def multi_days_time_seq(time_seq_dict, period_length=7):
"""
:param time_seq_dict:
:param period_length: 考察的时间，单位为天数， 日期之间未必连续,因为在不连续的日期之间补上24 * 缺少的天数个0
如时间序列中的日期为
:return:
"""
sorted_time_seq_tuple = sorted(time_seq_dict.items(), reverse=False)
sorted_time_seq = np.array([])
# print("=======================stary===============================")
# print("len of sorted_time_seq_tuple: %s" % (len(sorted_time_seq_tuple)))
# 取出有序时间序列元组中的第一个日期，若它在记录起始日期之后，则计算它与起始日期的距离，并加上相应个数的0
ret_start_date = str(
sorted_time_seq_tuple[0]
[0]) # 这里的sorted_time_seq_tuple[0][0]为什么会是Numpy.int64类型呢
seq_start_date = days_offset(change_date_str_format(PERIOD_START),
-1 * (DAY_RANGE - 1))
days_gap = differate_one_day_more(seq_start_date, ret_start_date)
if days_gap >= 0: # 说明有序时间序列中的起始日期与记录起始日期不同
tmp_time_seq = np.zeros(HOURS_IN_ONE_DAY * (days_gap + 1))
sorted_time_seq = np.append(sorted_time_seq, tmp_time_seq)
# print("add %s days zero" % days_gap)
# print("ret_start_date: %s, seq_start_date: %s, days_gap: %s" % (ret_start_date, seq_start_date, days_gap))
for index in range(len(sorted_time_seq_tuple)):
date_str = str(sorted_time_seq_tuple[index][0])
time_seq = sorted_time_seq_tuple[index][1]
if index:
day_before = str(sorted_time_seq_tuple[index - 1][0])
days_gap = differate_one_day_more(day_before, date_str)
# print("day_before: ", day_before, "date_str: ", date_str, "days_gap: ", days_gap)
# 两个相邻的时间序列之间相差不止一天， 那么缺少的这些天都是24个0，即这两个日期之间的日期中的查询次数都是0
if days_gap >= 0:
tmp_time_seq = np.zeros(HOURS_IN_ONE_DAY * days_gap)
sorted_time_seq = np.append(sorted_time_seq, tmp_time_seq)
sorted_time_seq = np.append(sorted_time_seq, time_seq)
# print("date_str: %s, len of sorted_time_seq: %s" % (date_str, len(sorted_time_seq)))
# 取出有序时间序列元组中的最后一个日期，若它在记录截止日期之前，则计算它与截止日期的距离，并加上相应个数的0
ret_end_date = str(
sorted_time_seq_tuple[-1]
[0]) # 这里的sorted_time_seq_tuple[0][0]为什么会是Numpy.int64类型呢
seq_end_date = change_date_str_format(PERIOD_START)
days_gap = differate_one_day_more(ret_end_date, seq_end_date)
if days_gap >= 0: # 说明有序时间序列中的截止日期与记录截止日期不同
tmp_time_seq = np.zeros(HOURS_IN_ONE_DAY * (days_gap + 1))
sorted_time_seq = np.append(sorted_time_seq, tmp_time_seq)
# print("ret_end_date: %s, seq_end_date: %s, days_gap: %s" % (ret_end_date, seq_end_date, days_gap))
# print("192 len of sorted_time_seq: %s" % (len(sorted_time_seq)))
return sorted_time_seq
def date_older_than_start_date(date_str):
seq_start_date = days_offset(change_date_str_format(START_DAY),
-1 * DAY_RANGE)
if differate_one_day_more(seq_start_date, date_str) >= 0:
return False
return True
def date_younger_than_start_date(date_str):
seq_end_date = change_date_str_format(START_DAY)
if differate_one_day_more(seq_end_date, date_str) >= 0:
return True
return False
def count_alive_days(domain_list, domain_bad):
"""
统计域名的存活时间，从注册日到当前日期；和域名的最近修改时间
并写入csv文件中
:param domain_list:
:param domain_bad:
:return:
"""
db = db_whois_dict[domain_bad]
data_dict_list = []
time_format = "%Y%m%d"
today = datetime.now().strftime(time_format)
for domain in domain_list:
print("=================================================")
print("domain: %s" % (domain))
query_body = {DOMAIN_2ND_FIELD: domain}
recs = db[whois_mongo_index].find(query_body)
print("recs.count(): %s" % recs.count())
if recs.count() > 0:
rec = recs[0]
create_date = rec.get(CREATE_DATE, "")
update_date = rec.get(UPDATE_DATE, "")
expiry_date = rec.get(EXPIRY_DATE, "")
try:
print("creat_date: %s, update_date: %s, expiry_date: %s" %
(len(create_date), len(update_date), len(expiry_date)))
print("creat_date: %s, update_date: %s, expiry_date: %s" %
(create_date, update_date, expiry_date))
# create_date = format_date_string(create_date) if len(create_date) > 0 else today
if len(create_date) > 0:
create_date = format_date_string(create_date)
else:
create_date = today
# update_date = format_date_string(update_date) if len(update_date) > 0 else today
if len(update_date) > 0:
update_date = format_date_string(update_date)
else:
update_date = today
# expiry_date = format_date_string(expiry_date) if len(expiry_date) > 0 else today
if len(expiry_date) > 0:
expiry_date = format_date_string(expiry_date)
else:
expiry_date = today
print("12232 create: ", create_date, " expiry_date: ",
expiry_date, " update: ", update_date)
days_gap1 = differate_one_day_more(create_date, today) + 1
days_gap2 = differate_one_day_more(update_date, today) + 1
days_gap3 = differate_one_day_more(create_date,
expiry_date) + 1
print("days_gap1:%s, days_gap2: %s, days_gap3: %s" %
(days_gap1, days_gap2, days_gap3))
# 信息不全，如果days_gap1、days_gap2、days_gap3都为0，则说明三者信息都不足，则忽略此域名
if days_gap1 or days_gap2 or days_gap3:
data_dict_list.append({
DOMAIN_2ND_FIELD: domain,
REGISTER_DAYS: days_gap1,
UPDATE_DAYS: days_gap2,
ALIVE_DAYS: days_gap3
})
except Exception as e:
print("error: ", e)
columns = [DOMAIN_2ND_FIELD, REGISTER_DAYS, ALIVE_DAYS, UPDATE_DAYS]
days_gap2csv(data_dict_list, columns, domain_bad)