def check_diff(self):
idx_list = [
idx.id for idx in History.select(History.id).order_by(
History.id.desc()).limit(self.cfg.DiffCount())
]
ip_diff_add_sql = IP.select(fn.Count(fn.Distinct(
IP.ip))).join(Item).where(IP.add == idx_list[0]).scalar()
ip_diff_purge_sql = IP.select(fn.Count(fn.Distinct(
IP.ip))).join(Item).where(IP.purge == idx_list[0]).scalar()
domain_diff_add_sql = Domain.select(fn.Count(fn.Distinct(Domain.domain)))\
.join(Item).where(Domain.add == idx_list[0]).scalar()
domain_diff_purge_sql = Domain.select(fn.Count(fn.Distinct(Domain.domain)))\
.join(Item).where(Domain.purge == idx_list[0]).scalar()
url_diff_add_sql = URL.select(fn.Count(fn.Distinct(URL.url)))\
.join(Item).where(URL.add == idx_list[0]).scalar()
url_diff_purge_sql = URL.select(fn.Count(fn.Distinct(URL.url)))\
.join(Item).where(URL.purge == idx_list[0]).scalar()
if ip_diff_add_sql or ip_diff_purge_sql or domain_diff_add_sql or \
domain_diff_purge_sql or url_diff_add_sql or url_diff_purge_sql:
History.update(dump=True).where(
History.id == idx_list[0]).execute()
return True
else:
# History.update(dump=False).where(History.id == idx_list[0]).execute()
return False
def cleaner(self):
logger.info('cleaner run')
# history = History.select(History.id).order_by(History.id.desc()).limit(self.cfg.DiffCount())
# Item.delete().where(~(Item.purge << history)).execute()
history_del = History.select(History.id).order_by(History.id.desc()).offset(self.cfg.DiffCount())
Item.delete().where(Item.purge << history_del).execute()
IP.delete().where(IP.purge << history_del).execute()
Domain.delete().where(Domain.purge << history_del).execute()
URL.delete().where(URL.purge << history_del).execute()
def test_01_insert_init_data(self):
"""sample domain insertion test"""
_domains = dict(domains)
for name, url in _domains.items():
d = Domain(domain=name, url=url)
d.save()
for d in Domain.get_by_filters():
_domains.pop(d.domain, None)
self.assertEqual(_domains, {})
def _domain_rollback_sql(self, rollback, bt):
rb_list = self.idx_list[:rollback]
if bt == 'ignore':
domain_sql = Domain.select(fn.Distinct(Domain.domain))\
.where(~(Domain.add << rb_list) & ((Domain.purge >> None) | (Domain.purge << rb_list)))
return domain_sql
elif bt == 'ip' or bt == 'default' or bt == 'domain' or bt == 'domain-mask':
domain_sql = Domain.select(fn.Distinct(Domain.domain))\
.join(Item).where((Item.blockType == bt) & ~(Domain.add << rb_list) &
((Domain.purge >> None) | (Domain.purge << rb_list)))
return domain_sql
def _domain_diff_sql(self, diff, bt, stat):
if stat and (bt == 'ip' or bt == 'default' or bt == 'domain' or bt == 'domain-mask'):
domain_sql = Domain.select(fn.Distinct(Domain.domain)).join(Item)\
.where(Item.blockType == bt, Domain.add == self.idx_list[diff])
return domain_sql
elif not stat and (bt == 'ip' or bt == 'default' or bt == 'domain' or bt == 'domain-mask'):
domain_sql = Domain.select(fn.Distinct(Domain.domain)).join(Item)\
.where(Item.blockType == bt, Domain.purge == self.idx_list[diff])
return domain_sql
elif stat and bt == 'ignore':
domain_sql = Domain.select(fn.Distinct(Domain.domain)).where(Domain.add == self.idx_list[diff])
return domain_sql
elif not stat and bt == 'ignore':
domain_sql = Domain.select(fn.Distinct(Domain.domain)).where(Domain.purge == self.idx_list[diff])
return domain_sql
def cleaner(self):
private_nets = [
'0.%', '127.%', '192.168.%', '10.%', '172.16.%', '172.17.%',
'172.18.%', '172.19.%', '172.20.%', '172.21.%', '172.22.%',
'172.23.%', '172.24.%', '172.25.%', '172.26.%', '172.27.%',
'172.28.%', '172.29.%', '172.30.%', '172.31.%'
]
logger.info('Dump cleaner run')
# history = History.select(History.id).order_by(History.id.desc()).limit(self.cfg.DiffCount())
# Item.delete().where(~(Item.purge << history)).execute()
history_clear = History.select(History.id).order_by(
History.id.desc()).offset(self.cfg.DiffCount())
item_del = Item.delete().where(Item.purge << history_clear).execute()
logger.info('Item deleted: %d', item_del)
ip_del = IP.delete().where(IP.purge << history_clear).execute()
logger.info('IP deleted: %d', ip_del)
domain_del = Domain.delete().where(
Domain.purge << history_clear).execute()
logger.info('Domain deleted: %d', domain_del)
url_del = URL.delete().where(URL.purge << history_clear).execute()
logger.info('URL deleted: %d', url_del)
history_rm = History.select(History.id).order_by(
History.id.desc()).offset(self.cfg.HistoryCount())
hist_del = History.delete().where(History.id << history_rm).execute()
logger.info('History deleted: %d', hist_del)
for net in private_nets:
ip_count = IP.delete().where(IP.ip % net).execute()
if ip_count:
logger.info('IP error LIKE %s, count %d', net, ip_count)
def test_03_regexp(self):
"""regexp process test"""
d = Domain(domain='dummy', url='dummy')
d.save()
s = Snapshot(
domain_id=d.id,
pulled_at=datetime.utcnow(),
html="find me with a regexp!!",
)
s.save()
dc = DomainCheck(domain_id=d.id,
name="dummy_check",
regexp="find (me|you)")
dc.save()
run_regexp(snapshot_id=s.id)
scd = SnapshotCheckData.get_by_filters(check_id=dc.id)[0]
self.assertEqual(json.loads(scd.check_value), ["me"])
def test_02_collector(self):
"""collector process test"""
done_snapshots = Snapshot.count_by_filters()
for d in Domain.get_by_filters():
ok = run_collector(domain_id=d.id)
if ok:
done_snapshots += 1
total_snapshots = Snapshot.count_by_filters()
self.assertEqual(done_snapshots, total_snapshots)
def run(self):
"""
Main function for thread that will
"""
