def get_miaoshao_goods_info_list(self, data):
'''
得到秒杀商品有用信息
:param data: 待解析的data
:return: 有用信息list
'''
miaosha_goods_list = []
for item in data:
tmp = {}
tmp['miaosha_time'] = {
'miaosha_begin_time':
timestamp_to_regulartime(int(item.get('start_time'))),
'miaosha_end_time':
timestamp_to_regulartime(int(item.get('end_time'))),
}
stock = item.get('stock', 0)
# 卷皮商品的goods_id
tmp['goods_id'] = item.get('goods_id')
# 限时秒杀库存信息
tmp['stock_info'] = {
'activity_stock':
int(item.get('stock', 0) * (item.get('rate', 0) / 100)),
'stock':
item.get('stock', 0),
}
# 原始价格
tmp['price'] = round(float(item.get('oprice', '0')), 2)
tmp['taobao_price'] = round(float(item.get('cprice', '0')), 2)
miaosha_goods_list.append(tmp)
return miaosha_goods_list
async def get_tiantiantejia_goods_list(self, data):
'''
将tmp_data转换为需求的list
:param body:
:return: a list
'''
if data != []:
# 处理得到需要的数据
try:
tejia_goods_list = [{
'goods_id':
item.get('baseinfo', {}).get('itemId', ''),
'start_time':
timestamp_to_regulartime(
int(item.get('baseinfo', {}).get('ostime', '')[0:10])),
'end_time':
timestamp_to_regulartime(
int(item.get('baseinfo', {}).get('oetime', '')[0:10])),
} for item in data]
except Exception as e:
self.my_lg.exception(e)
tejia_goods_list = []
else:
tejia_goods_list = []
return tejia_goods_list
def _get_pintuan_goods_info(self):
'''
模拟构造得到data的url, 得到近期所有的限时拼团商品信息
:return:
'''
pintuan_goods_id_list = []
for page in range(0, 100):
tmp_url = 'https://tuan.juanpi.com/pintuan/get_goods_list?page={0}&pageSize=20&cid=pinhaohuo_sx&show_type=wap'.format(
str(page)
)
print('正在抓取的页面地址为: ', tmp_url)
body = MyRequests.get_url_body(url=tmp_url, headers=self.headers)
if body == '': body = '{}'
try:
tmp_data = json.loads(body)
tmp_data = tmp_data.get('data', {}).get('goods', [])
except:
print('json.loads转换tmp_data时出错!')
tmp_data = []
# print(tmp_data)
sleep(.5)
if tmp_data == []:
print('该tmp_url得到的goods为空list, 此处跳过!')
break
tmp_pintuan_goods_id_list = [{
'goods_id': item.get('goods_id', ''),
'begin_time': timestamp_to_regulartime(int(item.get('start_time', ''))),
'end_time': timestamp_to_regulartime(int(item.get('end_time', ''))),
'all_sell_count': str(item.get('join_number_int', '')),
'page': page,
} for item in tmp_data]
# print(tmp_pintuan_goods_id_list)
for item in tmp_pintuan_goods_id_list:
if item.get('goods_id', '') not in [item2.get('goods_id', '') for item2 in pintuan_goods_id_list]:
pintuan_goods_id_list.append(item)
print('该pintuan_goods_id_list的总个数为: ', len(pintuan_goods_id_list))
print(pintuan_goods_id_list)
return pintuan_goods_id_list
def _get_goods_schedule(self, data):
'''
获取商品销售时间段
:param data:
:return:
'''
# print(data.get('skudata', {}).get('info', {}))
# print(data.get('skudata', {}))
begin_time = data.get('skudata', {}).get('info', {}).get(
'start_time') # 取这个时间段才是正确的销售时间, 之前baseInfo是虚假的
end_time = data.get('skudata', {}).get('info', {}).get('end_time')
if begin_time is None or end_time is None:
schedule = []
else:
schedule = [{
'begin_time': timestamp_to_regulartime(begin_time),
'end_time': timestamp_to_regulartime(end_time),
}]
return schedule
def json_to_dict(self, tmp_data):
try:
data = json.loads(tmp_data)
# pprint(data)
times = [str(timestamp_to_regulartime(int(item))) for item in data.get('times', [])]
data = data.get('items', [])
# print(data)
# print(times)
except:
print('json.loads转换data的时候出错，data为空')
data = []
return data
def get_miaoshao_goods_info_list(self, data):
'''
得到秒杀商品有用信息
:param data: 待解析的data
:return: 有用信息list
'''
miaosha_goods_list = []
for item in data:
# pprint(item)
tmp = {}
# 秒杀开始时间和结束时间
tmp['miaosha_time'] = {
'miaosha_begin_time':
timestamp_to_regulartime(int(
