async def _handle_al_goods_is_delete(self, goods_id) -> dict:
'''
处理商品无法查看或者下架的
:return:
'''
try:
sql_cli = SqlServerMyPageInfoSaveItemPipeline()
is_in_db = sql_cli._select_table(sql_str=al_select_str_1,
params=(str(goods_id), ))
# self.lg.info(str(is_in_db))
except Exception:
self.lg.error('数据库连接失败!' + self.error_base_record, exc_info=True)
return await self._data_error_init()
self.result_data = {}
# 初始化下架商品的属性
tmp_data_s = await self._init_al_pull_off_shelves_goods()
if is_in_db != []:
# 表示该goods_id以前已被插入到db中, 于是只需要更改其is_delete的状态即可
sql_cli._update_table_2(sql_str=al_update_str_1,
params=(goods_id),
logger=self.lg)
self.lg.info('@@@ 该商品goods_id原先存在于db中, 此处将其is_delete=1')
# 用来判断原先该goods是否在db中
tmp_data_s['before'] = True
else:
# 表示该goods_id没存在于db中
self.lg.info('@@@ 该商品已下架[但未存在于db中], ** 此处将其插入到db中...')
tmp_data_s['before'] = False
return tmp_data_s
class JDUpdater(AsyncCrawler):
"""jd常规商品更新"""
def __init__(self, *params, **kwargs):
AsyncCrawler.__init__(
self,
*params,
**kwargs,
log_print=True,
log_save_path=MY_SPIDER_LOGS_PATH + '/jd/实时更新/'
)
self.sql_cli = None
self.goods_index = 1
# 并发量
self.concurrency = 10
async def _get_db_old_data(self):
self.sql_cli = SqlServerMyPageInfoSaveItemPipeline()
result = None
try:
result = list(self.sql_cli._select_table(sql_str=jd_select_str_1))
except TypeError:
self.lg.error('TypeError错误, 原因数据库连接失败...(可能维护中)')
await _print_db_old_data(logger=self.lg, result=result)
return result
async def _get_new_jd_obj(self, index):
if index % 10 == 0: # 不能共享一个对象了, 否则驱动访问会异常!
try:
del self.jd
except:
pass
collect()
self.jd = JdParse(logger=self.lg)
async def _get_tmp_item(self, site_id, goods_id):
tmp_item = []
if site_id == 7 or site_id == 8: # 从数据库中取出时，先转换为对应的类型
tmp_item.append(0)
elif site_id == 9:
tmp_item.append(1)
elif site_id == 10:
tmp_item.append(2)
tmp_item.append(goods_id)
return tmp_item
async def _update_one_goods_info(self, db_goods_info_obj, index):
'''
更新单个jd商品信息
:param db_goods_info_obj:
:param index:
:return:
'''
res = False
await self._get_new_jd_obj(index=index)
self.sql_cli = await _get_new_db_conn(db_obj=self.sql_cli, index=index, logger=self.lg)
if self.sql_cli.is_connect_success:
self.lg.info('------>>>| 正在更新的goods_id为({0}) | --------->>>@ 索引值为({1})'.format(
db_goods_info_obj.goods_id,
index))
tmp_item = await self._get_tmp_item(
site_id=db_goods_info_obj.site_id,
goods_id=db_goods_info_obj.goods_id,)
data = self.jd.get_goods_data(goods_id=tmp_item)
if data.get('is_delete', 1) == 1:
self.lg.info('该商品已下架...')
self.sql_cli._update_table_2(
sql_str=jd_update_str_2,
params=(str(get_shanghai_time()), tmp_item[1],),
logger=self.lg)
await async_sleep(1.2)
index += 1
self.goods_index = index
return db_goods_info_obj.goods_id, index
data = self.jd.deal_with_data(goods_id=tmp_item)
if data != {}:
data = get_goods_info_change_data(
target_short_name='jd',
logger=self.lg,
data=data,
db_goods_info_obj=db_goods_info_obj,)
self.jd.to_right_and_update_data(data, pipeline=self.sql_cli)
else: # 表示返回的data值为空值
pass
else: # 表示返回的data值为空值
self.lg.error('数据库连接失败，数据库可能关闭或者维护中')
pass
index += 1
self.goods_index = index
collect()
await async_sleep(1.2) # 避免被发现使用代理
return db_goods_info_obj.goods_id, index
async def _update_db(self):
while True:
self.lg = await self._get_new_logger(logger_name=get_uuid1())
result = await self._get_db_old_data()
if result is None:
pass
else:
self.goods_index = 1
tasks_params_list = TasksParamsListObj(tasks_params_list=result, step=self.concurrency)
self.jd = JdParse(logger=self.lg)
index = 1
while True:
try:
slice_params_list = tasks_params_list.__next__()
# self.lg.info(str(slice_params_list))
except AssertionError: # 全部提取完毕, 正常退出
break
tasks = []
for item in slice_params_list:
db_goods_info_obj = JDDbGoodsInfoObj(item=item, logger=self.lg)
self.lg.info('创建 task goods_id: {}'.format(db_goods_info_obj.goods_id))
tasks.append(self.loop.create_task(self._update_one_goods_info(
db_goods_info_obj=db_goods_info_obj,
index=index)))
index += 1
await _get_async_task_result(tasks=tasks, logger=self.lg)
self.lg.info('全部数据更新完毕'.center(100, '#'))
if get_shanghai_time().hour == 0: # 0点以后不更新
await async_sleep(60 * 60 * 5.5)
else:
await async_sleep(5.5)
try:
del self.jd
except:
pass
collect()
def __del__(self):
try:
del self.lg
except: pass
try:
del self.loop
except:pass
collect()
def get_goods_data(self, goods_id):
'''
模拟构造得到data的url
:param goods_id:
:return: data 类型dict
'''
self.msg = '------>>>| 对应的手机端地址为: ' + 'https://h5.m.taobao.com/awp/core/detail.htm?id=' + str(
goods_id)
self.lg.info(self.msg)
# 获取主接口的body
last_url = self._get_last_url(goods_id=goods_id)
data = Requests.get_url_body(url=last_url,
headers=self.headers,
params=None,
timeout=14,
high_conceal=True,
ip_pool_type=self.ip_pool_type)
if data == '':
self.lg.error('出错goods_id: {0}'.format((goods_id)))
return self._data_error_init()
try:
data = re.compile(r'mtopjsonp1\((.*)\)').findall(data)[
0] # 贪婪匹配匹配所有
# self.lg.info(str(data))
except IndexError:
self.lg.error('data为空! 出错goods_id: {0}'.format(goods_id))
return self._data_error_init()
data = json_2_dict(json_str=data, logger=self.lg)
if data == {}:
self.lg.error('出错goods_id: {0}'.format(str(goods_id)))
return self._data_error_init()
# pprint(data)
if data.get('data', {}).get('trade', {}).get('redirectUrl', '') != '' \
and data.get('data', {}).get('seller', {}).get('evaluates') is None:
'''
## 表示该商品已经下架, 原地址被重定向到新页面
'''
self.lg.info('@@@@@@ 该商品已经下架...')
_ = SqlServerMyPageInfoSaveItemPipeline()
if _.is_connect_success:
_._update_table_2(sql_str=tb_update_str_3,
params=(goods_id, ),
logger=self.lg)
try:
del _
except:
pass
tmp_data_s = self.init_pull_off_shelves_goods()
self.result_data = {}
return tmp_data_s
# 处理商品被转移或者下架导致页面不存在的商品
if data.get('data').get('seller', {}).get('evaluates') is None:
self.lg.info('data为空, 地址被重定向, 该商品可能已经被转移或下架')
return self._data_error_init()
data['data']['rate'] = '' # 这是宝贝评价
data['data']['resource'] = '' # 买家询问别人
data['data']['vertical'] = '' # 也是问和回答
data['data']['seller']['evaluates'] = '' # 宝贝描述, 卖家服务, 物流服务的评价值...
result_data = data['data']
# 处理result_data['apiStack'][0]['value']
# self.lg.info(result_data.get('apiStack', [])[0].get('value', ''))
result_data_apiStack_value = result_data.get('apiStack',
[])[0].get('value', {})
# 将处理后的result_data['apiStack'][0]['value']重新赋值给result_data['apiStack'][0]['value']
result_data['apiStack'][0][
'value'] = self._wash_result_data_apiStack_value(
goods_id=goods_id,
result_data_apiStack_value=result_data_apiStack_value)
# 处理mockData
mock_data = result_data['mockData']
mock_data = json_2_dict(json_str=mock_data, logger=self.lg)
if mock_data == {}:
self.lg.error('出错goods_id: {0}'.format(goods_id))
return self._data_error_init()
mock_data['feature'] = ''
# pprint(mock_data)
result_data['mockData'] = mock_data
# self.lg.info(str(result_data.get('apiStack', [])[0])) # 可能会有{'name': 'esi', 'value': ''}的情况
if result_data.get('apiStack', [])[0].get('value', '') == '':
self.lg.info(
"result_data.get('apiStack', [])[0].get('value', '')的值为空....")
result_data['trade'] = {}
return self._data_error_init()
else:
result_data['trade'] = result_data.get('apiStack', [])[0].get(
'value', {}).get('trade', {}) # 用于判断该商品是否已经下架的参数
# pprint(result_data['trade'])
self.result_data = result_data
# pprint(self.result_data)
return result_data
class GX8899Spider(Crawler):
def __init__(self, logger=None):
super(GX8899Spider, self).__init__(
ip_pool_type=IP_POOL_TYPE,
log_print=True,
logger=logger,
log_save_path=MY_SPIDER_LOGS_PATH + '/gx8899/_/',
is_use_driver=True,
driver_executable_path=PHANTOMJS_DRIVER_PATH
)
self._set_sort_type_name()
self.my_pipeline = SqlServerMyPageInfoSaveItemPipeline()
self.update_sql = 'update dbo.sina_weibo set head_img_url=%s, modify_time=%s where id=%s'
self.id_list = []
self.update_index = 0
def _set_sort_type_name(self):
'''
设置抓取的分类名
:return:
'''
self.sort_type_name_list = [
# 'weixin',
# 'nansheng',
# 'nvsheng',
'fengjing',
'jingxuan',
'wupin',
'oumei',
'weimei',
'heibai',
'baqi',
'xiaoqingxin',
'yijing',
'beiying',
'chouyan',
'sumiao',
'gexing',
'xiaohai',
'qiche',
'zhiwu',
'shouhui',
'weshen',
'mingxing',
'jianzhu',
'renwu',
]
def _get_gx8899_all_img_url(self):
self.lg.info('即将开始采集gx8899...')
