def deal_with_data(self, goods_list):
'''
处理并存储相关拼团商品的数据
:param goods_list:
:return:
'''
mia = MiaPintuanParse()
my_pipeline = SqlServerMyPageInfoSaveItemPipeline()
if my_pipeline.is_connect_success:
db_goods_id_list = [item[0] for item in list(my_pipeline.select_mia_pintuan_all_goods_id())]
# print(db_goods_id_list)
for item in goods_list:
if item.get('goods_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('goods_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['sub_title'] = item.get('sub_title', '')
goods_data['pintuan_begin_time'], goods_data['pintuan_end_time'] = self.get_pintuan_begin_time_and_pintuan_end_time(pintuan_time=goods_data['pintuan_time'])
goods_data['pid'] = item.get('pid')
# pprint(goods_data)
# print(goods_data)
mia.insert_into_mia_pintuan_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_pintuan_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
pintuan_end_time = json.loads(item[1]).get('end_time')
pintuan_end_time = int(
str(
time.mktime(
time.strptime(pintuan_end_time,
'%Y-%m-%d %H:%M:%S')))[0:10])
# print(miaosha_end_time)
data = {}
mia_pintuan = MiaPintuanParse()
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(pintuan_end_time) == 0:
tmp_sql_server.delete_mia_pintuan_expired_goods_id(
goods_id=item[0])
print(
'过期的goods_id为(%s)' % item[0],
', 拼团开始时间为(%s), 删除成功!' %
json.loads(item[1]).get('begin_time'))
elif self.is_recent_time(pintuan_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/groupon/common_list/' + str(
item[2]) + '/0/'
# print(tmp_url)
body = MyRequests.get_url_body(url=tmp_url,
headers=self.headers,
had_referer=True)
if 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!')
if tmp_data.get('data_list', []) == []:
print('得到的data_list为[]!')
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_pintuan_expired_goods_id(
goods_id=item[0])
print('下架的goods_id为(%s)' % item[0], ', 删除成功!')
pass
else:
data_list = [{
'goods_id':
item_2.get('sku', ''),
'sub_title':
item_2.get('intro', ''),
} for item_2 in tmp_data.get('data_list', [])]
# pprint(data_list)
pintuan_goods_all_goods_id = [
item_1.get('goods_id', '')
for item_1 in data_list
]
# print(pintuan_goods_all_goods_id)
'''
蜜芽拼团不对内部下架的进行操作，一律都更新未过期商品 (根据pid来进行更新多次研究发现出现商品还在拼团，误删的情况很普遍)
'''
if item[0] not in pintuan_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_pintuan_expired_goods_id(goods_id=item[0])
# print('下架的goods_id为(%s)' % item[0], ', 删除成功!')
# pass
# 一律更新
mia_pintuan.get_goods_data(
goods_id=item[0])
goods_data = mia_pintuan.deal_with_data()
if goods_data == {}: # 返回的data为空则跳过
pass
else:
goods_data['goods_id'] = str(item[0])
if goods_data[
'pintuan_time'] == {}: # 当没有拼团时间时，就表示已下架拼团(未让其正常更新进数据库, 我把拼团开始结束时间都设置为当前时间)
'''
时区处理，时间处理到上海时间
'''
tz = pytz.timezone(
'Asia/Shanghai') # 创建时区对象
now_time = datetime.datetime.now(
tz)
# 处理为精确到秒位，删除时区信息
now_time = re.compile(r'\..*').sub(
'', str(now_time))
# 将字符串类型转换为datetime类型
now_time = datetime.datetime.strptime(
now_time, '%Y-%m-%d %H:%M:%S')
goods_data[
'pintuan_begin_time'], goods_data[
'pintuan_end_time'] = (
now_time, now_time)
else:
goods_data[
'pintuan_begin_time'], goods_data[
'pintuan_end_time'] = self.get_pintuan_begin_time_and_pintuan_end_time(
pintuan_time=goods_data[
'pintuan_time'])
# pprint(goods_data)
# print(goods_data)
mia_pintuan.update_mia_pintuan_table(
data=goods_data,
pipeline=tmp_sql_server)
sleep(MIA_SPIKE_SLEEP_TIME) # 放慢速度
else: # 未下架的
for item_2 in data_list:
if item_2.get('goods_id',
'') == item[0]:
mia_pintuan.get_goods_data(
goods_id=item[0])
goods_data = mia_pintuan.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[
'sub_title'] = item_2.get(
'sub_title', '')
if goods_data[
'pintuan_time'] == {}: # 当没有拼团时间时，就表示已下架拼团
'''
时区处理，时间处理到上海时间
'''
tz = pytz.timezone(
'Asia/Shanghai'
) # 创建时区对象
now_time = datetime.datetime.now(
tz)
# 处理为精确到秒位，删除时区信息
now_time = re.compile(
r'\..*').sub(
'', str(now_time))
# 将字符串类型转换为datetime类型
now_time = datetime.datetime.strptime(
now_time,
'%Y-%m-%d %H:%M:%S')
goods_data[
'pintuan_begin_time'], goods_data[
'pintuan_end_time'] = (
now_time,
now_time)
else:
goods_data[
'pintuan_begin_time'], goods_data[
'pintuan_end_time'] = self.get_pintuan_begin_time_and_pintuan_end_time(
pintuan_time=
goods_data[
'pintuan_time']
)
# pprint(goods_data)
# print(goods_data)
mia_pintuan.update_mia_pintuan_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()
