def save_backup(self, doc_source: str, doc_source_name: str, doc_name: str) -> bool:
"""保存文件备份状态
Args:
doc_source (str): 文章获取源
doc_source_name (str): 文章源
doc_name (str): 文章名字
Returns:
bool: 是否成功
"""
file_msg = f"{doc_source}/{doc_source_name}/{doc_name}"
try:
filter_dict = {
"backup_type": self.backup_type,
"doc_source": doc_source,
"doc_source_name": doc_source_name,
"doc_name": doc_name,
}
update_data = {"$set": {**filter_dict, **{"ts": int(time.time())}}}
self.bak_coll.update_one(
filter=filter_dict, update=update_data, upsert=True
)
LOGGER.info(f"Backup({self.backup_type}): 文章 {file_msg} 状态保存成功！")
except Exception as e:
LOGGER.error(f"Backup({self.backup_type}): 文章 {file_msg} 状态保存失败！{e}")
def send(self, send_data) -> bool:
"""
下发到Bark终端
:param send_data: 下发内容字典，字段开发者自定义
:return:
"""
doc_name = send_data["doc_name"]
# doc_source = send_data["doc_source"]
doc_link = send_data["doc_link"]
doc_cus_des = send_data["doc_cus_des"]
doc_source_name = send_data["doc_source_name"]
doc_id = send_data["doc_id"]
is_send = self.is_send(doc_id=doc_id)
send_status = True
notice_msg = f"{doc_cus_des}👉{doc_source_name}_{doc_name}：{doc_link} 分发到 {self.send_type}"
if not is_send:
url = self.compose(send_data)
resp = requests.post(url)
if resp.status_code == 200 and json.loads(
resp.text)["code"] == 200:
# 将状态持久化到数据库
self.sl_coll.insert_one({
"send_type": self.send_type,
"doc_id": doc_id,
"ts": int(time.time()),
})
# 下发成功
LOGGER.info(f"{notice_msg} 成功！")
else:
errmsg = json.loads(resp.text)["code"]
LOGGER.error(f"{notice_msg} 失败：{errmsg}")
send_status = False
return send_status
def delete_backup(
self, doc_source: str, doc_source_name: str, doc_name: str
) -> bool:
"""删除文件分备份状态
Args:
doc_source (str): 文章获取源
doc_source_name (str): 文章源
doc_name (str): 文章名字
Returns:
bool: 是否成功
"""
file_msg = f"{doc_source}/{doc_source_name}/{doc_name}"
try:
self.bak_coll.delete_one(
{
"backup_type": self.backup_type,
"doc_source": doc_source,
"doc_source_name": doc_source_name,
"doc_name": doc_name,
}
)
LOGGER.info(f"Backup({self.backup_type}): 文章 {file_msg} 状态删除成功！")
except Exception as e:
LOGGER.error(f"Backup({self.backup_type}): 文章 {file_msg} 状态删除失败！{e}")
def create_app():
"""
建立web应用
url: http://flask.pocoo.org/docs/1.0/quickstart/
:return:
"""
flask_app = Flask(__name__)
with flask_app.app_context():
# 项目内部配置
mongodb_base = MongodbManager.get_mongo_base(
mongodb_config=Config.MONGODB_CONFIG)
app_loop = asyncio.get_event_loop()
flask_app.config["app_config"] = Config
flask_app.config["app_logger"] = LOGGER
flask_app.config["app_loop"] = app_loop
flask_app.config["mongodb_base"] = mongodb_base
# 每次启动先保证公众号名称爬虫运行成功
# spider = run_wechat_name_spider(loop=app_loop)
# if spider.success_counts == 1:
# # 爬虫运行成功
# LOGGER.info("Wechat spider started successfully :)")
LOGGER.info("API started successfully :)")
flask_app.register_blueprint(bp_api)
return flask_app
def delete(self, doc_source: str, doc_source_name: str, doc_name: str) -> bool:
"""删除某个文件
Args:
doc_source (str): 文章获取源
doc_source_name (str): 文章源
doc_name (str): 文章名字
Returns:
bool: 是否成功
"""
file_path = f"{doc_source}/{doc_source_name}/{doc_name}.html"
db_res = mongodb_delete_many_data(
coll_conn=self.liuli_backup_coll,
filter_dict={
"doc_source": doc_source,
"doc_source_name": doc_source_name,
"doc_name": doc_name,
},
)
op_res = True
if db_res["status"]:
LOGGER.info(f"Backup({self.backup_type}): {file_path} 删除成功！")
# 删除当前文章状态
self.delete_backup(
doc_source=doc_source,
doc_source_name=doc_source_name,
doc_name=doc_name,
)
else:
op_res = False
LOGGER.error(
f"Backup({self.backup_type}): {file_path} 删除失败！{db_res['info']}"
)
return op_res
def run_liuli_schedule(ll_config_name: str = "default"):
"""调度启动函数
Args:
task_config (dict): 调度任务配置
"""
ll_config_path = os.path.join(Config.LL_CONFIG_DIR,
f"{ll_config_name}.json")
with open(ll_config_path, "r", encoding="utf-8") as load_f:
ll_config = json.load(load_f)
