def start_subject_statistics_schedule(self):
"""
定时启动题目分析任务
:param self:
:return:
"""
try:
logger.info('START(%s)' % self.request.id)
category_list = ReportSubjectStatisticsMiddle.sync_distinct('category')
ReportSubjectStatisticsMiddle().get_sync_collection(
read_preference=ReadPreference.PRIMARY).drop()
task_dt = datetime.now()
logger.info('-- category -- %s' % str(category_list))
if not category_list:
category_list = [{
'subject_cid': '$subject_cid',
'province_code': '$province_code',
'city_code': '$city_code'
}]
for category in category_list:
logger.info('-- start -- %s' % str(category))
RedisCache.hdel(KEY_CACHE_REPORT_CONDITION, str(category))
start_split_subject_stat_task.delay(category, task_dt)
except Exception:
logger.error(str(traceback.format_exc()))
logger.info(' END (%s)' % self.request.id)
def start_extract_subjects(self,
subject_choice_rule: SubjectChoiceRules,
times=1):
try:
logger.info('START(%s): rules_cid=%s' %
(self.request.id, subject_choice_rule.cid))
cut_off_datetime = datetime.datetime.now()
# 抽题
def extract():
for i in range(times):
extract_subjects(subject_choice_rule)
while True:
extract()
count = SubjectBanks.sync_count(
dict(rule_cid=subject_choice_rule.cid),
read_preference=ReadPreference.PRIMARY)
if count >= 4096:
break
# 删除就题库
SubjectBanks.sync_delete_many(
dict(rule_cid=subject_choice_rule.cid,
choice_dt={'$lt': cut_off_datetime}))
logger.info('END(%s): rules_cid=%s' %
(self.request.id, subject_choice_rule.cid))
except Exception:
logger.error(traceback.format_exc())
finally:
# 删除题库数量缓存
RedisCache.hdel(KEY_EXTRACTING_SUBJECT_QUANTITY,
subject_choice_rule.cid)
# 删除抽题状态
RedisCache.hdel(KEY_PREFIX_EXTRACTING_SUBJECT_RULE,
subject_choice_rule.cid)
# 删除数量缓存
RedisCache.delete(
'%s_%s' %
(KEY_PREFIX_SUBJECT_BANKS_COUNT, subject_choice_rule.cid))