str(item.get('begin_time'))[0:10])),
'miaosha_end_time':
timestamp_to_regulartime(int(str(item.get('end_time'))[0:10])),
}
# 折800商品地址
tmp['zid'] = item.get('zid')
# 是否包邮
# tmp['is_baoyou'] = item.get('is_baoyou', 0)
# 限时秒杀的库存信息
tmp['stock_info'] = {
'activity_stock': item.get('activity_stock',
0), # activity_stock为限时抢的剩余数量
'stock': item.get('stock', 0), # stock为限时秒杀的总库存
}
# 原始价格
tmp['price'] = float(item.get('list_price'))
# 秒杀的价格, float类型
tmp['taobao_price'] = float(item.get('price'))
# 子标题
tmp['sub_title'] = item.get('description', '')
miaosha_goods_list.append(tmp)
# pprint(miaosha_goods_list)
return miaosha_goods_list
def get_miaoshao_goods_info_list(self, data):
'''
得到秒杀商品有用信息
:param data: 待解析的data
:return: 有用信息list
'''
miaosha_goods_list = []
for item in data:
tmp = {}
miaosha_begin_time = str(timestamp_to_regulartime(int(item.get('data', {}).get('start_time'))))
tmp_hour = miaosha_begin_time[-8:-6]
if tmp_hour in PINDUODUO_MIAOSHA_SPIDER_HOUR_LIST:
if tmp_hour in PINDUODUO_MIAOSHA_BEGIN_HOUR_LIST:
'''
# 这些起始的点秒杀时间只有30分钟
'''
miaosha_end_time = str(timestamp_to_regulartime(int(item.get('data', {}).get('start_time')) + 60*30))
else:
miaosha_end_time = str(timestamp_to_regulartime(int(item.get('data', {}).get('start_time')) + 60*60))
tmp['miaosha_time'] = {
'miaosha_begin_time': miaosha_begin_time,
'miaosha_end_time': miaosha_end_time,
}
# 卷皮商品的goods_id
tmp['goods_id'] = str(item.get('data', {}).get('goods_id'))
# 限时秒杀库存信息
tmp['stock_info'] = {
'activity_stock': int(item.get('data', {}).get('all_quantity', 0) - item.get('data', {}).get('sold_quantity', 0)),
'stock': item.get('data', {}).get('all_quantity', 0),
}
# 原始价格
tmp['price'] = round(float(item.get('data', {}).get('normal_price', '0'))/100, 2)
tmp['taobao_price'] = round(float(item.get('data', {}).get('price', '0'))/100, 2)
miaosha_goods_list.append(tmp)
else:
pass
return miaosha_goods_list
def get_pintuan_goods_info(self):
'''
模拟构造得到data的url，得到近期所有的限时拼团商品信息
:return: None
'''
goods_list = []
'''
方法一: 蘑菇街手机版拼团商品列表获取签名暂时无法破解，所以不用手机端的方法来获取数据
'''
# mw_appkey = '100028'
# mw_t = str(time.time().__round__()) + str(randint(100, 999)) # time.time().__round__() 表示保留到个位
# mw_uuid = '956bf265-90a4-45b0-bfa8-31040782f99e'
# mw_ttid = 'NMMain%40mgj_h5_1.0'
#
# _ = str(time.time().__round__()) + str(randint(100, 999)) # time.time().__round__() 表示保留到个位
#
# data = {
# "pid": "93745",
# "platform": "m",
# "cKey": "mwp_mait",
# "fcid": "",
# }
#
# params = {
# 'data': data
# }
#
# # https://api.mogujie.com/h5/mwp.darwin.get/3/?mw-appkey=100028&mw-t=1517647409632&mw-uuid=956bf265-90a4-45b0-bfa8-31040782f99e&mw-ttid=NMMain%40mgj_h5_1.0&mw-sign=abde92f778e47bce98a3ed25fd71eb1a&data=%7B%22pid%22%3A%2293745%22%2C%22platform%22%3A%22m%22%2C%22cKey%22%3A%22mwp_mait%22%2C%22fcid%22%3A%22%22%7D&callback=mwpCb1&_=1517647409648
# # https://api.mogujie.com/h5/mwp.darwin.get/3/?mw-appkey=100028&mw-t=1517647893930&mw-uuid=956bf265-90a4-45b0-bfa8-31040782f99e&mw-ttid=NMMain%40mgj_h5_1.0&callback=mwpCb1&_=1517647893748&data=pid&data=platform&data=cKey&data=fcid
#
# tmp_url = 'https://api.mogujie.com/h5/mwp.darwin.get/3/?mw-appkey={0}&mw-t={1}&mw-uuid={2}&mw-ttid={3}&callback=mwpCb1&_={4}'.format(
# mw_appkey, mw_t, mw_uuid, mw_ttid, _
# )
#
# # 设置代理ip
# ip_object = MyIpPools()
# self.proxies = ip_object.get_proxy_ip_from_ip_pool() # {'http': ['xx', 'yy', ...]}
# self.proxy = self.proxies['http'][randint(0, len(self.proxies) - 1)]
#
# tmp_proxies = {
# 'http': self.proxy,
# }
#
# try:
# response = requests.post(tmp_url, headers=self.headers, data=data, proxies=tmp_proxies, timeout=13) # 在requests里面传数据，在构造头时，注意在url外头的&xxx=也得先构造
