def testExternalDbBuffer(self):
backup_db_addr = "/Users/superCat/Desktop/PycharmProjectPortable/sync/ResultSitesList.db"
table = "24/10/2015 Gardening"
backup_db = ExternalSiteDB(table, db_addr=backup_db_addr)
total = backup_db.site_count(False)
count = 0
db_addr = "/Users/superCat/Desktop/PycharmProjectPortable/sync/"
stop_event = Event()
db_buffer = ExternalTempDataDiskBuffer("TempDB.db", None, stop_event, dir_path=db_addr)
db_buffer._input_convert_tuple = False
while count < total:
results = backup_db.get_next_patch_no_rollover(count, 5000)
print(count)
db_buffer.write(results)
count += 5000
def __init__(self,
stop_event: Event,
input_queue: Queue = None,
output_queue: Queue = None,
max_worker: int = 10,
dir_path="",
is_debug=False,
**kwargs):
self._is_debug = is_debug
FeedbackInterface.__init__(self, **kwargs)
ExternalTempInterface.__init__(self)
# do not use predefined queue here
# self._input_q = input_queue
# self._output_q = output_queue
self._stop_event = stop_event
self._internal_stop_event = Event()
self._max_worker = max_worker
self._job_done = 0
self._job_done_shadow = 0
self._job_done_lock = RLock()
self._input_period = 0.0001 # time to sample data into the buffer
self._max_sample_results = 100000
self._min_sampling_duration = 0.0001
self._sample_batch_size = 5000
self._sample_batch_timeout = 60
if is_debug:
self._min_buff_delete_threshold = 10000 # default is 100000
else:
self._min_buff_delete_threshold = 100000
self._speed_penalty_count = 0
self._finished = False
manager, self._output_q = get_queue_client(
QueueManager.MachineSettingCrawler,
QueueManager.Method_Whois_Output)
self._db_buffer = ExternalTempDataDiskBuffer(
"whois_check.db",
self,
self._internal_stop_event,
buf_size=self._max_worker * 50,
terminate_callback=WhoisChecker.terminate_callback,
dir_path=dir_path)
self._populate_with_state() # FeedbackInterface
if not is_debug:
log_period = 120
else:
log_period = 10
self._progress_logger = ProgressLogger(log_period, self,
self._internal_stop_event)
def __init__(self, stop_event: Event, input_queue: Queue=None, output_queue: Queue=None,
max_worker: int=10, dir_path="", is_debug=False, **kwargs):
self._is_debug = is_debug
FeedbackInterface.__init__(self, **kwargs)
ExternalTempInterface.__init__(self)
# do not use predefined queue here
# self._input_q = input_queue
# self._output_q = output_queue
self._stop_event = stop_event
self._internal_stop_event = Event()
self._max_worker = max_worker
self._job_done = 0
self._job_done_shadow = 0
self._job_done_lock = RLock()
self._input_period = 0.0001 # time to sample data into the buffer
self._max_sample_results = 100000
self._min_sampling_duration = 0.0001
self._sample_batch_size = 5000
self._sample_batch_timeout = 60
if is_debug:
self._min_buff_delete_threshold = 10000 # default is 100000
else:
self._min_buff_delete_threshold = 100000
self._speed_penalty_count = 0
self._finished = False
manager, self._output_q = get_queue_client(QueueManager.MachineSettingCrawler, QueueManager.Method_Whois_Output)
self._db_buffer = ExternalTempDataDiskBuffer("whois_check.db", self, self._internal_stop_event, buf_size=self._max_worker*50,
terminate_callback=WhoisChecker.terminate_callback, dir_path=dir_path)
self._populate_with_state() # FeedbackInterface
if not is_debug:
log_period = 120
else:
log_period = 10
self._progress_logger = ProgressLogger(log_period, self, self._internal_stop_event)
def __init__(self,
db_ref: str,
db_dir: str,
input_queue: Queue,
output_queue: Queue,
stop_event: Event,
matrix: CrawlMatrix,
accounts: list,
force_mode=False,
force_mode_offset=0,
force_mode_total=0,
**kwargs):
FeedbackInterface.__init__(self, **kwargs)
self._stop_event = stop_event
self._matrix = matrix
self._db_ref = db_ref
self._input_queue = input_queue
self._output_queue = output_queue
self._pool_input = Queue()
self._pool = FilterPool(self._pool_input,
self._output_queue,
self._queue_lock,
self._stop_event,
self._matrix,
accounts=accounts)
self._db_buffer = ExternalTempDataDiskBuffer(
self._db_ref,
self,
self._stop_event,
dir_path=db_dir,
buf_size=2500,
output_f=5000) # control how data flow speed,
# it can keep input:output ratio = 1:1 at max 10 milion data row per hour
#FeedbackInterface.__init__(self, **kwargs)
ExternalTempInterface.__init__(self)
self._populate_with_state()
if force_mode:
new_state = _FilterState(progress=force_mode_offset,
all_data=force_mode_total)
self.populate_with_state(new_state)
def __init__(self, db_ref: str, db_dir: str, input_queue: Queue, output_queue: Queue, stop_event: Event,
matrix: CrawlMatrix, accounts: list, force_mode=False, force_mode_offset=0, force_mode_total=0, **kwargs):
FeedbackInterface.__init__(self, **kwargs)
self._stop_event = stop_event
self._matrix = matrix
self._db_ref = db_ref
self._input_queue = input_queue
self._output_queue = output_queue
self._pool_input = Queue()
self._pool = FilterPool(self._pool_input, self._output_queue, self._queue_lock, self._stop_event, self._matrix,
accounts=accounts)
self._db_buffer = ExternalTempDataDiskBuffer(self._db_ref, self, self._stop_event, dir_path=db_dir,
buf_size=2500, output_f=5000) # control how data flow speed,
# it can keep input:output ratio = 1:1 at max 10 milion data row per hour
#FeedbackInterface.__init__(self, **kwargs)
ExternalTempInterface.__init__(self)
self._populate_with_state()
if force_mode:
new_state = _FilterState(progress=force_mode_offset, all_data=force_mode_total)
self.populate_with_state(new_state)
class FilterController(FeedbackInterface, ExternalTempInterface):
"""
This Controller can be memory controlled since it implement FeedbackInterface
"""
def __init__(self,
db_ref: str,
db_dir: str,
input_queue: Queue,
output_queue: Queue,
stop_event: Event,
matrix: CrawlMatrix,
accounts: list,
force_mode=False,
force_mode_offset=0,
force_mode_total=0,
**kwargs):
FeedbackInterface.__init__(self, **kwargs)
self._stop_event = stop_event
self._matrix = matrix
self._db_ref = db_ref
self._input_queue = input_queue
self._output_queue = output_queue
self._pool_input = Queue()
self._pool = FilterPool(self._pool_input,
self._output_queue,
self._queue_lock,
self._stop_event,
self._matrix,
accounts=accounts)
self._db_buffer = ExternalTempDataDiskBuffer(
self._db_ref,
self,
self._stop_event,
dir_path=db_dir,
buf_size=2500,
output_f=5000) # control how data flow speed,
# it can keep input:output ratio = 1:1 at max 10 milion data row per hour
#FeedbackInterface.__init__(self, **kwargs)
ExternalTempInterface.__init__(self)
self._populate_with_state()
if force_mode:
new_state = _FilterState(progress=force_mode_offset,
all_data=force_mode_total)
self.populate_with_state(new_state)
@staticmethod
def get_input_parameters(db_ref: str, db_dir: str, input_queue: Queue,
output_queue: Queue, stop_event: Event,
matrix: CrawlMatrix, accounts: list,
force_mode: bool, force_mode_offset: int,
force_mode_total: int):
return {
"db_ref": db_ref,
"db_dir": db_dir,
"input_queue": input_queue,
"output_queue": output_queue,
"stop_event": stop_event,
"matrix": matrix,
"accounts": accounts,
"force_mode": force_mode,
"force_mode_offset": force_mode_offset,
"force_mode_total": force_mode_total,
}
def _sample_data_buffer_input(self):
while not self._stop_event.is_set():
data_list = []
# if isinstance(self._queue_lock, multiprocessing.RLock):
while not self._input_queue.empty():
#with self._queue_lock:
data = self._input_queue.get()
if isinstance(data, tuple) and len(data) == 2:
data_list.append(data)
if len(data_list) > 0:
self._db_buffer.append_to_buffer(data_list,
convert_tuple=False)
data_list.clear()
time.sleep(1)
def _sample_data_buffer_output(self):
for item in self._db_buffer:
if self._stop_event.is_set():
break
if isinstance(item, OnSiteLink):
self._pool_input.put(
