def sync_data2mongodb(logger=None):
logger.info(Colored.green("[Data2MongoDB]: 正在从 redis 同步数据到 mongodb----"))
items = rdb.spop(STORE_TOPIC, count=500)
index = 0
wait_publish_ids = []
for item in items:
try:
json_dict = json.loads(item)
collection = json_dict.get('data_table', "").split(AT)[1]
del json_dict['data_table']
_id = json_dict['_id']
if _id == '1d33c55f57dd60237f2657028814b638':
continue
if insert_data(collection, json_dict):
logger.info(
Colored.blue(
"[Data2MongoDB]: db:{1} >> content 的 id==>{0}".format(
json_dict.get('_id'), collection)))
wait_publish_ids.append(
(PUBLISHED.format(collection), json_dict['_id']))
index += 1
except Exception as e:
rdb.sadd(STORE_TOPIC, item)
logger.error(Colored.red("[Data2MongoDB]: 发生错误>>>{0},".format(e)),
exc_info=True)
try:
pipeline = rdb_publish.pipeline(transaction=False)
for key, value in wait_publish_ids:
pipeline.sadd(key, value)
pipeline.execute()
except Exception as e:
logger.error(Colored.red(
"[Data2MongoDB]: 推送到 redis 发生错误>>>{0},".format(e)),
exc_info=True)
logger.info(Colored.blue("[Data2MongoDB]: 当前轮次共同步条数==>{0}".format(index)))
def _parameter_expression_handler(self, expression, parameters):
"""
参数表达式处理
:param expression: 参数表达式
:return: 参数表达式的结果
"""
expression_result = {}
self.logger.info(
Colored.blue(
"[Liz2Bird-_parameter_expression_handler()]: 参数处理模板 =>{0}".
format(expression)))
try:
# 表达式中包含dot
if DOT in expression and 'http' not in expression:
left, rights = expression.split(DOT, 1)
# 包含process处理
if left == PROCESSES:
number = rights[0]
process = self.queue.find(PROCESSES_TOPIC.format(
parameters['spider_name']),
way=WayType.HGET,
key=number)
process_template = self.middlewares[
2].middle_origin2engine(process)[0]
expression_result = HTTPProcessTemplate(**process_template)
# 包含result处理
elif left == RESULTS:
number, *parameter_name = rights.split(DOT)
if len(parameter_name) == 0:
result = self.queue.find(RESULTS_TOPIC.format(
parameters['spider_name']),
way=WayType.HGET,
key=number)
result_template = self.middlewares[
2].middle_origin2engine(result)[0]
expression_result = HTTPResultTemplate(
**result_template)
else:
value = parameters.get(parameter_name[0], None)
expression_result = {parameter_name[0]: value}
# 包含$处理
elif left == DOLLAR:
parameter_name = rights
value = parameters.get(parameter_name, None)
expression_result = {parameter_name: value}
else:
expression_result = {expression: expression}
except Exception as e:
self.logger.error(Colored.red(
"[Liz2Bird-_parameter_expression_handler()]: 出现处理错误>>>{0}".