fz = []
for sort_type_name in self.sort_type_name_list:
tmp = self._get_one_sort_type_name_page_info(sort_type_name)
if tmp != []:
fz += tmp
self.lg.info('@@@ 全部头像抓取完毕!')
self.fz = fz
return fz
def _get_new_wait_2_handle_id_list(self):
'''
获取新的带处理的
:return:
'''
sql_str = '''
select top 1000 id
from dbo.sina_weibo
where sina_type = 'bilibili' and modify_time is null
'''
if self.id_list == []:
self.lg.info('@@@ 重新获取id_list...')
self.my_pipeline = SqlServerMyPageInfoSaveItemPipeline()
try:
wait = self.my_pipeline._select_table(sql_str=sql_str)
self.id_list = [i[0] for i in wait]
except TypeError or IndexError:
sleep(8)
return []
else:
pass
return self.id_list
@fz_set_timeout(6)
def oo(self, id, img_url):
try:
self.my_pipeline._update_table_2(
sql_str=self.update_sql,
params=(img_url, get_shanghai_time(), id),
logger=self.lg
)
except Exception:
return False
return True
def _get_one_sort_type_name_page_info(self, sort_type_name):
'''
得到一个分类的某页信息
:return:
'''
base_url = 'http://m.gx8899.com/{0}/'.format(sort_type_name)
headers = {
'Connection': 'keep-alive',
'Upgrade-Insecure-Requests': '1',
'User-Agent': get_random_pc_ua(),
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8',
# 'Referer': 'http://m.gx8899.com/weixin/',
'Accept-Encoding': 'gzip, deflate',
'Accept-Language': 'zh-CN,zh;q=0.9',
}
index = 0
res = []
while True:
if index == 0:
url = base_url
index += 1 # 第二页index_2开始
else:
url = base_url + 'index_{0}.html'.format(index)
self.lg.info('正在抓取{0}'.format(url))
# 太慢, 改用phantomjs
# body = self._get_loop_run_result(url=url, headers=headers)
if index % 15 == 0:
try:
del self.driver
except:
pass
gc.collect()
self.driver = BaseDriver(executable_path=PHANTOMJS_DRIVER_PATH, logger=self.lg, ip_pool_type=self.ip_pool_type)
self.lg.info('[+] phantomjs已重置!')
body = self.driver.get_url_body(url=url)
# self.lg.info(str(body))
if re.compile(r'<title>404 - 找不到文件或目录。</title>').findall(body) != []:
break
need = Selector(text=body).css('div#con_tabone_1 li.last a:last-child ::attr(href)').extract()
pprint(need)
if need == []:
self.lg.error('获取到的need为空list!出错地址:{0}'.format(url))
continue
for article_url in need:
_ = self._get_one_article_page_info(article_url)
if _ != []:
res += _
self.lg.info('#### 已更新{0}个id !'.format(self.update_index))
index += 1
return res
def _get_one_article_page_info(self, url):
'''
得到一个推荐地址里面所有图片list
:param url:
:return:
'''
headers = {
'Connection': 'keep-alive',
'Cache-Control': 'max-age=0',
'Upgrade-Insecure-Requests': '1',
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/68.0.3440.106 Safari/537.36',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8',
'Accept-Encoding': 'gzip, deflate',
'Accept-Language': 'zh-CN,zh;q=0.9',
}
# body = self._get_loop_run_result(url=url, headers=headers)
body = self.driver.get_url_body(url=url)
if body == '':
self.lg.info('获取到img list为空list!出错地址:{}'.format(url))
return []
need = Selector(text=body).css('div.content p img ::attr(src)').extract()
# pprint(need)
# self.lg.info(str(need))
if need != []:
self.lg.info('[+] crawl子地址success')
else:
self.lg.info('[-] crawl子地址fail')
# 数据更新操作
for img_url in need:
try:
random_id_index = randint(0, len(self._get_new_wait_2_handle_id_list())-1)
except:
sleep(5)
continue
res = self.oo(
id=self.id_list[random_id_index],
img_url=img_url,
)
if res:
self.id_list.pop(random_id_index)
self.update_index += 1
return need
async def _get_one_page_body(self, url, headers):
'''
异步获取body
:param url:
:param headers:
:return:
'''
body = await AioHttp.aio_get_url_body(url=url, headers=headers, ip_pool_type=self.ip_pool_type)
return body
def _get_loop_run_result(self, **kwargs):
loop = get_event_loop()
result = loop.run_until_complete(self._get_one_page_body(
url=kwargs.get('url', ''),
headers=kwargs.get('headers', {})
))
return result
def __del__(self):
try:
del self.driver
del self.lg
except:
pass
gc.collect()
class YXGoodsInfoMonitorSpider(AsyncCrawler):
"""cp goods info monitor"""
def __init__(self):
AsyncCrawler.__init__(
self,
ip_pool_type=IP_POOL_TYPE,
log_print=True,
log_save_path=MY_SPIDER_LOGS_PATH + '/cp/yx_goods_monitor/',
)
self.req_num_retries = 5
self.concurrency = 100
self.concurrent_type = 1
self.sql_cli = SqlServerMyPageInfoSaveItemPipeline()
self.init_sql_str()
async def _fck_run(self):
while True:
self.db_res = await self.get_db_res()
await self.get_all_goods_info_and_handle_by_goods_id_list(
goods_id_list=[item[0] for item in self.db_res], )
await async_sleep(10.)
async def get_db_res(self) -> list:
"""
获取目标goods_id_list
:return:
"""
db_res = []
try:
db_res = list(self.sql_cli._select_table(sql_str=self.sql_tr0, ))
except Exception:
self.lg.error('遇到错误:', exc_info=True)
assert db_res != []
return db_res
async def get_all_goods_info_and_handle_by_goods_id_list(
self, goods_id_list: list):
"""
根据goods_id_list获取所有goods info并处理
:return:
"""
async def get_tasks_params_list() -> list:
tasks_params_list = []
for goods_id in goods_id_list:
tasks_params_list.append({'goods_id': goods_id})
return tasks_params_list
def get_create_task_msg(k) -> str:
return 'create task[where goods_id: {}] ...'.format(
k['goods_id'], )
def get_now_args(k) -> list:
return [
k['goods_id'],
]
assert goods_id_list != []
all_res = await get_or_handle_target_data_by_task_params_list(
loop=self.loop,
tasks_params_list=await get_tasks_params_list(),
func_name_where_get_create_task_msg=get_create_task_msg,
func_name=self.get_goods_info_by_goods_id,
func_name_where_get_now_args=get_now_args,
func_name_where_handle_one_res=self.handle_one_res,
func_name_where_add_one_res_2_all_res=
default_add_one_res_2_all_res2,
one_default_res={},
step=self.concurrency,
logger=self.lg,
concurrent_type=self.concurrent_type,
)
return all_res
def handle_one_res(self, one_res) -> None:
"""
处理单个结果
:return:
"""
# 每次重连
self.sql_cli = SqlServerMyPageInfoSaveItemPipeline()
if not self.sql_cli.is_connect_success:
return
# pprint(one_res)
for item in one_res:
try:
goods_id = item.get('goods_id', '')
assert goods_id != ''
# 会员价
tb_price0 = item.get('tb_price0', 0.)
assert tb_price0 != 0.
# 优点价
tb_price1 = item.get('tb_price1', 0.)
