feature:全球政治经济新闻
This commit is contained in:
liuwei
7
plugins/global_news/__init__.py
Normal file
7
plugins/global_news/__init__.py
Normal file
@@ -0,0 +1,7 @@
|
||||
# 从当前包的main模块导入GlobalNewsPlugin类
|
||||
from .main import GlobalNewsPlugin
|
||||
|
||||
# 提供get_plugin函数，返回插件实例
|
||||
def get_plugin():
|
||||
"""获取插件实例"""
|
||||
return GlobalNewsPlugin()
|
||||
6
plugins/global_news/config.toml
Normal file
6
plugins/global_news/config.toml
Normal file
@@ -0,0 +1,6 @@
|
||||
enable = true
|
||||
command = ["全球新闻", "国际新闻", "环球新闻", "政经新闻", "政治经济新闻"]
|
||||
command-format = """
|
||||
🌍全球新闻指令：
|
||||
全球新闻 - 获取最新的全球政治经济新闻
|
||||
"""
|
||||
191
plugins/global_news/main.py
Normal file
191
plugins/global_news/main.py
Normal file
@@ -0,0 +1,191 @@
|
||||
import logging
|
||||
import asyncio
|
||||
import threading
|
||||
import time # 添加这一行
|
||||
from typing import Dict, Any, List, Optional, Tuple
|
||||
|
||||
from wcferry import Wcf
|
||||
|
||||
from plugin_common.message_plugin_interface import MessagePluginInterface
|
||||
from plugin_common.plugin_interface import PluginStatus
|
||||
from utils.decorator.plugin_decorators import plugin_stats_decorator
|
||||
from utils.robot_cmd.robot_command import Feature, PermissionStatus, GroupBotManager
|
||||
from utils.decorator.points_decorator import plugin_points_cost
|
||||
from utils.ai.dify_news_analyze import dify_news_title_analyze
|
||||
from utils.markdown_to_image import convert_md_str_to_image
|
||||
|
||||
# 导入新闻抓取函数
|
||||
from .news_crawler import nbc, cnn, abc, fox, bbc
|
||||
|
||||
|
||||
class GlobalNewsPlugin(MessagePluginInterface):
|
||||
"""全球政治经济新闻插件"""
|
||||
|
||||
@property
|
||||
def name(self) -> str:
|
||||
return "全球政治经济新闻"
|
||||
|
||||
@property
|
||||
def version(self) -> str:
|
||||
return "1.0.0"
|
||||
|
||||
@property
|
||||
def description(self) -> str:
|
||||
return "提供全球政治经济新闻，支持多个国际新闻源"
|
||||
|
||||
@property
|
||||
def author(self) -> str:
|
||||
return "Trae AI"
|
||||
|
||||
@property
|
||||
def command_prefix(self) -> Optional[str]:
|
||||
return "" # 不需要前缀，直接匹配命令
|
||||
|
||||
@property
|
||||
def commands(self) -> List[str]:
|
||||
return self._commands
|
||||
|
||||
def __init__(self):
|
||||
super().__init__()
|
||||
self._news_tasks = {} # 存储正在进行的新闻抓取任务
|
||||
|
||||
def initialize(self, context: Dict[str, Any]) -> bool:
|
||||
"""初始化插件"""
|
||||
self.LOG = logging.getLogger(f"Plugin.{self.name}")
|
||||
self.LOG.info(f"正在初始化 {self.name} 插件...")