format(e)),
exc_info=True)
raise e
finally:
return expression_result
async def single_download(self, task: Task):
headers = {'User-Agent': ua.random}
http_request = task.request
if http_request['referer']:
headers.setdefault('Referer', http_request['referer'])
elif 'referer' in task.parameters:
headers.setdefault('Referer', task.parameters["referer"])
if http_request['extra_headers']:
headers.update(http_request['extra_headers'])
method = http_request['method']
posts_data = None
if http_request['post_data']:
posts_data = http_request['post_data']
if http_request['use_proxy']:
use_proxy = http_request['use_proxy']
else:
use_proxy = task.parameters['use_proxy']
if http_request['timeout']:
timeout = int(http_request['timeout'])
else:
timeout = int(task.parameters['timeout'])
url = http_request['start_url']
if 'cookies' in task.parameters and task.parameters['cookies']:
cookies = task.parameters['cookies']
else:
cookies = None
self.logger.info(
Colored.green("[Downloader()]: 目前下载URL =>{0}".format(url)))
retry_times = 0
max_time = int(http_request['sleep_time'])
sleep_time = random.uniform(1, max_time)
max_retry = random.randint(4, 15)
while retry_times < max_retry:
proxy = get_random_proxy() if use_proxy else None
if proxy is not None:
self.logger.info(
Colored.green(
"[Downloader()]: 目前使用代理 =>{0}".format(proxy)))
try:
request = Any
ssl_context = httpx.create_ssl_context()
session = httpx.AsyncClient(proxies=proxy, verify=False)
if http_request.get('use_session',
None) or task.parameters.get(
'use_session', None):
self.session_container.setdefault(
task.parameters['spider_name'], session)
session = self.session_container.get(
task.parameters['spider_name'])
async with session:
if method == GET:
request: Coroutine = session.get(url,
headers=headers,
timeout=timeout,
cookies=cookies,
follow_redirects=True)
elif method == POST:
request: Coroutine = session.post(
url,
data=posts_data,
headers=headers,
timeout=timeout,
cookies=cookies,
follow_redirects=True)
response: Response = await request
if response.status_code != httpx.codes.OK:
response.raise_for_status()
if response is None:
return None
byte_content: bytes = await response.aread()
# 返回请求内容
process_type = http_request['return_type']
self.logger.info(
Colored.green(
"[Downloader()]: 请求返回类型>>{0}".format(process_type)))
encoding = chardet.detect(byte_content)['encoding']
time.sleep(sleep_time)
if process_type:
if process_type == TEXT:
return decode_content(byte_content, encoding)
elif process_type == JSON:
text = decode_content(byte_content, encoding)
return json.loads(text)
elif process_type == CSS:
return bs(byte_content, 'html5lib')
text = decode_content(byte_content, encoding).replace(
'\x00', '').strip().encode('utf-8')
parser = HTMLParser(encoding=encoding, remove_comments=True)
return etree.fromstring(text, parser)
except Exception as e:
retry_times += 1
self.logger.error(Colored.red(
"[Downloader()]: 目前下载URL =>{0} 重试次数[{1}/{2}] ERROR: {3}".
format(url, retry_times, max_retry, e)),
exc_info=True)
sleep_time *= random.random() * self.sleep_factor
time.sleep(sleep_time)
def _result_template_handler(self, result_template, data, parameters):
"""
结果template的处理
:param result_template: 结果template
:param data: 数据
:return: 结果template的普通映射结果, 以及generator 的映射结果
"""
temp_results = {}
temp_generator_results = {}
temp_parameters = {k: v for k, v in parameters.items()}
try:
# 处理result 模板的各个字段
field_dict = result_template.field_dict
if result_template.global_parameter is not None and isinstance(
result_template.global_parameter, str):
global_parameters = [
p.strip()
for p in result_template.global_parameter.split(AND)
]
self.logger.info(
Colored.blue(
"[Liz2Bird-_result_template_handler()]: 全局参数>>>>{0}".
format(', '.join(global_parameters))))
else:
global_parameters = []
for field, expression in field_dict.items():
if field in EXCLUDE_FIELD:
continue
self.logger.info(
Colored.blue(
"Liz2Bird-_result_template_handler()]: result 处理字典>>>> (<{0}::{1}>)"
.format(field, expression)))