except AssertionError:
continue
for item in self.db_res:
db_goods_id = item[0]
db_tb_price = float(item[1]).__round__(2)
site_id = item[3]
modify_time = item[4]
goods_url = item[5]
if goods_id == db_goods_id:
if tb_price1 != db_tb_price:
# 先对比会员价
if tb_price0 != db_tb_price:
# 再对比优点价格, 两者都不同才进行修正, 否则pass(原因cp 部分会员价显示错误)
now_time = get_shanghai_time()
cp_url = 'https://m.yiuxiu.com/Product/Info/{}'.format(
goods_id)
self.lg.info(
'goods_id: {}, 优点价格: {}, 会员价格: {}, db_tb_price: {}, site_id: {}, modify_time: {}, cp_url: {}, goods_url: {}'
.format(
goods_id,
tb_price1,
tb_price0,
db_tb_price,
site_id,
modify_time,
cp_url,
goods_url,
))
self.sql_cli._update_table_2(
sql_str=self.sql_tr1,
params=(
now_time,
now_time,
now_time,
goods_id,
),
logger=self.lg,
)
else:
pass
break
else:
continue
return None
@catch_exceptions_with_class_logger(default_res={})
def get_goods_info_by_goods_id(self, goods_id: str) -> dict:
"""
根据goods_id获取商品信息
:param goods_id:
:return:
"""
def parse_body() -> dict:
"""
解析
:return:
"""
nonlocal body
# 多规格的最低价
# 会员价yx 部分显示错误, 改用big 价格加上优点, 两个一起用
tb_price_sel = {
'method': 'css',
'selector': 'div.goodsPriceTips span:nth-child(2) ::text',
}
big_price_sel = {
'method': 'css',
'selector': 'div.goodsPrice big ::text',
}
yd_sel = {
'method': 're',
'selector': '<span class=\"yiudianPrice\">\+(\d+)优点</span>'
}
tb_price0 = parse_field(
parser=tb_price_sel,
target_obj=body,
)
assert tb_price0 != ''
big_price = parse_field(
parser=big_price_sel,
target_obj=body,
)
assert big_price != ''
yd = parse_field(
parser=yd_sel,
target_obj=body,
)
assert yd != ''
# 会员价
tb_price0 = float(tb_price0).__round__(2)
# 优点价
tb_price1 = (float(big_price) + float(yd) / 100).__round__(2)
return {
'goods_id': goods_id,
'tb_price0': tb_price0,
'tb_price1': tb_price1,
}
headers = get_random_headers(
user_agent_type=1,
connection_status_keep_alive=False,
)
headers.update({
'authority': 'm.yiuxiu.com',
'sec-fetch-mode': 'navigate',
'sec-fetch-user': '******',
'sec-fetch-site': 'none',
})
url = 'https://m.yiuxiu.com/Product/Info/{}'.format(goods_id)
body = Requests.get_url_body(
url=url,
headers=headers,
ip_pool_type=self.ip_pool_type,
num_retries=self.req_num_retries,
proxy_type=PROXY_TYPE_HTTPS,
)
assert body != ''
# self.lg.info(body)
res = parse_body()
self.lg.info('[{}] goods_id: {}'.format(
'+' if res != {} else '-',
goods_id,
))
return res
def init_sql_str(self):
self.sql_tr0 = '''
select MainGoodsID, TaoBaoPrice, Price, SiteID, ModfiyTime, GoodsUrl
from dbo.GoodsInfoAutoGet
where MainGoodsID is not null
and IsDelete=0
'''
self.sql_tr1 = '''
update dbo.GoodsInfoAutoGet
set is_spec_change=1,
spec_trans_time=%s,
ModfiyTime=%s,
IsPriceChange=1,
sku_info_trans_time=%s,
PriceChangeInfo=SKUInfo
where MainGoodsID=%s
'''
def __del__(self):
try:
del self.lg
del self.loop
except:
pass
collect()
def get_ali_1688_data(self, goods_id):
if goods_id == '':
return self._data_error_init()
wait_to_deal_with_url = 'https://m.1688.com/offer/' + str(
goods_id) + '.html'
self.my_lg.info(
'------>>>| 待处理的阿里1688地址为: {0}'.format(wait_to_deal_with_url))
self.error_base_record = '出错goods_id:{0}'.format(goods_id)
body = self.my_phantomjs.use_phantomjs_to_get_url_body(
url=wait_to_deal_with_url, css_selector='div.d-content')
# self.my_lg.info(str(body))
if body == '':
self.my_lg.error('获取到的body为空str!请检查!' + self.error_base_record)
return self._data_error_init()
tmp_body = body
try:
pull_off_shelves = Selector(
text=body).css('div.d-content p.info::text').extract_first()
except:
pull_off_shelves = ''
if pull_off_shelves == '该商品无法查看或已下架': # 表示商品已下架, 同样执行插入数据操作
try:
tmp_my_pipeline = SqlServerMyPageInfoSaveItemPipeline()
is_in_db = tmp_my_pipeline._select_table(
sql_str=al_select_str_1, params=(str(goods_id), ))
# self.my_lg.info(str(is_in_db))
except Exception:
self.my_lg.error('数据库连接失败!' + self.error_base_record,
exc_info=True)
return self._data_error_init()
if is_in_db != []: # 表示该goods_id以前已被插入到db中, 于是只需要更改其is_delete的状态即可
tmp_my_pipeline._update_table_2(sql_str=al_update_str_1,
params=(goods_id),
logger=self.my_lg)
self.my_lg.info('@@@ 该商品goods_id原先存在于db中, 此处将其is_delete=1')
tmp_data_s = self.init_pull_off_shelves_goods() # 初始化下架商品的属性
tmp_data_s['before'] = True # 用来判断原先该goods是否在db中
self.result_data = {}
return tmp_data_s
else: # 表示该goods_id没存在于db中
self.my_lg.info('@@@ 该商品已下架[但未存在于db中], ** 此处将其插入到db中...')
tmp_data_s = self.init_pull_off_shelves_goods() # 初始化下架商品的属性
tmp_data_s['before'] = False
self.result_data = {}
return tmp_data_s
body = re.compile(r'{"beginAmount"(.*?)</script></div></div>').findall(
body)
if body != []:
body = body[0]
body = r'{"beginAmount"' + body
# self.my_lg.info(str(body))
body = json_2_dict(json_str=body)
# pprint(body)
if body.get('discountPriceRanges') is not None:
self.result_data = self._wash_discountPriceRanges(body=body)
return self.result_data
else:
self.my_lg.error('data为空!' + self.error_base_record)
return self._data_error_init()
else:
self.my_lg.info('解析ing..., 该商品正在参与火拼, 此处为火拼价, 为短期活动价格!')
body = re.compile(
r'{"activityId"(.*?)</script></div></div>').findall(tmp_body)
if body != []:
body = body[0]
body = r'{"activityId"' + body
# self.my_lg.info(str(body))
body = json_2_dict(json_str=body)
# pprint(body)
if body.get('discountPriceRanges') is not None:
self.result_data = self._wash_discountPriceRanges(
body=body)
self.is_activity_goods = True
return self.result_data
else:
self.my_lg.error('data为空!' + self.error_base_record)
return self._data_error_init()
else:
self.my_lg.error('这个商品对应活动属性未知, 此处不解析, 设置为跳过!' +
self.error_base_record)
return self._data_error_init()
def run_forever():
while True:
# ** 不能写成全局变量并放在循环中, 否则会一直记录到同一文件中
my_lg = set_logger(
log_file_name=MY_SPIDER_LOGS_PATH + '/网易严选/实时更新/' +
str(get_shanghai_time())[0:10] + '.txt',
console_log_level=INFO,
file_log_level=ERROR,
)
#### 实时更新数据
sql_cli = SqlServerMyPageInfoSaveItemPipeline()
try:
result = list(sql_cli._select_table(sql_str=yx_select_str_1))
except TypeError:
my_lg.error('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:
_block_print_db_old_data(result=result, logger=my_lg)
index = 1
# 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
yanxuan = YanXuanParse(logger=my_lg)
for item in result: # 实时更新数据
if index % 5 == 0:
try:
del yanxuan
except:
pass
yanxuan = YanXuanParse(logger=my_lg)
collect()
sql_cli = _block_get_new_db_conn(db_obj=sql_cli,
index=index,
logger=my_lg,
remainder=10)
if sql_cli.is_connect_success:
my_lg.info(
'------>>>| 正在更新的goods_id为(%s) | --------->>>@ 索引值为(%s)'
% (str(item[1]), str(index)))
yanxuan._get_goods_data(goods_id=item[1])
data = yanxuan._deal_with_data()
db_goods_info_obj = YXDbGoodsInfoObj(item=item,
logger=my_lg)
if data != {}:
if data.get('is_delete') == 1:
# 单独处理下架商品
my_lg.info('@@@ 该商品已下架...')
sql_cli._update_table_2(
sql_str=yx_update_str_2,
params=(db_goods_info_obj.goods_id, ),
logger=my_lg,
)
sleep(TMALL_REAL_TIMES_SLEEP_TIME)
continue
else:
data = get_goods_info_change_data(
target_short_name='yx',
logger=my_lg,
data=data,
db_goods_info_obj=db_goods_info_obj,
)
yanxuan.to_right_and_update_data(data,
pipeline=sql_cli)
else: # 表示返回的data值为空值
my_lg.info('------>>>| 休眠8s中...')
sleep(8)
else: # 表示返回的data值为空值
my_lg.error('数据库连接失败，数据库可能关闭或者维护中')
sleep(5)
pass
index += 1
collect()
sleep(TMALL_REAL_TIMES_SLEEP_TIME)
my_lg.info('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(60)
collect()
def get_goods_data(self, goods_id):
'''
得到data
:param goods_id:
:return: data 类型dict
'''
if goods_id == []:
self.result_data = {}
return {}
type = goods_id[0] # 天猫类型
# self.my_lg.info(str(type))
goods_id = goods_id[1] # 天猫goods_id
tmp_url = 'https://detail.m.tmall.com/item.htm?id=' + str(goods_id)
self.my_lg.info('------>>>| 得到的移动端地址为: %s' % tmp_url)
self.headers.update({'Referer': tmp_url})
last_url = self._get_last_url(goods_id=goods_id)
body = MyRequests.get_url_body(url=last_url, headers=self.headers, params=None, timeout=14)
if body == '':
self.my_lg.error('出错goods_id: {0}'.format((goods_id)))
self.result_data = {}
return {}
try:
assert body != '', '获取到的body为空值, 此处跳过! 出错type %s: , goods_id: %s' % (str(type), goods_id)
data = re.compile('mtopjsonp3\((.*)\)').findall(body)[0] # 贪婪匹配匹配所有
except (AssertionError, IndexError) as e:
self.my_lg.exception(e)
self.result_data = {}
return {}
if data != '':
data = json_2_dict(json_str=data, logger=self.my_lg)
if data == {}:
self.my_lg.error('出错type: %s, goods_id: %s' % (str(type), str(goods_id)))
self.result_data = {} # 重置下，避免存入时影响下面爬取的赋值
return {}
# pprint(data)
if data.get('data', {}).get('trade', {}).get('redirectUrl', '') != '' \
and data.get('data', {}).get('seller', {}).get('evaluates') is None:
'''
## 表示该商品已经下架, 原地址被重定向到新页面
'''
self.my_lg.info('@@@@@@ 该商品已经下架...')
_ = SqlServerMyPageInfoSaveItemPipeline()
if _.is_connect_success:
_._update_table_2(sql_str=tm_update_str_3, params=(goods_id,), logger=self.my_lg)
try:
del _
except:
pass
tmp_data_s = self.init_pull_off_shelves_goods(type)
self.result_data = {}
return tmp_data_s
# 处理商品被转移或者下架导致页面不存在的商品
if data.get('data', {}).get('seller', {}).get('evaluates') is None:
self.my_lg.error('data为空, 地址被重定向, 该商品可能已经被转移或下架, 出错type: %s, goods_id: %s' % (str(type), str(goods_id)))
self.result_data = {} # 重置下，避免存入时影响下面爬取的赋值
return {}
data['data']['rate'] = '' # 这是宝贝评价
data['data']['resource'] = '' # 买家询问别人
data['data']['vertical'] = '' # 也是问和回答
data['data']['seller']['evaluates'] = '' # 宝贝描述, 卖家服务, 物流服务的评价值...