|
||||
|
||||
# 保存上下文对象
|
||||
self.wcf = context.get("wcf")
|
||||
self.event_system = context.get("event_system")
|
||||
self.message_util = context.get("message_util")
|
||||
|
||||
self._commands = self._config.get("GlobalNews", {}).get("command", ["全球新闻", "国际新闻", "环球新闻", "政经新闻"])
|
||||
self.command_format = self._config.get("GlobalNews", {}).get("command-format", "全球新闻 - 获取最新的全球政治经济新闻")
|
||||
self.enable = self._config.get("GlobalNews", {}).get("enable", True)
|
||||
|
||||
self.LOG.info(f"[{self.name}] 插件初始化完成，指令：{self._commands}")
|
||||
return True
|
||||
|
||||
def start(self) -> bool:
|
||||
"""启动插件"""
|
||||
self.LOG.info(f"[{self.name}] 插件已启动")
|
||||
self.status = PluginStatus.RUNNING
|
||||
return True
|
||||
|
||||
def stop(self) -> bool:
|
||||
"""停止插件"""
|
||||
self.LOG.info(f"[{self.name}] 插件已停止")
|
||||
self.status = PluginStatus.STOPPED
|
||||
return True
|
||||
|
||||
def can_process(self, message: Dict[str, Any]) -> bool:
|
||||
"""检查是否可以处理该消息"""
|
||||
if not self.enable:
|
||||
return False
|
||||
|
||||
content = str(message.get("content", "")).strip()
|
||||
command = content.split(" ")[0]
|
||||
|
||||
return command in self._commands
|
||||
|
||||
@plugin_stats_decorator(plugin_name="全球政治经济新闻")
|
||||
@plugin_points_cost(5, "全球新闻消耗积分", Feature.NEWS)
|
||||
def process_message(self, message: Dict[str, Any]) -> Tuple[bool, Optional[str]]:
|
||||
"""处理消息"""
|
||||
content = str(message.get("content", "")).strip()
|
||||
self.LOG.info(f"插件执行： {self.name}：{content}")
|
||||
sender = message.get("sender")
|
||||
roomid = message.get("roomid", "")
|
||||
wcf: Wcf = message.get("wcf")
|
||||
gbm: GroupBotManager = message.get("gbm")
|
||||
|
||||
# 检查权限
|
||||
if roomid and gbm.get_group_permission(roomid, Feature.NEWS) == PermissionStatus.DISABLED:
|
||||
return False, "没有权限"
|
||||
|
||||
# 生成唯一任务ID
|
||||
task_id = f"{sender}_{roomid}_{int(time.time())}"
|
||||
|
||||
# 发送等待消息
|
||||
wcf.send_text("🌍正在获取全球新闻，请稍候...",
|
||||
(roomid if roomid else sender), sender)
|
||||
|
||||
# 启动异步任务
|
||||
self._start_news_task(task_id, sender, roomid, wcf)
|
||||
|
||||
return True, "新闻获取任务已启动"
|
||||
|
||||
def _start_news_task(self, task_id: str, sender: str, roomid: str, wcf: Wcf):
|
||||
"""启动异步新闻获取任务"""
|
||||
thread = threading.Thread(
|
||||
target=self._fetch_news_thread,
|
||||
args=(task_id, sender, roomid, wcf)
|
||||
)
|
||||
thread.daemon = True
|
||||
thread.start()
|
||||
self._news_tasks[task_id] = thread
|
||||
self.LOG.info(f"启动新闻获取任务: {task_id}")
|
||||
|
||||
def _fetch_news_thread(self, task_id: str, sender: str, roomid: str, wcf: Wcf):
|
||||
"""在单独的线程中运行异步新闻获取任务"""
|
||||
try:
|
||||
loop = asyncio.new_event_loop()
|
||||
asyncio.set_event_loop(loop)
|
||||
news_result = loop.run_until_complete(self._fetch_news_async())
|
||||
loop.close()
|
||||
|
||||
# 处理结果
|
||||
if news_result:
|
||||
# 发送新闻图片
|
||||
receiver = roomid if roomid else sender
|
||||
wcf.send_image(news_result, receiver)
|
||||
wcf.send_text("🌍全球新闻获取完成！", receiver, sender)
|
||||
else:
|
||||
wcf.send_text("❌获取新闻失败，请稍后再试",
|
||||
(roomid if roomid else sender), sender)
|
||||
except Exception as e:
|
||||
self.LOG.error(f"新闻获取任务出错: {e}")
|
||||