# body = response.content.decode('utf-8')
# print(body)
# except Exception:
# print('requests.get()请求超时....')
# print('data为空!')
# self.result_data = {} # 重置下，避免存入时影响下面爬取的赋值
# return {}
'''
方法二: 通过pc端来获取拼团商品列表
'''
self.my_phantomjs = MyPhantomjs()
for key in self.fcid_dict:
print('正在抓取的分类为: ', key)
for index in range(1, 100):
if index % 5 == 0:
try: del self.my_phantomjs
except: pass
gc.collect()
self.my_phantomjs = MyPhantomjs()
fcid = self.fcid_dict[key]
tmp_url = 'http://list.mogujie.com/search?page={0}&fcid={1}&algoKey=pc_tuan_book_pop&cKey=pc-tuan'.format(
str(index), fcid
)
# requests请求数据被过滤(起初能用)，改用phantomjs
# body = MyRequests.get_url_body(url=tmp_url, headers=self.headers, had_referer=True)
body = self.my_phantomjs.use_phantomjs_to_get_url_body(url=tmp_url)
# print(body)
try:
body = re.compile(r'<pre.*?>(.*?)</pre>').findall(body)[0]
tmp_data = json.loads(body)
except:
print('json.loads转换body时出错, 请检查')
continue
if tmp_data.get('result', {}).get('wall', {}).get('docs', []) == []:
# 表示拼团数据为空则跳出循环
break
# pprint(tmp_data)
# print(tmp_data)
tmp_item_list = tmp_data.get('result', {}).get('wall', {}).get('docs', [])
# print(tmp_item_list)
# pprint(tmp_item_list)
begin_time_timestamp = int(time.time()) # 开始拼团的时间戳
item_list = [{
'goods_id': item.get('tradeItemId', ''),
'pintuan_time': {
'begin_time': timestamp_to_regulartime(timestamp=begin_time_timestamp),
'end_time': timestamp_to_regulartime(self.get_pintuan_end_time(begin_time_timestamp, item.get('leftTimeOrg', ''))),
},
'all_sell_count': str(item.get('salesVolume', 0)),
'fcid': fcid,
'page': index,
'sort': key,
} for item in tmp_item_list]
print(item_list)
for item_1 in item_list:
goods_list.append(item_1)
sleep(MOGUJIE_SLEEP_TIME)
# 处理goods_list数据
print(goods_list)
self.deal_with_data(goods_list)
sleep(5)
def run_forever(self):
'''
实时更新数据
:return:
'''
tmp_sql_server = SqlServerMyPageInfoSaveItemPipeline()
sql_str = r'select goods_id, miaosha_time, event_time, goods_url from dbo.mogujie_xianshimiaosha where site_id=22'
try:
result = list(tmp_sql_server._select_table(sql_str=sql_str))
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
mogujie_miaosha = MoGuJieMiaoShaParse()
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_table(
sql_str=self.delete_sql_str, params=(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]
item_list = self.get_item_list(event_time=str(item[2]))
if item_list == '':
# 可能网络状况导致, 先跳过
pass
elif item_list == []:
print('该商品已被下架限时秒杀活动，此处将其删除')
tmp_sql_server._delete_table(
sql_str=self.delete_sql_str, params=(item[0]))
print('下架的goods_id为(%s)' % item[0], ', 删除成功!')