result_data = data['data']
# 处理result_data['apiStack'][0]['value']
# self.my_lg.info(result_data.get('apiStack', [])[0].get('value', ''))
result_data_apiStack_value = result_data.get('apiStack', [])[0].get('value', {})
# 将处理后的result_data['apiStack'][0]['value']重新赋值给result_data['apiStack'][0]['value']
result_data['apiStack'][0]['value'] = self._wash_result_data_apiStack_value(
goods_id=goods_id,
result_data_apiStack_value=result_data_apiStack_value
)
# 处理mockData
mock_data = result_data['mockData']
mock_data = json_2_dict(json_str=mock_data, logger=self.my_lg)
if mock_data == {}:
self.my_lg.error('出错type: {0}, goods_id: {1}'.format(type, goods_id))
self.result_data = {} # 重置下，避免存入时影响下面爬取的赋值
return {}
mock_data['feature'] = ''
# pprint(mock_data)
result_data['mockData'] = mock_data
# self.my_lg.info(str(result_data.get('apiStack', [])[0])) # 可能会有{'name': 'esi', 'value': ''}的情况
if result_data.get('apiStack', [])[0].get('value', '') == '':
self.my_lg.error("result_data.get('apiStack', [])[0].get('value', '')的值为空....出错type: %s, goods_id: %s" % (str(type), goods_id))
result_data['trade'] = {}
self.result_data = {} # 重置下，避免存入时影响下面爬取的赋值
return {}
else:
result_data['trade'] = result_data.get('apiStack', [])[0].get('value', {}).get('trade', {}) # 用于判断该商品是否已经下架的参数
# pprint(result_data['trade'])
result_data['type'] = type
result_data['goods_id'] = goods_id
self.result_data = result_data
# pprint(self.result_data)
return result_data
else:
self.my_lg.error('data为空! 出错type: %s, goods_id: %s' % (str(type), str(goods_id)))
self.result_data = {} # 重置下，避免存入时影响下面爬取的赋值
return {}
class CommentRealTimeUpdateSpider(object):
def __init__(self):
self._set_logger()
self.msg = ''
self.debugging_api = self._init_debugging_api()
self._set_func_name_dict()
self.sql_str = cm_update_str_1
if self._init_debugging_api().get(2):
self.my_lg.info('初始化 1688 phantomjs中...')
self.ali_1688 = ALi1688CommentParse(logger=self.my_lg)
if self._init_debugging_api().get(3) is True \
or self._init_debugging_api().get(4) is True\
or self._init_debugging_api().get(6) is True:
self.my_lg.info('初始化 天猫 phantomjs中...')
self.tmall = TmallCommentParse(logger=self.my_lg)
if self._init_debugging_api().get(7) is True \
or self._init_debugging_api().get(8) is True\
or self._init_debugging_api().get(9) is True\
or self._init_debugging_api().get(10) is True:
self.my_lg.info('初始化 京东 phantomjs中...')
self.jd = JdCommentParse(logger=self.my_lg)
def _set_logger(self):
self.my_lg = set_logger(
log_file_name=MY_SPIDER_LOGS_PATH + '/all_comment/实时更新/' + str(get_shanghai_time())[0:10] + '.txt',
console_log_level=INFO,
file_log_level=ERROR
)
def _init_debugging_api(self):
'''
用于设置待抓取的商品的site_id
:return: dict
'''
return {
1: True,
2: True,
3: True,
4: True,
6: True,
7: True,
8: True,
9: True,
10: True,
11: False,
12: False,
13: False,
25: False,
}
def _set_func_name_dict(self):
self.func_name_dict = {
'taobao': 'self._update_taobao_comment({0}, {1}, {2})',
'ali': 'self._update_ali_1688_comment({0}, {1}, {2})',
'tmall': 'self._update_tmall_comment({0}, {1}, {2})',
'jd': 'self._update_jd_comment({0}, {1}, {2})',
'zhe_800': 'self._update_zhe_800_comment({0}, {1}, {2})',
'juanpi': 'self._update_juanpi_comment({0}, {1}, {2})',
'pinduoduo': 'self._update_pinduoduo_comment({0}, {1}, {2})',
'vip': 'self._update_vip_comment({0}, {1}, {2})',
}
def _just_run(self):
while True:
#### 更新数据
self._comment_pipeline = SqlServerMyPageInfoSaveItemPipeline()
# and GETDATE()-a.modify_time>1
try:
result = list(self._comment_pipeline._select_table(sql_str=cm_select_str_1, logger=self.my_lg))
except TypeError:
self.my_lg.error('TypeError错误, 原因数据库连接失败...(可能维护中)')
continue
self.my_lg.info('------>>> 下面是数据库返回的所有符合条件的goods_id <<<------')
self.my_lg.info(str(result))
self.my_lg.info('--------------------------------------------------------')
self.my_lg.info('待更新个数: {0}'.format(len(result)))
self.my_lg.info('即将开始实时更新数据, 请耐心等待...'.center(100, '#'))
# 1.淘宝 2.阿里 3.天猫 4.天猫超市 5.聚划算 6.天猫国际 7.京东 8.京东超市 9.京东全球购 10.京东大药房 11.折800 12.卷皮 13.拼多多 14.折800秒杀 15.卷皮秒杀 16.拼多多秒杀 25.唯品会
for index, item in enumerate(result): # item: ('xxxx':goods_id, 'y':site_id)
if not self.debugging_api.get(item[1]):
self.my_lg.info('api为False, 跳过! 索引值[%s]' % str(index))
continue
if index % 20 == 0:
try: del self._comment_pipeline
except: pass
self._comment_pipeline = SqlServerMyPageInfoSaveItemPipeline()
switch = {
1: self.func_name_dict.get('taobao'), # 淘宝
2: self.func_name_dict.get('ali'), # 阿里1688
3: self.func_name_dict.get('tmall'), # 天猫
4: self.func_name_dict.get('tmall'), # 天猫超市
6: self.func_name_dict.get('tmall'), # 天猫国际
7: self.func_name_dict.get('jd'), # 京东
8: self.func_name_dict.get('jd'), # 京东超市
9: self.func_name_dict.get('jd'), # 京东全球购
10: self.func_name_dict.get('jd'), # 京东大药房
11: self.func_name_dict.get('zhe_800'), # 折800
12: self.func_name_dict.get('juanpi'), # 卷皮
13: self.func_name_dict.get('pinduoduo'), # 拼多多
25: self.func_name_dict.get('vip'), # 唯品会
}
# 动态执行
exec_code = compile(switch[item[1]].format(index, item[0], item[1]), '', 'exec')
exec(exec_code)
sleep(1.1)
def _update_taobao_comment(self, index, goods_id, site_id):
'''
处理淘宝的商品comment
:param index: 索引
:param goods_id:
:param site_id:
:return:
'''
if self.debugging_api.get(site_id):
self.my_lg.info('------>>>| 淘宝\t\t索引值(%s)' % str(index))
taobao = TaoBaoCommentParse(logger=self.my_lg)
_r = taobao._get_comment_data(goods_id=str(goods_id))
if _r.get('_comment_list', []) != []:
if self._comment_pipeline.is_connect_success:
self._comment_pipeline._update_table_2(
sql_str=self.sql_str,
params=self._get_db_update_params(item=_r),
logger=self.my_lg)
else:
self.my_lg.info('该商品_comment_list为空list! 此处跳过!')
try:
del taobao
except:
self.my_lg.info('del taobao失败!')
gc.collect()
else:
pass
def _update_ali_1688_comment(self, index, goods_id, site_id):
'''
处理阿里1688的商品comment
:param index: 索引
:param goods_id:
:param site_id:
:return:
'''
if self.debugging_api.get(site_id):
self.my_lg.info('------>>>| 阿里1688\t\t索引值(%s)' % str(index))
if index % 5 == 0:
try: del self.ali_1688
except: self.my_lg.error('del ali_1688失败!')
gc.collect()
self.ali_1688 = ALi1688CommentParse(logger=self.my_lg)
_r = self.ali_1688._get_comment_data(goods_id=goods_id)
if _r.get('_comment_list', []) != []:
if self._comment_pipeline.is_connect_success:
self._comment_pipeline._update_table_2(
sql_str=self.sql_str,
params=self._get_db_update_params(item=_r),
logger=self.my_lg)
else:
self.my_lg.info('该商品_comment_list为空list! 此处跳过!')
else:
pass
def _update_tmall_comment(self, index, goods_id, site_id):
'''
处理tmall商品的comment
:param index:
:param goods_id:
:param site_id:
:return:
'''
if self.debugging_api.get(site_id):
self.my_lg.info('------>>>| 天猫\t\t索引值(%s)' % str(index))
if site_id == 3:
_type = 0
elif site_id == 4:
_type = 1
elif site_id == 6:
_type = 2
else:
return None
if index % 5 == 0:
try:
del self.tmall
except:
self.my_lg.info('del tmall失败!')
gc.collect()
self.tmall = TmallCommentParse(logger=self.my_lg)
_r = self.tmall._get_comment_data(type=_type, goods_id=str(goods_id))
if _r.get('_comment_list', []) != []:
if self._comment_pipeline.is_connect_success:
self._comment_pipeline._update_table_2(
sql_str=self.sql_str,
params=self._get_db_update_params(item=_r),
logger=self.my_lg)
else:
self.my_lg.info('该商品_comment_list为空list! 此处跳过!')
gc.collect()
else:
pass
def _update_jd_comment(self, index, goods_id, site_id):
'''
处理京东商品的comment
:param index:
:param goods_id:
:param site_id:
:return:
'''
if self.debugging_api.get(site_id):
self.my_lg.info('------>>>| 京东\t\t索引值(%s)' % str(index))
if index % 5 == 0:
try: del self.jd
except: self.my_lg.info('del jd失败!')
gc.collect()
self.jd = JdCommentParse(logger=self.my_lg)
_r = self.jd._get_comment_data(goods_id=str(goods_id))
if _r.get('_comment_list', []) != []:
if self._comment_pipeline.is_connect_success:
self._comment_pipeline._update_table_2(
sql_str=self.sql_str,
params=self._get_db_update_params(item=_r),
logger=self.my_lg)
else:
self.my_lg.info('该商品_comment_list为空list! 此处跳过!')