wcf.send_text(f"❌获取新闻出错: {str(e)}",
|
||||
(roomid if roomid else sender), sender)
|
||||
finally:
|
||||
# 清理任务
|
||||
if task_id in self._news_tasks:
|
||||
del self._news_tasks[task_id]
|
||||
|
||||
async def _fetch_news_async(self) -> str:
|
||||
"""异步获取所有新闻源的新闻"""
|
||||
try:
|
||||
# 创建所有新闻源的任务
|
||||
tasks = [
|
||||
self._run_in_executor(nbc),
|
||||
self._run_in_executor(cnn),
|
||||
self._run_in_executor(abc),
|
||||
self._run_in_executor(fox),
|
||||
self._run_in_executor(bbc)
|
||||
]
|
||||
|
||||
# 并行执行所有任务
|
||||
results = await asyncio.gather(*tasks)
|
||||
|
||||
# 合并结果
|
||||
news_titles = "\n".join(results)
|
||||
|
||||
# 使用AI分析新闻
|
||||
markdown_news = await self._run_in_executor(
|
||||
dify_news_title_analyze, news_titles
|
||||
)
|
||||
|
||||
# 转换为图片
|
||||
image_path = await self._run_in_executor(
|
||||
convert_md_str_to_image, markdown_news, "news_output.png"
|
||||
)
|
||||
|
||||
return image_path
|
||||
except Exception as e:
|
||||
self.LOG.error(f"异步获取新闻失败: {e}")
|
||||
return ""
|
||||
|
||||
async def _run_in_executor(self, func, *args):
|
||||
"""在线程池中运行同步函数"""
|
||||
loop = asyncio.get_event_loop()
|
||||
return await loop.run_in_executor(None, func, *args)
|
||||
307
plugins/global_news/news_crawler.py
Normal file
307
plugins/global_news/news_crawler.py
Normal file
@@ -0,0 +1,307 @@
|
||||
# -*- coding: utf-8 -*-
|
||||
"""
|
||||
Program: Global News Crawler
|
||||
Author: Trae AI (based on MrCrawL's work)
|
||||
Created Date: 2024-05-01
|
||||
"""
|
||||
import requests
|
||||
from time import localtime, sleep
|
||||
from lxml import etree
|
||||
import logging
|
||||
from datetime import datetime
|
||||
import time
|
||||
|
||||
# 配置日志
|
||||
logging.basicConfig(
|
||||
level=logging.INFO,
|
||||
format='%(asctime)s - %(levelname)s - %(message)s',
|
||||
handlers=[
|
||||
logging.FileHandler(f'global_news_{datetime.now().strftime("%Y%m%d")}.log'),
|
||||
logging.StreamHandler()
|
||||
]
|
||||
)
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
# 请求配置
|
||||
HEADERS = {
|
||||
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
|
||||
}
|
||||
TIMEOUT = 10
|
||||
MAX_RETRIES = 3
|
||||
NEWS_LIMIT = 30
|
||||
|
||||
|
||||
def get_time():
|
||||
date_ = f'{str(localtime().tm_year).zfill(4)}-{str(localtime().tm_mon).zfill(2)}-{str(localtime().tm_mday).zfill(2)}'
|
||||
return date_
|
||||
|
||||
|
||||
def title_tidy(title_list):
|
||||
t_index = []
|
||||
for i in range(1, len(title_list)):
|
||||
if title_list[i] == title_list[i - 1]: t_index.append(i)
|
||||
t_index.reverse()
|
||||
for i in range(len(t_index)): title_list.pop(t_index[i])
|
||||
return title_list
|
||||
|
||||
|
||||
def safe_request(url, retry_count=0):
|
||||
"""安全的请求方法，包含重试机制"""
|
||||
try:
|
||||
response = requests.get(url, headers=HEADERS, timeout=TIMEOUT)
|
||||
response.raise_for_status()
|
||||
return response
|
||||
except requests.RequestException as e:
|
||||
if retry_count < MAX_RETRIES:
|
||||
logger.warning(f"请求失败，正在进行第{retry_count + 1}次重试: {url}")
|
||||
sleep(1)
|
||||
return safe_request(url, retry_count + 1)
|
||||
else:
|
||||
logger.error(f"请求失败: {url}, 错误: {str(e)}")
|
||||
return None
|
||||
|
||||
|
||||
def nbc():