pass
else:
# 该event_time中现有的所有goods_id的list
miaosha_goods_all_goods_id = [
item_1.get('iid', '') 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_table(
sql_str=self.delete_sql_str,
params=(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('iid', '') == item[0]:
spider_url = item[3]
mogujie_miaosha.get_goods_data(
goods_id=spider_url)
goods_data = mogujie_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])
# price设置为原价
try:
tmp_price_list = sorted([
round(
float(
item_4.get(
'normal_price',
'')), 2)
for item_4 in goods_data[
'price_info_list']
])
price = Decimal(
tmp_price_list[-1]
).__round__(2) # 商品原价
goods_data['price'] = price
except:
print('设置price为原价时出错!请检查')
continue
goods_data['miaosha_time'] = {
'miaosha_begin_time':
timestamp_to_regulartime(
int(
item_2.get(
'startTime', 0))),
'miaosha_end_time':
timestamp_to_regulartime(
int(
item_2.get(
'endTime', 0))),
}
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)
mogujie_miaosha.update_mogujie_xianshimiaosha_table(
data=goods_data,
pipeline=tmp_sql_server)
sleep(MOGUJIE_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()
def run_forever(self):
'''
实时更新数据
:return:
'''
tmp_sql_server = SqlServerMyPageInfoSaveItemPipeline()
sql_str = r'select goods_id, miaosha_time, fcid, page from dbo.mogujie_pintuan where site_id=23'
try:
result = list(tmp_sql_server._select_table(sql_str=sql_str))
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
self.my_phantomjs = MyPhantomjs()
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 = {}
mogujie_pintuan = MoGuJieParse()
if index % 8 == 0:
try:
del self.my_phantomjs
except:
pass
gc.collect()
self.my_phantomjs = MyPhantomjs()
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_table(
sql_str=self.delete_sql_str, params=(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]
tmp_url = 'http://list.mogujie.com/search?page={0}&fcid={1}&algoKey=pc_tuan_book_pop&cKey=pc-tuan'.format(
item[3], item[2])
# print(tmp_url)
# requests请求不到数据，涉及证书认证，直接用phantomjs
# body = MyRequests.get_url_body(url=tmp_url, headers=self.headers, had_referer=True)
body = self.my_phantomjs.use_phantomjs_to_get_url_body(
url=tmp_url)
# print(body)
if body == '':
print('获取到的body为空值! 此处跳过')
else:
try:
body = re.compile(
r'<pre.*?>(.*?)</pre>').findall(body)[0]
tmp_data = json.loads(body)
# pprint(tmp_data)
except:
print('json.loads转换body时出错, 请检查')
tmp_data = {}
if tmp_data.get('result',
{}).get('wall',
{}).get('docs', []) == []:
print('得到的docs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_table(
sql_str=self.delete_sql_str,
params=(item[0]))
print('下架的goods_id为(%s)' % item[0], ', 删除成功!')
pass
else:
tmp_item_list = tmp_data.get('result', {}).get(
'wall', {}).get('docs', [])
# print(tmp_item_list)
# pprint(tmp_item_list)
begin_time_timestamp = int(
time.time()) # 开始拼团的时间戳
item_list = [{
'goods_id':
item.get('tradeItemId', ''),
'pintuan_time': {
'begin_time':
timestamp_to_regulartime(
timestamp=begin_time_timestamp),
'end_time':
timestamp_to_regulartime(
self.get_pintuan_end_time(
begin_time_timestamp,
item.get('leftTimeOrg', ''))),
},
'all_sell_count':
str(item.get('salesVolume', 0)),
} for item in tmp_item_list]
# print(item_list)
pintuan_goods_all_goods_id = [
item_1.get('goods_id', '')
for item_1 in item_list
]
# print(pintuan_goods_all_goods_id)
'''
内部已经下架的(内部下架的其实并未真实下架，还在卖的，所以我就更新其商品信息数据，不更新上下架时间)
'''
if item[0] not in pintuan_goods_all_goods_id:
