Upload free_ask_internet.py

free_ask_internet.py

@@ -26,19 +26,16 @@ def extract_url_content(url):
     return {"url": url, "content": content}
 
 
-def search_web_ref(query: str, lang="zh-CN", debug=False):
+def search_web_ref(query:str, debug=False):
+ 
     content_list = []
 
     try:
+
         safe_string = urllib.parse.quote_plus(":all !general " + query)
 
         searxng_url = os.environ.get('SEARXNG_URL')
-        params = {
-            "q": safe_string,
-            "language": "zh-CN" if lang.startswith("zh") else "en-US",
-            "time_range": "day"  # 限制当天结果
-        }
-        response = requests.get(searxng_url, params=params)
+        response = requests.get(searxng_url + '?q=' + safe_string + '&format=json')
         response.raise_for_status()
         search_results = response.json()
  
@@ -46,6 +43,7 @@ def search_web_ref(query: str, lang="zh-CN", debug=False):
             print("JSON Response:")
             pprint(search_results)
         pedding_urls = []
+
         conv_links = []
 
         if search_results.get('results'):
@@ -58,15 +56,15 @@ def search_web_ref(query: str, lang="zh-CN", debug=False):
                 if url:
                     url_parsed = urlparse(url)
                     domain = url_parsed.netloc
-                    icon_url = url_parsed.scheme + '://' + url_parsed.netloc + '/favicon.ico'
+                    icon_url =  url_parsed.scheme + '://' + url_parsed.netloc + '/favicon.ico'
                     site_name = tldextract.extract(url).domain
  
                 conv_links.append({
-                    'site_name': site_name,
-                    'icon_url': icon_url,
-                    'title': name,
-                    'url': url,
-                    'snippet': snippet
+                    'site_name':site_name,
+                    'icon_url':icon_url,
+                    'title':name,
+                    'url':url,
+                    'snippet':snippet
                 })
 
             results = []
@@ -74,28 +72,29 @@ def search_web_ref(query: str, lang="zh-CN", debug=False):
 
             executor = ThreadPoolExecutor(max_workers=10) 
             for url in pedding_urls:
-                futures.append(executor.submit(extract_url_content, url))
+                futures.append(executor.submit(extract_url_content,url))
             try:
                 for future in futures:
                     res = future.result(timeout=5)
                     results.append(res)
             except concurrent.futures.TimeoutError:
                 print("任务执行超时")
-                executor.shutdown(wait=False, cancel_futures=True)
+                executor.shutdown(wait=False,cancel_futures=True)
 
             for content in results:
                 if content and content.get('content'):
+                    
                     item_dict = {
-                        "url": content.get('url'),
+                        "url":content.get('url'),
                         "content": content.get('content'),
-                        "length": len(content.get('content'))
+                        "length":len(content.get('content'))
                     }
                     content_list.append(item_dict)
                 if debug:
                     print("URL: {}".format(url))
                     print("=================")
  
-        return content_list
+        return  content_list
     except Exception as ex:
         raise ex
 
@@ -214,7 +213,7 @@ def chat(prompt, model: str, stream=True, debug=False):
         print(total_content)
 
 def ask_internet(query: str, model: str, debug=False):
-    content_list = search_web_ref(query, lang="en-US", debug=debug)  # 确保使用英文
+    content_list = search_web_ref(query,debug=debug)
     if debug:
         print(content_list)
     prompt = gen_prompt(query, content_list, context_length_limit=6000, debug=debug)