FilteredDomainData(domain=item.link,
found=int(time.time())))
def begin_filtering(self):
self._start_sending_feedback()
self._db_buffer.start_input_output_cycle(
) # start input and output data to/from file
input_t = threading.Thread(target=self._sample_data_buffer_input)
output_t = threading.Thread(target=self._sample_data_buffer_output)
input_t.start() # start sampling data
output_t.start()
self._pool.start()
if self._pool.is_alive():
self._pool.join()
self._db_buffer.terminate()
if input_t.is_alive():
input_t.join()
if output_t.is_alive():
output_t.join()
self._end_sending_feedback()
def populate_with_state(self, state):
"""
FeedbackInterface, subclass implement this method to
:param state: the state from previous
:return:
"""
if isinstance(state, _FilterState):
self._pool.set_job_done(state.progress)
self._db_buffer.set_progress(state.progress)
def get_state(self):
"""
FeedbackInterface, subclass this so that the controller can gather state info, which in turn will feedback into next iteration
:return:
"""
return _FilterState(progress=self._pool.get_job_done(),
all_data=self._db_buffer.get_total_record())
def get_callback_data(self):
"""
FeedbackInterface, subclass this so that any callback data can be gathered by the controller
:return:
"""
return None
def is_programme_finshed(self):
"""
FeedbackInterface, indicate if the programme has finished execution, if not it goes to next iteration
:return:
"""
return self._stop_event.is_set()
def get_external_count_finished(self) -> int:
"""
ExternalTempInterface, get the number of job done in ExternalTempDataDiskBuffer
:return:
"""
return self._pool.get_job_done()
def set_internal_count(self, count: int):
"""
ExternalTempInterface, set the number of job done in ExternalTempDataDiskBuffer
:param count:
:return:
"""
self._pool.set_job_done(count)
class WhoisChecker(FeedbackInterface, ExternalTempInterface,
ProgressLogInterface):
"""
usage:
checker = WhoisChecker(...)
checker.run_farm()
"""
def __init__(self,
stop_event: Event,
input_queue: Queue = None,
output_queue: Queue = None,
max_worker: int = 10,
dir_path="",
is_debug=False,
**kwargs):
self._is_debug = is_debug
FeedbackInterface.__init__(self, **kwargs)
ExternalTempInterface.__init__(self)
# do not use predefined queue here
# self._input_q = input_queue
# self._output_q = output_queue
self._stop_event = stop_event
self._internal_stop_event = Event()
self._max_worker = max_worker
self._job_done = 0
self._job_done_shadow = 0
self._job_done_lock = RLock()
self._input_period = 0.0001 # time to sample data into the buffer
self._max_sample_results = 100000
self._min_sampling_duration = 0.0001
self._sample_batch_size = 5000
self._sample_batch_timeout = 60
if is_debug:
self._min_buff_delete_threshold = 10000 # default is 100000
else:
self._min_buff_delete_threshold = 100000
self._speed_penalty_count = 0
self._finished = False
manager, self._output_q = get_queue_client(
QueueManager.MachineSettingCrawler,
QueueManager.Method_Whois_Output)
self._db_buffer = ExternalTempDataDiskBuffer(
"whois_check.db",
self,
self._internal_stop_event,
buf_size=self._max_worker * 50,
terminate_callback=WhoisChecker.terminate_callback,
dir_path=dir_path)
self._populate_with_state() # FeedbackInterface
if not is_debug:
log_period = 120
else:
log_period = 10
self._progress_logger = ProgressLogger(log_period, self,
self._internal_stop_event)
@staticmethod
def terminate_callback():
ErrorLogger.log_error("WhoisChecker", StopIteration("terminated."))
@staticmethod
def get_input_parameters(input_queue: Queue,
output_queue: Queue,
stop_event: Event,
max_worker: int = 20) -> dict:
return {
"input_queue": input_queue,
"output_queue": output_queue,
"stop_event": stop_event,
"max_worker": max_worker
}
def get_job_done_count(self):
with self._job_done_lock:
job_done = self._job_done
return job_done
def _add_job_done_one(self):
with self._job_done_lock:
self._job_done += 1
def _put_output_result_in_queue(self, domain_data: OnSiteLink):
if not self._stop_event.is_set(
) or not self._internal_stop_event.is_set():
try:
self._output_q.put(
(domain_data.link, domain_data.response_code))
except Exception as inner_ex:
if self._output_q is None:
manager, self._output_q = get_queue_client(
QueueManager.MachineSettingCrawler,
QueueManager.Method_Whois_Output)
ErrorLogger.log_error(
"WhoisChecker",
inner_ex,
addtional="failed to put result in queue.")
time.sleep(0.01)
self._put_output_result_in_queue(domain_data)
def _check_whois_v1(self, domain_data: OnSiteLink):
root_domain = domain_data.link
try:
if root_domain.startswith("http"):
root_domain = LinkChecker.get_root_domain(domain_data.link)[1]
real_response_code = domain_data.response_code
whois = LinkChecker.check_whois(root_domain) # check whois record
if whois[0]:
if whois[2]: # domain is expired
real_response_code = ResponseCode.Expired
else:
real_response_code = ResponseCode.MightBeExpired
if real_response_code == ResponseCode.Expired:
#if ResponseCode.domain_might_be_expired(real_response_code):
domain_data.link = root_domain
domain_data.response_code = real_response_code
#return_obj = OnSiteLink(root_domain, real_response_code, domain_data.link_level, OnSiteLink.TypeOutbound)
# if isinstance(self._queue_lock, multiprocessing.RLock):
with self._queue_lock:
self._output_q.put(
(domain_data.link, domain_data.response_code))
except Exception as ex:
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker", ex,
"_check_whois() " + root_domain)
finally:
self._add_job_done_one()
def _check_whois(self, domain_data: OnSiteLink):
root_domain = domain_data.link.lower()
try:
if not self._is_debug:
if root_domain.startswith("http"):
root_domain = LinkChecker.get_root_domain(
domain_data.link)[1]
is_available, is_redemption = LinkChecker.is_domain_available_whois(
root_domain) # check whois record
if is_available or is_redemption:
if is_available:
real_response_code = ResponseCode.Expired
else:
real_response_code = ResponseCode.MightBeExpired
domain_data.link = root_domain
domain_data.response_code = real_response_code
#return_obj = OnSiteLink(root_domain, real_response_code, domain_data.link_level, OnSiteLink.TypeOutbound)
self._put_output_result_in_queue(domain_data)
else:
self._put_output_result_in_queue(domain_data)
except Exception as ex:
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker", ex,
"_check_whois() " + root_domain)
finally:
self._add_job_done_one()
def _check_whois_with_dns(self, page: OnSiteLink):
real_response_code = ResponseCode.DNSError
skip_whois_check = False
try:
if not self._is_debug:
root_result = LinkChecker.get_root_domain(page.link)
root_domain = root_result[1]
sub_domain = root_result[4]
suffix = root_result[5]
if len(sub_domain
) == 0 or suffix not in TldUtility.TOP_TLD_LIST:
skip_whois_check = True
else:
if LinkChecker.is_domain_DNS_OK(
sub_domain): # check DNS first
real_response_code = ResponseCode.NoDNSError
skip_whois_check = True
elif not sub_domain.startswith("www."):
if LinkChecker.is_domain_DNS_OK("www." + root_domain):
real_response_code = ResponseCode.NoDNSError
skip_whois_check = True
page.response_code = real_response_code
page.link_type = OnSiteLink.TypeOutbound
page.link = root_domain
except Exception as ex:
ErrorLogger.log_error("WhoisChecker", ex,
"_check_whois_with_dns() " + page.link)
skip_whois_check = True
finally:
if not skip_whois_check and real_response_code == ResponseCode.DNSError:
self._check_whois(page)
else:
self._add_job_done_one()
def _sample_data(self):
ref_time = time.time()
manager, result_queue = get_queue_client(
QueueManager.MachineSettingCrawler,
QueueManager.Method_Whois_Input)
if result_queue is None:
ErrorLogger.log_error(
"ExternalSiteChecker.WhoisChecker._sample_data()",
ValueError("result queue is None, cannot get data."))