else:
pass
def _update_zhe_800_comment(self, index, goods_id, site_id):
'''
处理折800商品的comment
:param index:
:param goods_id:
:param site_id:
:return:
'''
if self.debugging_api.get(site_id):
pass
else:
pass
def _update_juanpi_comment(self, index, goods_id, site_id):
'''
处理卷皮商品的comment
:param index:
:param goods_id:
:param site_id:
:return:
'''
if self.debugging_api.get(site_id):
pass
else:
pass
def _update_pinduoduo_comment(self, index, goods_id, site_id):
'''
处理拼多多的comment
:param index:
:param goods_id:
:param site_id:
:return:
'''
if self.debugging_api.get(site_id):
pass
else:
pass
def _update_vip_comment(self, index, goods_id, site_id):
'''
处理唯品会的comment
:param index:
:param goods_id:
:param site_id:
:return:
'''
if self.debugging_api.get(site_id):
pass
else:
pass
def _get_db_update_params(self, item):
return (
item['modify_time'],
dumps(item['_comment_list'], ensure_ascii=False),
item['goods_id'],
)
def __del__(self):
try:
del self.my_lg
del self.msg
del self.debugging_api
except:
pass
try: del self._comment_pipeline
except: pass
try:
del self.tmall
except:
pass
gc.collect()
class MIUpdater(AsyncCrawler):
def __init__(self, *params, **kwargs):
AsyncCrawler.__init__(
self,
*params,
**kwargs,
log_print=True,
log_save_path=MY_SPIDER_LOGS_PATH + '/蜜芽/秒杀实时更新/',
ip_pool_type=IP_POOL_TYPE,
)
self.delete_sql_str = mia_delete_str_3
self.concurrency = 8 # 并发量
self.tmp_sql_server = None
self.goods_index = 1
async def _get_pc_headers(self) -> dict:
return {
'Accept':
'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
# 'Accept-Encoding:': 'gzip',
'Accept-Language': 'zh-CN,zh;q=0.8',
'Cache-Control': 'max-age=0',
'Connection': 'keep-alive',
'Host': 'm.mia.com',
'User-Agent': get_random_pc_ua(), # 随机一个请求头
}
async def _get_db_old_data(self):
self.tmp_sql_server = SqlServerMyPageInfoSaveItemPipeline()
result = None
try:
self.tmp_sql_server._delete_table(sql_str=mia_delete_str_4)
await async_sleep(5)
result = list(
self.tmp_sql_server._select_table(sql_str=mia_select_str_3))
except TypeError:
self.lg.error('TypeError错误, 原因数据库连接失败...(可能维护中)')
await _print_db_old_data(logger=self.lg, result=result)
return result
async def _get_miaosha_end_time(self, miaosha_time):
miaosha_end_time = json.loads(miaosha_time).get('miaosha_end_time')
miaosha_end_time = int(
str(
time.mktime(
time.strptime(miaosha_end_time,
'%Y-%m-%d %H:%M:%S')))[0:10])
return miaosha_end_time
async def _get_new_mia_obj(self, index):
if index % 10 == 0: # 不能共享一个对象了, 否则驱动访问会异常!
try:
del self.mia_miaosha
except:
pass
collect()
self.mia_miaosha = MiaParse()
async def _update_one_goods_info(self, item, index) -> tuple:
'''
单个更新
:param item:
:param index:
:return:
'''
res = False
goods_id = item[0]
miaosha_time = item[1]
pid = item[2]
miaosha_end_time = await self._get_miaosha_end_time(miaosha_time)
await self._get_new_mia_obj(index)
self.tmp_sql_server = await _get_new_db_conn(
db_obj=self.tmp_sql_server,
index=index,
logger=self.lg,
remainder=30)
if self.tmp_sql_server.is_connect_success:
is_recent_time = await self._is_recent_time(miaosha_end_time)
if is_recent_time == 0:
res = self.tmp_sql_server._update_table_2(
sql_str=mia_update_str_6,
params=(goods_id, ),
logger=self.lg)
self.lg.info('过期的goods_id为({}), 限时秒杀开始时间为({}), 删除成功!'.format(
goods_id,
json_2_dict(miaosha_time).get('miaosha_begin_time')))
await async_sleep(.5)
self.goods_index = index + 1
return goods_id, res
elif is_recent_time == 2:
self.goods_index = index + 1
return goods_id, res
else: # 返回1，表示在待更新区间内
self.lg.info(
'------>>>| 正在更新的goods_id为({0}) | --------->>>@ 索引值为({1})'.
format(goods_id, index))
tmp_url = 'https://m.mia.com/instant/seckill/seckillPromotionItem/' + str(
pid)
body = Requests.get_url_body(url=tmp_url,
headers=self.headers,
had_referer=True,
ip_pool_type=self.ip_pool_type)
# print(body)
body = '' if body == '' or body == '[]' else body
try:
tmp_data = json_2_dict(body, default_res={})
assert tmp_data != {}, 'tmp_data为空dict!'
except AssertionError:
self.lg.error('遇到错误:', exc_info=True)
self.goods_index = index + 1
await async_sleep(.3)
return goods_id, res
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
]
# self.lg.info(str(miaosha_goods_all_goods_id))
if goods_id not in miaosha_goods_all_goods_id: # 内部已经下架的
self.lg.info('该商品已被下架限时秒杀活动，此处将其删除')
res = self.tmp_sql_server._update_table_2(
sql_str=mia_update_str_6,
params=(goods_id, ),
logger=self.lg)
self.lg.info('下架的goods_id为({}), 删除成功!'.format(goods_id))
self.goods_index = index + 1
await async_sleep(.3)
return goods_id, res
else: # 未下架的
res = await self._one_update(item_list=item_list,
goods_id=goods_id,
tmp_data=tmp_data)
else: # 表示返回的data值为空值
self.lg.info('数据库连接失败，数据库可能关闭或者维护中')
await async_sleep(MIA_SPIKE_SLEEP_TIME) # 放慢速度
self.goods_index = index + 1
collect()
return goods_id, res
async def _update_db(self) -> None:
'''
秒杀实时更新
:return:
'''
while True:
self.lg = await self._get_new_logger(logger_name=get_uuid1())
result = await self._get_db_old_data()
if result is None:
pass
else:
self.goods_index = 1
tasks_params_list = TasksParamsListObj(
tasks_params_list=result, step=self.concurrency)
self.mia_miaosha = MiaParse()
index = 1
while True:
try:
slice_params_list = tasks_params_list.__next__()
# self.lg.info(str(slice_params_list))
except AssertionError: # 全部提取完毕, 正常退出
break
tasks = []
for item in slice_params_list:
self.lg.info('创建 task goods_id: {}'.format(item[0]))
tasks.append(
self.loop.create_task(
self._update_one_goods_info(item=item,
index=index)))
index += 1
await _get_async_task_result(tasks=tasks, logger=self.lg)
self.lg.info('全部数据更新完毕'.center(100, '#'))
if get_shanghai_time().hour == 0: # 0点以后不更新
await async_sleep(60 * 60 * 5.5)
else:
await async_sleep(2.5 * 60)
try:
del self.mia_miaosha
except:
pass
collect()
async def _one_update(self, **kwargs) -> bool:
'''
未下架的更新
:param kwargs:
:return:
'''
res = False
item_list = kwargs.get('item_list')
goods_id = kwargs.get('goods_id')
tmp_data = kwargs.get('tmp_data')
begin_time, end_time = await self._get_begin_time_and_end_time(tmp_data
)
for item_2 in item_list:
if item_2.get('item_id', '') == goods_id:
self.mia_miaosha.get_goods_data(goods_id=goods_id)
goods_data = self.mia_miaosha.deal_with_data()
if goods_data == {}: # 返回的data为空则跳过
pass
else:
goods_data['goods_id'] = str(goods_id)
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'] = get_miaosha_begin_time_and_miaosha_end_time(
miaosha_time=goods_data['miaosha_time'])
res = self.mia_miaosha.update_mia_xianshimiaosha_table(
data=goods_data, pipeline=self.tmp_sql_server)
break
else:
pass
return res
async def _get_begin_time_and_end_time(self, tmp_data) -> tuple:
begin_time = tmp_data.get('p_info', {}).get('start_time', '')
end_time = tmp_data.get('p_info', {}).get('end_time', '')
# 把str字符串类型转换为时间戳的形式
begin_time = int(
time.mktime(time.strptime(begin_time, '%Y/%m/%d %H:%M:%S')))
end_time = int(
time.mktime(time.strptime(end_time, '%Y/%m/%d %H:%M:%S')))
return begin_time, end_time
async def _is_recent_time(self, timestamp) -> int:
'''
判断是否在指定的日期差内
:param timestamp: 时间戳
:return: 0: 已过期恢复原价的 1: 待更新区间内的 2: 未来时间的
'''
time_1 = int(timestamp)
time_2 = datetime_to_timestamp(get_shanghai_time()) # 当前的时间戳
diff_time = time_1 - time_2
if diff_time < -86400: # (为了后台能同步下架)所以设置为 24个小时
# if diff_time < 0: # (原先的时间)结束时间 与当前时间差 <= 0
return 0 # 已过期恢复原价的
elif diff_time > 0:
return 1 # 表示是昨天跟今天的也就是待更新的
else: # 表示过期但是处于等待的数据不进行相关先删除操作(等<=24小时时再2删除)
return 2
def __del__(self):
try:
del self.lg
except:
pass
try:
del self.loop
except:
pass
try:
del self.mia_miaosha
except:
pass
collect()
def run_forever():
while True:
# ** 不能写成全局变量并放在循环中, 否则会一直记录到同一文件中
my_lg = set_logger(
log_file_name=MY_SPIDER_LOGS_PATH + '/网易严选/实时更新/' + str(get_shanghai_time())[0:10] + '.txt',
console_log_level=INFO,
file_log_level=ERROR
)
#### 实时更新数据
tmp_sql_server = SqlServerMyPageInfoSaveItemPipeline()
try:
result = list(tmp_sql_server._select_table(sql_str=yx_select_str_1))
except TypeError:
my_lg.error('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:
my_lg.info('------>>> 下面是数据库返回的所有符合条件的goods_id <<<------')
my_lg.info(str(result))
my_lg.info('--------------------------------------------------------')
my_lg.info('总计待更新个数: {0}'.format(len(result)))
my_lg.info('即将开始实时更新数据, 请耐心等待...'.center(100, '#'))
index = 1
# 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
yanxuan = YanXuanParse(logger=my_lg)
for item in result: # 实时更新数据
if index % 5 == 0:
try:
del yanxuan
except:
pass
yanxuan = YanXuanParse(logger=my_lg)
gc.collect()
if index % 10 == 0: # 每10次重连一次，避免单次长连无响应报错
my_lg.info('正在重置，并与数据库建立新连接中...')
tmp_sql_server = SqlServerMyPageInfoSaveItemPipeline()
my_lg.info('与数据库的新连接成功建立...')
