feature:全球政治经济新闻

base/func_english_news.py

@@ -6,11 +6,35 @@ Created Date: 2024-01-21
 Last Modified: 2024-03-24
 Modified by: MrCrawL
 """
+from utils.ai.dify_news_analyze import dify_news_title_analyze
+from utils.markdown_to_image import convert_md_str_to_image
+
 '''Existing problem: text with hyperlink won't be saved'''
 
 import requests
 from time import localtime, sleep
 from lxml import etree
+import logging
+from datetime import datetime
+
+# 配置日志
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(levelname)s - %(message)s',
+    handlers=[
+        logging.FileHandler(f'news_crawler_{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():
@@ -18,7 +42,6 @@ def get_time():
     return date_
 
 
-# delete duplicated
 def title_tidy(title_list):
     t_index = []
     for i in range(1, len(title_list)):
@@ -28,7 +51,6 @@ def title_tidy(title_list):
     return title_list
 
 
-# tidy text, seems a little bit redundant
 def text_tidy(p_text):
     text_ = p_text.replace('’', "'")
     text_ = text_.replace(' \n\n', ' ')
@@ -46,218 +68,275 @@ def text_tidy(p_text):
     return text_
 
 
-def save(text, file_name, mode='w', encoding='utf-8'):
+def safe_request(url, retry_count=0):
-    with open(f'{file_name}.txt', mode, encoding=encoding) as f: f.write(text)
+    """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():
-    url = 'https://www.nbcnews.com/'
+    logger.info("开始获取NBC新闻")
-    res = requests.get(url)
+    try:
-    html = etree.HTML(res.text)
+        url = 'https://www.nbcnews.com/'
-    href = html.xpath('//h2/a/@href')
+        response = safe_request(url)
-    href = title_tidy(href)
+        if not response:
-    # quant = int(input(f'There are {len(href)} pieces detected. How many would you download:'))
+            return "获取NBC新闻失败"
-    # if quant > len(href) or quant < 1:
+
-    #     print("Outnumber!")
+        html = etree.HTML(response.text)
-    #     quit()
+        href = html.xpath('//h2/a/@href')
-    count = 0
+        href = title_tidy(href)
-    # save('', f'NBC_news_title_{get_time()}')
+
-    # save('', f'NBC_news_text_{get_time()}')
+        msg = ''
-    msg =''
+        count = 0
-    for i in range(30):
+
-        url = href[i]
+        for url in href[:NEWS_LIMIT]:
-        sleep(0.1)  # delete to speed up
+            try:
-        res = requests.get(url)
+                response = safe_request(url)
-        html = etree.HTML(res.text)
+                if not response:
-        title = html.xpath('//h1/text()')
+                    continue
-        if len(title) == 0:
+
-            print(f'Video or other news. Link: {url}')
+                html = etree.HTML(response.text)
-            continue
+                title = html.xpath('//h1/text()')
-        title = title[0]
+
-        author = html.xpath('//span[@class="byline-name"]/a/text() | //span[@class="byline-name" and not(a)]/text()')
+                if not title:
-        author = ', '.join(author)
+                    logger.warning(f'跳过视频或其他类型新闻: {url}')
-        text = html.xpath('//p[@class=""]/text()')
+                    continue
-        text = '\n\n'.join(text)
+
-        text = text_tidy(text)
+                title = title[0]
-        count += 1
+                msg += f'Title: {title}. Link: {url}\n'
-        # save(f'Title: {title}\nLink: {url}\n\n', f'NBC_news_title_{get_time()}', 'a')  # news title
+                count += 1
-        # save(f'Title: {title}\n\nOrigin: {url}\n\nAuthor: {author}\n\n\n', f'NBC_news_text_{get_time()}', 'a')
+                sleep(0.1)
-        # save(f'{text}' + '\n\n------------------------------\n\n', f'NBC_news_text_{get_time()}', 'a')
+
-        # print(f'Title: {title}. Link: {href[i]}.')
+            except Exception as e:
-        msg += f'Title: {title}. Link: {href[i]}.\n'
+                logger.error(f"处理新闻失败: {url}, 错误: {str(e)}")
-    return msg
+                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():
-    head = 'https://www.cnn.com'
+    logger.info("开始获取CNN新闻")
-    res = requests.get(head + '/')
+    try:
-    html = etree.HTML(res.text)
+        head = 'https://www.cnn.com'
-    href = html.xpath('//a[@data-link-type="article"]/@href')
+        response = safe_request(head + '/')
-    href = title_tidy(href)
+        if not response:
-    # quant = int(input(f'{len(href)} data detected. How many would you like to download:'))
+            return "获取CNN新闻失败"
-    # if quant > len(href) or quant < 1:
+
-    #     print("Outnumber!")
+        html = etree.HTML(response.text)
-    #     quit()
+        href = html.xpath('//a[@data-link-type="article"]/@href')
-    count = 0
+        href = title_tidy(href)
-    msg = ''
+
-    # save('', f'CNN_news_title_{get_time()}')
+        msg = ''
-    # save('', f'CNN_news_text_{get_time()}')
+        count = 0
-    for i in range(30):
+
-        url = head + href[i]
+        for url in href[:NEWS_LIMIT]:
-        sleep(0.1)  # delete to speed up
+            try:
-        res = requests.get(url)