schdule_time_list = ll_config["schedule"].get("period_list",
["00:10", "12:10", "21:10"])
for each in schdule_time_list:
schedule.every().day.at(each).do(run_liuli_task, ll_config)
name: str = ll_config["name"]
author: str = ll_config["author"]
start_info = f"Schedule({Config.SCHEDULE_VERSION}) task({name}@{author}) started successfully :)"
LOGGER.info(start_info)
schdule_msg = f"Task({name}@{author}) schedule time:\n " + "\n ".join(
schdule_time_list)
LOGGER.info(schdule_msg)
# 启动就执行一次
run_liuli_task(ll_config)
while True:
schedule.run_pending()
time.sleep(1)
def delete(self, doc_source: str, doc_source_name: str,
doc_name: str) -> bool:
"""删除某个文件
Args:
doc_source (str): 文章获取源
doc_source_name (str): 文章源
doc_name (str): 文章名字
Returns:
bool: 是否成功
"""
file_path = f"{doc_source}/{doc_source_name}/{doc_name}.html"
op_res = True
try:
contents = self.repo.get_contents(file_path)
_ = self.repo.delete_file(contents.path, f"Remove {file_path}",
contents.sha)
LOGGER.info(f"Backup({self.backup_type}): {file_path} 删除成功！")
# 删除当前文章状态
self.delete_backup(
doc_source=doc_source,
doc_source_name=doc_source_name,
doc_name=doc_name,
)
except Exception as e:
op_res = False
LOGGER.error(f"Backup({self.backup_type}): {file_path} 删除失败！{e}")
return op_res
def save(self, backup_data: dict) -> bool:
"""执行备份动作
Args:
backup_data (dict): 备份数据
Returns:
bool: 是否成功
"""
# 以下字段必须存在
doc_source = backup_data["doc_source"]
doc_source_name = backup_data["doc_source_name"]
doc_name = backup_data["doc_name"]
# 源文件
doc_html = backup_data["doc_html"]
file_msg = f"{doc_source}/{doc_source_name}/{doc_name}"
file_path = f"{file_msg}.html"
is_backup = self.is_backup(
doc_source=doc_source,
doc_source_name=doc_source_name,
doc_name=doc_name,
)
# 未备份过或者强制备份下将继续执行
if not is_backup or self.force_backup:
filter_dict = {
"doc_source": doc_source,
"doc_source_name": doc_source_name,
"doc_name": doc_name,
}
update_data = {
"$set": {
**filter_dict,
**{"ts": int(time.time()), "content": text_compress(doc_html)},
}
}
db_update_res = mongodb_update_data(
coll_conn=self.liuli_backup_coll,
filter_dict=filter_dict,
update_data=update_data,
upsert=True,
)
if db_update_res["status"]:
msg = f"Backup({self.backup_type}): {file_path} 上传成功！"
# 保存当前文章状态
self.save_backup(
doc_source=doc_source,
doc_source_name=doc_source_name,
doc_name=doc_name,
)
else:
msg = f"Backup({self.backup_type}): {file_path} 上传失败！{db_update_res['info']}"
else:
msg = f"Backup({self.backup_type}): {file_path} 已存在！"
LOGGER.info(msg)
def send_post_request(url, data: dict = None, **kwargs) -> dict:
"""
发起post请求
:param url: 请求目标地址
:param data: 请求参数
:param kwargs:
:return:
"""
try:
resp_dict = requests.post(url, data=json.dumps(data), **kwargs).json()
except Exception as e:
resp_dict = {}
LOGGER.error(f"请求出错：{e}")
return resp_dict
def send_factory(send_type: str, send_config: dict, send_data: dict) -> bool:
"""
分发器工厂函数
:param send_type: 下发终端类型
:param send_config: 下发终端配置
:param send_data: 下发内容字典，字段开发者自定义
:return:
"""
send_status = False
try:
send_module = import_module(f"src.sender.{send_type}_sender")
send_status = send_module.send(send_config, send_data)
except ModuleNotFoundError:
LOGGER.error(f"目标终端类型不存在 {send_type} - {send_config} - {send_data}")
return send_status
def collect_factory(collect_type: str, collect_config: dict) -> bool:
"""
采集器工厂函数
:param collect_type: 采集器类型
:param collect_config: 采集器配置
:return:
"""
collect_status = False
try:
collect_module = import_module(f"src.collector.{collect_type}")
collect_status = collect_module.run(collect_config)
except ModuleNotFoundError as e:
LOGGER.error(f"采集器类型不存在 {collect_type} - {collect_config} -{e}")
except Exception as e:
LOGGER.error(f"采集器执行出错 {collect_type} - {collect_config} - {e}")
return collect_status
def send_get_request(url: str, params: dict = None, timeout: int = 3, **kwargs):
"""发起GET请求
Args:
url (str): 目标地址
params (dict, optional): 请求参数. Defaults to None.