# It might be good to create a process rather than a thread per file, and to create a multithreading environment
# to process the list rather than a single thread for all of it (test under big file environment).
# One json per line
my_list = self.buf_file.splitlines()
for line in my_list:
# Ignoring if there is an empty entrance in the list
if line:
# load json
json_line = json.loads(line)
# we are going to look for creative_size, if it does not exist we will get the information from ad_width
# and ad_height
creative_size = find_key("creative_size", json_line)
if not creative_size:
value_width = find_key("ad_width", json_line)
value_height = find_key("ad_height", json_line)
if value_width and value_height:
creative_size = [value_width[0] + "x" + value_height[0]]
# We are going to look for the keys page_url and Referer
referer = find_key("Referer", json_line)
url = find_key("page_url", json_line)
# If the three elements were found, introduce them in the DB.
if creative_size and referer and url:
with transaction() as session:
added = False
# Check the existence of the entrance before introducing a repetitive one
if not session.query(Domain).filter(Domain.url==url[0]).first():
session.add(Domain(url[0]))
added = True
if not session.query(Referer).filter(Referer.url==referer[0]).first():
session.add(Referer(referer[0]))
added = True
session.flush()
# If one of the previous tables has a new entry. No need to check of existence in here.
if added:
session.add(Information(domain_url=url[0], referer_url=referer[0], creative_size=creative_size[0]))
elif not session.query(Information).filter(Information.domain_url==url[0])\
.filter(Information.referer_url==referer[0])\
.filter(Information.creative_size==creative_size[0]).first():
session.add(Information(domain_url=url[0], referer_url=referer[0], creative_size=creative_size[0]))
print "Database updated with information from file: %s" % self.f_name
def check_diff(self):
idx_list = [idx.id for idx in History.select(History.id).where(History.diff == True)
.order_by(History.id.desc()).limit(self.cfg.DiffCount())]
ip_diff_add_sql = IP.select(fn.Count(fn.Distinct(IP.ip))).join(Item).where(IP.add == idx_list[0]).scalar()
ip_diff_purge_sql = IP.select(fn.Count(fn.Distinct(IP.ip))).join(Item).where(IP.purge == idx_list[0]).scalar()
domain_diff_add_sql = Domain.select(fn.Count(fn.Distinct(Domain.domain)))\
.join(Item).where(Domain.add == idx_list[0]).scalar()
domain_diff_purge_sql = Domain.select(fn.Count(fn.Distinct(Domain.domain)))\
.join(Item).where(Domain.purge == idx_list[0]).scalar()
url_diff_add_sql = URL.select(fn.Count(fn.Distinct(URL.url)))\
.join(Item).where(URL.add == idx_list[0]).scalar()
url_diff_purge_sql = URL.select(fn.Count(fn.Distinct(URL.url)))\
.join(Item).where(URL.purge == idx_list[0]).scalar()
if ip_diff_add_sql or ip_diff_purge_sql or domain_diff_add_sql or \
domain_diff_purge_sql or url_diff_add_sql or url_diff_purge_sql:
return True
else:
History.update(diff=False).where(History.id == idx_list[0]).execute()
return False
def run(*args, domain_id='', **kwargs):
"""Collector execution, get domains HTML running an HTTP GET request
Keyword arguments:
domain_id -- domain_id that should be retrieved
"""
logging.info("start collector run")
domain = Domain.get_by_filters(id=domain_id)
if domain:
domain = domain[0]
headers = {
#custom headers
}
proxies = {
#custom proxies
}
params = {
#custom query string
}
res = None
try:
res = requests.get(
domain.url,
headers=headers,
proxies=proxies,
data=params,
timeout=
10 #avoid hangs! -> https://requests.readthedocs.io/en/master/user/advanced/#timeouts
)
except:
#bad things can happen with users' input and i/o operations
#here a good error callback function will be super!