if tmp_sql_server.is_connect_success:
my_lg.info('------>>>| 正在更新的goods_id为(%s) | --------->>>@ 索引值为(%s)' % (str(item[1]), str(index)))
yanxuan._get_goods_data(goods_id=item[1])
data = yanxuan._deal_with_data()
if data != {}:
data['goods_id'] = item[1]
data['shelf_time'], data['delete_time'] = get_shelf_time_and_delete_time(
tmp_data=data,
is_delete=item[2],
shelf_time=item[5],
delete_time=item[6])
if data.get('is_delete') == 1: # 单独处理下架商品
my_lg.info('@@@ 该商品已下架...')
tmp_sql_server._update_table_2(sql_str=yx_update_str_2, params=(item[1],), logger=my_lg)
sleep(TMALL_REAL_TIMES_SLEEP_TIME)
continue
else:
data['_is_price_change'], data['_price_change_info'] = _get_price_change_info(
old_price=item[3],
old_taobao_price=item[4],
new_price=data['price'],
new_taobao_price=data['taobao_price']
)
try:
old_sku_info = format_price_info_list(price_info_list=json_2_dict(item[7]), site_id=30)
except AttributeError: # 处理已被格式化过的
old_sku_info = item[7]
data['_is_price_change'], data['sku_info_trans_time'] = get_sku_info_trans_record(
old_sku_info=old_sku_info,
new_sku_info=format_price_info_list(data['price_info_list'], site_id=30),
is_price_change=item[8] if item[8] is not None else 0
)
yanxuan.to_right_and_update_data(data, pipeline=tmp_sql_server)
else: # 表示返回的data值为空值
my_lg.info('------>>>| 休眠8s中...')
sleep(8)
else: # 表示返回的data值为空值
my_lg.error('数据库连接失败，数据库可能关闭或者维护中')
sleep(5)
pass
index += 1
gc.collect()
sleep(TMALL_REAL_TIMES_SLEEP_TIME)
my_lg.info('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(60)
gc.collect()
async def get_coupon_url_list_by_goods_id_list(self,
slice_params_list) -> list:
"""
根据给与的goods_id_list来获取对应的coupon_url_list
:return:
"""
def get_create_task_msg(k) -> str:
return 'create task[where goods_id: {}, site_id: {}] ...'.format(
k['goods_id'],
k['site_id'],
)
def get_now_args(k) -> list:
return [
k['goods_id'],
]
all_res = await get_or_handle_target_data_by_task_params_list(
loop=self.loop,
tasks_params_list=slice_params_list,
func_name_where_get_create_task_msg=get_create_task_msg,
func_name=self.get_tm_coupon_url_from_lq5u,
func_name_where_get_now_args=get_now_args,
func_name_where_handle_one_res=None,
func_name_where_add_one_res_2_all_res=
default_add_one_res_2_all_res2,
one_default_res='',
step=self.concurrency,
logger=self.lg,
concurrent_type=self.concurrent_type,
func_timeout=25,
)
res = []
for item in all_res:
if item != '':
res.append(item)
# 修改对应的goods_id的coupon_check_time
sql_str = 'update dbo.GoodsInfoAutoGet set coupon_check_time=%s where GoodsID=%s'
sql_cli = SqlServerMyPageInfoSaveItemPipeline()
for item in slice_params_list:
goods_id = item['goods_id']
coupon_check_time_change_res = False
try:
coupon_check_time_change_res = sql_cli._update_table_2(
sql_str=sql_str,
params=(
get_shanghai_time(),
goods_id,
),
logger=self.lg,
)
except Exception:
self.lg.error('遇到错误:', exc_info=True)
self.lg.info('[{}] update goods_id: {} coupon_check_time'.format(
'+' if coupon_check_time_change_res else '-',
goods_id,
))
try:
del sql_cli
except:
pass
try:
del all_res
except:
pass
collect()
return res
class ZWMSpider(AsyncCrawler):
def __init__(self):
AsyncCrawler.__init__(
self,
ip_pool_type=IP_POOL_TYPE,
log_print=True,
log_save_path=MY_SPIDER_LOGS_PATH + '/zwm/_/',
)
self.init_zwm_pwd()
self.concurrency = 20
self.num_retries = 6
self.max_transaction_details_page_num = 20 # 交易截止抓取页
self.max_business_settlement_records_page_num = 20 # 商户结算记录截止抓取页
self.max_business_manage_page_num = 80 # 商户及门店管理截止抓取页(单数据也超过此数量就得进行修改)
self.login_cookies_dict = {}
self.sleep_time = 5
def init_zwm_pwd(self):
ori_data = ''
with open(ZWM_PWD_PATH, 'r') as f:
for line in f:
ori_data += line.replace('\n', '').replace(' ', '')
data = json_2_dict(
json_str=ori_data,
logger=self.lg,
default_res={},)
self.zwm_username, self.zwm_pwd = data['username'], data['pwd']
assert self.zwm_username != '' and self.zwm_pwd != ''
async def _fck_run(self) -> None:
while True:
try:
login_res = await self._login()
assert login_res is True, '登录失败, 退出后续同步操作!'
# 获取所有交易明细(自己有接口, 不需要了)
# all_transaction_details = await self._get_all_transaction_details()
# pprint(all_transaction_details)
# self.lg.info('len_all_transaction_details: {}'.format(len(all_transaction_details)))
# await self._wash_and_save_all_transaction_details(target_list=all_transaction_details)
# 获取所有商户结算记录
self.lg.info('获取所有商户结算记录...')
all_business_settlement_records = await self._get_all_business_settlement_records_by_something()
# pprint(all_business_settlement_records)
self.lg.info('len_now_business_settlement_records: {}'.format(len(all_business_settlement_records)))
await self._wash_save_all_business_settlement_records(target_list=all_business_settlement_records)
self.lg.info('\n')
# 获取所有商户及门店管理记录
self.lg.info('获取所有商户及门店管理记录 ...')
all_business_manage_records = await self._get_all_business_manage_records_by_something()
# pprint(all_business_manage_records)
self.lg.info('len_all_business_manage_records: {}'.format(len(all_business_manage_records)))
await self._wash_save_all_business_manage_records(target_list=all_business_manage_records)
self.lg.info('\n')
except Exception:
self.lg.error('遇到错误:', exc_info=True)
self.lg.info('## 同步完成 ##')
self.lg.info('休眠 {} minutes ...'.format(self.sleep_time))
# 定时
await async_sleep(60 * self.sleep_time)
async def _login(self) -> bool:
"""
登录
:return:
"""
headers = await self._get_random_pc_headers()
headers.update({
'Referer': 'https://agent.yrmpay.com/JHAdminConsole/loginNew.jsp',
})
file_load = {
'loginName': self.zwm_username,
'userPassword': self.zwm_pwd,
}
m = MultipartEncoder(fields=file_load)
# self.lg.info(m)
headers.update({
'Content-Type': m.content_type
})
login_url = 'https://agent.yrmpay.com/JHAdminConsole/foreigncard/permissionsLogin.do'
with session() as _session:
try:
response = _session.post(
url=login_url,
headers=headers,
data=m,
proxies=self._get_proxies(),)
login_res = json_2_dict(
json_str=response.text,
default_res={},
logger=self.lg, ).get('message', '')
assert login_res == '登录成功', '登录失败!'
self.lg.info(login_res)
self.login_cookies_dict = response.cookies.get_dict()
assert self.login_cookies_dict != {}, 'self.login_cookies_dict != 空dict!'