if not (self._stop_event.is_set()
or self._internal_stop_event.is_set()):
self._sample_data()
else:
while not (self._stop_event.is_set()
or self._internal_stop_event.is_set()):
data_list = []
counter = 0
while not result_queue.empty():
data = None
try:
data = result_queue.get()
except Exception as ex:
ErrorLogger.log_error("WhoisChecker._sampling_data",
ex)
if result_queue is None:
manager, result_queue = get_queue_client(
QueueManager.MachineSettingCrawler,
QueueManager.Method_Whois_Input)
if isinstance(data, OnSiteLink):
counter += 1
data_list.append((data.link, data.response_code))
elif isinstance(data, tuple) and len(data) == 2:
#print("External Site checker: recieved:", data)
counter += 1
data_list.append(data)
if isinstance(data, list):
data_list += data
counter += 1
if counter >= self._sample_batch_size:
break
current_time = time.time()
if current_time - ref_time >= self._sample_batch_timeout:
break
time.sleep(self._min_sampling_duration)
ref_time = time.time()
if len(data_list) > 0:
#print("whois checker input data in db_buff: ", len(data_list))
self._db_buffer.append_to_buffer(data_list,
convert_tuple=False)
data_list.clear()
else:
pass
time.sleep(self._input_period)
def sample_gen(self):
while not self._stop_event.is_set(
) or not self._internal_stop_event.is_set():
try:
if not self._db_buffer.reset_event.is_set():
for next_item in self._db_buffer:
time.sleep(0.0001)
return next_item
print("going to next cycle in sample_gen.")
time.sleep(1)
except Exception as ex:
print("sample_gen error:", ex)
def run_farm(self):
try:
self._start_sending_feedback()
input_t = threading.Thread(target=self._sample_data)
input_t.start() # start sampling data
self._progress_logger.start()
self._db_buffer.start_input_output_cycle(
) # start input and output data to/from file
pool = ThreadPool(processes=self._max_worker)
# pool.imap_unordered(self._check_whois_with_dns, self._db_buffer, chunksize=1)
pool.imap_unordered(self._check_whois_with_dns,
iter(self.sample_gen, None),
chunksize=1)
while not self._stop_event.is_set(
) or not self._internal_stop_event.is_set():
time.sleep(1)
if self._stop_event.is_set():
self._internal_stop_event.set()
input_t.join()
self._progress_logger.join()
self._db_buffer.terminate()
if self._stop_event.is_set():
self._finished = True
self._end_sending_feedback()
except Exception as ex:
if self._stop_event.is_set():
self._finished = True
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker", ex,
"run_farm() index at:" + str(self._job_done))
def get_external_count_finished(self):
"""
ExternalTempInterface
:return:
"""
return self.get_job_done_count()
def set_internal_count(self, count: int):
"""
ExternalTempInterface
:param count:
:return:
"""
self._job_done = count
def populate_with_state(self, state):
"""
FeedbackInterface, subclass implement this method to
:param state: the state from previous
:return:
"""
if isinstance(state, WhoisCheckerState):
self._job_done = state.progress_count
self._db_buffer.set_progress(state.progress_count)
def get_state(self):
"""
FeedbackInterface, subclass this so that the controller can gather state info, which in turn will feedback into next iteration
:return:
"""
if self._internal_stop_event.is_set():
return WhoisCheckerState(0, 0)
else:
return WhoisCheckerState(self.get_job_done_count(),
self._db_buffer.get_total_record())
def get_callback_data(self):
"""
FeedbackInterface, subclass this so that any callback data can be gathered by the controller
:return:
"""
return None
def is_programme_finshed(self):
"""
FeedbackInterface, indicate if the programme has finished execution, if not it goes to next iteration
:return:
"""
return self._finished
def is_progamme_need_restart(self):
"""
FeedbackInterface, indicate if the programme needs to be restarted, if not it goes to next iteration
"""
return self._internal_stop_event.is_set()
def get_file_name(self) -> str:
"""
ProgressLogInterface, the file name used to save in file system.
:return:
"""
return "whois_check_progress.csv"
def get_column_names(self) -> []:
"""
ProgressLogInterface, the column name for each prograss entry in get_prograss(), all in str format
:return: array contains column names, length should match the length of prograss entries
"""
return ["Done", "Total"]
def reset(self):
self._job_done = 0
self._job_done_shadow = 0
self._speed_penalty_count = 0
def get_progress(self) -> []:
"""
ProgressLogInterface, get the prograss data in tuple format, so that it can be used to complie to standard format
:return: array contains prograss data, which has the exact length of column names in get_column_names()
"""
total_record = self._db_buffer.get_total_record()
if (self._job_done == self._job_done_shadow and self._job_done > 0
) or (self._job_done > self._min_buff_delete_threshold * 0.9
and total_record > self._min_buff_delete_threshold):
self._speed_penalty_count += 1
if self._speed_penalty_count >= 2:
ErrorLogger.log_error(
"WhoisChecker.get_progress()",
TimeoutError("progress is stucked, restarted internal."),
self._db_buffer._file_name)
print("going to clear cache")
self._db_buffer.clear_cache()
self.reset()
total_record = 0
self._db_buffer.start_input_output_cycle()
else:
print("no need to clear cache.")
self._job_done_shadow = self._job_done
self._speed_penalty_count = 0
return [self._job_done, total_record]
def get_limit(self) -> int:
"""
ProgressLogInterface, the number of samples you want to collect.
:return: max number of samples
"""
return self._max_sample_results
class WhoisChecker(FeedbackInterface, ExternalTempInterface, ProgressLogInterface):
"""
usage:
checker = WhoisChecker(...)
checker.run_farm()
"""
def __init__(self, stop_event: Event, input_queue: Queue=None, output_queue: Queue=None,
max_worker: int=10, dir_path="", is_debug=False, **kwargs):
self._is_debug = is_debug
FeedbackInterface.__init__(self, **kwargs)
ExternalTempInterface.__init__(self)
# do not use predefined queue here
# self._input_q = input_queue
# self._output_q = output_queue
self._stop_event = stop_event
self._internal_stop_event = Event()
self._max_worker = max_worker
self._job_done = 0
self._job_done_shadow = 0
self._job_done_lock = RLock()
self._input_period = 0.0001 # time to sample data into the buffer
self._max_sample_results = 100000
self._min_sampling_duration = 0.0001
self._sample_batch_size = 5000
self._sample_batch_timeout = 60
if is_debug:
self._min_buff_delete_threshold = 10000 # default is 100000
else:
self._min_buff_delete_threshold = 100000
self._speed_penalty_count = 0
self._finished = False
manager, self._output_q = get_queue_client(QueueManager.MachineSettingCrawler, QueueManager.Method_Whois_Output)
self._db_buffer = ExternalTempDataDiskBuffer("whois_check.db", self, self._internal_stop_event, buf_size=self._max_worker*50,
terminate_callback=WhoisChecker.terminate_callback, dir_path=dir_path)
self._populate_with_state() # FeedbackInterface
if not is_debug:
log_period = 120
else:
log_period = 10
self._progress_logger = ProgressLogger(log_period, self, self._internal_stop_event)
@staticmethod
def terminate_callback():
ErrorLogger.log_error("WhoisChecker", StopIteration("terminated."))