logging.exception(f"Error fething {domain.url}")
snapshot_id = None
if res:
s = Snapshot(domain_id=domain_id,
pulled_at=datetime.utcnow(),
html=res.text)
s.save()
domain.last_snapshot_at = datetime.utcnow()
domain.save()
snapshot_id = s.id
SnapshotMetadata(
snapshot_id=s.id,
domain_id=domain_id,
request_time=res.elapsed.total_seconds(),
request_status=res.status_code,
).save()
return snapshot_id
def message_handler(self, channel, method, properties, body):
if self.verbose:
logger.info("received %r" % (body,))
message = json.loads(body)
domain = Domain.normalize(message.get('domain'))
response = message.get('response')
query_time = message.get('time', time.time())
# parse dns message if there was one
if response:
try:
response = wdns.parse_message(response.decode('base64'))
except wdns.MessageParseException, e:
response = None
def message_handler(self, channel, method, properties, body):
if self.verbose:
logger.info("received %r" % (body, ))
message = json.loads(body)
domain = Domain.normalize(message.get('domain'))
response = message.get('response')
query_time = message.get('time', time.time())
# parse dns message if there was one
if response:
try:
response = wdns.parse_message(response.decode('base64'))
except wdns.MessageParseException, e:
response = None
def statistics_show(self, diff=0, stdout=False):
date_time = datetime.fromtimestamp(int(Dump.get(Dump.param == 'lastDumpDate')
.value)).strftime('%Y-%m-%d %H:%M:%S')
message = 'vigruzki.rkn.gov.ru update: ' + date_time + '\n'
url_add_sql = self._url_diff_sql(diff, 'ignore', 1)
message += '\nURLs added: \n\n'
for url_add in url_add_sql:
message += url_add.url + '\n'
ip_add_sql = self._ip_diff_sql(diff, 'ignore', 1)
message += '\nIPs added: \n\n'
for ip_add in ip_add_sql:
if ip_add.mask < 32:
message += ip_add.ip + '/' + str(ip_add.mask)
else:
message += ip_add.ip + '\n'
domain_add_sql = self._domain_diff_sql(diff, 'ignore', 1)
message += '\nDOMAINs added: \n\n'
for domain_add in domain_add_sql:
message += domain_add.domain + '\n'
url_del_sql = self._url_diff_sql(diff, 'ignore', 0)
message += '\nURLs deleted: \n\n'
for url_del in url_del_sql:
message += url_del.url + '\n'
ip_del_sql = self._ip_diff_sql(diff, 'ignore', 0)
message += '\nIPs deleted: \n\n'
for ip_del in ip_del_sql:
if ip_del.mask < 32:
message += ip_del.ip + '/' + str(ip_del.mask)
else:
message += ip_del.ip + '\n'
domain_del_sql = self._domain_diff_sql(diff, 'ignore', 0)
message += '\nDOMAINs deleted: \n\n'
for domain_del in domain_del_sql:
message += domain_del.domain + '\n'
rb_list = self.idx_list[:diff]
domain_count = Domain.select(fn.Count(fn.Distinct(Domain.domain)))\
.where(~(Domain.add << rb_list) & ((Domain.purge >> None) | (Domain.purge << rb_list))).scalar()
url_count = URL.select(fn.Count(fn.Distinct(URL.url)))\
.where(~(URL.add << rb_list) & ((URL.purge >> None) | (URL.purge << rb_list))).scalar()
ip_count = IP.select(fn.Count(fn.Distinct(IP.ip)))\
.where(~(IP.add << rb_list) & ((IP.purge >> None) | (IP.purge << rb_list))).scalar()
id_count = Item.select(fn.Count(fn.Distinct(Item.content_id)))\
.where(~(Item.add << rb_list) & ((Item.purge >> None) | (Item.purge << rb_list))).scalar()
message += '\nURLs count: ' + str(url_count) + '\n'
message += 'IPs count: ' + str(ip_count) + '\n'
message += 'DOMAINs count: ' + str(domain_count) + '\n'
message += 'Item count: ' + str(id_count) + '\n'
if stdout:
print(message)
return False
else:
return message
def parse_dump(self):
if not os.path.exists(self.path_py + '/dump.xml'):
logger.info('dump.xml not found: s%', self.path_py + '/dump.xml')
return 0
logger.info('dump.xml already exists.')
tree_xml = ElementTree().parse(self.path_py + '/dump.xml')
dt = datetime.strptime(tree_xml.attrib['updateTime'][:19],
'%Y-%m-%dT%H:%M:%S')
update_time = int(time.mktime(dt.timetuple()))
Dump.update(value=update_time).where(
Dump.param == 'lastDumpDate').execute()
logger.info('Got updateTime: %s.', update_time)
dt = datetime.strptime(tree_xml.attrib['updateTimeUrgently'][:19],
'%Y-%m-%dT%H:%M:%S')
update_time_urgently = int(time.mktime(dt.timetuple()))
Dump.update(value=update_time_urgently).where(
Dump.param == 'lastDumpDateUrgently').execute()
logger.info('Got updateTimeUrgently: %s.', update_time_urgently)
list_xml = tree_xml.findall(".//*[@id]")
id_set_dump = set()
id_set_db = set()
for content_xml in list_xml:
# print(content_xml.tag, content_xml.attrib, content_xml.text)
id_set_dump.add(int(content_xml.attrib['id']))
select_content_id_db = Item.select(
Item.content_id).where(Item.purge >> None)
for content_db in select_content_id_db:
id_set_db.add(content_db.content_id)
common_id_set = id_set_dump.intersection(id_set_db)
delete_id_set = id_set_db.difference(common_id_set)
add_id_set = id_set_dump.difference(common_id_set)
# print(delete_id_set)
# print(add_id_set)
if len(delete_id_set) > 0:
with self.transact.atomic():
for del_item in delete_id_set:
logger.info('Full delete Item, IP, Domain, URL id: %s.',
del_item)
Item.update(purge=self.code_id).where(
Item.content_id == del_item,
Item.purge >> None).execute()