timeout (int, optional): 超时时间. Defaults to 3.
Returns:
[type]: [description]
"""
try:
resp = requests.get(url, params, timeout=timeout, **kwargs)
except Exception as e:
resp = None
LOGGER.exception(f"请求出错 - {url} - {str(e)}")
return resp
def get_html_by_requests(url: str, params: dict = None, timeout: int = 3, **kwargs):
"""发起GET请求，获取文本
Args:
url (str): 目标网页
params (dict, optional): 请求参数. Defaults to None.
timeout (int, optional): 超时时间. Defaults to 3.
"""
resp = send_get_request(url=url, params=params, timeout=timeout, **kwargs)
text = None
try:
content = resp.content
charset = cchardet.detect(content)
text = content.decode(charset["encoding"])
except Exception as e:
LOGGER.exception(f"请求内容提取出错 - {url} - {str(e)}")
return text
def send_text_card(self, send_data):
"""
发送卡片消息
:param send_data:
:return:
"""
doc_name = send_data["doc_name"]
doc_source = send_data["doc_source"]
doc_link = send_data["doc_link"]
doc_content = send_data["doc_content"]
doc_cus_des = send_data["doc_cus_des"]
doc_source_name = send_data["doc_source_name"]
doc_keywords = send_data["doc_keywords"]
doc_ts = send_data["doc_ts"]
doc_date = time.strftime("%Y-%m-%d", time.localtime(doc_ts))
doc_des_info = f"亲，来自 {doc_source} 源的 {doc_source_name} 有更新啦! \n\n{doc_content}\n\n文章关键字：{doc_keywords}"
doc_des = f'<div class="black">{doc_date} | {doc_cus_des}</div>\n<div class="normal">{doc_des_info}</div>\n来自[2c]👉技术支持❤️'
data = {
"toparty": 1,
"msgtype": "textcard",
"agentid": self.wecom_agent_id,
"textcard": {
"title": f"[{doc_source_name}]{doc_name}",
"description": doc_des,
"url": doc_link,
"btntxt": "更多",
},
"safe": 0,
}
data = json.dumps(data, ensure_ascii=False)
try:
resp_dict = requests.post(
url=self.url,
data=data.encode("utf-8").decode("latin1"),
headers={
"Content-Type": "application/json"
},
).json()
return resp_dict
except Exception as e:
resp_dict = {}
LOGGER.error(f"请求出错：{e}")
return resp_dict
def backup_factory(backup_type: str, init_config: dict) -> BackupBase:
"""
备份器工厂函数
:param backup_type: 备份类型
:param init_config: 备份配置
:return:
"""
backup_ins = None
try:
backup_class_name = f"{backup_type}_backup"
backup_module = import_module(f"src.backup.{backup_class_name}")
# 备份类实例化
backup_ins = getattr(
backup_module,
string_camelcase(backup_class_name))(init_config=init_config)
except ModuleNotFoundError as e:
LOGGER.error(f"目标备份类型不存在 {backup_type} - {init_config} - {e}")
return backup_ins
def send_post_request(url: str, data: dict = None, timeout: int = 5, **kwargs) -> dict:
"""发起post请求
Args:
url (str): 目标地址
data (dict, optional): 请求参数. Defaults to None.
timeout (int, optional): 超时时间. Defaults to 5.