# print('该商品已被下架限时秒杀活动，此处将其删除')
# tmp_sql_server._delete_table(sql_str=self.delete_sql_str, params=(item[0]))
# print('下架的goods_id为(%s)' % item[0], ', 删除成功!')
# pass
mogujie_pintuan.get_goods_data(
goods_id=item[0])
goods_data = mogujie_pintuan.deal_with_data(
)
if goods_data == {}:
pass
else:
# 规范化
print('+++ 内部下架，其实还在售卖的商品更新')
tmp_price_info_list = goods_data[
'price_info_list']
price_info_list = [{
'spec_value':
item_4.get('spec_value'),
'pintuan_price':
item_4.get('detail_price'),
'detail_price':
'',
'normal_price':
item_4.get('normal_price'),
'img_url':
item_4.get('img_url'),
'rest_number':
item_4.get('rest_number'),
} for item_4 in tmp_price_info_list]
goods_data['goods_id'] = item[0]
goods_data[
'price_info_list'] = price_info_list
# pprint(goods_data)
# print(goods_data)
mogujie_pintuan.update_mogujie_pintuan_table_2(
data=goods_data,
pipeline=tmp_sql_server)
sleep(MOGUJIE_SLEEP_TIME) # 放慢速度
else: # 未下架的
for item_2 in item_list:
if item_2.get('goods_id',
'') == item[0]:
mogujie_pintuan.get_goods_data(
goods_id=item[0])
goods_data = mogujie_pintuan.deal_with_data(
)
if goods_data == {}: pass
else:
# 规范化
tmp_price_info_list = goods_data[
'price_info_list']
price_info_list = [
{
'spec_value':
item_4.get(
'spec_value'),
'pintuan_price':
item_4.get(
'detail_price'),
'detail_price':
'',
'normal_price':
item_4.get(
'normal_price'),
'img_url':
item_4.get('img_url'),
'rest_number':
item_4.get(
'rest_number'),
} for item_4 in
tmp_price_info_list
]
goods_data['goods_id'] = item[
0]
goods_data[
'price_info_list'] = price_info_list
goods_data[
'pintuan_time'] = item_2.get(
'pintuan_time', {})
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[
'all_sell_count'] = item_2.get(
'all_sell_count', '')
# pprint(goods_data)
# print(goods_data)
mogujie_pintuan.update_mogujie_pintuan_table(
data=goods_data,
pipeline=tmp_sql_server)
sleep(
MOGUJIE_SLEEP_TIME) # 放慢速度
else:
pass
else:
print('数据库连接失败，此处跳过!')
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()
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:
sql_str = r'select goods_id, miaosha_time, pid from dbo.mia_xianshimiaosha where site_id=20'
db_goods_id_list = [
item[0]
for item in list(my_pipeline._select_table(sql_str=sql_str))
]
# 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':
timestamp_to_regulartime(begin_time),
'miaosha_end_time':
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 deal_with_data(self):
'''
处理得到规范的data数据
:return: result 类型 dict
'''
data = self.result_data
if data != {}:
# 店铺名称
shop_name = data['shop_name']
# 掌柜
account = ''
# 商品名称
title = data['title']
# 子标题
sub_title = data['sub_title']
# 商品标签属性名称
detail_name_list = data['detail_name_list']
# 要存储的每个标签对应规格的价格及其库存
price_info_list = data['price_info_list']
# 所有示例图片地址
all_img_url = data['all_img_url']
# 详细信息标签名对应属性
p_info = data['p_info']
# pprint(p_info)
# div_desc
div_desc = data['div_desc']
'''
用于判断商品是否已经下架
'''
is_delete = 0
all_rest_number = 0
for item in price_info_list:
all_rest_number += item.get('rest_number', 0)
if all_rest_number == 0:
is_delete = 1
# 当官方下架时间< int(time.time()) 则商品已下架 is_delete = 1
if int(data.get('sell_time', {}).get('end_time', '')) < int(
time.time()):
print('该商品已经过期下架...! 进行逻辑删除 is_delete=1')
is_delete = 1
# print(is_delete)
# 上下架时间
schedule = [{
'begin_time':
timestamp_to_regulartime(
int(data.get('sell_time', {}).get('begin_time', ''))),
'end_time':
timestamp_to_regulartime(
int(data.get('sell_time', {}).get('end_time', ''))),
}]
# 销售总量
all_sell_count = ''
# 商品价格和淘宝价
# pprint(data['price_info_list'])
try:
tmp_price_list = sorted([
round(float(item.get('detail_price', '')), 2)
for item in data['price_info_list']
])
price = tmp_price_list[-1] # 商品价格
taobao_price = tmp_price_list[0] # 淘宝价
except IndexError:
print('获取price和taobao_price时出错, 请检查!'