Domain.update(purge=self.code_id).where(
Domain.content_id == del_item,
Domain.purge >> None).execute()
URL.update(purge=self.code_id).where(
URL.content_id == del_item,
URL.purge >> None).execute()
IP.update(purge=self.code_id).where(
IP.content_id == del_item, IP.purge >> None).execute()
if len(add_id_set) > 0:
include_time = str()
urgency_type = int()
entry_type = int()
block_type = str()
hash_value = str()
with self.transact.atomic():
for new_item in add_id_set:
logger.info('New Item, IP, Domain, URL id: %s.', new_item)
new_item_xml = tree_xml.find(".//content[@id='" +
str(new_item) + "']")
for data_xml in new_item_xml.iter():
if data_xml.tag == 'content':
content_id = int(data_xml.attrib['id'])
try:
urgency_type = int(
data_xml.attrib['urgencyType'])
except KeyError:
urgency_type = 0
include_time = self.date_time_xml_to_db(
data_xml.attrib['includeTime'])
try:
block_type = data_xml.attrib['blockType']
except KeyError:
block_type = 'default'
entry_type = int(data_xml.attrib['entryType'])
hash_value = data_xml.attrib['hash']
if data_xml.tag == 'decision':
decision_date = data_xml.attrib['date']
decision_number = data_xml.attrib['number']
decision_org = data_xml.attrib['org']
item_new = Item(content_id=content_id,
includeTime=include_time,
urgencyType=urgency_type,
entryType=entry_type,
blockType=block_type,
hashRecord=hash_value,
decision_date=decision_date,
decision_num=decision_number,
decision_org=decision_org,
add=self.code_id)
item_new.save()
if data_xml.tag == 'url':
if not self.only_ascii(data_xml.text):
url_split = str(data_xml.text).split(':')
url = url_split[0] + ':' + urllib.parse.quote(
url_split[1])
else:
url = data_xml.text
URL.create(item=item_new.id,
content_id=content_id,
url=url,
add=self.code_id)
if data_xml.tag == 'domain':
if not self.only_ascii(data_xml.text):
domain = (str(
data_xml.text).encode('idna')).decode()
else:
domain = data_xml.text
Domain.create(item=item_new.id,
content_id=content_id,
domain=domain,
add=self.code_id)
if data_xml.tag == 'ip':
ip = data_xml.text
IP.create(item=item_new.id,
content_id=content_id,
ip=ip,
add=self.code_id)
if data_xml.tag == 'ipSubnet':
net = data_xml.text.split('/')
ip = net[0]
mask = net[1]
IP.create(item=item_new.id,
content_id=content_id,
ip=ip,
mask=mask,
add=self.code_id)
url_db_set = set()
url_xml_set = set()
ip_db_set = set()
ip_xml_set = set()
sub_ip_xml_set = set()
sub_ip_db_set = set()
domain_db_set = set()
domain_xml_set = set()
data_update = False
with self.transact.atomic():
for item_xml in list_xml:
for data_xml in item_xml.iter():
# print(data_xml.tag, data_xml.attrib, data_xml.text)
if data_xml.tag == 'content':
content_id = int(data_xml.attrib['id'])
hash_value = data_xml.attrib['hash']
item_db = Item.get(Item.content_id == content_id,
Item.purge >> None)
if hash_value != item_db.hashRecord:
logger.info('Hashes not equal, update hash id: %s',
content_id)
try:
urgency_type = int(
data_xml.attrib['urgencyType'])
except KeyError:
urgency_type = 0
include_time = self.date_time_xml_to_db(
data_xml.attrib['includeTime'])
try:
block_type = data_xml.attrib['blockType']
except KeyError:
block_type = 'default'
entry_type = int(data_xml.attrib['entryType'])
item_db.hashRecord = hash_value
# Item.update(purge=None).where(Item.content_id == content_id).execute()
data_update = True
else:
data_update = False
break
if data_xml.tag == 'decision':
decision_date = data_xml.attrib['date']
decision_number = data_xml.attrib['number']
decision_org = data_xml.attrib['org']
# print(item_db)
if str(item_db.includeTime) != include_time:
logger.info('content_id: %s.', content_id)
logger.info('XML includeTime: %s.', include_time)
logger.info('DB includeTime: %s.',
item_db.includeTime)
item_db.includeTime = include_time
# Item.update(includeTime=include_time).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if item_db.urgencyType != urgency_type:
logger.info('content_id: %s.', content_id)
logger.info('XML urgencyType: %s.', urgency_type)
logger.info('DB urgencyType: %s.',
item_db.urgencyType)
item_db.urgencyType = urgency_type
# Item.update(urgencyType=urgency_type).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if item_db.blockType != block_type:
logger.info('content_id: %s.', content_id)
logger.info('XML blockType: %s.', block_type)
logger.info('DB blockType: %s.', item_db.blockType)
item_db.blockType = block_type
# Item.update(blockType=block_type).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if item_db.entryType != entry_type:
logger.info('content_id: %s.', content_id)
logger.info('XML entryType: %s.', entry_type)
logger.info('DB entryType: %s.', item_db.entryType)
item_db.entryType = entry_type
# Item.update(entryType=entry_type).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if str(item_db.decision_date) != decision_date:
logger.info('content_id: %s.', content_id)
logger.info('XML date: %s.', decision_date)
logger.info('DB date: %s.',
str(item_db.decision_date))
item_db.decision_date = decision_date
# Item.update(decision_date=decision_date).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if item_db.decision_num != decision_number:
logger.info('content_id: %s.', content_id)
logger.info('XML number: %s.', decision_number)
logger.info('DB number: %s.', item_db.decision_num)
item_db.decision_num = decision_number
# Item.update(decision_num=decision_number).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if item_db.decision_org != decision_org:
logger.info('content_id: %s.', content_id)
logger.info('XML org: %s.', decision_org)
logger.info('DB org: %s.', item_db.decision_org)
item_db.decision_org = decision_org
# Item.update(decision_org=decision_org).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if data_xml.tag == 'url':
if not self.only_ascii(data_xml.text):
url_split = str(data_xml.text).split(':')
url = url_split[0] + ':' + urllib.parse.quote(
url_split[1])
else:
url = data_xml.text
url_xml_set.add(url)
if data_xml.tag == 'domain':
if not self.only_ascii(data_xml.text):
domain = (str(
data_xml.text).encode('idna')).decode()
else:
domain = data_xml.text
domain_xml_set.add(domain)
if data_xml.tag == 'ip':
ip_xml_set.add(data_xml.text)
if data_xml.tag == 'ipSubnet':
sub_ip_xml_set.add(data_xml.text)
if data_update:
url_db = URL.select().where(URL.item == item_db.id,
URL.purge >> None)
for url_item in url_db:
url_db_set.add(url_item.url)
if url_db_set != url_xml_set:
common_url_set = url_xml_set.intersection(url_db_set)
delete_url_set = url_db_set.difference(common_url_set)
add_url_set = url_xml_set.difference(common_url_set)
if len(delete_url_set) > 0:
logger.info('Delete id %s URL: %s', content_id,
delete_url_set)
for delete_url in delete_url_set:
URL.update(purge=self.code_id).where(
URL.item == item_db.id,
URL.url == delete_url,
URL.purge >> None).execute()
if len(add_url_set) > 0:
logger.info('Add id %s URL: %s', content_id,
add_url_set)
for add_url in add_url_set:
URL.create(item=item_db.id,
content_id=item_db.content_id,
url=add_url,
add=self.code_id)
url_db_set.clear()
url_xml_set.clear()
domain_db = Domain.select().where(
Domain.item == item_db.id, Domain.purge >> None)
for domain_item in domain_db:
domain_db_set.add(domain_item.domain)
if domain_db_set != domain_xml_set:
common_domain_set = domain_xml_set.intersection(
domain_db_set)
delete_domain_set = domain_db_set.difference(
common_domain_set)
add_domain_set = domain_xml_set.difference(
common_domain_set)
if len(delete_domain_set) > 0:
logger.info('Delete id %s Domain: %s', content_id,
delete_domain_set)
for delete_domain in delete_domain_set:
Domain.update(purge=self.code_id).where(
Domain.item == item_db.id,
Domain.domain == delete_domain,
Domain.purge >> None).execute()
if len(add_domain_set) > 0:
logger.info('Add id %s Domain: %s', content_id,
add_domain_set)
for add_domain in add_domain_set:
Domain.create(item=item_db.id,
content_id=item_db.content_id,
domain=add_domain,
add=self.code_id)
domain_db_set.clear()
domain_xml_set.clear()
ip_db = IP.select().where(IP.item == item_db.id,
IP.mask == 32, IP.purge >> None)
for ip_item in ip_db:
ip_db_set.add(ip_item.ip)
if ip_db_set != ip_xml_set:
common_ip_set = ip_xml_set.intersection(ip_db_set)
delete_ip_set = ip_db_set.difference(common_ip_set)
add_ip_set = ip_xml_set.difference(common_ip_set)
if len(delete_ip_set) > 0:
logger.info('Delete id %s ip: %s', content_id,
delete_ip_set)
for delete_ip in delete_ip_set:
IP.update(purge=self.code_id).where(
IP.item == item_db.id, IP.ip == delete_ip,
IP.mask == 32, IP.purge >> None).execute()
if len(add_ip_set) > 0:
logger.info('Add id %s ip: %s', content_id,
add_ip_set)
for add_ip in add_ip_set:
IP.create(item=item_db.id,
content_id=item_db.content_id,
ip=add_ip,
add=self.code_id)
ip_db_set.clear()
ip_xml_set.clear()
sub_ip_db = IP.select().where(IP.item == item_db.id,
IP.mask < 32,
IP.purge >> None)
for sub_ip_item in sub_ip_db:
sub_ip_db_set.add(
str(sub_ip_item.ip) + '/' + str(sub_ip_item.mask))
if sub_ip_db_set != sub_ip_xml_set:
common_sub_ip_set = sub_ip_xml_set.intersection(
sub_ip_db_set)
delete_sub_ip_set = sub_ip_db_set.difference(
common_sub_ip_set)
add_sub_ip_set = sub_ip_xml_set.difference(
common_sub_ip_set)
if len(delete_sub_ip_set) > 0:
logger.info('Delete id %s subnet: %s', content_id,
delete_sub_ip_set)
for delete_sub_ip in delete_sub_ip_set:
del_subnet = str(delete_sub_ip).split('/')
del_ip = del_subnet[0]
del_mask = del_subnet[1]
IP.update(purge=self.code_id).where(
IP.item == item_db.id, IP.ip == del_ip,
IP.mask == del_mask,
IP.purge >> None).execute()
if len(add_sub_ip_set) > 0:
logger.info('Add id %s subnet: %s', content_id,
add_sub_ip_set)
for add_sub_ip in add_sub_ip_set:
add_subnet = str(add_sub_ip).split('/')
add_ip = add_subnet[0]
add_mask = add_subnet[1]
IP.create(item=item_db.id,
content_id=item_db.content_id,
ip=add_ip,
mask=add_mask,
add=self.code_id)
item_db.save()
sub_ip_db_set.clear()
sub_ip_xml_set.clear()
if self.check_diff():
self.cleaner()
return 1
else:
logger.info('no updates')
# print('no updates')
return 2
def parse_dump(self):
if not os.path.exists(self.path_py + '/dump.xml'):
logger.info('dump.xml not found: s%', self.path_py + '/dump.xml')
return 0
logger.info('dump.xml already exists.')
