:
self.lock.acquire(blocking=True)
# 'application/octet-stream; charset=binary'
# Acquire a pointer
ptr = _magic['open'](flags)
# Acquire a pointer
_magic['load'](ptr, self.magic_file)
self.errno = _magic['errno'](ptr)
if self.errno == 0:
# Achieve our results as a list
_typ, _enc = MAGIC_LIST_RE.split(self._tostr(_magic['buffer'](
ptr,
self._tobytes(content),
len(content),
)))
# Acquire our errorstr (if one exists)
self.errstr = _magic['error'](ptr)
_typs = []
for n, _typ_re in enumerate(TYPE_PARSE_RE.finditer(_typ)):
if n is 0:
_typs.append((0, _typ_re.group('mtype')))
else:
_typs.append(
(
int(_typ_re.group('offset')),
_typ_re.group('mtype'),
),
)
_enc_re = ENCODING_PARSE_RE.match(_enc)
if _enc_re:
_enc = _enc_re.group('encoding')
mr = MimeResponse(mime_type=_typs, mime_encoding=_enc)
# return our object
return mr
except TypeError:
# This occurs if buffer check returns None
# Acquire our errorstr (if one exists)
self.errstr = _magic['error'](ptr)
if self.errstr:
# an error occured; return None
return None
# If we get here, we didn't even get an error
return MimeResponse(
mime_type=DEFAULT_MIME_TYPE,
mime_encoding=DEFAULT_MIME_ENCODING,
)
finally:
if ptr is not None:
# Release our pointer
_magic['close'](ptr)
# Release our lock
self.lock.release()
# We failed if we got here, return nothing
return None
def from_file(self, path, uncompress=False, fullscan=False):
"""
detect the type based on the content specified.
path - the file to process
uncompress - Try to look inside compressed files.
fullscan - Scan entire file and extract as much details as possible
"""
if not HAS_LIBMAGIC:
return None
if not path:
return None
# our pointer to our libmagic object
ptr = None
# our flags
flags = self.flags
if fullscan:
flags |= MAGIC_CONTINUE
if uncompress:
flags |= MAGIC_COMPRESS
# Default Response
res = ""
try:
self.lock.acquire(blocking=True)
# 'application/octet-stream; charset=binary'
# Acquire a pointer
ptr = _magic['open'](flags)
# Acquire a pointer
_magic['load'](ptr, self.magic_file)
self.errno = _magic['errno'](ptr)
if self.errno == 0:
res = self._tostr(_magic['file'](ptr, self._tobytes(path)))
# Acquire our errorstr (if one exists)
self.errstr = _magic['error'](ptr)
# Achieve our results as a list
_typ, _enc = MAGIC_LIST_RE.split(res)
_typs = []
for n, _typ_re in enumerate(TYPE_PARSE_RE.finditer(_typ)):
if n is 0:
_typs.append((0, _typ_re.group('mtype')))
else:
_typs.append(
(
int(_typ_re.group('offset')),
_typ_re.group('mtype')
),
)
_enc_re = ENCODING_PARSE_RE.match(_enc)
if _enc_re:
_enc = _enc_re.group('encoding')
mr = MimeResponse(
mime_type=_typs,
mime_encoding=_enc,
extension=self.extension_from_filename(path),
)
# return our object
return mr
except TypeError:
# This occurs if buffer check returns None
# Acquire our errorstr (if one exists)
self.errstr = _magic['error'](ptr)
if self.errstr:
# an error occured; return None
return None
# If we get here, we didn't even get an error
return MimeResponse(
mime_type=DEFAULT_MIME_TYPE,
mime_encoding=DEFAULT_MIME_ENCODING,
extension=self.extension_from_filename(path),
)
except ValueError:
# This occurs during our regular expression extraction which
# couldn't accomplish it's feat because an error string (such as
# no file exists) was returned as part of the response instead
# of our expected mime type. Store this error
self.errstr = res
# Now return None
return None
finally:
if ptr is not None:
# Release our pointer
_magic['close'](ptr)
# Release our lock
self.lock.release()
# We failed if we got here, return nothing
return None
def from_filename(self, filename):
"""
detect the type based on the filename (and/or extension) specified.
"""
if not filename:
# Invalid
return None
mime_type = next((m for m in MIME_TYPES if m[1].match(filename)), None)
if mime_type:
return MimeResponse(
mime_type=mime_type[0],
mime_encoding=mime_type[2],
extension=self.extension_from_filename(filename),
)
# No match; default response
return MimeResponse(
mime_type=DEFAULT_MIME_TYPE,
mime_encoding=DEFAULT_MIME_ENCODING,
extension=self.extension_from_filename(filename),
)
def from_bestguess(self, path):
"""
First attempts to look at the files contents (if it can), if not it
falls back to looking at the filename and always returns it's best
guess.
"""
mr = self.from_file(path)
if mr is None or mr.type() in \
(DEFAULT_MIME_TYPE, DEFAULT_MIME_EMTPTY_FILE):
_mr = self.from_filename(path)
if _mr is None:
if mr is None:
# Not parseable
return None
# Otherwise return a stream and intentionally
# do not count the empty file (if detected)
return MimeResponse(
mime_type=DEFAULT_MIME_TYPE,
mime_encoding=DEFAULT_MIME_ENCODING,
extension=self.extension_from_filename(path),
)
elif mr is not None and _mr == DEFAULT_MIME_TYPE:
return mr
return _mr
return mr
def extension_from_mime(self, mime_type):
"""
takes a mime type and returns the file extension that bests matches
it. This function returns an empty string if the mime type can't
be looked up correctly, otherwise it returns the matching extension.
"""
if not mime_type:
# Invalid; but return an empty string
return ''
# iterate over our list and return on our first match
return next((m[3] for m in MIME_TYPES if mime_type == m[0]), '')
def extension_from_filename(self, filename):
"""
Takes a filename and extracts the extension from it (if possible).
This function is a little like the os.path.splitext() which can
take a file like: test.jpg and return the .jpg. But it can not
however handle names like text.jpeg.gz (to which it would just
return .gz in this case.
This function (specifically written for newsreap/usenet parsing)
will attempt to extract a worthy extension (all of it) from the
filename. Hence:
extension_from_filename('myfile.pdf.vol03+4.par2')
will return '.pdf.vol03+4.par2'
"""
if not filename:
# Invalid; but return an empty string
return ''
result = EXTENSION_SEARCH_RE.search(filename)
if result:
ext = result.group('ext')
if ext is not None:
return ext
# Nothing found if we reach here
return ''
def _tostr(self, s, encoding='utf-8'):
if s is None:
return None
if isinstance(s, str):
return s
try: # keep Python 2 compatibility
return str(s, encoding)
except TypeError:
return str(s)
def _tobytes(self, b, encoding='utf-8'):
if b is None:
return None
if isinstance(b, bytes):
return b
try: # keep Python 2 compatibility
return bytes(b, encoding)
except TypeError:
return bytes(b)