# 起始下划线，则直接eval运行
if field.startswith(UNDER_LINE):
expression_result = eval(expression)
elif isinstance(expression, str):
# 起始//采用format_data获得结果
if expression.startswith(DIAGONAL):
expression_result = format_data(data, expression)
# 包含::的 css 提取数据
elif COLON in expression:
expression_result = format_data(data, expression)
# 包含>采用format_data获得结果
elif GREATER_THAN in expression:
expression_result = format_data(data, expression)
# 包含@的正则匹配
elif AT in expression:
exp, temp_data_field = expression.split(AT)
data_field, reg_idx = temp_data_field.split(HASH)
temp_data = temp_parameters.get(data_field)
if isinstance(temp_data, list):
expression_result = []
for temp_da in temp_data:
expression_one = format_data(
temp_da, exp, reg_idx)
expression_result.append(expression_one)
else:
expression_result = format_data(
temp_data, exp, reg_idx)
# 包含$符号的拼接字段
elif expression.startswith(DOLLAR):
at_parameters_name = PARAMETER.findall(expression)
one_parameters = {}
dict_parameter = {}
for at_parameter_name in at_parameters_name:
parameter_value = temp_parameters.get(
at_parameter_name, None)
if isinstance(parameter_value, list):
dict_parameter.setdefault(
at_parameter_name, parameter_value)
else:
one_parameters.setdefault(
at_parameter_name, parameter_value)
# 处理 format 表达式
extend_parameters_group = []
one_round = 0
if len(dict_parameter) > 0:
for key, value_list in dict_parameter.items():
# 扩展参数组
for index, value in enumerate(value_list):
if index == one_round:
copy_one = deepcopy(one_parameters)
copy_one.update({key: value})
extend_parameters_group.append(
copy_one)
one_round += 1
else:
extend_parameters_group[index].update(
{key: value})
if len(extend_parameters_group) > 0:
expression_result = [
expression.replace(
DOLLAR, "").format(**parameters_group)
for parameters_group in extend_parameters_group
]
else:
expression_result = expression.replace(
DOLLAR, "").format(**one_parameters)
elif LEFT_BRACE in expression and RIGHT_BRACE in expression:
# 包含{xxx}的参数，进行计算
at_parameters_name = PARAMETER.findall(expression)
total_parameters = {}
for at_parameter_name in at_parameters_name:
if at_parameter_name in global_parameters:
site = HASH_MAP.format(
temp_parameters['site_name'])
parameter_value = self.queue.find(
"{site}:{spider}:{key}".format(
site=site,
spider=temp_parameters['spider_name'],
key=at_parameter_name),
way=WayType.GET,
key=at_parameter_name)
if isinstance(parameter_value, bytes):
parameter_value = parameter_value.decode(
'utf-8')
total_parameters.setdefault(
at_parameter_name, parameter_value)
else:
parameter_value = temp_parameters.get(
at_parameter_name, None)
total_parameters.setdefault(
at_parameter_name, parameter_value)
expression_result = eval(
expression.format(**total_parameters))
else:
# 否则直接返回原结果
expression_result = expression
else:
# 否则直接返回原结果
expression_result = expression
# 更新 global 参数
if field in global_parameters:
if isinstance(expression_result, str):
expression_result_ = expression_result.encode('utf-8')
else:
expression_result_ = expression_result
site = HASH_MAP.format(temp_parameters['site_name'])
self.queue.commit_data("{site}:{spider}:{key}".format(
site=site,
spider=temp_parameters['spider_name'],
key=field),
key=field,
value=expression_result_,
way=WayType.SET)
self.logger.info(
Colored.white(
"Liz2Bird-_result_template_handler()]: 提交全局参数>>>> (<{0}: {1}>)"
.format(field, expression_result)))
# 更新模板结果, result模板字典
if isinstance(expression_result, list):
temp_generator_results.update({field: expression_result})
else:
temp_results.update({field: expression_result})
temp_parameters.update({field: expression_result})
except Exception as e:
self.logger.error(Colored.red(
"[Liz2Bird-_result_template_handler()]: 出现处理错误>>>{0}....(@ $ _ $ @)----"
.format(e)),
exc_info=True)
return temp_results, temp_generator_results
def _process(self, content, task: Task):
"""
处理下载内容
:param content: 下载的内容
:param task: task 任务
:return: 处理后的结果集合
"""
if content is None:
return None