) # 商品下架时, detail_price为空str, 所以会IndexError报错
print('@@@@@@ 此处对该商品进行逻辑删除! @@@@@@')
self.result_data = {}
price = 0.
taobao_price = 0.
is_delete = 1
# return {}
result = {
'shop_name': shop_name, # 店铺名称
'account': account, # 掌柜
'title': title, # 商品名称
'sub_title': sub_title, # 子标题
'price': price, # 商品价格
'taobao_price': taobao_price, # 淘宝价
# 'goods_stock': goods_stock, # 商品库存
'detail_name_list': detail_name_list, # 商品标签属性名称
# 'detail_value_list': detail_value_list,# 商品标签属性对应的值
'price_info_list': price_info_list, # 要存储的每个标签对应规格的价格及其库存
'all_img_url': all_img_url, # 所有示例图片地址
'p_info': p_info, # 详细信息标签名对应属性
'div_desc': div_desc, # div_desc
'schedule': schedule, # 商品特价销售时间段
'all_sell_count': all_sell_count, # 销售总量
'is_delete': is_delete # 用于判断商品是否已经下架
}
# pprint(result)
# print(result)
# wait_to_send_data = {
# 'reason': 'success',
# 'data': result,
# 'code': 1
# }
# json_data = json.dumps(wait_to_send_data, ensure_ascii=False)
# print(json_data)
self.result_data = {}
return result
else:
print('待处理的data为空的dict, 该商品可能已经转移或者下架')
self.result_data = {}
return {}
def deal_with_data(self, *params):
'''
处理并存储相关秒杀商品数据
:param params: 相关参数
:return:
'''
item_list = params[0]
chuchujie = ChuChuJie_9_9_Parse()
my_pipeline = SqlServerMyPageInfoSaveItemPipeline()
if my_pipeline.is_connect_success:
sql_str = r'select goods_id, miaosha_time, gender, page, goods_url from dbo.chuchujie_xianshimiaosha where site_id=24'
db_goods_id_list = [
item[0]
for item in list(my_pipeline._select_table(sql_str=sql_str))
]
# print(db_goods_id_list)
# my_phantomjs = MyPhantomjs()
# my_phantomjs.init_phantomjs()
# index = 1
for item in item_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 = item.get('goods_id', '')
tmp_url = 'https://m.chuchujie.com/details/detail.html?id=' + str(
goods_id)
chuchujie.get_goods_data(goods_id=goods_id)
goods_data = chuchujie.deal_with_data()
if goods_data == {}: # 返回的data为空则跳过
pass
elif goods_data.get('is_delete',
0) == 1: # is_delete=1(即库存为0)则跳过
print('------>>>| 该商品库存为0，已被抢光!')
pass
else: # 否则就解析并且插入
my_phantomjs = MyPhantomjs()
my_phantomjs.init_phantomjs()
# 获取剩余时间
tmp_body = my_phantomjs.use_phantomjs_to_get_url_body(
url=tmp_url, css_selector='p#activityTime span')
# print(tmp_body)
try:
del my_phantomjs
except:
pass
gc.collect()
if tmp_body == '': # 获取手机版的页面完整html失败
sleep(.4)
pass
else:
# p#activityTime span
_t = Selector(text=tmp_body).css(
'p#activityTime span::text').extract_first()
_t = re.compile(r'剩余').sub('', _t)
# print(_t)
if _t == '' or _t is None:
print('获取到的_t为空值, 严重错误! 请检查!')
miaosha_end_time = self.get_miaosha_end_time(_t)
goods_data['goods_url'] = tmp_url
goods_data['goods_id'] = str(goods_id)
goods_data['sub_title'] = item.get('sub_title', '')
goods_data['miaosha_time'] = {
'miaosha_begin_time':
timestamp_to_regulartime(int(time.time())),
'miaosha_end_time':
timestamp_to_regulartime(
int(miaosha_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['gender'] = str(item.get('gender', '0'))
goods_data['page'] = item.get('page')
# pprint(goods_data)
# print(goods_data)
chuchujie.insert_into_chuchujie_xianshimiaosha_table(
data=goods_data, pipeline=my_pipeline)
# sleep(CHUCHUJIE_SLEEP_TIME) # 放慢速度 由于初始化用了phantomjs时间久，于是就不睡眠
# index += 1
else:
print('数据库连接失败，此处跳过!')