tree_xml = ElementTree().parse(self.path_py + '/dump.xml')
dt = datetime.strptime(tree_xml.attrib['updateTime'][:19], '%Y-%m-%dT%H:%M:%S')
update_time = int(time.mktime(dt.timetuple()))
Dump.update(value=update_time).where(Dump.param == 'lastDumpDate').execute()
logger.info('Got updateTime: %s.', update_time)
dt = datetime.strptime(tree_xml.attrib['updateTimeUrgently'][:19], '%Y-%m-%dT%H:%M:%S')
update_time_urgently = int(time.mktime(dt.timetuple()))
Dump.update(value=update_time_urgently).where(Dump.param == 'lastDumpDateUrgently').execute()
logger.info('Got updateTimeUrgently: %s.', update_time_urgently)
list_xml = tree_xml.findall(".//*[@id]")
id_set_dump = set()
id_set_db = set()
for content_xml in list_xml:
# print(content_xml.tag, content_xml.attrib, content_xml.text)
id_set_dump.add(int(content_xml.attrib['id']))
select_content_id_db = Item.select(Item.content_id).where(Item.purge >> None)
for content_db in select_content_id_db:
id_set_db.add(content_db.content_id)
common_id_set = id_set_dump.intersection(id_set_db)
delete_id_set = id_set_db.difference(common_id_set)
add_id_set = id_set_dump.difference(common_id_set)
# print(delete_id_set)
# print(add_id_set)
if len(delete_id_set) > 0:
with self.transact.atomic():
for del_item in delete_id_set:
logger.info('Full delete Item, IP, Domain, URL id: %s.', del_item)
Item.update(purge=self.code_id).where(Item.content_id == del_item, Item.purge >> None).execute()
Domain.update(purge=self.code_id).where(Domain.content_id == del_item,
Domain.purge >> None).execute()
URL.update(purge=self.code_id).where(URL.content_id == del_item, URL.purge >> None).execute()
IP.update(purge=self.code_id).where(IP.content_id == del_item, IP.purge >> None).execute()
if len(add_id_set) > 0:
include_time = str()
urgency_type = int()
entry_type = int()
block_type = str()
hash_value = str()
with self.transact.atomic():
for new_item in add_id_set:
logger.info('New Item, IP, Domain, URL id: %s.', new_item)
new_item_xml = tree_xml.find(".//content[@id='" + str(new_item) + "']")
for data_xml in new_item_xml.iter():
if data_xml.tag == 'content':
content_id = int(data_xml.attrib['id'])
try:
urgency_type = int(data_xml.attrib['urgencyType'])
except KeyError:
urgency_type = 0
include_time = self.date_time_xml_to_db(data_xml.attrib['includeTime'])
try:
block_type = data_xml.attrib['blockType']
except KeyError:
block_type = 'default'
entry_type = int(data_xml.attrib['entryType'])
hash_value = data_xml.attrib['hash']
if data_xml.tag == 'decision':
decision_date = data_xml.attrib['date']
decision_number = data_xml.attrib['number']
decision_org = data_xml.attrib['org']
item_new = Item(content_id=content_id, includeTime=include_time,
urgencyType=urgency_type, entryType=entry_type, blockType=block_type,
hashRecord=hash_value, decision_date=decision_date,
decision_num=decision_number, decision_org=decision_org,
add=self.code_id)
item_new.save()
if data_xml.tag == 'url':
if not self.only_ascii(data_xml.text):
url_split = str(data_xml.text).split(':')
url = url_split[0] + ':' + urllib.parse.quote(url_split[1])
else:
url = data_xml.text
URL.create(item=item_new.id, content_id=content_id, url=url, add=self.code_id)
if data_xml.tag == 'domain':
if not self.only_ascii(data_xml.text):
domain = (str(data_xml.text).encode('idna')).decode()
else:
domain = data_xml.text
Domain.create(item=item_new.id, content_id=content_id, domain=domain, add=self.code_id)
if data_xml.tag == 'ip':
ip = data_xml.text
IP.create(item=item_new.id, content_id=content_id, ip=ip, add=self.code_id)
if data_xml.tag == 'ipSubnet':
net = data_xml.text.split('/')
ip = net[0]
mask = net[1]
IP.create(item=item_new.id, content_id=content_id, ip=ip, mask=mask, add=self.code_id)
url_db_set = set()
url_xml_set = set()
ip_db_set = set()
ip_xml_set = set()
sub_ip_xml_set = set()
sub_ip_db_set = set()
domain_db_set = set()
domain_xml_set = set()
data_update = False
with self.transact.atomic():
for item_xml in list_xml:
for data_xml in item_xml.iter():
# print(data_xml.tag, data_xml.attrib, data_xml.text)
if data_xml.tag == 'content':
content_id = int(data_xml.attrib['id'])
hash_value = data_xml.attrib['hash']
item_db = Item.get(Item.content_id == content_id, Item.purge >> None)
if hash_value != item_db.hashRecord:
logger.info('Hashes not equal, update hash id: %s', content_id)
try:
urgency_type = int(data_xml.attrib['urgencyType'])
except KeyError:
urgency_type = 0
include_time = self.date_time_xml_to_db(data_xml.attrib['includeTime'])
try:
block_type = data_xml.attrib['blockType']
except KeyError:
block_type = 'default'
entry_type = int(data_xml.attrib['entryType'])
item_db.hashRecord = hash_value
# Item.update(purge=None).where(Item.content_id == content_id).execute()
data_update = True
else:
data_update = False
break
if data_xml.tag == 'decision':
decision_date = data_xml.attrib['date']