|
||||
logger.info("开始获取NBC新闻")
|
||||
try:
|
||||
url = 'https://www.nbcnews.com/'
|
||||
response = safe_request(url)
|
||||
if not response:
|
||||
return "获取NBC新闻失败"
|
||||
|
||||
html = etree.HTML(response.text)
|
||||
href = html.xpath('//h2/a/@href')
|
||||
href = title_tidy(href)
|
||||
|
||||
msg = ''
|
||||
count = 0
|
||||
|
||||
for url in href[:NEWS_LIMIT]:
|
||||
try:
|
||||
response = safe_request(url)
|
||||
if not response:
|
||||
continue
|
||||
|
||||
html = etree.HTML(response.text)
|
||||
title = html.xpath('//h1/text()')
|
||||
|
||||
if not title:
|
||||
logger.warning(f'跳过视频或其他类型新闻: {url}')
|
||||
continue
|
||||
|
||||
title = title[0]
|
||||
msg += f'Title: {title}. Link: {url}\n'
|
||||
count += 1
|
||||
sleep(0.1)
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"处理新闻失败: {url}, 错误: {str(e)}")
|
||||
continue
|
||||
|
||||
logger.info(f"NBC新闻获取完成，共获取{count}条")
|
||||
return msg
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"获取NBC新闻失败: {str(e)}")
|
||||
return "获取新闻失败，请查看日志了解详情"
|
||||
|
||||
|
||||
def cnn():
|
||||
logger.info("开始获取CNN新闻")
|
||||
try:
|
||||
head = 'https://www.cnn.com'
|
||||
response = safe_request(head + '/')
|
||||
if not response:
|
||||
return "获取CNN新闻失败"
|
||||
|
||||
html = etree.HTML(response.text)
|
||||
href = html.xpath('//a[@data-link-type="article"]/@href')
|
||||
href = title_tidy(href)
|
||||
|
||||
msg = ''
|
||||
count = 0
|
||||
|
||||
for url in href[:NEWS_LIMIT]:
|
||||
try:
|
||||
full_url = head + url
|
||||
response = safe_request(full_url)
|
||||
if not response:
|
||||
continue
|
||||
|
||||
html = etree.HTML(response.text)
|
||||
title = html.xpath('//h1[@data-editable="headlineText"]/text()')
|
||||
|
||||
if not title:
|
||||
logger.warning(f'跳过视频或其他类型新闻: {full_url}')
|
||||
continue
|
||||
|
||||
title = title[0].strip()
|
||||
msg += f'Title: {title}. Link: {full_url}\n'
|
||||
count += 1
|
||||
sleep(0.1)
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"处理新闻失败: {full_url}, 错误: {str(e)}")
|
||||
continue
|
||||
|
||||
logger.info(f"CNN新闻获取完成，共获取{count}条")
|
||||
return msg
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"获取CNN新闻失败: {str(e)}")
|
||||
return "获取新闻失败，请查看日志了解详情"
|
||||
|
||||
|
||||
def abc():
|
||||
logger.info("开始获取ABC新闻")
|
||||
try:
|
||||
head = 'https://abcnews.go.com' # 移除末尾的斜杠
|
||||
response = safe_request(head)
|
||||
if not response:
|
||||
return "获取ABC新闻失败"
|
||||
|
||||
html = etree.HTML(response.text)
|
||||
href1 = html.xpath('//div[@class="HeadlinesTrio"]/a/@href')
|
||||
href2 = html.xpath(
|
||||
'//div[@class="title card"]/a[@class="AnchorLink"]/@href | //div[@class="title"]/a[@class="AnchorLink"]/@href')
|
||||
href3 = html.xpath('//a[@target="_self"]/@href')
|
||||
href4 = html.xpath('//a[@class="AnchorLink VideoTile"]/@href')
|
||||
href = title_tidy(href1 + href2 + href3 + href4)
|
||||
|
||||
msg = ''
|
||||
count = 0
|
||||
|
||||
for url in href[:NEWS_LIMIT]:
|
||||
try:
|
||||
# 处理URL格式
|
||||
if url.startswith('http'):
|
||||
full_url = url
|
||||
elif url.startswith('//'):
|
||||
full_url = 'https:' + url
|
||||
else:
|
||||
full_url = head + ('' if url.startswith('/') else '/') + url
|
||||
|
||||