# pprint(self.login_cookies_dict)
except Exception:
self.lg.error('遇到错误:', exc_info=True)
return False
return True
async def _wash_save_all_business_manage_records(self, target_list: list):
"""
清洗并存储所有未存储的 or 更新所有已存储的business manage records
:param target_list:
:return:
"""
all_res = []
for item in target_list:
try:
now_time = get_shanghai_time()
create_time, modify_time, approval_status_change_time = now_time, now_time, now_time
agent_name = item['agentName']
top_agent_name = item['topAgentName']
shop_type = item['merType']
is_high_quality_shop = item['isHighQualityMer']
if is_high_quality_shop == '否':
is_high_quality_shop = 0
elif is_high_quality_shop == '是':
is_high_quality_shop = 1
else:
raise ValueError('is_high_quality_shop value: {} 异常!'.format(is_high_quality_shop))
shop_id = item.get('jhmid', '')
assert shop_id != ''
shop_chat_name = item.get('merchantName', '')
assert shop_chat_name != ''
phone_num = item.get('phone', '')
assert phone_num != ''
shop_chant_num = int(item['merchantNum'])
sale = item['sale']
is_real_time = 0 if item['isRealTime'] == '未开通' else 1
approve_date = date_parse(item['approveDate'])
rate = Decimal(item['rate']).__round__(4)
account_type = item['accType']
apply_time = date_parse(item['applyTime'])
# 可为空值
process_context = item.get('processContext', '')
is_non_contact = 0 if item['isNonContact'] == '未开通' else 1
approval_status = item['approvalStatus']
if approval_status == '待审核':
approval_status = 1
elif approval_status == '审核通过':
approval_status = 0
elif approval_status == '退回':
approval_status = 2
else:
raise ValueError('approval_status value: {} 异常'.format(approval_status))
# 用其原值为定值不变, 且唯一
unique_id = item['id']
except Exception:
self.lg.error('遇到错误:', exc_info=True)
continue
zwm_item = ZWMBusinessManageRecordItem()
zwm_item['unique_id'] = unique_id
zwm_item['create_time'] = create_time
zwm_item['modify_time'] = modify_time
zwm_item['agent_name'] = agent_name
zwm_item['top_agent_name'] = top_agent_name
zwm_item['shop_type'] = shop_type
zwm_item['is_high_quality_shop'] = is_high_quality_shop
zwm_item['shop_id'] = shop_id
zwm_item['shop_chat_name'] = shop_chat_name
zwm_item['phone_num'] = phone_num
zwm_item['shop_chant_num'] = shop_chant_num
zwm_item['sale'] = sale
zwm_item['is_real_time'] = is_real_time
zwm_item['approve_date'] = approve_date
zwm_item['rate'] = rate
zwm_item['account_type'] = account_type
zwm_item['apply_time'] = apply_time
zwm_item['process_context'] = process_context
zwm_item['is_non_contact'] = is_non_contact
zwm_item['approval_status'] = approval_status
zwm_item['approval_status_change_time'] = approval_status_change_time
all_res.append(dict(zwm_item))
# 查看
# if shop_id == 'YRMPAY100038574':
# if phone_num == '18192242001':
# if shop_chat_name == '哇哇叫':
# pprint(dict(zwm_item))
# pprint(all_res)
await self._insert_or_update_shop_manage_records_table(all_res=all_res)
try:
del all_res
except:
pass
return None
async def _insert_or_update_shop_manage_records_table(self, all_res: list):
"""
插入or update原数据
:param all_res:
:return:
"""
self.sql_cli = SqlServerMyPageInfoSaveItemPipeline()
try:
db_data = self.sql_cli._select_table(
sql_str=zwm_select_str_2,
params=None,
logger=self.lg, )
# pprint(db_data)
db_unique_id_list = [item[0] for item in db_data]
assert db_unique_id_list != [], 'db_unique_id_list != []'
self.lg.info('len_db_unique_id_list: {}'.format(len(db_unique_id_list)))
except Exception:
self.sql_cli = SqlServerMyPageInfoSaveItemPipeline()
self.lg.error('遇到错误:', exc_info=True)
return None
new_add_count = 0
for item in all_res:
unique_id = item['unique_id']
if unique_id not in db_unique_id_list:
# 插入
self.lg.info('inserting unique_id: {} ...'.format(unique_id))
params = await self._get_insert_item_params2(item=item)
try:
res = self.sql_cli._insert_into_table_2(
sql_str=zwm_insert_str_2,
params=params,
logger=self.lg)
if res:
new_add_count += 1
except Exception:
self.lg.error('遇到错误:', exc_info=True)
continue
else:
db_old_approval_status, db_old_approval_status_change_time = await self._get_dd_old_approval_status_and_approval_status_change_time(
db_data=db_data,
unique_id=unique_id,)
item['approval_status_change_time'] = await self._get_new_approval_status_change_time(
db_old_approval_status=db_old_approval_status,
db_old_approval_status_change_time=db_old_approval_status_change_time,
new_approval_status=item['approval_status'],
new_approval_status_change_time=item['approval_status_change_time'])
# 更新
self.lg.info('updating unique_id: {} ...'.format(unique_id))
params = await self._get_update_item_params(item=item)
try:
res = self.sql_cli._update_table_2(
sql_str=zwm_update_str_1,
params=params,
logger=self.lg)
except Exception:
self.lg.error('遇到错误:', exc_info=True)
continue
if not self.sql_cli.is_connect_success:
self.sql_cli = SqlServerMyPageInfoSaveItemPipeline()
else:
pass
try:
del db_data
del db_unique_id_list
except:
pass
self.lg.info('table.zwm_buss_manage_records新增个数: {}'.format(new_add_count))
async def _get_new_approval_status_change_time(self,
db_old_approval_status,
db_old_approval_status_change_time,
new_approval_status,
new_approval_status_change_time):
"""
获取新的approval_status_change_time
:return:
"""
if db_old_approval_status_change_time is not None:
new_approval_status_change_time = db_old_approval_status_change_time \
if db_old_approval_status == new_approval_status \
else get_shanghai_time()
else:
pass
return new_approval_status_change_time
async def _get_dd_old_approval_status_and_approval_status_change_time(self, db_data: list, unique_id: str) -> tuple:
"""
获取db 原先的approval_status
:param db_data:
:param unique_id:
:return:
"""
for item in db_data:
if unique_id == item[0]:
return item[1], item[2]
else:
continue
async def _get_all_business_manage_records_by_something(self,):
"""
获取所有商户及门店管理记录
:return:
"""
async def get_tasks_params_list(max_business_manage_page_num) -> list:
"""获取tasks_params_list"""
tasks_params_list = []
for page_num in range(1, max_business_manage_page_num):
tasks_params_list.append({
'page_num': page_num,
})
return tasks_params_list
def get_create_task_msg(k) -> str:
return 'create task[where page_num: {}]...'.format(k['page_num'])
def get_now_args(k) -> list:
return [
k['page_num'],
]
res = await get_or_handle_target_data_by_task_params_list(
loop=self.loop,
tasks_params_list=await get_tasks_params_list(
max_business_manage_page_num=self.max_business_manage_page_num),
func_name_where_get_create_task_msg=get_create_task_msg,
func_name=self._get_one_page_business_manage_records_by_something,
func_name_where_get_now_args=get_now_args,
func_name_where_handle_one_res=None,
func_name_where_add_one_res_2_all_res=default_add_one_res_2_all_res,
one_default_res=[],
step=self.concurrency,
logger=self.lg,
get_all_res=True,)
return res
@catch_exceptions_with_class_logger(default_res=[])
def _get_one_page_business_manage_records_by_something(self,
page_num: int,
start_date: str = None,
end_date: str = None,):
"""
获取单页商户及门店管理记录
:param page_num:
:param start_date: 默认设置前一个月27号, eg: '2019-01-27 00:00'
:param end_date: eg: '2019-07-20 09:39'
:return:
"""
# todo 获取最开始->至今的, 即采集所有, 避免老店铺的审核状态变动, 而后台无法同步状态, 审核时间
# start_date = str(self.get_1_on_the_month() if start_date is None else start_date).split(' ')[0] + ' 00:00'
start_date = '2018-01-01 00:00'
end_date = (str(get_shanghai_time()) if end_date is None else end_date)[0:16]
self.lg.info('start_date: {}, end_date: {}'.format(start_date, end_date))
headers = self.get_random_pc_headers()
headers.update({
'Accept': '*/*',
'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
'Referer': 'https://agent.yrmpay.com/JHAdminConsole/merchantMaterial/page.do',
'X-Requested-With': 'XMLHttpRequest',
})
params = (
('_dc', get_now_13_bit_timestamp()),
)
data = {
'merchantCode': '',
'accType': '',
'phone': '',
'approveDate': '',
'merchantName': '',
'processStatus': '',
'startTime': start_date,
'endTime': end_date,
'agentName': '',
'page': str(page_num),
'start': str((page_num - 1) * 100), # 开始位置0, 100, 200
'limit': '100',
}
url = 'https://agent.yrmpay.com/JHAdminConsole/merchantMaterial/materialList.do'
body = Requests.get_url_body(
method='post',
url=url,
headers=headers,
params=params,
cookies=self.login_cookies_dict,
data=data,
ip_pool_type=self.ip_pool_type,
num_retries=self.num_retries,)
assert body != '', 'body不为空值!'
res = json_2_dict(
json_str=body,
logger=self.lg,
default_res={}).get('materialList', [])
self.lg.info('[{}] page_num: {}'.format(
'+' if res != [] else '-',
page_num,
))
return res
async def _wash_save_all_business_settlement_records(self, target_list):
"""
清洗并存储 未被存储的所有商户结算记录
:param target_list:
:return:
"""
self.sql_cli = SqlServerMyPageInfoSaveItemPipeline()
try:
db_data = self.sql_cli._select_table(
sql_str=zwm_select_str_1,
params=None,
logger=self.lg,)
# pprint(db_data)
db_unique_id_list = [item[0] for item in db_data]
assert db_unique_id_list != [], 'db_unique_id_list != []'
self.lg.info('len_db_unique_id_list: {}'.format(len(db_unique_id_list)))
except Exception:
self.sql_cli = SqlServerMyPageInfoSaveItemPipeline()
self.lg.error('遇到错误:', exc_info=True)
return None
all_res = []
for item in target_list:
# pprint(item)
try:
create_time = get_shanghai_time()
shop_name = item.get('merName', '')
assert shop_name != ''
shop_id = item.get('mid', '')
assert shop_id != ''
agent_name = item['agentName']
top_agent_name = item['topAgentName']
date_settle_type = item['settleType']
trans_amount = item.get('transAmt', '')
assert trans_amount != ''
trans_amount = Decimal(trans_amount).__round__(2)
service_charge = Decimal(item['mda']).__round__(2)
accounting_amount = Decimal(item['mnamt']).__round__(2)
# 正常情况为: '20190704', 异常为'20190824-20190824'
txn_day = item['txnDay']
if re.compile('-').findall(txn_day) != []:
txn_day = txn_day.split('-')[0]
else:
pass
trans_date = date_parse(txn_day)
trans_status = item['status']
if trans_status == '已结算':
trans_status = 0
else:
raise ValueError('trans_status: {}, 未知交易状态!'.format(trans_status))
settle_type = item['type']
settle_date = date_parse(item['minDay'])
# 生成唯一标识码
unique_id = get_uuid3(
target_str=shop_id + str(date_settle_type) + str(trans_amount) + \
str(service_charge) + str(trans_date) + \
str(settle_type) + str(settle_date),)
except Exception:
self.lg.error('遇到错误:', exc_info=True)
continue
if unique_id in db_unique_id_list:
# self.lg.info('该record[unique_id: {}]已存在!'.format(unique_id))
continue
settle_record_item = ZWMBusinessSettlementRecordItem()