@staticmethod
def get_input_parameters(input_queue: Queue, output_queue: Queue, stop_event: Event, max_worker: int=20) -> dict:
return {"input_queue": input_queue, "output_queue": output_queue, "stop_event": stop_event,
"max_worker": max_worker}
def get_job_done_count(self):
with self._job_done_lock:
job_done = self._job_done
return job_done
def _add_job_done_one(self):
with self._job_done_lock:
self._job_done += 1
def _put_output_result_in_queue(self, domain_data: OnSiteLink):
if not self._stop_event.is_set() or not self._internal_stop_event.is_set():
try:
self._output_q.put((domain_data.link, domain_data.response_code))
except Exception as inner_ex:
if self._output_q is None:
manager, self._output_q = get_queue_client(QueueManager.MachineSettingCrawler, QueueManager.Method_Whois_Output)
ErrorLogger.log_error("WhoisChecker", inner_ex, addtional="failed to put result in queue.")
time.sleep(0.01)
self._put_output_result_in_queue(domain_data)
def _check_whois_v1(self, domain_data: OnSiteLink):
root_domain = domain_data.link
try:
if root_domain.startswith("http"):
root_domain = LinkChecker.get_root_domain(domain_data.link)[1]
real_response_code = domain_data.response_code
whois = LinkChecker.check_whois(root_domain) # check whois record
if whois[0]:
if whois[2]: # domain is expired
real_response_code = ResponseCode.Expired
else:
real_response_code = ResponseCode.MightBeExpired
if real_response_code == ResponseCode.Expired:
#if ResponseCode.domain_might_be_expired(real_response_code):
domain_data.link = root_domain
domain_data.response_code = real_response_code
#return_obj = OnSiteLink(root_domain, real_response_code, domain_data.link_level, OnSiteLink.TypeOutbound)
# if isinstance(self._queue_lock, multiprocessing.RLock):
with self._queue_lock:
self._output_q.put((domain_data.link, domain_data.response_code))
except Exception as ex:
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker", ex, "_check_whois() " + root_domain)
finally:
self._add_job_done_one()
def _check_whois(self, domain_data: OnSiteLink):
root_domain = domain_data.link.lower()
try:
if not self._is_debug:
if root_domain.startswith("http"):
root_domain = LinkChecker.get_root_domain(domain_data.link)[1]
is_available, is_redemption = LinkChecker.is_domain_available_whois(root_domain) # check whois record
if is_available or is_redemption:
if is_available:
real_response_code = ResponseCode.Expired
else:
real_response_code = ResponseCode.MightBeExpired
domain_data.link = root_domain
domain_data.response_code = real_response_code
#return_obj = OnSiteLink(root_domain, real_response_code, domain_data.link_level, OnSiteLink.TypeOutbound)
self._put_output_result_in_queue(domain_data)
else:
self._put_output_result_in_queue(domain_data)
except Exception as ex:
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker", ex, "_check_whois() " + root_domain)
finally:
self._add_job_done_one()
def _check_whois_with_dns(self, page: OnSiteLink):
real_response_code = ResponseCode.DNSError
skip_whois_check = False
try:
if not self._is_debug:
root_result = LinkChecker.get_root_domain(page.link)
root_domain = root_result[1]
sub_domain = root_result[4]
suffix = root_result[5]
if len(sub_domain) == 0 or suffix not in TldUtility.TOP_TLD_LIST:
skip_whois_check = True
else:
if LinkChecker.is_domain_DNS_OK(sub_domain): # check DNS first
real_response_code = ResponseCode.NoDNSError
skip_whois_check = True
elif not sub_domain.startswith("www."):
if LinkChecker.is_domain_DNS_OK("www." + root_domain):
real_response_code = ResponseCode.NoDNSError
skip_whois_check = True
page.response_code = real_response_code
page.link_type = OnSiteLink.TypeOutbound
page.link = root_domain
except Exception as ex:
ErrorLogger.log_error("WhoisChecker", ex, "_check_whois_with_dns() " + page.link)
skip_whois_check = True
finally:
if not skip_whois_check and real_response_code == ResponseCode.DNSError:
self._check_whois(page)
else:
self._add_job_done_one()
def _sample_data(self):
ref_time = time.time()
manager, result_queue = get_queue_client(QueueManager.MachineSettingCrawler, QueueManager.Method_Whois_Input)
if result_queue is None:
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker._sample_data()", ValueError("result queue is None, cannot get data."))
if not (self._stop_event.is_set() or self._internal_stop_event.is_set()):
self._sample_data()
else:
while not (self._stop_event.is_set() or self._internal_stop_event.is_set()):
data_list = []
counter = 0
while not result_queue.empty():
data = None
try:
data = result_queue.get()
except Exception as ex:
ErrorLogger.log_error("WhoisChecker._sampling_data", ex)
if result_queue is None:
manager, result_queue = get_queue_client(QueueManager.MachineSettingCrawler,
QueueManager.Method_Whois_Input)
if isinstance(data, OnSiteLink):
counter += 1
data_list.append((data.link, data.response_code))
elif isinstance(data, tuple) and len(data) == 2:
#print("External Site checker: recieved:", data)
counter += 1
data_list.append(data)
if isinstance(data, list):
data_list += data
counter += 1
if counter >= self._sample_batch_size:
break
current_time = time.time()
if current_time - ref_time >= self._sample_batch_timeout:
break
time.sleep(self._min_sampling_duration)
ref_time = time.time()
if len(data_list) > 0:
#print("whois checker input data in db_buff: ", len(data_list))
self._db_buffer.append_to_buffer(data_list, convert_tuple=False)
data_list.clear()
else:
pass
time.sleep(self._input_period)
def sample_gen(self):
while not self._stop_event.is_set() or not self._internal_stop_event.is_set():
try:
if not self._db_buffer.reset_event.is_set():
for next_item in self._db_buffer:
time.sleep(0.0001)
return next_item
print("going to next cycle in sample_gen.")
time.sleep(1)
except Exception as ex:
print("sample_gen error:", ex)
def run_farm(self):
try:
self._start_sending_feedback()
input_t = threading.Thread(target=self._sample_data)
input_t.start() # start sampling data
self._progress_logger.start()
self._db_buffer.start_input_output_cycle() # start input and output data to/from file
pool = ThreadPool(processes=self._max_worker)
# pool.imap_unordered(self._check_whois_with_dns, self._db_buffer, chunksize=1)
pool.imap_unordered(self._check_whois_with_dns, iter(self.sample_gen, None), chunksize=1)
while not self._stop_event.is_set() or not self._internal_stop_event.is_set():
time.sleep(1)
if self._stop_event.is_set():
self._internal_stop_event.set()
input_t.join()
self._progress_logger.join()
self._db_buffer.terminate()
if self._stop_event.is_set():
self._finished = True
self._end_sending_feedback()
except Exception as ex:
if self._stop_event.is_set():
self._finished = True
ErrorLogger.log_error("ExternalSiteChecker.WhoisChecker", ex, "run_farm() index at:" + str(self._job_done))
def get_external_count_finished(self):
"""
ExternalTempInterface
:return:
"""
return self.get_job_done_count()
def set_internal_count(self, count: int):
"""
ExternalTempInterface
:param count:
:return:
"""
self._job_done = count
def populate_with_state(self, state):
"""
FeedbackInterface, subclass implement this method to
:param state: the state from previous
:return:
"""
if isinstance(state, WhoisCheckerState):
self._job_done = state.progress_count
self._db_buffer.set_progress(state.progress_count)
def get_state(self):
"""
FeedbackInterface, subclass this so that the controller can gather state info, which in turn will feedback into next iteration
:return:
"""
if self._internal_stop_event.is_set():
return WhoisCheckerState(0, 0)
else:
return WhoisCheckerState(self.get_job_done_count(), self._db_buffer.get_total_record())
def get_callback_data(self):
"""
FeedbackInterface, subclass this so that any callback data can be gathered by the controller
:return:
"""
return None
def is_programme_finshed(self):
"""
FeedbackInterface, indicate if the programme has finished execution, if not it goes to next iteration
:return:
"""
return self._finished
def is_progamme_need_restart(self):
"""
FeedbackInterface, indicate if the programme needs to be restarted, if not it goes to next iteration
"""
return self._internal_stop_event.is_set()
def get_file_name(self) -> str:
"""
ProgressLogInterface, the file name used to save in file system.
:return:
"""
return "whois_check_progress.csv"
def get_column_names(self) -> []:
"""
ProgressLogInterface, the column name for each prograss entry in get_prograss(), all in str format
:return: array contains column names, length should match the length of prograss entries
"""
return ["Done", "Total"]
def reset(self):
self._job_done = 0
self._job_done_shadow = 0
self._speed_penalty_count = 0
def get_progress(self) -> []:
"""
ProgressLogInterface, get the prograss data in tuple format, so that it can be used to complie to standard format
:return: array contains prograss data, which has the exact length of column names in get_column_names()
"""
total_record = self._db_buffer.get_total_record()
if (self._job_done == self._job_done_shadow and self._job_done > 0) or (self._job_done > self._min_buff_delete_threshold * 0.9 and total_record > self._min_buff_delete_threshold):
self._speed_penalty_count += 1
if self._speed_penalty_count >= 2:
ErrorLogger.log_error("WhoisChecker.get_progress()", TimeoutError("progress is stucked, restarted internal."), self._db_buffer._file_name)
print("going to clear cache")
self._db_buffer.clear_cache()
self.reset()
total_record = 0
self._db_buffer.start_input_output_cycle()
else:
print("no need to clear cache.")
self._job_done_shadow = self._job_done
self._speed_penalty_count = 0
return [self._job_done, total_record]
def get_limit(self) -> int:
"""
ProgressLogInterface, the number of samples you want to collect.
:return: max number of samples
"""
return self._max_sample_results
class FilterController(FeedbackInterface, ExternalTempInterface):
"""
This Controller can be memory controlled since it implement FeedbackInterface
"""
def __init__(self, db_ref: str, db_dir: str, input_queue: Queue, output_queue: Queue, stop_event: Event,
matrix: CrawlMatrix, accounts: list, force_mode=False, force_mode_offset=0, force_mode_total=0, **kwargs):
FeedbackInterface.__init__(self, **kwargs)
self._stop_event = stop_event
self._matrix = matrix
self._db_ref = db_ref
self._input_queue = input_queue
self._output_queue = output_queue
self._pool_input = Queue()
self._pool = FilterPool(self._pool_input, self._output_queue, self._queue_lock, self._stop_event, self._matrix,
accounts=accounts)
self._db_buffer = ExternalTempDataDiskBuffer(self._db_ref, self, self._stop_event, dir_path=db_dir,
buf_size=2500, output_f=5000) # control how data flow speed,
# it can keep input:output ratio = 1:1 at max 10 milion data row per hour
#FeedbackInterface.__init__(self, **kwargs)
ExternalTempInterface.__init__(self)
self._populate_with_state()
if force_mode:
new_state = _FilterState(progress=force_mode_offset, all_data=force_mode_total)
self.populate_with_state(new_state)
@staticmethod
def get_input_parameters(db_ref: str, db_dir: str, input_queue: Queue, output_queue: Queue, stop_event: Event,
matrix: CrawlMatrix, accounts: list, force_mode: bool, force_mode_offset:int,force_mode_total:int):
return {"db_ref": db_ref, "db_dir": db_dir, "input_queue": input_queue, "output_queue": output_queue, "stop_event": stop_event,
"matrix": matrix, "accounts": accounts,
"force_mode": force_mode, "force_mode_offset": force_mode_offset, "force_mode_total": force_mode_total,}
def _sample_data_buffer_input(self):
while not self._stop_event.is_set():
data_list = []
# if isinstance(self._queue_lock, multiprocessing.RLock):
while not self._input_queue.empty():
#with self._queue_lock:
data = self._input_queue.get()
if isinstance(data, tuple) and len(data) == 2:
data_list.append(data)
if len(data_list) > 0:
self._db_buffer.append_to_buffer(data_list, convert_tuple=False)
data_list.clear()
time.sleep(1)
def _sample_data_buffer_output(self):
for item in self._db_buffer:
if self._stop_event.is_set():
break
if isinstance(item, OnSiteLink):
self._pool_input.put(FilteredDomainData(domain=item.link, found=int(time.time())))
def begin_filtering(self):
self._start_sending_feedback()
self._db_buffer.start_input_output_cycle() # start input and output data to/from file
input_t = threading.Thread(target=self._sample_data_buffer_input)
output_t = threading.Thread(target=self._sample_data_buffer_output)
input_t.start() # start sampling data
output_t.start()
self._pool.start()
if self._pool.is_alive():
self._pool.join()
self._db_buffer.terminate()
if input_t.is_alive():
input_t.join()
if output_t.is_alive():
output_t.join()
self._end_sending_feedback()
def populate_with_state(self, state):
"""
FeedbackInterface, subclass implement this method to
:param state: the state from previous
:return:
"""
if isinstance(state, _FilterState):
self._pool.set_job_done(state.progress)
self._db_buffer.set_progress(state.progress)
def get_state(self):
"""
FeedbackInterface, subclass this so that the controller can gather state info, which in turn will feedback into next iteration
:return:
"""
return _FilterState(progress=self._pool.get_job_done(), all_data=self._db_buffer.get_total_record())
def get_callback_data(self):
"""
FeedbackInterface, subclass this so that any callback data can be gathered by the controller
:return:
"""
return None
def is_programme_finshed(self):
"""
FeedbackInterface, indicate if the programme has finished execution, if not it goes to next iteration
:return:
"""
return self._stop_event.is_set()
def get_external_count_finished(self) -> int:
"""
ExternalTempInterface, get the number of job done in ExternalTempDataDiskBuffer
:return:
"""
return self._pool.get_job_done()
def set_internal_count(self, count: int):
"""
ExternalTempInterface, set the number of job done in ExternalTempDataDiskBuffer
:param count:
:return:
"""
self._pool.set_job_done(count)
def __init__(self, full_link: str="", data_source: SiteTempDataSrcInterface=None,
controller: SiteCheckerController=None,
max_level=10, max_page=1000, delegate=None, output_buff_size=2000,
output_queue=None, output_all_external=False, result_delegate=None,
memory_control_terminate_event=None, check_robot_text=True,
**kwargs):
"""
:param full_link: The full link of a domain, e.g: https://www.google.co.uk
:param domain: domain to crawl
:param max_level: stop crawling if it reaches this level
:param max_page: maximum pages to check within a site, also stop crawling
:param delegate: if this is not None, then it will send the latest result of external domain of ResponseCode==404 or 999
:param result_delegate: send site_info upon finish
:param memory_control_terminate_event: if this is not None and being set, it will be able to terminate an external memory controlled process.