+                full_url = head + url
-        html = etree.HTML(res.text)
+                response = safe_request(full_url)
-        title = html.xpath('//h1[@data-editable="headlineText"]/text()')
+                if not response:
-        if len(title) == 0:
+                    continue
-            print(f'Video or other news. Link: {url}')
+
-            continue
+                html = etree.HTML(response.text)
-        title = title[0].strip()
+                title = html.xpath('//h1[@data-editable="headlineText"]/text()')
-        author = html.xpath('//span[@class="byline__name"]/text()')
+
-        author = ', '.join(author)
+                if not title:
-        text = html.xpath('//p[@class="paragraph inline-placeholder"]/text()')
+                    logger.warning(f'跳过视频或其他类型新闻: {full_url}')
-        for k in range(len(text)): text[k].strip()
+                    continue
-        text = ''.join(text)
+
-        text = text_tidy(text)
+                title = title[0].strip()
-        count += 1
+                msg += f'Title: {title}. Link: {full_url}\n'
-        # save(f'Title: {title}\nLink: {url}\n\n', f'CNN_news_title_{get_time()}', 'a')  # news title
+                count += 1
-        # save(f'Title: {title}\n\nOrigin: {url}\n\nAuthor: {author}\n\n\n', f'CNN_news_text_{get_time()}', 'a')
+                sleep(0.1)
-        # save(f'{text}' + '\n\n------------------------------\n\n', f'CNN_news_text_{get_time()}', 'a')
+
-        # print(f'Title: {title}. Link: {url}')
+            except Exception as e:
-        msg +=f'Title: {title}. Link: {url}\n'
+                logger.error(f"处理新闻失败: {full_url}, 错误: {str(e)}")
-    # print(f'Files saved with {count} articles available.')
+                continue
-    return msg
+
+        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():
-    head = 'https://abcnews.go.com/'
+    logger.info("开始获取ABC新闻")
-    res = requests.get(head)
+    try:
-    html = etree.HTML(res.text)
+        head = 'https://abcnews.go.com'  # 移除末尾的斜杠
-    href1 = html.xpath('//div[@class="HeadlinesTrio"]/a/@href')
+        response = safe_request(head)
-    href2 = html.xpath('//div[@class="title card"]/a[@class="AnchorLink"]/@href | //div[@class="title"]/a[@class="AnchorLink"]/@href')
+        if not response:
-    href3 = html.xpath('//a[@target="_self"]/@href')
+            return "获取ABC新闻失败"
-    href4 = html.xpath('//a[@class="AnchorLink VideoTile"]/@href')
+
-    href = href1 + href2 + href3 + href4
+        html = etree.HTML(response.text)
-    href = title_tidy(href)
+        href1 = html.xpath('//div[@class="HeadlinesTrio"]/a/@href')
-    # quant = int(input(f'{len(href)} data detected. How many would you like to download:'))
+        href2 = html.xpath(
-    # if quant > len(href) or quant < 1:
+            '//div[@class="title card"]/a[@class="AnchorLink"]/@href | //div[@class="title"]/a[@class="AnchorLink"]/@href')
-    #     print("Outnumber!")
+        href3 = html.xpath('//a[@target="_self"]/@href')
-    #     quit()
+        href4 = html.xpath('//a[@class="AnchorLink VideoTile"]/@href')
-    count = 0
+        href = title_tidy(href1 + href2 + href3 + href4)
-    msg = ''
+
-    # save('', f'ABC_news_title_{get_time()}')
+        msg = ''
-    # save('', f'ABC_news_text_{get_time()}')
+        count = 0
-    for i in range(30):
+
-        url = href[i]
+        for url in href[:NEWS_LIMIT]:
-        sleep(0.1)  # delete to speed up
+            try:
-        res = requests.get(url)
+                # 处理URL格式
-        html = etree.HTML(res.text)
+                if url.startswith('http'):
-        title = html.xpath('//div[@data-testid="prism-headline"]/h1/text()')
+                    full_url = url
-        if len(title) == 0:
+                elif url.startswith('//'):
-            print(f'Video or other news. Link: {url}')
+                    full_url = 'https:' + url
-            continue
+                else:
-        title = title[0]
+                    full_url = head + ('' if url.startswith('/') else '/') + url
-        author = html.xpath('//a[@data-testid="prism-linkbase"]/text()')
+
-        author = ', '.join(author)
+                response = safe_request(full_url)
-        text = html.xpath('//div[@data-testid="prism-article-body"]/p/text()')
+                if not response:
-        text = '\n\n'.join(text)
+                    continue
-        text = text_tidy(text)
+
-        count += 1
+                html = etree.HTML(response.text)
-        # save(f'Title: {title}\nLink: {url}\n\n', f'ABC_news_title_{get_time()}', 'a')  # news title
+                title = html.xpath('//div[@data-testid="prism-headline"]/h1/text()')
-        # save(f'Title: {title}\n\nOrigin: {url}\n\nAuthor: {author}\n\n\n', f'ABC_news_text_{get_time()}', 'a')
+
-        # save(f'{text}' + '\n\n------------------------------\n\n', f'ABC_news_text_{get_time()}', 'a')
+                if not title:
-        # print(f'Title: {title}. Link: {url}')