pass
try:
del chuchujie
except:
pass
gc.collect()
def run_forever(self):
'''
实时更新数据
:return:
'''
tmp_sql_server = SqlServerMyPageInfoSaveItemPipeline()
sql_str = r'select goods_id, miaosha_time, pid from dbo.mia_xianshimiaosha where site_id=20'
try:
result = list(tmp_sql_server._select_table(sql_str=sql_str))
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_table(
sql_str=self.delete_sql_str, params=(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_table(
sql_str=self.delete_sql_str,
params=(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':
timestamp_to_regulartime(
begin_time),
'miaosha_end_time':
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()
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:
'''
print(60 * '*')
event_time = param[0]
print('秒杀开始时间:', timestamp_to_regulartime(event_time), '\t',
'对应时间戳为: ', event_time)
print(60 * '*')
item_list = param[1]
mogujie = MoGuJieMiaoShaParse()
my_pipeline = SqlServerMyPageInfoSaveItemPipeline()
if my_pipeline.is_connect_success:
sql_str = r'select goods_id, miaosha_time, event_time, goods_url from dbo.mogujie_xianshimiaosha where site_id=22'
db_goods_id_list = [
item[0]
for item in list(my_pipeline._select_table(sql_str=sql_str))
]
# print(db_goods_id_list)
for item in item_list:
if item.get('iid', '') 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('iid', ''))
tmp_url = item.get('link', '')
try:
object_id = re.compile(r'objectId=(.*?)&').findall(
tmp_url)[0]
except IndexError: # 表示匹配到的地址不是秒杀商品的地址
print('+++++++ 这个url不是秒杀的url: ', tmp_url)
continue
tmp_url = 'https://shop.mogujie.com/rushdetail/{0}?objectId={1}&type=rush'.format(
goods_id, object_id)
tmp_ = mogujie.get_goods_id_from_url(tmp_url)
mogujie.get_goods_data(goods_id=tmp_)
goods_data = mogujie.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_data['goods_url'] = tmp_url
goods_data['goods_id'] = str(goods_id)
# price设置为原价
try:
tmp_price_list = sorted([
round(float(item_4.get('normal_price', '')), 2)
for item_4 in goods_data['price_info_list']
])
price = Decimal(tmp_price_list[-1]).__round__(
2) # 商品原价
goods_data['price'] = price
except:
print('设置price为原价时出错!请检查')
continue
goods_data['miaosha_time'] = {
'miaosha_begin_time':
timestamp_to_regulartime(
int(item.get('startTime', 0))),
'miaosha_end_time':
timestamp_to_regulartime(
int(item.get('endTime', 0))),
}
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['event_time'] = str(event_time)
# pprint(goods_data)
# print(goods_data)
mogujie.insert_into_mogujie_xianshimiaosha_table(
data=goods_data, pipeline=my_pipeline)
sleep(MOGUJIE_SLEEP_TIME) # 放慢速度
else:
print('数据库连接失败，此处跳过!')
pass
try:
del mogujie
except:
pass
gc.collect()
async def get_one_page_goods_list(self, **kwargs):
'''
获取单页面的goods_list
:param kwargs:
:return: item_list 类型list
'''
my_phantomjs = kwargs.get('my_phantomjs')
key = kwargs.get('key', '')
tab = kwargs.get('tab', '')
index = kwargs.get('index')
i_time = time.time()
tmp_url = 'http://s.h5.jumei.com/yiqituan/tab_list?tab={0}&page={1}&per_page=20'.format(
tab, str(index))
# 常规requests被过滤, aiohttp成功, 测试发现：设置时间短抓取较快
# body = await MyAiohttp.aio_get_url_body(url=tmp_url, headers=self.headers, timeout=JUMEIYOUPIN_PINTUAN_API_TIMEOUT)
# 改用phantomjs，aiohttp太慢
body = my_phantomjs.use_phantomjs_to_get_url_body(url=tmp_url)
try:
body = re.compile('<pre .*?>(.*)</pre>').findall(body)[0]
except:
pass
await asyncio.sleep(1)
# self.my_lg.info(body)
self.msg = '正在抓取第' + str(index) + '页...' + ' ☭ 用时: ' + str(
time.time() - i_time)
self.my_lg.info(self.msg)
item_list = []
if body == '':
self.msg = '获取到的body为空str!' + ' 出错地址: ' + tmp_url
self.my_lg.error(self.msg)