:return:
"""
FeedbackInterface.__init__(self, **kwargs)
#super(SiteChecker, self).__init__(**kwargs)
if full_link is None or len(full_link) == 0:
raise ValueError()
original_path = ""
try:
paras = urlsplit(full_link)
self.scheme, self.domain, original_path = paras[0], paras[1], paras[2]
except:
pass
domain_data = LinkChecker.get_root_domain(full_link, False)
self.root_domain = domain_data[1]
self.sub_domain = domain_data[4]
self.domain_suffix = domain_data[5]
self.sub_domain_no_local = self.sub_domain.strip(self.domain_suffix)
if self.scheme == "":
self.scheme = "http"
if self.domain == "":
self.domain = self.root_domain
self.orginal_link = full_link
self.domain_link = LinkChecker.get_valid_link(self.root_domain, full_link, self.scheme)
self.max_level = max_level
self.max_page = max_page
self.page_count = 0 # keep track page done
self._page_count_shadow = 0 # track previous count
self._all_page_count_shadow = 0 #track previous count in datasource
self.internal_page_count = 0
self.internal_page_last_count = 0
self.page_allocated = 0
self.current_level = 0 # if this = 0, it is root domain/home_page
self._stop_event = Event()
valid_file_name = SiteTempDataSrcInterface.get_valid_file_name(self.domain_link)
self._external_db_buffer = ExternalTempDataDiskBuffer(valid_file_name+".ext.db", self,
stop_event=self._stop_event,
buf_size=int(output_buff_size/2),
dir_path=get_db_buffer_default_dir(),
convert_output=False)
self._external_db_buffer.append_to_buffer([(self.root_domain, ResponseCode.DNSError),], convert_tuple=False)
self._memory_control_terminate_event = memory_control_terminate_event
self.task_control_lock = threading.RLock()
if data_source is None:
#self.data_source = SiteTempDataDisk(self.root_domain, ref_obj=self)
self.data_source = SiteTempDataDiskWithBuff(ref=self.domain_link, output_buff_size=output_buff_size, ref_obj=self)
else:
self.data_source = data_source # a list of OnSiteLink
self.delegate = delegate
if LinkChecker.might_be_link_html_page(original_path):
self.data_source.append(OnSiteLink(self.domain_link, response_code=ResponseCode.LinkOK, link_level=1)) # add the root domain as a starting point
self.data_source.append(OnSiteLink(self.scheme + "://www."+self.sub_domain, ResponseCode.LinkOK, link_level=1))
self.data_source.append(OnSiteLink(self.scheme + "://" + self.domain, ResponseCode.LinkOK, link_level=1))
self.cache_list = [] # internal page cache
self.page_need_look_up_temp = 0
self.cache_list.append(self.domain_link)
if "www." not in self.sub_domain:
self.cache_list.append(self.scheme + "://www."+self.sub_domain)
self.cache_list.append(self.scheme + "://" + self.domain)
self.page_need_look_up = self.data_source.count_all()
self.cache_size = 500 # create a small cache list to avoid going to check link in file system with lots of read and write
self._double_check_cache_lock = threading.RLock()
self._double_check_cache = deque(maxlen=self.cache_size)
self.external_cache_list = []
self.external_cache_size = 500 # cache that hold external sites
self.external_links_checked = 0
self.add_internal_page_OK_only = True
self.output_queue = output_queue
self.output_all_external = output_all_external
self.controller = controller
self.result_delegate = result_delegate
self.page_count_lock = threading.RLock()
self.internal_page_count_lock = threading.RLock()
self.level_lock = threading.RLock()
self.page_look_up_lock = threading.RLock()
self.external_link_check_lock = threading.RLock()
self._finihsed = False
self.task_control_max = 1
self.agent = "VegeBot (we follow your robots.txt settings before crawling, you can slow down the bot by change the Crawl-Delay parameter in the settings." \
"if you have an enquiry, please email to: [email protected])"
self.agent_from = "*****@*****.**"
if check_robot_text:
self.robot_agent = LinkChecker.get_robot_agent(self.sub_domain, protocol=self.scheme)
else:
self.robot_agent = None
self.site_crawl_delay = 0.60
if isinstance(self.robot_agent, Rules):
delay_temp = self.robot_agent.delay(self.agent)
if delay_temp is not None and delay_temp != self.site_crawl_delay:
self.site_crawl_delay = delay_temp
self.task_control_counter = 1
self._speed_penalty_count = 0
self._speed_penalty_threshold = 10
self._progress_logging_speed = 120
self._output_period = 120
self._output_batch_size = 100
self._death_wish_sent = False
SiteChecker._is_lxml_parser_exist()
self._output_thread = None
self._output_queue = None
self.progress_logger = ProgressLogger(self._progress_logging_speed, self, self._stop_event)
self._status = "Start"
self._populate_with_state() # restore laste known state
class SiteChecker(FeedbackInterface, SiteTempDataSrcRefInterface, ProgressLogInterface, ExternalTempInterface):
full_link_key = "full_link"
datasource_key = "data_source"
controller_ley = "controller"
max_level_key = "max_level"
max_page_key = "max_page"
output_queue_key = "output_queue"
_use_lxml_parser = False
def __init__(self, full_link: str="", data_source: SiteTempDataSrcInterface=None,
controller: SiteCheckerController=None,
max_level=10, max_page=1000, delegate=None, output_buff_size=2000,
output_queue=None, output_all_external=False, result_delegate=None,
memory_control_terminate_event=None, check_robot_text=True,
**kwargs):
"""
:param full_link: The full link of a domain, e.g: https://www.google.co.uk
:param domain: domain to crawl
:param max_level: stop crawling if it reaches this level
:param max_page: maximum pages to check within a site, also stop crawling
:param delegate: if this is not None, then it will send the latest result of external domain of ResponseCode==404 or 999
:param result_delegate: send site_info upon finish
:param memory_control_terminate_event: if this is not None and being set, it will be able to terminate an external memory controlled process.