+                    logger.warning(f'跳过视频或其他类型新闻: {full_url}')
-        msg +=f'Title: {title}. Link: {url}\n'
+                    continue
-    # print(f'Files saved with {count} articles available.')
+
-    return msg
+                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():
-    head = 'https://www.foxnews.com/'
+    logger.info("开始获取FOX新闻")
-    res = requests.get(head)
+    try:
-    html = etree.HTML(res.text)
+        head = 'https://www.foxnews.com/'
-    href = html.xpath('//h3[@class="title"]/a/@href')
+        response = safe_request(head)
-    href = title_tidy(href)
+        if not response:
-    # quant = int(input(f'{len(href)} data detected. How many would you like to download:'))
+            return "获取FOX新闻失败"
-    # if quant > len(href) or quant < 1:
+
-    #     print("Outnumber!")
+        html = etree.HTML(response.text)
-    #     quit()
+        href = html.xpath('//h3[@class="title"]/a/@href')
-    count = 0
+        href = title_tidy(href)
-    msg =''
+
-    # save('', f'FOX_news_title_{get_time()}')
+        msg = ''
-    # save('', f'FOX_news_text_{get_time()}')
+        count = 0
-    for i in range(30):
+
-        if href[i][0:4] != 'http': href[i] = 'https:' + href[i]
+        for url in href[:NEWS_LIMIT]:
-        url = href[i]
+            try:
-        sleep(0.1)  # delete to speed up
+                if url[0:4] != 'http':
-        res = requests.get(url)
+                    url = 'https:' + url
-        html = etree.HTML(res.text)
+
-        title = html.xpath('//h1[@itemprop="headline"]/text()')
+                response = safe_request(url)
-        if len(title) == 0:
+                if not response:
-            print(f'Video or other news. Link: {url}')
+                    continue
-            continue
+
-        title = title[0]
+                html = etree.HTML(response.text)
-        author = html.xpath('//a[@rel="author"]/strong/text()')
+                title = html.xpath('//h1[@itemprop="headline"]/text()')
-        author = ', '.join(author)
+
-        text = html.xpath('//div[@itemprop="articleBody"]/p/text()')
+                if not title:
-        text = '\n\n'.join(text)
+                    logger.warning(f'跳过视频或其他类型新闻: {url}')
-        text = text_tidy(text)
+                    continue
-        count += 1
+
-        # save(f'Title: {title}\nLink: {url}\n\n', f'FOX_news_title_{get_time()}', 'a')  # news title
+                title = title[0]
-        # save(f'Title: {title}\n\nOrigin: {url}\n\nAuthor: {author}\n\n\n', f'FOX_news_text_{get_time()}', 'a')
+                msg += f'Title: {title}. Link: {url}\n'
-        # save(f'{text}' + '\n\n------------------------------\n\n', f'FOX_news_text_{get_time()}', 'a')
+                count += 1
-        # print(f'Title: {title}. Link: {url}')
+                sleep(0.1)
-        msg +=f'Title: {title}. Link: {url}\n'
+
-    # print(f'Files saved with {count} articles available.')
+            except Exception as e:
-    return msg
+                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():
-    head = 'https://www.bbc.com'
+    logger.info("开始获取BBC新闻")
-    res = requests.get(head + '/')
+    try:
-    html = etree.HTML(res.text)
+        head = 'https://www.bbc.com'
-    href = html.xpath('//h2[@data-testid="card-headline"]/../../../../../@href | //h2[@data-testid="card-headline"]/../../../../@href')
+        response = safe_request(head + '/')
-    href = title_tidy(href)
+        if not response:
-    # quant = int(input(f'{len(href)} data detected. How many would you like to download:'))
+            return "获取BBC新闻失败"
-    # if quant > len(href) or quant < 1:
-    #     print("Outnumber!")
-    #     quit()
-    count = 0
-    msg =''
-    # save('', f'BBC_news_title_{get_time()}')
-    # save('', f'BBC_news_text_{get_time()}')
-    for i in range(30):
-        if href[i][0:4] == 'http': continue
-        url = head + href[i]
-        sleep(0.1)  # delete to speed up
-        print(url)
-        res = requests.get(url)
-        html = etree.HTML(res.text)
-        title = html.xpath('//div[@data-component="headline-block"]/h1/text()')
-        if len(title) == 0:
-            # print(f'Video or other news. Link: {url}')
-            continue
-        title = title[0]
-        # author = html.xpath('//div[@data-testid="byline"]/div/span[@data-testid="byline-name"]/text()')
-        # author = ', '.join(author)
-        # text = html.xpath('//div[@data-component="text-block"]/p/b/text() | //div[@data-component="text-block"]/p/text()')
-        # text = '\n\n'.join(text)
-        # text = text_tidy(text)
-        count += 1
-        # save(f'Title: {title}\nLink: {url}\n\n', f'BBC_news_title_{get_time()}', 'a')  # news title
-        # save(f'Title: {title}\n\nOrigin: {url}\n\nAuthor: {author}\n\n\n', f'BBC_news_text_{get_time()}', 'a')
-        # save(f'{text}' + '\n\n------------------------------\n\n', f'BBC_news_text_{get_time()}', 'a')
-        # print(f'Title: {title}. Link: {url}')
 