else:
one_data = await self.json_2_dict(json_str=body)
if one_data == {}:
self.msg = '出错地址: ' + tmp_url
self.my_lg.error(self.msg)
else:
if one_data.get('data', []) == []:
pass
else:
tmp_item_list = one_data.get('data', [])
for item in tmp_item_list: # 由于await 不能理解列表表达式，就采用常规做法
if item.get('status', '') != 'soldout':
item_list.append({
'goods_id':
item.get('item_id', ''),
'pintuan_time': {
'begin_time':
timestamp_to_regulartime(
item.get('start_time', '0')),
'end_time':
timestamp_to_regulartime(
item.get('end_time', '0')),
},
'type':
item.get('type', ''),
'sort':
key,
'page':
index,
'tab':
tab,
})
# self.my_lg.info(str(item_list))
return item_list
def get_spike_hour_goods_info(self):
'''
模拟构造得到data的url，得到近期所有的限时秒杀商品信息
:return:
'''
base_session_id = BASE_SESSION_ID
while base_session_id < MAX_SESSION_ID:
print('待抓取的session_id为: ', base_session_id)
data = self._get_one_session_id_data(
base_session_id=base_session_id)
sleep(.2)
if data.get('data', {}).get('blocks', []) == []: # session_id不存在
pass
else: # 否则session_id存在
try:
_ = str(
data.get('data',
{}).get('blocks',
[])[0].get('deal',
{}).get('begin_time',
''))[:10]
if _ != '':
pass
elif data.get('data', {}).get('blocks', [])[0].get(
'showcase', {}) != {}: # 未来时间
print('*** 未来时间 ***')
# pprint(data.get('data', {}))
_ = str(
data.get('data', {}).get('blocks', [])[1].get(
'deal', {}).get('begin_time', ''))[:10]
else:
raise Exception
begin_times_timestamp = int(
_) # 将如 "2017-09-28 10:00:00"的时间字符串转化为时间戳，然后再将时间戳取整
except Exception as e:
print('遇到严重错误: ', e)
continue
print('秒杀时间为: ',
timestamp_to_regulartime(begin_times_timestamp))
if self.is_recent_time(
timestamp=begin_times_timestamp): # 说明秒杀日期合法
try:
data = [
item_s.get('deal', {}) for item_s in data.get(
'data', {}).get('blocks', [])
]
except Exception as e:
print('遇到严重错误: ', e)
continue
# pprint(data)
if data != []: # 否则说明里面有数据
miaosha_goods_list = self.get_miaoshao_goods_info_list(
data=data)
# pprint(miaosha_goods_list)
zhe_800 = Zhe800Parse()
my_pipeline = SqlServerMyPageInfoSaveItemPipeline()
if my_pipeline.is_connect_success:
sql_str = r'select goods_id, miaosha_time, session_id from dbo.zhe_800_xianshimiaosha where site_id=14'
db_goods_id_list = [
item[0] for item in list(
my_pipeline._select_table(sql_str=sql_str))
]
for item in miaosha_goods_list:
if item.get('zid', '') in db_goods_id_list:
print('该goods_id已经存在于数据库中, 此处跳过')
pass
else:
tmp_url = 'https://shop.zhe800.com/products/' + str(
item.get('zid', ''))
goods_id = zhe_800.get_goods_id_from_url(
tmp_url)
zhe_800.get_goods_data(goods_id=goods_id)
goods_data = zhe_800.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_data['stock_info'] = item.get(
'stock_info')
goods_data['goods_id'] = str(
item.get('zid'))
goods_data['spider_url'] = tmp_url
goods_data['username'] = '******'
goods_data['price'] = item.get('price')
goods_data['taobao_price'] = item.get(
'taobao_price')
goods_data['sub_title'] = item.get(
'sub_title')
# goods_data['is_baoyou'] = item.get('is_baoyou')
goods_data['miaosha_time'] = item.get(
'miaosha_time')
goods_data[
'miaosha_begin_time'], goods_data[
'miaosha_end_time'] = get_miaosha_begin_time_and_miaosha_end_time(
miaosha_time=item.get(
'miaosha_time'))
goods_data['session_id'] = str(
base_session_id)
# print(goods_data['miaosha_time'])
# print(goods_data)
zhe_800.insert_into_zhe_800_xianshimiaosha_table(
data=goods_data,
pipeline=my_pipeline)
sleep(ZHE_800_SPIKE_SLEEP_TIME) # 放慢速度
# sleep(2)
else:
pass
try:
del zhe_800
except:
pass
gc.collect()
else: # 说明这个sessionid没有数据
print('该sessionid没有相关key为jsons的数据')
# return {}
pass
else:
pass
base_session_id += 2