process_bytes = self.queue.find(PROCESSES_TOPIC.format(
task.parameters['spider_name']),
way=WayType.HGET,
key=task.process_id)
processed_content = content
if task.process_id is not None and process_bytes is None:
raise Exception('没有找到 process method：{0}！'.format(task.process_id))
elif process_bytes is not None:
process_template = self.middlewares[2].middle_origin2engine(
process_bytes)[0]
process_obj = HTTPProcessTemplate(**process_template)
# 处理 content
if process_obj.process_method is not None:
method = load_method(process_obj.process_method)
if process_obj.parameters is None:
processed_content = method(content)
else:
parameters = self._get_parameters(process_obj.parameters,
task.parameters)
processed_content = method(content, parameters)
result_bytes = self.queue.find(RESULTS_TOPIC.format(
task.parameters['spider_name']),
way=WayType.HGET,
key=task.result_id)
result_template = self.middlewares[2].middle_origin2engine(
result_bytes)[0]
result_obj = HTTPResultTemplate(**result_template)
self.logger.info(
Colored.green("[Liz2Bird-_process()]: 处理的 result 模板>>{0}".format(
result_obj.result_id)))
# 处理 request 中定义的 stopped，默认是位于 range time 的返回内
stopped = task.request['stopped']
# 目前支持 page 和range time两种方式
judge_flag = True if stopped is not None and 'page' in stopped else False
# range time 的范围
if HASH in task.parameters['range_time']:
s_date_str, e_date_str = task.parameters['range_time'].split('#')
else:
s_date_str = (datetime.datetime.now() -
datetime.timedelta(days=1)).strftime(DAILY_FORMAT)
e_date_str = ''
# 规则匹配内容并返回
self.logger.info(
Colored.green("[Liz2Bird-_process()]: 正在处理的 URL >> {0}".format(
task.request['start_url'])))
result_dict, result_generator_dict = self._result_template_handler(
result_obj, processed_content, task.parameters)
all_task_result = []
try:
if stopped is not None:
if stopped in result_dict:
if judge_flag:
# 处理 page 的 stopped
page = result_dict.get('page', None)
if int(page) > int(result_dict.get(stopped)):
return all_task_result
elif not is_in_range_time(result_dict.get(stopped),
s_date_str, e_date_str):
return all_task_result
if len(result_dict) > 0:
all_task_result.append(result_dict)
stopped_list = []
is_stopped = False
# 先确定 stopped 停止的字段
if stopped is not None and stopped in result_generator_dict:
stopped_value = result_generator_dict.get(stopped)
# 排序 stopped 字段index
stopped_args = sort_time(stopped_value)
for arg in stopped_args:
if not is_in_range_time(stopped_value[arg], s_date_str,
e_date_str):
is_stopped = True
break
else:
stopped_list.append((arg, stopped_value[arg]))
for key, value in result_generator_dict.items():
index = 0
if len(stopped_list) > 0:
# 如果 value 小于 stopped_list，则需要填充
if 1 == len(value) < len(stopped_list):
value = value * len(stopped_list)
elif len(value) != len(
stopped_list): # 如果不等于 stopped_list 无法正确填充，故抛出异常
raise Exception(
'key: {0}的 value 值在result_generator_dict中少于 stopped 字段的个数'
.format(key))
# 如果有停止列表，则按照排序的结果进行选择需要的字段
for arg, stopped_val in stopped_list:
all_task_result = self._generate_result(
key, all_task_result, index, value[arg],
stopped_val, stopped)
index += 1
else:
# 否则，迭代所有的字段，并返回结果
for gen_value in value:
all_task_result = self._generate_result(
key, all_task_result, index, gen_value)
index += 1
# 如果没有停止，则把刚才的 request template task 再次扔进去
if not is_stopped and task.request['is_multiple']:
current_request_bytes = self.queue.find(
REQUESTS_TOPIC.format(task.parameters['spider_name']),
way=WayType.HGET,
key=task.request['request_id'])
current_request_template = self.middlewares[
2].middle_origin2engine(current_request_bytes)[0]
current_request_obj = HTTPRequestTemplate(
**current_request_template)
repeat_task = self._build_task(current_request_obj,
None,
pre_task=task,
result_dict=result_dict)
self._queue_push([repeat_task])
self.logger.info(
Colored.green(
"[Liz2Bird-_process()]: repeat task再次放入>>{0}".format(
repeat_task['task_id'])))
except Exception as e:
self.logger.error(Colored.red(
"[Liz2Bird-_process()]: 处理出现错误>>>{0}=>{1}....(@ $ _ $ @)----".
format(e, task.request['url'])),
exc_info=True)
finally:
return all_task_result
def _build_task(self,
request: HTTPRequestTemplate,
template=None,
pre_task=None,
result_dict=None):
"""
构建 task
:param request: 请求模板
:param template: 母模板
:param pre_task: 前置的 task
:param result_dict: 结果字典
:return: task 对象
"""
# 处理 request 中的未知参数
parameters = {}
if template is not None:
rest_template_fields = {
key.lower(): template.get(key)
for key in template.keys()
if key not in [PROCESSES, RESULTS, REQUESTS]
}
parameters = rest_template_fields
task = request.obj2dict()
if pre_task is not None:
# 更新 parameters
parameters.update(pre_task.parameters)
self.logger.info(
Colored.green(
"[Liz2Bird-_build_task()]: 更新 pre_task 的参数=>{0}".format(
pre_task.task_id)))
# 更新参数组
if result_dict is not None:
parameters.update(result_dict)
self.logger.info(
Colored.green("[Liz2Bird-_build_task()]: 用 result_dict 更新参数"))
# 第一种情况：start_url 不存在
if request.start_url is None:
task['start_url'] = parameters["start_url"]
self.logger.info(
Colored.green("[Liz2Bird-_build_task()]: 采用起始链接>>>{0}".format(
parameters["start_url"])))
# 第二种情况：start_url 是某个 result 模板的解析结果
elif RESULTS in request.start_url:
req_url = self._parameter_expression_handler(
request.start_url, parameters)
task['start_url'] = req_url.popitem()[1]
self.logger.info(
Colored.green(
"[Liz2Bird-_build_task()]: 采用result 模板{0}，链接为>>{1}".format(
request.start_url, task['start_url'])))
# start_url 的参数替换处理
if request.parameters is not None:
url_placeholder = self._get_parameters(request.parameters,
parameters)
task['start_url'] = task['start_url'].format(**url_placeholder)
self.logger.info(
Colored.green(
"[Liz2Bird-_build_task()]: 更新参数组{0}，链接为>>{1}".format(
url_placeholder, task['start_url'])))
# post data 处理
if request.post_data is not None:
if not isinstance(request.post_data, dict):
raise Exception("post data 参数需要时 dict 类型")
post_data = {}
for post_data_key, post_data_value in request.post_data.items():
value = self._parameter_expression_handler(
post_data_value, parameters).popitem()[1]
post_data.setdefault(post_data_key, value)
task['post_data'] = post_data
# 处理 referer 的参数处理
if request.referer is not None and RESULTS in request.referer:
referer_url = self._parameter_expression_handler(
request.referer, parameters)
task['referer'] = referer_url.popitem()[1]
self.logger.info(
Colored.green(
"[Liz2Bird-_build_task()]: 采用result 模板{0}，链接为>>{1}".format(
request.referer, task['referer'])))
if task['start_url'] is None or 'None' in task['start_url']:
self.logger.error(
Colored.red(
"[Liz2Bird-_build_task()]: URL 中出现 None>>>start_url=>{0}".
format(task['start_url'])))
return None
return Task(task_id=None, request=task,
parameters=parameters).obj2dict()
async def _start(self, is_master=False):
"""
开始爬虫
:param is_master: 是否是主节点
:return:
"""
self.logger.info(
Colored.green("[Liz2Bird-start()]: Start Spider d=====(￣▽￣*)b"))
if is_master and self.spider_status == SpiderStatus.Null:
# 记录 spider 的映射名称
spiders = []
global_parameters = []
# 迭代并提取模板
for template in self.templates:
self.logger.info(
Colored.green("[Liz2Bird-start()]: 当前读取模板=>{0}".format(
template['SPIDER_NAME'])))
requests, processes, results = self.middlewares[
0].middle_origin2engine(template)
# 所有的 spider 列表
task_topic = TASK_TOPIC.format(template['SPIDER_NAME'])
spiders.append(task_topic)
# spider 对应的全局参数映射，作为后续删除
global_parameters.append(HASH_MAP.format(
template['SITE_NAME']))
# 放置一些映射关系
self._queue_push([r.obj2dict() for r in requests],
topic=REQUESTS_TOPIC.format(
template['SPIDER_NAME']),
way=WayType.HSET,
key_ids=[r.request_id for r in requests])
self._queue_push([p.obj2dict() for p in processes],
topic=PROCESSES_TOPIC.format(
template['SPIDER_NAME']),
way=WayType.HSET,
key_ids=[p.process_id for p in processes])
self._queue_push([r.obj2dict() for r in results],
topic=RESULTS_TOPIC.format(
template['SPIDER_NAME']),
way=WayType.HSET,
key_ids=[r.result_id for r in results])
# 如果对应的task spider 没有任务，则先清除旧的 key，然后初始化 task，否则不添加初始 task
count = self.queue.count(task_topic)
if count == 0:
# 去除全局参数的 key
values = self.queue.find(TEMPLATES_TOPIC,
way=WayType.HGET,
key=task_topic)
if values is not None:
self_global_parameters_ = self.middlewares[
2].middle_origin2engine(*[values])
spider = task_topic.split('_')[1]
self_global_parameters_ = [
'{0}:{1}*'.format(global_parameter, spider)
for global_parameter in self_global_parameters_
]
self.queue.delete_all(*self_global_parameters_)
# 放入第一个 request 任务
one_task = self._build_task(requests[0], template)
self._queue_push([one_task])
# 记录 spider 的映射名称
self._queue_push(global_parameters,
topic=TEMPLATES_TOPIC,
key_ids=spiders,
way=WayType.HSET)
# 非 master 节点，进行爬取
if not is_master:
try:
sleep_time = random.uniform(0, 2)
counter_task = []
while True:
# 弹出 task 并下载
total_count, tasks = self._task_pop()
if tasks is None or len(tasks) == 0:
# 线性增加等待时间
sleep_time += self.task_sleep * random.random()
self.logger.info(
Colored.green(
"[Liz2Bird-start()]: 无任务，休眠时间 =>{0}".format(
sleep_time)))
await asyncio.sleep(sleep_time)
continue
else:
sleep_time = random.uniform(0, 2)
# 第一个任务
task = tasks[0]
time.sleep(sleep_time)
try:
counter_task.append(task)
self.logger.info(
Colored.green(
"[Liz2Bird-start()]: 弹出下载Task，task_id>>{0}>>>{1}"
.format(task.task_id,
task.parameters['site_name'])))
download_content_list = await asyncio.gather(
*[self._download(t) for t in counter_task],
return_exceptions=True)
self.logger.info(
Colored.green("[Liz2Bird-start()]: 少于 4 个串行请求"))
# 处理结果
for i, download_content in enumerate(
download_content_list):
if download_content is None:
continue
task = counter_task[i]
self.logger.info(
Colored.green(
"[Liz2Bird-start()]: 处理Task，采用 process 模板=>{0}"
.format(task.process_id)))
# 处理 task 请求后产生的内容
new_task_results = self._process(
download_content, task)
# next request 的生成
next_request_obj = None
if task.next_request is not None:
self.logger.info(
Colored.green(
"[Liz2Bird-start()]: Next Request，采用 request 模板=>{0}"
.format(task.next_request)))
next_request_bytes = self.queue.find(
REQUESTS_TOPIC.format(
task.parameters['spider_name']),
way=WayType.HGET,
key=task.next_request)
next_request_template = self.middlewares[
2].middle_origin2engine(
next_request_bytes)[0]
next_request_obj = HTTPRequestTemplate(
**next_request_template)
# 同一级请求的任务
if task.parallel_request is not None:
self.logger.info(
Colored.green(
"[Liz2Bird-start()]: Parallel Request，采用 request 模板=>{0}"
.format(task.parallel_request)))
parallel_request_bytes = self.queue.find(
REQUESTS_TOPIC.format(
task.parameters['spider_name']),
way=WayType.HGET,
key=task.parallel_request)
parallel_request_template = \
self.middlewares[2].middle_origin2engine(parallel_request_bytes)[0]
parallel_request_obj = HTTPRequestTemplate(
**parallel_request_template)
new_task = self._build_task(
parallel_request_obj,
None,
pre_task=task,
result_dict=new_task_results[0])
self._queue_push([new_task])
if new_task is not None:
self.logger.info(
Colored.green(
"[Liz2Bird-start()]: Parallel Task，task_id模板=>{0}"
.format(new_task['task_id'])))
# 下一级请求任务
for new_result in new_task_results:
# 分类放置处理后的内容
if next_request_obj is not None:
new_task = self._build_task(
next_request_obj,
None,
pre_task=task,
result_dict=new_result)
self._queue_push([new_task])
if new_task is not None:
self.logger.info(
Colored.green(
"[Liz2Bird-start()]: 下一个Task，task_id模板=>{0}"
.format(new_task['task_id'])))
else:
self.logger.info(
Colored.green(
"[Liz2Bird-start()]: 生成结果，采用 result 模板=>{0}"
.format(task.result_id)))
result = self._build_result(
task, result_dict=new_result)
self.pipeline.content_process(result)
counter_task.clear()
except Exception as e:
# 添加任务重试机制
self.logger.error(Colored.red(
"[Liz2Bird-start()]: Task>>>{0}出现错误>>>{1}, 重试次数[{2}/3]...."
.format(task.task_id, e, task.retry_times)),
exc_info=True)
except Exception as e:
self.logger.error(Colored.red(
"[Liz2Bird-start()]: 出现错误>>>{0}....(@ $ _ $ @)----".format(
e)),
exc_info=True)