-        msg +=f'Title: {title}. Link: {url}\n'
+        html = etree.HTML(response.text)
-    # print(f'Files saved with {count} articles available.')
+        href = html.xpath(
-    return  msg
+            '//h2[@data-testid="card-headline"]/../../../../../@href | //h2[@data-testid="card-headline"]/../../../../@href')
+        href = title_tidy(href)
 
-if __name__ == '__main__':
+        msg = ''
-    # Hello, World! :)
+        count = 0
-    # news = input('Choose news site["nbc","cnn","abc","fox","bbc"]:').lower()
+
-    # if news == 'nbc': nbc()
+        for url in href[:NEWS_LIMIT]:
-    # elif news == 'cnn': cnn()
+            try:
-    # elif news == 'abc': abc()
+                if url[0:4] == 'http':
-    # elif news == 'fox': fox()
+                    continue
-    # elif news == 'bbc': bbc()
+
-    # else:
+                full_url = head + url
-    #     print('Oops! It seems a wrong input. Please retry...')
+                response = safe_request(full_url)
-    #     sleep(2)
+                if not response:
-    print(bbc())
+                    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"
+
+
+def all_english_news():
+    news_titles = ""
+    news_titles += nbc() + "\n"
+    news_titles += cnn() + "\n"
+    news_titles += abc() + "\n"
+    news_titles += fox() + "\n"
+    news_titles += bbc() + "\n"
+    markdown_news = dify_news_title_analyze(news_titles)
+    spath = convert_md_str_to_image(markdown_news, "news_output.png")
+    return spath

