def deal_with_data(self, *param):
'''
处理并存储相关秒杀商品的数据
:param param: 相关参数
:return:
'''
pid = param[0]
begin_time = int(
time.mktime(time.strptime(
param[1], '%Y/%m/%d %H:%M:%S'))) # 把str字符串类型转换为时间戳的形式
end_time = int(
time.mktime(time.strptime(param[2], '%Y/%m/%d %H:%M:%S')))
item_list = param[3]
mia = MiaParse()
my_pipeline = SqlServerMyPageInfoSaveItemPipeline()
if my_pipeline.is_connect_success:
db_goods_id_list = [
item[0] for item in list(
my_pipeline.select_mia_xianshimiaosha_all_goods_id())
]
# print(db_goods_id_list)
for item in item_list:
if item.get('item_id', '') in db_goods_id_list:
print('该goods_id已经存在于数据库中, 此处跳过')
pass
else:
goods_id = str(item.get('item_id', ''))
tmp_url = 'https://www.mia.com/item-' + str(
goods_id) + '.html'
mia.get_goods_data(goods_id=str(goods_id))
goods_data = mia.deal_with_data()
if goods_data == {}: # 返回的data为空则跳过
pass
else: # 否则就解析并且插入
goods_url = goods_data['goods_url']
if re.compile(r'://m.miyabaobei.hk/').findall(
goods_url) != '':
goods_url = 'https://www.miyabaobei.hk/item-' + str(
goods_id) + '.html'
else:
goods_url = 'https://www.mia.com/item-' + str(
goods_id) + '.html'
goods_data['goods_url'] = goods_url
goods_data['goods_id'] = str(goods_id)
goods_data['price'] = item.get('active_price')
goods_data['taobao_price'] = item.get(
'active_price') # 秒杀最低价
goods_data['sub_title'] = item.get('short_info', '')
goods_data['miaosha_time'] = {
'miaosha_begin_time':
self.timestamp_to_regulartime(begin_time),
'miaosha_end_time':
self.timestamp_to_regulartime(end_time),
}
goods_data['miaosha_begin_time'], goods_data[
'miaosha_end_time'] = self.get_miaosha_begin_time_and_miaosha_end_time(
miaosha_time=goods_data['miaosha_time'])
goods_data['pid'] = str(pid)
# pprint(goods_data)
# print(goods_data)
mia.insert_into_mia_xianshimiaosha_table(
data=goods_data, pipeline=my_pipeline)
sleep(MIA_SPIKE_SLEEP_TIME) # 放慢速度
else:
print('数据库连接失败，此处跳过!')
pass
try:
del mia
except:
pass
gc.collect()
def run_forever(self):
'''
实时更新数据
:return:
'''
tmp_sql_server = SqlServerMyPageInfoSaveItemPipeline()
try:
result = list(tmp_sql_server.select_mia_xianshimiaosha_all_goods_id())
except TypeError:
print('TypeError错误, 原因数据库连接失败...(可能维护中)')
result = None
if result is None:
pass
else:
print('------>>> 下面是数据库返回的所有符合条件的goods_id <<<------')
print(result)
print('--------------------------------------------------------')
print('即将开始实时更新数据, 请耐心等待...'.center(100, '#'))
index = 1
for item in result: # 实时更新数据
miaosha_end_time = json.loads(item[1]).get('miaosha_end_time')
miaosha_end_time = int(str(time.mktime(time.strptime(miaosha_end_time,'%Y-%m-%d %H:%M:%S')))[0:10])
# print(miaosha_end_time)
data = {}
# 释放内存, 在外面声明就会占用很大的, 所以此处优化内存的方法是声明后再删除释放
mia_miaosha = MiaParse()
if index % 50 == 0: # 每50次重连一次，避免单次长连无响应报错
print('正在重置，并与数据库建立新连接中...')
tmp_sql_server = SqlServerMyPageInfoSaveItemPipeline()
print('与数据库的新连接成功建立...')
if tmp_sql_server.is_connect_success:
if self.is_recent_time(miaosha_end_time) == 0:
tmp_sql_server.delete_mia_miaosha_expired_goods_id(goods_id=item[0])
print('过期的goods_id为(%s)' % item[0], ', 限时秒杀开始时间为(%s), 删除成功!' % json.loads(item[1]).get('miaosha_begin_time'))
elif self.is_recent_time(miaosha_end_time) == 2:
# break # 跳出循环
pass # 此处应该是pass,而不是break，因为数据库传回的goods_id不都是按照顺序的
else: # 返回1，表示在待更新区间内
print('------>>>| 正在更新的goods_id为(%s) | --------->>>@ 索引值为(%d)' % (item[0], index))
data['goods_id'] = item[0]
# print('------>>>| 爬取到的数据为: ', data)
tmp_url = 'https://m.mia.com/instant/seckill/seckillPromotionItem/' + str(item[2])
body = MyRequests.get_url_body(url=tmp_url, headers=self.headers, had_referer=True)
# print(body)
if body == '' or body == '[]':
print('获取到的body为空值! 此处跳过')
else:
try:
tmp_data = json.loads(body)
except:
tmp_data = {}
print('json.loads转换body时出错, 此处跳过!')
begin_time = tmp_data.get('p_info', {}).get('start_time', '')
end_time = tmp_data.get('p_info', {}).get('end_time', '')
begin_time = int(time.mktime(time.strptime(begin_time, '%Y/%m/%d %H:%M:%S'))) # 把str字符串类型转换为时间戳的形式
end_time = int(time.mktime(time.strptime(end_time, '%Y/%m/%d %H:%M:%S')))
item_list = tmp_data.get('item_list', [])
# 该pid中现有的所有goods_id的list
miaosha_goods_all_goods_id = [item_1.get('item_id', '') for item_1 in item_list]
if item[0] not in miaosha_goods_all_goods_id: # 内部已经下架的
print('该商品已被下架限时秒杀活动，此处将其删除')
tmp_sql_server.delete_mia_miaosha_expired_goods_id(goods_id=item[0])
print('下架的goods_id为(%s)' % item[0], ', 删除成功!')
pass
else: # 未下架的
for item_2 in item_list:
if item_2.get('item_id', '') == item[0]:
mia_miaosha.get_goods_data(goods_id=item[0])
goods_data = mia_miaosha.deal_with_data()
if goods_data == {}: # 返回的data为空则跳过
pass
else:
goods_data['goods_id'] = str(item[0])
goods_data['price'] = item_2.get('active_price')
goods_data['taobao_price'] = item_2.get('active_price')
goods_data['sub_title'] = item_2.get('short_info', '')
goods_data['miaosha_time'] = {
'miaosha_begin_time': self.timestamp_to_regulartime(begin_time),
'miaosha_end_time': self.timestamp_to_regulartime(end_time),
}
goods_data['miaosha_begin_time'], goods_data['miaosha_end_time'] = self.get_miaosha_begin_time_and_miaosha_end_time(miaosha_time=goods_data['miaosha_time'])
# pprint(goods_data)
# print(goods_data)
mia_miaosha.update_mia_xianshimiaosha_table(data=goods_data, pipeline=tmp_sql_server)
sleep(MIA_SPIKE_SLEEP_TIME) # 放慢速度
else:
pass
else: # 表示返回的data值为空值
print('数据库连接失败，数据库可能关闭或者维护中')
pass
index += 1
gc.collect()
print('全部数据更新完毕'.center(100, '#')) # sleep(60*60)
if get_shanghai_time().hour == 0: # 0点以后不更新
sleep(60 * 60 * 5.5)
else:
sleep(5)
gc.collect()