decision_number = data_xml.attrib['number']
decision_org = data_xml.attrib['org']
# print(item_db)
if str(item_db.includeTime) != include_time:
logger.info('content_id: %s.', content_id)
logger.info('XML includeTime: %s.', include_time)
logger.info('DB includeTime: %s.', item_db.includeTime)
item_db.includeTime = include_time
# Item.update(includeTime=include_time).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if item_db.urgencyType != urgency_type:
logger.info('content_id: %s.', content_id)
logger.info('XML urgencyType: %s.', urgency_type)
logger.info('DB urgencyType: %s.', item_db.urgencyType)
item_db.urgencyType = urgency_type
# Item.update(urgencyType=urgency_type).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if item_db.blockType != block_type:
logger.info('content_id: %s.', content_id)
logger.info('XML blockType: %s.', block_type)
logger.info('DB blockType: %s.', item_db.blockType)
item_db.blockType = block_type
# Item.update(blockType=block_type).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if item_db.entryType != entry_type:
logger.info('content_id: %s.', content_id)
logger.info('XML entryType: %s.', entry_type)
logger.info('DB entryType: %s.', item_db.entryType)
item_db.entryType = entry_type
# Item.update(entryType=entry_type).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if str(item_db.decision_date) != decision_date:
logger.info('content_id: %s.', content_id)
logger.info('XML date: %s.', decision_date)
logger.info('DB date: %s.', str(item_db.decision_date))
item_db.decision_date = decision_date
# Item.update(decision_date=decision_date).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if item_db.decision_num != decision_number:
logger.info('content_id: %s.', content_id)
logger.info('XML number: %s.', decision_number)
logger.info('DB number: %s.', item_db.decision_num)
item_db.decision_num = decision_number
# Item.update(decision_num=decision_number).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if item_db.decision_org != decision_org:
logger.info('content_id: %s.', content_id)
logger.info('XML org: %s.', decision_org)
logger.info('DB org: %s.', item_db.decision_org)
item_db.decision_org = decision_org
# Item.update(decision_org=decision_org).where(Item.content_id == content_id,
# Item.purge >> None).execute()
if data_xml.tag == 'url':
if not self.only_ascii(data_xml.text):
url_split = str(data_xml.text).split(':')
url = url_split[0] + ':' + urllib.parse.quote(url_split[1])
else:
url = data_xml.text
url_xml_set.add(url)
if data_xml.tag == 'domain':
if not self.only_ascii(data_xml.text):
domain = (str(data_xml.text).encode('idna')).decode()
else:
domain = data_xml.text
domain_xml_set.add(domain)
if data_xml.tag == 'ip':
ip_xml_set.add(data_xml.text)
if data_xml.tag == 'ipSubnet':
sub_ip_xml_set.add(data_xml.text)
if data_update:
url_db = URL.select().where(URL.item == item_db.id, URL.purge >> None)
for url_item in url_db:
url_db_set.add(url_item.url)
if url_db_set != url_xml_set:
common_url_set = url_xml_set.intersection(url_db_set)
delete_url_set = url_db_set.difference(common_url_set)
add_url_set = url_xml_set.difference(common_url_set)
if len(delete_url_set) > 0:
logger.info('Delete id %s URL: %s', content_id, delete_url_set)
for delete_url in delete_url_set:
URL.update(purge=self.code_id).where(URL.item == item_db.id, URL.url == delete_url,
URL.purge >> None).execute()
if len(add_url_set) > 0:
logger.info('Add id %s URL: %s', content_id, add_url_set)
for add_url in add_url_set:
URL.create(item=item_db.id, content_id=item_db.content_id, url=add_url,
add=self.code_id)
url_db_set.clear()
url_xml_set.clear()
domain_db = Domain.select().where(Domain.item == item_db.id, Domain.purge >> None)
for domain_item in domain_db:
domain_db_set.add(domain_item.domain)
if domain_db_set != domain_xml_set:
common_domain_set = domain_xml_set.intersection(domain_db_set)
delete_domain_set = domain_db_set.difference(common_domain_set)
add_domain_set = domain_xml_set.difference(common_domain_set)
if len(delete_domain_set) > 0:
logger.info('Delete id %s Domain: %s', content_id, delete_domain_set)
for delete_domain in delete_domain_set:
Domain.update(purge=self.code_id).where(Domain.item == item_db.id,
Domain.domain == delete_domain,
Domain.purge >> None).execute()
if len(add_domain_set) > 0:
logger.info('Add id %s Domain: %s', content_id, add_domain_set)
for add_domain in add_domain_set:
Domain.create(item=item_db.id, content_id=item_db.content_id, domain=add_domain,
add=self.code_id)
domain_db_set.clear()
domain_xml_set.clear()
ip_db = IP.select().where(IP.item == item_db.id, IP.mask == 32, IP.purge >> None)
for ip_item in ip_db:
ip_db_set.add(ip_item.ip)
if ip_db_set != ip_xml_set:
common_ip_set = ip_xml_set.intersection(ip_db_set)
delete_ip_set = ip_db_set.difference(common_ip_set)