response = safe_request(full_url)
|
||||
if not response:
|
||||
continue
|
||||
|
||||
html = etree.HTML(response.text)
|
||||
title = html.xpath('//div[@data-testid="prism-headline"]/h1/text()')
|
||||
|
||||
if not title:
|
||||
logger.warning(f'跳过视频或其他类型新闻: {full_url}')
|
||||
continue
|
||||
|
||||
title = title[0]
|
||||
msg += f'Title: {title}. Link: {full_url}\n'
|
||||
count += 1
|
||||
sleep(0.1)
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"处理新闻失败: {full_url}, 错误: {str(e)}")
|
||||
continue
|
||||
|
||||
logger.info(f"ABC新闻获取完成，共获取{count}条")
|
||||
return msg
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"获取ABC新闻失败: {str(e)}")
|
||||
return "获取新闻失败，请查看日志了解详情"
|
||||
|
||||
|
||||
def fox():
|
||||
logger.info("开始获取FOX新闻")
|
||||
try:
|
||||
head = 'https://www.foxnews.com/'
|
||||
response = safe_request(head)
|
||||
if not response:
|
||||
return "获取FOX新闻失败"
|
||||
|
||||
html = etree.HTML(response.text)
|
||||
href = html.xpath('//h3[@class="title"]/a/@href')
|
||||
href = title_tidy(href)
|
||||
|
||||
msg = ''
|
||||
count = 0
|
||||
|
||||
for url in href[:NEWS_LIMIT]:
|
||||
try:
|
||||
if url[0:4] != 'http':
|
||||
url = 'https:' + url
|
||||
|
||||
response = safe_request(url)
|
||||
if not response:
|
||||
continue
|
||||
|
||||
html = etree.HTML(response.text)
|
||||
title = html.xpath('//h1[@itemprop="headline"]/text()')
|
||||
|
||||
if not title:
|
||||
logger.warning(f'跳过视频或其他类型新闻: {url}')
|
||||
continue
|
||||
|
||||
title = title[0]
|
||||
msg += f'Title: {title}. Link: {url}\n'
|
||||
count += 1
|
||||
sleep(0.1)
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"处理新闻失败: {url}, 错误: {str(e)}")
|
||||
continue
|
||||
|
||||
logger.info(f"FOX新闻获取完成，共获取{count}条")
|
||||
return msg
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"获取FOX新闻失败: {str(e)}")
|
||||
return "获取新闻失败，请查看日志了解详情"
|
||||
|
||||
|
||||
def bbc():
|
||||
logger.info("开始获取BBC新闻")
|
||||
try:
|
||||
head = 'https://www.bbc.com'
|
||||
response = safe_request(head + '/')
|
||||
if not response:
|
||||
return "获取BBC新闻失败"
|
||||
|
||||
html = etree.HTML(response.text)
|
||||
href = html.xpath(
|
||||
'//h2[@data-testid="card-headline"]/../../../../../@href | //h2[@data-testid="card-headline"]/../../../../@href')
|
||||
href = title_tidy(href)
|
||||
|
||||
msg = ''
|
||||
count = 0
|
||||
|
||||
for url in href[:NEWS_LIMIT]:
|
||||
try:
|
||||
if url[0:4] == 'http':
|
||||
continue
|
||||
|
||||
full_url = head + url
|
||||
response = safe_request(full_url)
|
||||
if not response:
|
||||
continue
|
||||
|
||||
html = etree.HTML(response.text)
|
||||
title = html.xpath('//div[@data-component="headline-block"]/h1/text()')
|
||||
|
||||
if not title:
|
||||
logger.warning(f'跳过视频或其他类型新闻: {full_url}')
|
||||
continue
|
||||
|
||||
title = title[0]
|
||||
msg += f'Title: {title}. Link: {full_url}\n'
|
||||
count += 1
|
||||
sleep(0.1)
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"处理新闻失败: {full_url}, 错误: {str(e)}")
|
||||
continue
|
||||
|
||||
logger.info(f"BBC新闻获取完成，共获取{count}条")
|
||||
return msg
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"获取BBC新闻失败: {str(e)}")
|
||||
return "获取新闻失败，请查看日志了解详情"
|
||||
Reference in New Issue
Block a user