:return:
"""
FeedbackInterface.__init__(self, **kwargs)
#super(SiteChecker, self).__init__(**kwargs)
if full_link is None or len(full_link) == 0:
raise ValueError()
original_path = ""
try:
paras = urlsplit(full_link)
self.scheme, self.domain, original_path = paras[0], paras[1], paras[2]
except:
pass
domain_data = LinkChecker.get_root_domain(full_link, False)
self.root_domain = domain_data[1]
self.sub_domain = domain_data[4]
self.domain_suffix = domain_data[5]
self.sub_domain_no_local = self.sub_domain.strip(self.domain_suffix)
if self.scheme == "":
self.scheme = "http"
if self.domain == "":
self.domain = self.root_domain
self.orginal_link = full_link
self.domain_link = LinkChecker.get_valid_link(self.root_domain, full_link, self.scheme)
self.max_level = max_level
self.max_page = max_page
self.page_count = 0 # keep track page done
self._page_count_shadow = 0 # track previous count
self._all_page_count_shadow = 0 #track previous count in datasource
self.internal_page_count = 0
self.internal_page_last_count = 0
self.page_allocated = 0
self.current_level = 0 # if this = 0, it is root domain/home_page
self._stop_event = Event()
valid_file_name = SiteTempDataSrcInterface.get_valid_file_name(self.domain_link)
self._external_db_buffer = ExternalTempDataDiskBuffer(valid_file_name+".ext.db", self,
stop_event=self._stop_event,
buf_size=int(output_buff_size/2),
dir_path=get_db_buffer_default_dir(),
convert_output=False)
self._external_db_buffer.append_to_buffer([(self.root_domain, ResponseCode.DNSError),], convert_tuple=False)
self._memory_control_terminate_event = memory_control_terminate_event
self.task_control_lock = threading.RLock()
if data_source is None:
#self.data_source = SiteTempDataDisk(self.root_domain, ref_obj=self)
self.data_source = SiteTempDataDiskWithBuff(ref=self.domain_link, output_buff_size=output_buff_size, ref_obj=self)
else:
self.data_source = data_source # a list of OnSiteLink
self.delegate = delegate
if LinkChecker.might_be_link_html_page(original_path):
self.data_source.append(OnSiteLink(self.domain_link, response_code=ResponseCode.LinkOK, link_level=1)) # add the root domain as a starting point
self.data_source.append(OnSiteLink(self.scheme + "://www."+self.sub_domain, ResponseCode.LinkOK, link_level=1))
self.data_source.append(OnSiteLink(self.scheme + "://" + self.domain, ResponseCode.LinkOK, link_level=1))
self.cache_list = [] # internal page cache
self.page_need_look_up_temp = 0
self.cache_list.append(self.domain_link)
if "www." not in self.sub_domain:
self.cache_list.append(self.scheme + "://www."+self.sub_domain)
self.cache_list.append(self.scheme + "://" + self.domain)
self.page_need_look_up = self.data_source.count_all()
self.cache_size = 500 # create a small cache list to avoid going to check link in file system with lots of read and write
self._double_check_cache_lock = threading.RLock()
self._double_check_cache = deque(maxlen=self.cache_size)
self.external_cache_list = []
self.external_cache_size = 500 # cache that hold external sites
self.external_links_checked = 0
self.add_internal_page_OK_only = True
self.output_queue = output_queue
self.output_all_external = output_all_external
self.controller = controller
self.result_delegate = result_delegate
self.page_count_lock = threading.RLock()
self.internal_page_count_lock = threading.RLock()
self.level_lock = threading.RLock()
self.page_look_up_lock = threading.RLock()
self.external_link_check_lock = threading.RLock()
self._finihsed = False
self.task_control_max = 1
self.agent = "VegeBot (we follow your robots.txt settings before crawling, you can slow down the bot by change the Crawl-Delay parameter in the settings." \
"if you have an enquiry, please email to: [email protected])"
self.agent_from = "*****@*****.**"
if check_robot_text:
self.robot_agent = LinkChecker.get_robot_agent(self.sub_domain, protocol=self.scheme)
else:
self.robot_agent = None
self.site_crawl_delay = 0.60
if isinstance(self.robot_agent, Rules):
delay_temp = self.robot_agent.delay(self.agent)
if delay_temp is not None and delay_temp != self.site_crawl_delay:
self.site_crawl_delay = delay_temp
self.task_control_counter = 1
self._speed_penalty_count = 0
self._speed_penalty_threshold = 10
self._progress_logging_speed = 120
self._output_period = 120
self._output_batch_size = 100
self._death_wish_sent = False
SiteChecker._is_lxml_parser_exist()
self._output_thread = None
self._output_queue = None
self.progress_logger = ProgressLogger(self._progress_logging_speed, self, self._stop_event)
self._status = "Start"
self._populate_with_state() # restore laste known state
# self.data_source.additional_startup_procedures() # use the data set in self._populate_with_state() to start
# def _empty_external_links_db(self):
# if self.output_queue is not None:
def _put_result_in_output_queue_loop(self, item_list: list):
if not self._stop_event.is_set():
try:
self._output_queue.put(item_list, True, 2)
except Exception as ex:
if self._output_queue is None:
manager, self._output_queue = get_queue_client(QueueManager.MachineSettingCrawler,
QueueManager.Method_Whois_Input)
time.sleep(0.1)
ErrorLogger.log_error("SiteChecker._get_external_links_to_queue", self.sub_domain+" "+str(ex))
self._put_result_in_output_queue_loop(item_list)
def _get_external_links_to_queue(self):
ref_time = time.time()
manager, self._output_queue = get_queue_client(QueueManager.MachineSettingCrawler, QueueManager.Method_Whois_Input)
self.output_queue = self._output_queue # override output_queue
# if result_queue is None:
# ErrorLogger.log_error("SiteChecker._get_external_links_to_queue()", ValueError("result queue is none, cannot put item in queue."))
# else:
batch = list()
counter = 0
for item in self._external_db_buffer:
if self._stop_event.is_set() or self.external_links_checked >= self._external_db_buffer.count_all():
try:
manager.shutdown()
except:
pass
finally:
# print("exist _get_external_links_to_queue")
# if self._stop_event.is_set() and self.external_links_checked >= self._external_db_buffer.count_all():
break
elif isinstance(item, tuple):
# print("outputting item: ", str(item))
batch.append((item[0], item[1]))
counter += 1
if len(batch) > 0:
current_time = time.time()
if current_time - ref_time or len(batch) >= self._output_batch_size:
self._put_result_in_output_queue_loop(batch)
self.external_links_checked += len(batch)
ref_time = time.time()
batch.clear()
time.sleep(0.0001)
@staticmethod
def _is_lxml_parser_exist():
try:
import lxml
except ImportError:
SiteChecker._use_lxml_parser = False
else:
SiteChecker._use_lxml_parser = True
def use_lxml_parser(self):
return SiteChecker._use_lxml_parser
@staticmethod
def get_input_parameter_base(full_link: str, max_page: int, max_level: int, output_queue) -> dict:
return {SiteChecker.full_link_key: full_link, SiteChecker.max_page_key: max_page,
SiteChecker.max_level_key: max_level, SiteChecker.output_queue_key: output_queue}
def get_external_count_finished(self) -> int:
"""
ExternalTempInterface, get the number of job done in ExternalTempDataDiskBuffer
:return:
"""
return self.external_links_checked
def set_internal_count(self, count: int):
"""
ExternalTempInterface, set the number of job done in ExternalTempDataDiskBuffer
:param count:
:return:
"""
self.external_links_checked = count
def _set_task_control_max(self, concurrent_task: int):
if concurrent_task <= 0:
raise ValueError
self.task_control_max = concurrent_task
self.task_control_counter = concurrent_task
min_page_per_s = concurrent_task/20
self._speed_penalty_threshold = self._progress_logging_speed * min_page_per_s
if self.site_crawl_delay > 1/min_page_per_s:
ErrorLogger.log_error("SiteChecker._set_task_control_max()",
ValueError("site has crawl delay greater than mas delay."), self.domain_link)
self._status = "Stopped"
self.sudden_death()
def get_site_feedback(self) -> SeedSiteFeedback:
return SeedSiteFeedback(self.orginal_link, page_count=self.get_page_need_look_up())
def get_site_info(self) -> SiteInfo: # keep the original reference when sending back the site infomation
info = SiteInfo(self.orginal_link, self.data_source)
return info
def populate_with_state(self, state):
if state is not None and isinstance(state, SiteCheckerState):
self._status = "Restarted"
self.page_count = state.page_count
self.page_allocated = state.page_count
self.internal_page_count = state.internal_page_count
self.internal_page_last_count = state.internal_page_count
self.external_links_checked = state.external_page_count
self._external_db_buffer.set_progress(state.external_page_count)
self.page_need_look_up = state.page_need_look_up
self.current_level = state.current_level
self.progress_logger.set_reference(state.log_sample_index, state.log_started_time)
counter = 0
if self.data_source is not None:
try:
for item in self.data_source.get_next():
if counter >= self.cache_size:
break
if isinstance(item, OnSiteLink) and not LinkChecker.is_external_link(self.root_domain, item.link):
self.cache_list.append(item.link)