add_ip_set = ip_xml_set.difference(common_ip_set)
if len(delete_ip_set) > 0:
logger.info('Delete id %s ip: %s', content_id, delete_ip_set)
for delete_ip in delete_ip_set:
IP.update(purge=self.code_id).where(IP.item == item_db.id, IP.ip == delete_ip,
IP.mask == 32, IP.purge >> None).execute()
if len(add_ip_set) > 0:
logger.info('Add id %s ip: %s', content_id, add_ip_set)
for add_ip in add_ip_set:
IP.create(item=item_db.id, content_id=item_db.content_id, ip=add_ip,
add=self.code_id)
ip_db_set.clear()
ip_xml_set.clear()
sub_ip_db = IP.select().where(IP.item == item_db.id, IP.mask < 32, IP.purge >> None)
for sub_ip_item in sub_ip_db:
sub_ip_db_set.add(str(sub_ip_item.ip) + '/' + str(sub_ip_item.mask))
if sub_ip_db_set != sub_ip_xml_set:
common_sub_ip_set = sub_ip_xml_set.intersection(sub_ip_db_set)
delete_sub_ip_set = sub_ip_db_set.difference(common_sub_ip_set)
add_sub_ip_set = sub_ip_xml_set.difference(common_sub_ip_set)
if len(delete_sub_ip_set) > 0:
logger.info('Delete id %s subnet: %s', content_id, delete_sub_ip_set)
for delete_sub_ip in delete_sub_ip_set:
del_subnet = str(delete_sub_ip).split('/')
del_ip = del_subnet[0]
del_mask = del_subnet[1]
IP.update(purge=self.code_id).where(IP.item == item_db.id, IP.ip == del_ip,
IP.mask == del_mask, IP.purge >> None).execute()
if len(add_sub_ip_set) > 0:
logger.info('Add id %s subnet: %s', content_id, add_sub_ip_set)
for add_sub_ip in add_sub_ip_set:
add_subnet = str(add_sub_ip).split('/')
add_ip = add_subnet[0]
add_mask = add_subnet[1]
IP.create(item=item_db.id, content_id=item_db.content_id, ip=add_ip, mask=add_mask,
add=self.code_id)
item_db.save()
sub_ip_db_set.clear()
sub_ip_xml_set.clear()
if self.check_diff():
self.cleaner()
return 1
else:
logger.info('no updates')
self.cleaner()
return 2
def _domain_dedup_sql(self, diff, bt, stat):
rb_list_add = self.idx_list[:diff+1]
rb_list_purge = self.idx_list[:diff]
if stat and bt == 'ignore':
domain_diff_sql = Domain.select(fn.Distinct(Domain.domain)).where(Domain.add == self.idx_list[diff])
domain_dup_sql = Domain.select(fn.Distinct(Domain.domain))\
.where(~(Domain.add << rb_list_add) & ((Domain.purge >> None) | (Domain.purge << rb_list_add)) &
(Domain.domain << domain_diff_sql))
domain_dedup_sql = Domain.select(fn.Distinct(Domain.domain)).where((Domain.add == self.idx_list[diff]) &
~(Domain.domain << domain_dup_sql))
return domain_dedup_sql
elif not stat and bt == 'ignore':
domain_diff_sql = Domain.select(fn.Distinct(Domain.domain)).where(Domain.purge == self.idx_list[diff])
domain_dup_sql = Domain.select(fn.Distinct(Domain.domain))\
.where(~(Domain.add << rb_list_purge) & (Domain.purge >> None) &
(Domain.domain << domain_diff_sql))
domain_dedup_sql = Domain.select(fn.Distinct(Domain.domain)).where((Domain.purge == self.idx_list[diff]) &
~(Domain.domain << domain_dup_sql))
return domain_dedup_sql
elif stat and (bt == 'ip' or bt == 'default' or bt == 'domain' or bt == 'domain-mask'):
domain_diff_sql = Domain.select(fn.Distinct(Domain.domain)).join(Item)\
.where(Item.blockType == bt, Domain.add == self.idx_list[diff])
domain_dup_sql = Domain.select(fn.Distinct(Domain.domain)).join(Item)\
.where((Item.blockType == bt) & ~(Domain.add << rb_list_add) &
(Domain.purge >> None) & (Domain.domain << domain_diff_sql))
domain_dedup_sql = Domain.select(fn.Distinct(Domain.domain)).join(Item)\
.where((Item.blockType == bt) & (Domain.add == self.idx_list[diff]) &
~(Domain.domain << domain_dup_sql))
return domain_dedup_sql
elif not stat and (bt == 'ip' or bt == 'default' or bt == 'domain' or bt == 'domain-mask'):
domain_diff_sql = Domain.select(fn.Distinct(Domain.domain)).join(Item)\
.where(Item.blockType == bt, Domain.purge == self.idx_list[diff])
domain_dup_sql = Domain.select(fn.Distinct(Domain.domain)).join(Item)\
.where((Item.blockType == bt) & ~(Domain.add << rb_list_purge) &
(Domain.purge >> None) & (Domain.domain << domain_diff_sql))
domain_dedup_sql = Domain.select(fn.Distinct(Domain.domain)).join(Item)\
.where((Item.blockType == bt) & (Domain.purge == self.idx_list[diff]) &
~(Domain.domain << domain_dup_sql))
return domain_dedup_sql
if __name__ == "__main__":
domains = {
"helsinkitimes": "https://www.helsinkitimes.fi/",
"berlin": "https://www.berlin.de/en/news/",
"9news": "https://www.9news.com.au/sydney",
"fail!": "fail://fail.com",
}
domain_checks = {
"helsinkitimes": ["covid(\\d+) ", "govern(\\w+)"],
}
delete_tables()
create_tables()
domain_ids = []
for name, url in domains.items():
d = Domain(domain=name, url=url)
d.save()
domain_ids.append(d.id)
for idx, regexp in enumerate(domain_checks.get(name, [])):
DomainCheck(
domain_id=d.id,
name=f"{name}-{idx}",
regexp=regexp,
).save()
for x in range(n_pulls):
for d_id in domain_ids:
print(f"sending collector task for {d_id}")
send_task(topic=collector_topic, domain_id=d_id)
time.sleep(sleep_between_pulls)
