Better wiki redirect

Reviewed-on: #1

functions.py

@@ -5,7 +5,8 @@ def processmsg(msg, rcpt):
     if "youtube.com/watch" in msg:
         return msg.replace("youtube.com", "iv.datura.network")
     elif msg.startswith("!wiki"):
-        return sf.query_external_website("https://en.wikipedia.org/wiki/", msg.split(" ")[1])
+        cmd, query = msg.split(" ", 1)
+        return sf.query_external_website("https://en.wikipedia.org", "/wiki/" + query)
     elif "good bot" in msg:
         return "^_^"
 
@@ -13,7 +14,8 @@ def command(msg, rcpt):
     if msg.startswith("!help"):
         response = "chatbot commands:"  + "\n"
         response += "!help            Show this help page" + "\n"
-        response += "!ai [message]    Ask llama2"
+        response += "!ai [message]    Ask llama2" + "\n"
+        response += "!wiki [message]    Ask wiki"
         return response
     elif msg.startswith("!ai"):
         client = ollama.Client(host='https://ollama.krov.dmz.rs')

scraper_functions.py

@@ -1,12 +1,19 @@
 import requests
 from bs4 import BeautifulSoup
+from urllib.parse import quote
 
 def query_external_website(base_url, query):
     try:
-        page = requests.get(base_url + query)
+        page = requests.get(base_url + quote(query))
         soup = BeautifulSoup(page.content, "html.parser")
-        title = soup.find("span", class_="mw-page-title-main").text
+        title = soup.find(id="firstHeading").text
-        content = soup.find(id="mw-content-text").select("p")[2].text
+        mainContentElement = soup.find(id="mw-content-text")
-        return "\nTITLE:\n" + title + "\n\nCONTENT:\n" + content + "\n\nFULL LINK:\n" + base_url + query
+        if "This page is a redirect" in mainContentElement.text:
-    except:
+            redirectLink = mainContentElement.find(class_="redirectMsg").find_all("a")[0]["href"]
-        return "Can't parse search result :("
+            return query_external_website(base_url, redirectLink)
+        content = next((paragraph for paragraph in mainContentElement.select("p") if not paragraph.has_attr("class")), None)
+        if content == None:
+            raise Exception("Can't parse search result :(")
+        return "\nTITLE:\n" + title + "\n\nCONTENT:\n" + content.text + "\n\nFULL LINK:\n" + base_url + quote(query)
+    except Exception as e:
+        return e