def process_data(self, data: FilteredDomainData, **kwargs):
result_ok = False
if isinstance(data, FilteredDomainData):
try:
if len(data.domain_var) == 0:
data.domain_var = data.domain
links = ArchiveOrg.get_url_info(data.domain_var, min_size=self._min_page_size, limit=-100)
count = len(links)
data.archive = count
if count < self._min_profile:
pass
# raise ValueError("profile count is less than:" + str(self._min_profile))
result_ok = True
except Exception as ex:
if not self._is_throughput_debug:
pass
# ErrorLogger.log_error("ArchiveOrgFilter.process_data()", ex, data.domain_var)
finally:
with self._sync_lock:
self._job_done += 1
#with self._process_queue_lock:
if result_ok:
if not self._is_throughput_debug:
CsvLogger.log_to_file(self._log_file, [(data.domain, data.da, data.archive)]) # log this to file
self._output_queue.put(data)
# return data
else:
if self._is_throughput_debug:
self._output_queue.put(data)
# return None
else:
return None
def process_data(self, data: FilteredDomainData, **kwargs):
account = kwargs.get("Account")
# is_domain_good = False
is_spammed = False
try:
if isinstance(data, FilteredDomainData) and isinstance(account, SiteAccount):
majestic = MajesticCom(account)
if self._en_spam_check:
self._filter_domain_name(domain=data.domain)
# self._filter_anchor_text(majestic, data.domain)
# self._filter_ref_domains(majestic, data.domain)
if self._en_tf_check:
data = self._filter_tf_cf_backlink_ratio(majestic, data)
if not (data.tf >= self._min_tf and data.ref_domains >= self._min_ref_domains):
raise ValueError("tf or cf doesn't match. tf:" + str(data.tf) + " cf: " + str(data.cf) + " ref domain: " + str(data.ref_domains))
# if data.backlinks / data.ref_domains > self._max_backlink_to_ref_domain_ratio:
# raise MajesticSpamException("backlink to ref domain ratio is greater than {0:.1f}".format(self._max_backlink_to_ref_domain_ratio,))
if self._en_spam_check:
self._filter_anchor_text(majestic, data.domain)
self._filter_ref_domains(majestic, data.domain)
# is_domain_good = True
else:
raise ValueError("account is none in process_data")
except MajesticSpamException as mjx_ex:
is_spammed = True
data.exception = str(mjx_ex)
except Exception as ex:
data.exception = str(ex)
# ErrorLogger.log_error("MajesticFilter.process_data()", ex, str(data))
finally:
PrintLogger.print("Majestic processed: '" + str(data) + "' with: " + account.userID)
if isinstance(data, FilteredDomainData):
with self._sync_lock:
self._job_done += 1
if account is not None:
account.Available = True
# if data.cf >= self._min_cf and data.tf >= self._min_tf:
if data.tf >= self._min_tf and data.cf >= self._min_cf and data.ref_domains >= self._min_ref_domains:
# if data.tf >= self._min_tf and data.ref_domains >= self._min_ref_domains:
#print("Majatic output:", data)
# PrintLogger.print("domain: " + data.domain + " is good.")
if not self._is_throughput_debug:
if is_spammed:
CsvLogger.log_to_file(self._bad_log_file, [data.to_tuple()], dir_path=FilePath.get_temp_db_dir())
else:
CsvLogger.log_to_file(self._log_file, [data.to_tuple()], dir_path=FilePath.get_temp_db_dir()) # log this to file
self._output_queue.put(data)
return data
# elif is_spammed:
# if not self._is_throughput_debug:
# CsvLogger.log_to_file(self._bad_log_file, [data.to_tuple()], dir_path=FilePath.get_temp_db_dir())
# self._output_queue.put(data)
# return data
else:
if self._is_throughput_debug:
self._output_queue.put(data)
# return None
# print("domain: " + data.domain + " has exception:" + data.exception)
else:
pass
def process_data(self, data: FilteredDomainData, **kwargs):
#print("MozFilter processing: ", data)
account = kwargs.get("Account")
try:
if isinstance(data, FilteredDomainData) and isinstance(account, SiteAccount):
if TldUtility.is_top_tld(data.domain):
sleep_time =random.randint(self._min_sleep_time, self._max_wait)
time.sleep(sleep_time)
moz = MozCom(account)
if not self._is_throughput_debug:
ranking = moz.get_ranking_data(data.domain)
else:
ranking = 100
data.da = ranking
else:
pass
else:
raise ValueError("account is none in process_data")
except Exception as ex:
ErrorLogger.log_error("MozFilter", ex, "process_data() " + str(data) + " account: " + account.userID)
finally:
PrintLogger.print("Moz processed: " + str(data) + " with: " + account.userID)
if isinstance(data, FilteredDomainData):
with self._sync_lock:
self._job_done += 1
if account is not None:
account.Available = True
if data.da >= self._min_DA_value:
if not self._is_throughput_debug:
CsvLogger.log_to_file(self._log_file, [(data.domain, data.da)]) # log this to file
self._output_queue.put(data)
def testArchiveProfileFilter1(self):