# print("--restore: ", item)
counter += 1
except Exception as ex:
msg = "error in SiteChecker.populate_with_state(), trying to populate cache, " + self.root_domain
ErrorLogger.log_error("SiteChecker", ex, msg)
self.data_source.ref = state.datasource_ref
self.data_source.output_c = state.datasource_output_c
self.data_source.set_progress(state.datasource_index if state.datasource_index < state.page_count else state.page_count)
self.data_source.set_continue_lock(True)
def get_file_name(self):
return self.data_source.ref
def get_limit(self):
return 100000
def get_column_names(self):
return ["Page Index", "External", "All", "Status"]
def get_progress(self):
data_source_count = self.data_source.count_all()
if self.page_count - self._page_count_shadow <= self._speed_penalty_threshold: # determine if site is slow
self._speed_penalty_count += 1
if self._speed_penalty_count > 2:
self._status = "Stopped"
self.sudden_death()
else:
self._speed_penalty_count = 0
if self.page_count == self._page_count_shadow and data_source_count == self._all_page_count_shadow: # determine if site is stucked
self._status = "Stopped"
self.sudden_death()
self._page_count_shadow = self.page_count
self._all_page_count_shadow = data_source_count
return [self.page_count, self.external_links_checked, data_source_count, self._status]
def is_programme_finshed(self):
return self._finihsed
def get_callback_data(self):
with self.page_count_lock:
gap = self.internal_page_count - self.internal_page_last_count
self.internal_page_last_count = self.internal_page_count
seed_feedback = None
if self._finihsed:
seed_feedback = self.get_site_feedback()
return SiteFeedback(gap, self._finihsed, seed_feedback=seed_feedback, datasource_ref=self.data_source.ref)
def get_state(self):
return SiteCheckerState(page_count=self.page_count, page_need_look_up=self.page_need_look_up,
current_level=self.current_level, internal_page_count=self.internal_page_count,
external_page_count= self.external_links_checked,
datasource_index=self.data_source.temp_counter,
datasource_output_c=self.data_source.output_c,
datasource_ref=self.data_source.ref, log_started_time=self.progress_logger.begin_time,
log_sample_index=self.progress_logger.limit_counter,)
def additional_reset(self):
pass
def addtional_clear(self):
pass
def stop(self):
# natural stop
self._status = "Stopped"
self.progress_logger.report_progress()
self._stop_event.set()
if self.progress_logger.is_alive():
self.progress_logger.join()
def clear(self):
self.cache_list.clear()
self.addtional_clear()
def acquire_task(self, level: int, link: str):
tasked_acquired = True
if link.endswith('/'):
temp = link
else:
temp = link + '/'
with self.task_control_lock:
if len(self._double_check_cache) > 0:
if temp in self._double_check_cache:
print("duplicate link found:", link)
tasked_acquired = False
else:
if len(self._double_check_cache) >= self.cache_size:
self._double_check_cache.popleft()
self._double_check_cache.append(temp)
self.task_control_counter -= 1
self.page_allocated += 1
if tasked_acquired:
if level > self.current_level:
self.current_level = level
# time.sleep(self.site_crawl_delay)
return tasked_acquired
def release_task(self, new_page: int):
with self.task_control_lock:
if self.page_need_look_up == 1 and new_page == 0:
PrintLogger.print("set to stop data source")
self.data_source.set_continue_lock(False)
else:
self.page_count += 1
self.page_need_look_up += new_page
#self.external_links_checked += external_page_count
self.task_control_counter += 1
# was determine if it is internal or external page
self.internal_page_count += 1
if self.internal_page_count > self.max_page or self.current_level > self.max_level:
if self.data_source.can_continue():
PrintLogger.print("set stop: " + str(self.internal_page_count)+" level: "+str(self.current_level))
self.data_source.set_continue_lock(False)
def get_page_count(self):
with self.page_count_lock:
page_count = self.page_count
return page_count
def set_page_count(self, page_count: int):
with self.page_count_lock:
self.page_count = page_count
def set_internal_page_count(self, count: int):
with self.internal_page_count_lock:
self.internal_page_count += count
def get_internal_page_count(self):
with self.internal_page_count_lock:
count = self.internal_page_count
return count
def get_current_level(self):
with self.level_lock:
current_level = self.current_level
return current_level
def set_current_level(self, level):
with self.level_lock:
self.current_level = level
def get_page_need_look_up(self):
with self.page_look_up_lock:
page_look_up = self.page_need_look_up
#self.page_look_up_lock.release()
return page_look_up
def set_page_need_look_up(self, page_count):
with self.page_look_up_lock:
#time.sleep(0.1)
self.page_need_look_up = page_count
# self.page_look_up_lock.release()
def set_page_need_look_up_plus_more(self, count: int):
with self.page_look_up_lock:
self.page_need_look_up += count
def get_internal_page_progress_index(self)->int:
return self.get_page_count()
def set_internal_page_progress_index(self, index: int):
self.page_count = index
self.page_allocated = index
def is_idle(self):
idle = False
with self.task_control_lock:
page_need_look_up = self.get_page_need_look_up()
new_task_added = page_need_look_up - self.page_need_look_up_temp
has_new_task = True if new_task_added > 0 else False
#page_count = self.get_page_count()
if has_new_task:
self.page_need_look_up_temp = page_need_look_up
else:
if self.task_control_counter >= self.task_control_max:
idle = True
# print("is idle")
# else:
# print("is working")
return idle
def add_link_to_cache(self, link):
if len(self.cache_list) > self.cache_size:
return
else:
if link.endswith('/'):
self.cache_list.append(link)
else:
self.cache_list.append(link+'/')
def is_link_in_cache(self, link):
if link.endswith('/'):
temp = link
else:
temp = link + '/'
return True if temp in self.cache_list else False
def reset_as(self, domain: str, link: str=""): # reset the target domain
PrintLogger.print("crawl reset as: "+domain)
self.domain = domain
self.domain_link = self.scheme + "://" + self.domain
self.page_count = 0
self.current_level = 0
self.set_page_need_look_up(1)
# self.set_page_looked_up(0)
self.clear()
if len(link) == 0:
self.cache_list.append(self.domain_link)
self.data_source.re_target(self.domain_link, OnSiteLink(self.domain_link, response_code=ResponseCode.LinkOK, link_level=1))
#self.data_source.append(OnSiteLink(self.domain_link, response_code=ResponseCode.LinkOK, link_level=1))
else:
self.cache_list.append(link)
self.data_source.re_target(link, OnSiteLink(link, response_code=ResponseCode.LinkOK, link_level=1))
#self.data_source.append(OnSiteLink(link, response_code=ResponseCode.LinkOK, link_level=1))
self.additional_reset()
self.data_source.additional_startup_procedures()
def crawling(self): # call this method to start operation
self._start_sending_feedback()
self._output_thread = threading.Thread(target=self._get_external_links_to_queue)
if self.data_source.can_continue():
self.data_source.additional_startup_procedures() # use the data set in self._populate_with_state() to start
self._external_db_buffer.start_input_output_cycle()
self._output_thread.start()
self.progress_logger.start()
self.progress_logger.report_progress() # log first row
self._status = "Work"
self.begin_crawl()
# prefix = "www."
# page_count_limit = 2
# if self.page_count <= page_count_limit and prefix not in self.domain_link:
# new_domain = prefix + self.sub_domain
# self.reset_as(new_domain)
# self._status = "Work"
# self.begin_crawl()
# print("going to stop all.")
self.stop()
self.clear()
self.data_source.additional_finish_procedures()
# print("going to finish output buffer.")
self._external_db_buffer.terminate()
# print("going to stop output_thread.")
if self._output_thread.is_alive():
self._output_thread.join()
PrintLogger.print("finished naturally: "+self.domain_link)
# print("finished naturally.")
self._finihsed = True
#calling this at the end of operation
PrintLogger.print("send last response")
# print("send last response")
# print("send last response.")
self._end_sending_feedback()
if self._memory_control_terminate_event is not None:
self._memory_control_terminate_event.set()
def sudden_death(self):
if not self._finihsed:
self._finihsed = True
PrintLogger.print("start sudden death: "+self.orginal_link)
#self.stop()
self.stop()
self.clear()
self.data_source.set_continue_lock(False)
self.data_source.additional_finish_procedures()
self._external_db_buffer.terminate()
if isinstance(self._output_thread, threading.Thread):
if self._output_thread.is_alive():
self._output_thread.join()
#calling this at the end of operation
PrintLogger.print("send last response")
self._end_sending_feedback()
if self._memory_control_terminate_event is not None:
ErrorLogger.log_error("SiteChecker", TimeoutError("slow processing speed, terminated."), self.orginal_link)
self._memory_control_terminate_event.set()
def begin_crawl(self, level=0): # subclass this to make different behaviour
pass