base/func_news.py

@@ -11,6 +11,7 @@ import requests
 from lxml import etree
 
 from base import func_english_news
+from utils.ai.dify_news_analyze import dify_news_title_analyze
 
 
 class News(object):
@@ -70,10 +71,10 @@ class News(object):
                 post = response.json()
                 # 提取content列表 - 避免使用str作为变量名
                 content_list = post.get('data', {}).get('cards', [])
-                
+
                 if content_list and len(content_list) > 0:
                     news_items = content_list[0].get('content', [])
-                    
+
                     # 遍历列表，并格式化每个字典的title, url，然后添加到output字符串中
                     for index, article in enumerate(news_items, start=1):
                         if isinstance(article, dict) and 'word' in article:
@@ -81,34 +82,34 @@ class News(object):
                             # url = article.get('url', '')
                             # 使用f-string格式化字符串，并添加到output中
                             output += f"{index} :#{title}\n"
-                
+
                 # 输出最终的字符串
                 return output
             else:
                 self.LOG.error(f"获取百度新闻失败，状态码: {response.status_code}")
                 return "获取百度新闻失败，请稍后再试"
-                
+
         except Exception as e:
             self.LOG.error(f"获取百度新闻时出错: {e}")
             return f"获取百度新闻时出错: {e}"
 
-    def get_eng_news(self,website):
+    def get_eng_news(self, website):
         if website == 'nbc':
-           return func_english_news.nbc()
+            return func_english_news.nbc()
         elif website == 'cnn':
-            return  func_english_news.cnn()
+            return func_english_news.cnn()
         elif website == 'abc':
             return func_english_news.abc()
         elif website == 'fox':
             return func_english_news.fox()
         elif website == 'bbc':
-            return  func_english_news.bbc()
+            return func_english_news.bbc()
+
 
 
 if __name__ == "__main__":
     news = News()
-    print(news.get_baidu_news())
     # # msg = "@水牛-分身 今日百度新闻"
     # # q = re.sub(r"@.*?[\u2005|\s]", "", msg).replace(" ", "")
     # # print(q)
-    # print(news.get_eng_news('nbc'))
+    # print(news.get_eng_news('nbc'))

plugins/global_news/__init__.py

@@ -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()

plugins/global_news/config.toml

@@ -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
+"""

plugins/global_news/main.py

@@ -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)

plugins/global_news/news_crawler.py

@@ -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"