input_param ={"input_queue": queue.Queue(), "output_queue": queue.Queue(), "stop_event": Event(),
"throughput_debug": True, "worker_number": 2}
archive_filter = get_archive_filter(**input_param)
links = FileIO.FileHandler.read_lines_from_file("/Users/superCat/Desktop/PycharmProjectPortable/test/archive_domain_test.txt")
for link in links:
site = FilteredDomainData(domain=link)
archive_filter.process_data(data=site, )
def testMajesticFilter(self):
filter = get_majestic_filter(worker_number=1, input_queue=Queue(), output_queue=Queue(), stop_event=Event())
param = {"Account": majestic_account}
links = FileIO.FileHandler.read_lines_from_file("/Users/superCat/Desktop/PycharmProjectPortable/test/spam_test2.txt")
for link in links:
link = LinkChecker.get_root_domain(link)[1]
print("doing link:", link)
site = FilteredDomainData(domain=link)
filter.process_data(data=site, **param)
def _filter_tf_cf_backlink_ratio(self, majestic: MajesticCom, data: FilteredDomainData) -> FilteredDomainData:
ranking = majestic.get_cf_tf_list(["http://"+data.domain,
"www."+data.domain,
"http://www."+data.domain],
is_dev=DomainFinderSrc.IS_DEBUG)
if ranking is not None and len(ranking) > 0:
current_tf = 0
for item in ranking:
if isinstance(item, MajesticComStruct):
item_cf_tf_ratio = 999
data_cf_tf_ratio = 999
item_deviation = 999
data_deviation = 999
if item.tf > 0:
item_cf_tf_ratio1 = abs(1-item.cf/item.tf)
item_cf_tf_ratio2 = abs(1-item.tf/item.cf)
item_deviation = min([item_cf_tf_ratio1, item_cf_tf_ratio2])
else:
continue
if data.tf > 0:
data_cf_tf_ratio1 = abs(1-data.cf/data.tf)
data_cf_tf_ratio2 = abs(1-data.tf/data.cf)
data_deviation = min([data_cf_tf_ratio1, data_cf_tf_ratio2])
# data_deviation = abs(1-data_cf_tf_ratio)
if item.tf >= self._min_tf and item.cf >=self._min_cf and item_deviation < data_deviation and item_deviation <= self._cf_tf_deviation:
data.domain_var = item.domain
data.tf = item.tf
data.cf = item.cf
data.backlinks = item.backlinks
data.ref_domains = item.ref_domains
data.topic = item.topic
return data
def input_thread():
data_path = "/Users/superCat/Desktop/PycharmProjectPortable/test/03-09-2015-Bad-Results.csv"
input_c = 0
with open(data_path, mode='r', newline='') as csv_file:
rd = csv.reader(csv_file, delimiter=',')
for row in rd:
if input_c < site_count:
input_q.put(FilteredDomainData(domain=row[0]))
time.sleep(0.0001)
else:
break
input_c += 1
def testFilter(self):
manager = AccountManager()
manager.AccountList.append(majestic_account)
input_param = {
"input_queue": queue.Queue(),
"output_queue": queue.Queue(),
"stop_event": Event()
}
filter = MajesticFilter(manager=manager, **input_param)
param = {"Account": majestic_account}
links = FileIO.FileHandler.read_lines_from_file(
"/Users/superCat/Desktop/PycharmProjectPortable/test/spam_test1.txt"
)
for link in links:
site = FilteredDomainData(domain=link)
filter.process_data(data=site, **param)
def testWriteToDb(self):
db_path = "/Users/superCat/Desktop/PycharmProjectPortable/sync/FilteredSitesList"
good_db = "/Users/superCat/Desktop/PycharmProjectPortable/sync/Majestic_filtering_good.csv"
table = "20/12/2015 Legal"
db = FilteredResultDB(table=table, offset=0, db_addr=db_path)
count = 0
temp_sites = []
with open(good_db, mode='r', newline='') as csv_file:
rd = csv.reader(csv_file, delimiter=',')
for row in rd:
if int(row[10]) > 1450612100:
data = FilteredDomainData.from_tuple(row)
print(data.__dict__)
count += 1
temp_sites.append(data)
print("total:", count)
db.add_sites(temp_sites, skip_check=False)
db.close()
def testFilterAll1(self):
db_path = "/Users/superCat/Desktop/PycharmProjectPortable/sync/"
manager = AccountManager(db_path)
input_param ={"input_queue": queue.Queue(), "output_queue": queue.Queue(), "stop_event": Event(),
"throughput_debug": True, "worker_number": 2}
file_path = "/Users/superCat/Desktop/PycharmProjectPortable/sync/Moz_filtering.csv"
archive_filter = get_archive_filter(**input_param)
majestic_filter = get_majestic_filter(**input_param)
param = {"Account": majestic_account}
count = 0
with open(file_path, 'rt') as csvfile:
reader = csv.reader(csvfile, delimiter=',')
for row in reader:
link, da = row
site = FilteredDomainData(domain=link, da=int(da))
print(count, " process:", link)
archive_data = archive_filter.process_data(data=site, )
if archive_data is not None:
majestic_data = majestic_filter.process_data(data=site, **param)
print(majestic_data)
count += 1
def format_input(data):
if isinstance(data, FilteredDomainData):
return data
elif isinstance(data, tuple) and len(data) == 2:
return FilteredDomainData(domain=data[0])
def input_thread():
for i in range(site_count):
input_q.put(FilteredDomainData(domain="domain{0:d}.com".format(i,)))
time.sleep(0.0001)