Returns:
dict: [description]
"""
try:
resp_dict = requests.post(
url, data=json.dumps(data), timeout=timeout, **kwargs
).json()
except Exception as e:
resp_dict = {}
LOGGER.error(f"请求出错：{e}")
return resp_dict
def send(self, send_data) -> bool:
"""
下发到钉钉终端
:param send_data: 下发内容字典，字段开发者自定义
:return:
"""
doc_name = send_data["doc_name"]
doc_cus_des = send_data["doc_cus_des"]
doc_id = send_data["doc_id"]
doc_link = send_data["doc_link"]
doc_source_name = send_data["doc_source_name"]
is_send = self.is_send(doc_id=doc_id)
send_status = True
if not is_send:
# 开始进行下发
resp_dict = self.send_text_card(send_data=send_data)
notice_msg = f"{doc_cus_des}👉{doc_source_name}_{doc_name}：{doc_link} 分发到 {self.send_type}"
if resp_dict:
if resp_dict.get("errcode") == 0:
# 将状态持久化到数据库
self.sl_coll.insert_one({
"send_type": self.send_type,
"doc_id": doc_id,
"ts": time.time(),
})
# 下发成功
LOGGER.info(f"{notice_msg} 成功！")
send_status = True
else:
LOGGER.error(f"{notice_msg} 失败：{resp_dict.get('errmsg')}")
else:
LOGGER.error(f"{notice_msg} 失败!")
return send_status
def send(self, send_data) -> bool:
"""
下发到钉钉终端
:param send_data: 下发内容字典，字段开发者自定义
:return:
"""
doc_id = send_data["doc_id"]
doc_name = send_data["doc_name"]
doc_source = send_data["doc_source"]
doc_link = send_data["doc_link"]
doc_cus_des = send_data["doc_cus_des"]
doc_source_name = send_data["doc_source_name"]
doc_keywords = send_data["doc_keywords"]
is_send = self.is_send(doc_id=doc_id)
doc_date = send_data["doc_date"]
send_status = True
if not is_send:
# 开始进行下发
# data = {
# "msgtype": "link",
# "link": {
# "text": f"[liuli]{doc_source_name}: {doc_cus_des}\n亲，{doc_source} 源有更新\n",
# "title": doc_name,
# "picUrl": "",
# "messageUrl": doc_link,
# },
# }
data = {
"msgtype": "markdown",
"markdown": {
"text":
f"## [{doc_name}]({doc_link})\n\n**{doc_source_name}** | **{doc_date}** | **{doc_cus_des}** \n\n-----\n\n> 文章关键字：{doc_keywords}\n\n-----\n\n识别错误？点击[广告反馈](https://github.com/howie6879/liuli/issues/4) 👉来自[liuli](https://github.com/howie6879/liuli)技术支持❤️",
"title": f"亲，{doc_source} 源有更新啦!👉{doc_name} ",
},
}
resp_dict = send_post_request(
url=self.url,
data=data,
headers={"Content-Type": "application/json"})
notice_msg = f"{doc_cus_des}👉{doc_source_name}_{doc_name}：{doc_link} 分发到 {self.send_type}"
if resp_dict:
if resp_dict.get("errmsg") == "ok":
# 将状态持久化到数据库
self.sl_coll.insert_one({
"send_type": self.send_type,
"doc_id": doc_id,
"ts": int(time.time()),
})
# 下发成功
LOGGER.info(f"{notice_msg} 成功！")
else:
LOGGER.error(f"{notice_msg} 失败：{resp_dict.get('errmsg')}")
send_status = False
else:
LOGGER.error(f"{notice_msg} 失败!")
send_status = False
return send_status
def save(self, backup_data: dict) -> bool:
"""执行备份动作
Args:
backup_data (dict): 备份数据
Returns:
bool: 是否成功
"""
# 以下字段必须存在
doc_source = backup_data["doc_source"]
doc_source_name = backup_data["doc_source_name"]
doc_name = backup_data["doc_name"]
# 源文件
doc_html = backup_data["doc_html"]
file_path = f"{doc_source}/{doc_source_name}/{doc_name}.html"
is_backup = self.is_backup(
doc_source=doc_source,
doc_source_name=doc_source_name,
doc_name=doc_name,
)
# 在数据库存在就默认线上必定存在，希望用户不操作这个仓库造成状态不同步
if not is_backup or self.force_backup:
# 上传前做是否存在检测，没有备份过继续远程备份
# 已存在的但是数据库没有状态需要重新同步
try:
# 先判断文件是否存在
try:
contents = self.repo.get_contents(file_path)
# 存在就更新
self.repo.update_file(contents.path, f"Update {file_path}",
doc_html, contents.sha)
except Exception as e:
# 不存在就上传
self.repo.create_file(file_path, f"Add {file_path}",
doc_html)
LOGGER.info(f"Backup({self.backup_type}): {file_path} 上传成功！")
# 保存当前文章状态
self.save_backup(
doc_source=doc_source,
doc_source_name=doc_source_name,
doc_name=doc_name,