utils/ai/dify_news_analyze.py

@@ -0,0 +1,73 @@
+#
+# curl -X POST 'http://192.168.2.240/v1/chat-messages' \
+# --header 'Authorization: Bearer {api_key}' \
+# --header 'Content-Type: application/json' \
+# --data-raw '{
+#     "inputs": {},
+#     "query": "What are the specs of the iPhone 13 Pro Max?",
+#     "response_mode": "streaming",
+#     "conversation_id": "",
+#     "user": "abc-123",
+#     "files": [
+#       {
+#         "type": "image",
+#         "transfer_method": "remote_url",
+#         "url": "https://cloud.dify.ai/logo/logo-site.png"
+#       }
+#     ]
+# }'
+import json
+
+import requests
+
+
+def dify_news_title_analyze(content):
+    # 设置Authorization和URL
+    authorization = "Bearer app-rhhKkbvHd2IAQoGX7xTzXZJj"  # 请替换为真实的Authorization token
+    url = 'http://192.168.2.240/v1/chat-messages'
+
+    data = {
+        "response_mode": "blocking",
+        "conversation_id": "",
+        "inputs": {},
+        "query": content,
+        "user": "a-bot"
+    }
+
+    # 设置请求头
+    headers = {
+        "Content-Type": "application/json; charset=utf-8",
+        "Authorization": authorization
+    }
+
+    # 发送POST请求
+    response = requests.post(url, headers=headers, data=json.dumps(data), )
+    response.encoding = 'utf-8'
+
+    # 输出响应内容
+    print(response.status_code)
+    print(response.json())
+    return extract_content(response.json())
+
+
+def extract_content(data):
+    """解析API响应内容
+    Args:
+        data: API返回的响应数据，可以是字典或字符串
+    Returns:
+        str: 提取的answer内容
+    """
+    try:
+        # 如果是字符串，尝试解析为字典
+        if isinstance(data, str):
+            data = json.dumps(data)
+        # 如果是字典，直接获取answer
+        if isinstance(data, dict):
+            answer = data.get('answer', '')
+            if answer:
+                return answer
+                
+        return None
+    except Exception as e:
+        print(f"解析响应失败: {str(e)}")
+        return None

utils/robot_cmd/robot_command.py

@@ -43,7 +43,8 @@ class Feature(Enum):
     GROUP_ADD = 16, "加群提醒功能"
     DOUYIN_PARSER = 17, "抖音链接转视频功能"
     GROUP_MEMBER_CHANGE = 18, "群成员变更提醒功能"
-    KID_PHOTO_EXTRACT =19, "儿童照片提取转发功能"  # 小朋友照片提取功能
+    KID_PHOTO_EXTRACT = 19, "儿童照片提取转发功能"  # 小朋友照片提取功能
+    NEWS = 20, "全球政治经济新闻"
 
     def __new__(cls, value, description):
         obj = object.__new__(cls)
@@ -240,11 +241,11 @@ class GroupBotManager:
             str: 格式化的已启用功能列表字符串
         """
         enabled_features = []
-        
+
         # 检查群是否在列表中
         if group_id not in GroupBotManager.local_cache["group_list"]:
             return "该群未启用机器人功能"
-            
+
         # 遍历所有功能，检查哪些已启用且包含指令
         for feature in Feature:
             status = GroupBotManager.get_group_permission(group_id, feature)
@@ -255,18 +256,18 @@ class GroupBotManager:
                     "name": feature.name,
                     "description": feature.description
                 })
-        
+
         # 如果没有启用任何带指令的功能
         if not enabled_features:
             return "该群未启用任何带指令的功能"
-            
+
         # 构建格式化的字符串
         result = f"群功能菜单：\n"
         for feature in enabled_features:
             result += f"{feature['id']}.{feature['description']}\n"
-                
+
         return result
-        
+
     @staticmethod
     def get_group_list():
         """返回所有启用了群机器人的群组清单，格式为集合"""