settle_record_item['unique_id'] = unique_id
settle_record_item['create_time'] = create_time
settle_record_item['shop_name'] = shop_name
settle_record_item['shop_id'] = shop_id
settle_record_item['agent_name'] = agent_name
settle_record_item['top_agent_name'] = top_agent_name
settle_record_item['date_settle_type'] = date_settle_type
settle_record_item['trans_amount'] = trans_amount
settle_record_item['service_charge'] = service_charge
settle_record_item['accounting_amount'] = accounting_amount
settle_record_item['trans_date'] = trans_date
settle_record_item['trans_status'] = trans_status
settle_record_item['settle_type'] = settle_type
settle_record_item['settle_date'] = settle_date
all_res.append(dict(settle_record_item))
# pprint(all_res)
self.lg.info('未存储个数: {}'.format(len(all_res)))
await self._save_all_business_settlement_records(all_res=all_res)
try:
del all_res
except:
pass
return None
async def _save_all_business_settlement_records(self, all_res) -> None:
"""
存储新增的商家提现记录
:param all_res:
:return:
"""
new_add_count = 0
for item in all_res:
# 处理未存储的新数据
unique_id = item['unique_id']
self.lg.info('saving unique_id: {} ...'.format(unique_id))
params = await self._get_insert_item_params(item=item)
try:
res = self.sql_cli._insert_into_table_2(
sql_str=zwm_insert_str_1,
params=params,
logger=self.lg)
if res:
new_add_count += 1
except Exception:
self.lg.error('遇到错误:', exc_info=True)
continue
if not self.sql_cli.is_connect_success:
self.sql_cli = SqlServerMyPageInfoSaveItemPipeline()
else:
pass
self.lg.info('新增个数: {}'.format(new_add_count))
return None
async def _get_insert_item_params(self, item) -> tuple:
"""
待插入对象
:param item:
:return:
"""
return tuple([
item['unique_id'],
item['create_time'],
item['shop_name'],
item['shop_id'],
item['agent_name'],
item['top_agent_name'],
item['date_settle_type'],
item['trans_amount'],
item['service_charge'],
item['accounting_amount'],
item['trans_date'],
item['trans_status'],
item['settle_type'],
item['settle_date'],
])
async def _get_insert_item_params2(self, item) -> tuple:
"""
待插入对象, zwm_buss_manage_records table
:param item:
:return:
"""
return tuple([
item['unique_id'],
item['create_time'],
item['modify_time'],
item['agent_name'],
item['top_agent_name'],
item['shop_type'],
item['is_high_quality_shop'],
item['shop_id'],
item['shop_chat_name'],
item['phone_num'],
item['shop_chant_num'],
item['sale'],
item['is_real_time'],
item['approve_date'],
item['rate'],
item['account_type'],
item['apply_time'],
item['process_context'],
item['is_non_contact'],
item['approval_status'],
item['approval_status_change_time'],
])
async def _get_update_item_params(self, item: dict) -> tuple:
"""
更新对象, zwm_buss_manage_records table
:param item:
:return:
"""
return tuple([
item['modify_time'],
item['agent_name'],
item['top_agent_name'],
item['shop_type'],
item['is_high_quality_shop'],
item['shop_id'],
item['shop_chat_name'],
item['phone_num'],
item['shop_chant_num'],
item['sale'],
item['is_real_time'],
item['approve_date'],
item['rate'],
item['account_type'],
item['apply_time'],
item['process_context'],
item['is_non_contact'],
item['approval_status'],
item['approval_status_change_time'],
item['unique_id'],
])
async def _wash_and_save_all_transaction_details(self, target_list: list):
"""
清洗并存储所有交易明细
:param target_list:
:return:
"""
pass
async def _get_all_business_settlement_records_by_something(self):
"""
获取所有商户结算记录
:return:
"""
async def get_tasks_params_list(max_business_settlement_records_page_num) -> list:
"""获取tasks_params_list"""
tasks_params_list = []
for page_num in range(1, max_business_settlement_records_page_num):
tasks_params_list.append({
'page_num': page_num,
})
return tasks_params_list
def get_create_task_msg(k) -> str:
return 'create task[where page_num: {}]...'.format(k['page_num'])
def get_now_args(k) -> list:
return [
k['page_num'],
]
res = await get_or_handle_target_data_by_task_params_list(
loop=self.loop,
tasks_params_list=await get_tasks_params_list(
max_business_settlement_records_page_num=self.max_business_settlement_records_page_num),
func_name_where_get_create_task_msg=get_create_task_msg,
func_name=self._get_one_page_business_settlement_records_by_something,
func_name_where_get_now_args=get_now_args,
func_name_where_add_one_res_2_all_res=default_add_one_res_2_all_res,
one_default_res=[],
step=self.concurrency,
logger=self.lg,
get_all_res=True,)
return res
@catch_exceptions_with_class_logger(default_res=[])
def _get_one_page_business_settlement_records_by_something(self,
page_num :int,
start_date: str=None,
end_date: str=None,
mid: str='',
agent_name: str='') -> list:
"""
得到单页商户结算记录
:param page_num:
:param start_date: 默认设置前一个月27号, eg: '2019-07-01'
:param end_date: eg: '2019-07-16'
:param mid: 商户编号
:param agent_name: 顶级机构名称
:return:
"""
start_date = str(self.get_1_on_the_month() if start_date is None else start_date).split(' ')[0]
# start_date = '2018-01-01'
end_date = (str(get_shanghai_time()) if end_date is None else end_date).split(' ')[0]
self.lg.info('start_date: {}, end_date: {}'.format(start_date, end_date))
headers = self.get_random_pc_headers()
headers.update({
'Accept': '*/*',
'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
'Referer': 'https://agent.yrmpay.com/JHAdminConsole/merSettle/querySettleJsp.do',
'X-Requested-With': 'XMLHttpRequest',
})
params = (
('_dc', get_now_13_bit_timestamp()),
)
data = {
'startDate': start_date,
'endDate': end_date,
'mid': mid,
'agentName': agent_name,
'loginAgentId': self.zwm_username[0:8], # 前8位
'page': str(page_num),
'start': str((page_num - 1) * 100), # 开始位置, 0, 100, 200
'limit': '100',
}
url = 'https://agent.yrmpay.com/JHAdminConsole/merSettle/queryMerSettleList.do'
body = Requests.get_url_body(
method='post',
url=url,
headers=headers,
params=params,
cookies=self.login_cookies_dict,
data=data,
ip_pool_type=self.ip_pool_type,
num_retries=self.num_retries,)
# self.lg.info(body)
assert body != '', 'body不为空值!'
res = json_2_dict(
json_str=body,
logger=self.lg,
default_res={}).get('data', [])
self.lg.info('[{}] page_num: {}'.format(
'+' if res != [] else '-',
page_num,
))
return res
async def _get_all_transaction_details(self) -> list:
"""
获取所有交易流水
:return:
"""
async def _get_tasks_params_list() -> list:
"""获取tasks_params_list"""
tasks_params_list = []
for page_num in range(1, self.max_transaction_details_page_num):
tasks_params_list.append({
'page_num': page_num,
})
return tasks_params_list
tasks_params_list = await _get_tasks_params_list()
tasks_params_list_obj = TasksParamsListObj(
tasks_params_list=tasks_params_list,
step=self.concurrency,)
all_res = []
while True:
try:
slice_params_list = tasks_params_list_obj.__next__()
except AssertionError:
break
tasks = []
for k in slice_params_list:
page_num = k['page_num']
self.lg.info('create task[where page_num: {}]...'.format(page_num))
func_args = [
page_num,
]
tasks.append(self.loop.create_task(
unblock_func(
func_name=self._get_one_page_transaction_details_by_something,
func_args=func_args,
logger=self.lg,)))
one_res = await async_wait_tasks_finished(tasks=tasks)
try:
del tasks
except:
pass
for i in one_res:
for j in i:
all_res.append(j)
return all_res
@catch_exceptions_with_class_logger(default_res=[])
def _get_one_page_transaction_details_by_something(self,
page_num: int,
start_date: str=None,
end_date: str=None,
transaction_status: str='',
mer_name: str='',
order_no: str='',
mid: str='',
agent_name: str='',
pay_channel: str ='',
sale_name: str='',) -> list:
"""
获取单页交易流水
:param page_num: 开始页面, eg: 1, 2, 3
:param start_date: eg: '2019-07-16 00:00'
:param end_data: eg: '2019-07-16 10:02'
:param transaction_status: 交易状态 | 选择全部: '' or 交易成功: '1' or 退款成功: '3'
:param mer_name: 待查询的商户名称
:param order_no: 订单号
:param mid: 商户编号
:param agent_name: 顶级机构名称
:param pay_channel: 支付渠道 | 请选择: '' or 微信: '50' or 支付宝: '51' or 微信条码: '55' or 支付宝条码: '56' or 微信小程序: '67'
:param sale_name: 销售名称
:return:
"""
res = []
start_date = self.get_0_00_on_the_day() if start_date is None else start_date
end_date = str(get_shanghai_time()) if end_date is None else end_date
headers = self.get_random_pc_headers()
headers.update({
'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
'Accept': '*/*',
'Referer': 'https://agent.yrmpay.com/JHAdminConsole/limafuReport/transflow.do',
'X-Requested-With': 'XMLHttpRequest',
})
params = (
('_dc', get_now_13_bit_timestamp()),
)
data = {
'startDate': start_date,
'endDate': end_date,
'type': '2',
'status': transaction_status,
'payChannel': pay_channel,
'orderNo': order_no,
'merName': mer_name,
'mid': mid,
'agentName': agent_name,
'saleName': sale_name,
'page': str(page_num),
'start': str((page_num - 1) * 20), # 开始位置, 0, 20, 40
'limit': '20',
}
url = 'https://agent.yrmpay.com/JHAdminConsole/limafuReport/querylimafuTransFlow.do'
body = Requests.get_url_body(
method='post',
url=url,
headers=headers,
params=params,
cookies=self.login_cookies_dict,
data=data,
ip_pool_type=self.ip_pool_type,
num_retries=self.num_retries,)
assert body != '', 'body不为空值!'
res = json_2_dict(
json_str=body,
logger=self.lg,
default_res={}).get('data', [])
self.lg.info('[{}] page_num: {}'.format(
'+' if res != [] else '-',
page_num,
))
return res
def get_0_00_on_the_day(self) -> str:
"""
获取当天的0点
:return:
"""
now_time = get_shanghai_time()
return str(datetime(
year=now_time.year,
month=now_time.month,
day=now_time.day))
def get_1_on_the_month(self) -> str:
"""
获取当月的第一天
:return:
"""
now_time = get_shanghai_time()
# 避免月底流水无法获取
day = 5
now_month = now_time.month
if now_month > 1:
now_month -= 1
else:
# now_month为1月份
now_month = 12
return str(datetime(
year=now_time.year,
month=now_month,
day=day,))
def _get_proxies(self) -> dict:
"""
获取代理
:return:
"""
proxies = Requests._get_proxies(ip_pool_type=self.ip_pool_type, )
assert proxies != {}, 'proxies不为空dict!'
return proxies
async def _get_random_pc_headers(self) -> dict:
"""
:return:
"""
return self.get_random_pc_headers()
@staticmethod
def get_random_pc_headers() -> dict:
headers = get_random_headers(
upgrade_insecure_requests=False,
cache_control='',)
headers.update({
'Origin': 'https://agent.yrmpay.com',
'accept-language': 'zh-CN,zh;q=0.9,en;q=0.8',
# 'Content-Type': 'multipart/form-data; boundary=----WebKitFormBoundarytSJCAoaErjNY4IbM',
'accept': 'text/plain, */*; q=0.01',
'X-Requested-With': 'XMLHttpRequest',
})
return headers
def __del__(self):
try:
del self.lg
del self.login_cookies_dict
except:
pass
try:
del self.loop
except:
pass
collect()