)
except GithubException as e:
LOGGER.error(
f"Backup({self.backup_type}): {file_path} 上传失败！{e}")
else:
LOGGER.info(f"Backup({self.backup_type}): {file_path} 已存在！")
def post(self):
from flask_restful import request
LOGGER.info(f'Received request {request}')
LOGGER.info(f'Request JSON data: {request.get_json()}')
message_attrs = parse_request(SlackMessageSchema, request)
token = self.get_token()
slack_client = self.setup_slack_client(token)
response = slack_client.chat_postMessage(**message_attrs)
LOGGER.info(f'Slack client response: {response}')
return make_response(
jsonify(code=response.status_code, data=response.data),
response.status_code)
def send(self, send_data) -> bool:
"""
下发到Telegram
:param send_data: 下发内容字典，字段开发者自定义
:return:
"""
doc_id = send_data["doc_id"]
doc_name = send_data["doc_name"]
doc_link = send_data["doc_link"]
doc_cus_des = send_data["doc_cus_des"]
doc_source_name = send_data["doc_source_name"]
is_send = self.is_send(doc_id=doc_id)
send_status = True
if not is_send:
message = TG_BOT_MSG_TEMPLATE.format_map(send_data)
data = {
"chat_id": self.chat_id,
"text": message,
"parse_mode": "HTML",
"disable_web_page_preview": "yes",
}
resp_dict = send_post_request(
url=self.url,
data=data,
headers={"Content-Type": "application/json"},
timeout=5,
)
notice_msg = f"{doc_cus_des}👉{doc_source_name}_{doc_name}：{doc_link} 分发到 {self.send_type}"
if resp_dict:
if resp_dict.get("ok") is True:
# 将状态持久化到数据库
self.sl_coll.insert_one({
"send_type": self.send_type,
"doc_id": doc_id,
"ts": int(time.time()),
})
# 下发成功
LOGGER.info(f"{notice_msg} 成功！")
else:
LOGGER.error(f"{notice_msg} 失败：{resp_dict.get('errmsg')}")
send_status = False
else:
LOGGER.error(f"{notice_msg} 失败!")
send_status = False
return send_status
def run_liuli_task(ll_config: dict):
"""执行调度任务
Args:
ll_config (dict): Liuli 任务配置
"""
try:
# 防止内部函数篡改
ll_config_data = deepcopy(ll_config)
# 文章源, 用于基础查询条件
doc_source: str = ll_config_data["doc_source"]
basic_filter = {"basic_filter": {"doc_source": doc_source}}
# 采集器配置
collector_conf: dict = ll_config_data["collector"]
# 处理器配置
processor_conf: dict = ll_config_data["processor"]
# 分发器配置
sender_conf: dict = ll_config_data["sender"]
sender_conf.update(basic_filter)
# 备份器配置
backup_conf: dict = ll_config_data["backup"]
backup_conf.update(basic_filter)
# 采集器执行
LOGGER.info("采集器开始执行!")
for collect_type, collect_config in collector_conf.items():
collect_factory(collect_type, collect_config)
LOGGER.info("采集器执行完毕!")
# 采集器执行
LOGGER.info("处理器(after_collect): 开始执行!")
for each in processor_conf["after_collect"]:
func_name = each.get("func")
# 注入查询条件
each.update(basic_filter)
LOGGER.info(f"处理器(after_collect): {func_name} 正在执行...")
processor_dict[func_name](**each)
LOGGER.info("处理器(after_collect): 执行完毕!")
# 分发器执行
LOGGER.info("分发器开始执行!")
send_doc(sender_conf)
LOGGER.info("分发器执行完毕!")
# 备份器执行
LOGGER.info("备份器开始执行!")
backup_doc(backup_conf)
LOGGER.info("备份器执行完毕!")
except Exception as e:
LOGGER.error(f"执行失败！{e}")
from src.utils import LOGGER
def schedule_task():
"""
更新持久化订阅的公众号最新文章
:return:
"""
# 抓取最新的文章，然后持久化到数据库
update_wechat_doc()
# 更新广告标签
update_ads_tag()
# 文章分发
send_doc()
if __name__ == "__main__":
# 初次启动执行即可
run_wechat_name_spider()
# 每日抓取公众号最新文章并更新广告标签
schedule.every().day.at("07:10").do(schedule_task)
schedule.every().day.at("11:10").do(schedule_task)
schedule.every().day.at("16:10").do(schedule_task)
schedule.every().day.at("20:10").do(schedule_task)
schedule.every().day.at("23:10").do(schedule_task)
LOGGER.info("Schedule started successfully :)")
LOGGER.info("Schedule time:\n 07:10 \n 11:10 \n 16:10 \n 20:10 \n 23:10")
while True:
schedule.run_pending()
time.sleep(1)
