Use BeautifulSoup for HTML parsing.

[plomlombot-irc.git] / plomlombot.py

diff --git a/plomlombot.py b/plomlombot.py
index 5fcd517a95f563cfba7a00146d1e6359780a0472..3e4734fa2ec2403aa093768623e1ca9b189d5cd8 100755 (executable)
--- a/plomlombot.py
+++ b/plomlombot.py
@@ -6,8 +6,8 @@ import datetime
 import select
 import time
 import re
-import urllib.request
-import html
+import requests
+import bs4
 
 # Defaults, may be overwritten by command line arguments.
 SERVER = "irc.freenode.net"
@@ -66,7 +66,11 @@ class IO:
                 self._pingtest(send_ping)
                 return None
             self.last_pong = time.time()
-            received_runes = self.socket.recv(1024).decode("UTF-8")
+            received_bytes = self.socket.recv(1024)
+            try:
+                received_runes = received_bytes.decode("UTF-8")
+            except UnicodeDecodeError:
+                received_runes = received_bytes.decode("latin1")
             if len(received_runes) == 0:
                 print("SOCKET CONNECTION BROKEN")
                 raise ExceptionForRestart
@@ -99,29 +103,25 @@ def lineparser_loop(io, nickname):
     def act_on_privmsg(tokens):
 
         def url_check(msg):
-            matches = re.findall("(https?://[^\s]+)", msg)
+
+            def notice(msg):
+                io.send_line("NOTICE " + target + " :" + msg)
+
+            matches = re.findall("(https?://[^\s>]+)", msg)
             for i in range(len(matches)):
                 url = matches[i]
                 try:
-                    webpage = urllib.request.urlopen(url, timeout=15)
-                except (urllib.error.HTTPError, urllib.error.URLError,
-                        UnicodeError) as error:
-                    print("TROUBLE FOLLOWING URL: " + str(error))
-                    continue
-                charset = webpage.info().get_content_charset()
-                if not charset:
-                    charset = "utf-8"
-                content_type = webpage.info().get_content_type()
-                if content_type not in ('text/html', 'text/xml',
-                                        'application/xhtml+xml'):
-                    print("TROUBLE INTERPRETING URL: bad content type "
-                          + content_type)
+                    r = requests.get(url, timeout=15)
+                except (requests.exceptions.TooManyRedirects,
+                        requests.exceptions.ConnectionError,
+                        requests.exceptions.InvalidSchema) as error:
+                    notice("TROUBLE FOLLOWING URL: " + str(error))
                     continue
-                content = webpage.read().decode(charset)
-                title = str(content).split('<title>')[1].split('</title>')[0]
-                title = html.unescape(title)
-                io.send_line("PRIVMSG " + target + " :page title for url: "
-                             + title)
+                title = bs4.BeautifulSoup(r.text).title
+                if title:
+                    notice("PAGE TITLE FOR URL: " + title.string)
+                else:
+                    notice("PAGE HAS NO TITLE TAG")
 
         sender = ""
         for rune in tokens[0]: