Use BeautifulSoup for HTML parsing.

[plomlombot-irc.git] / plomlombot.py

diff --git a/plomlombot.py b/plomlombot.py
index f34d97a821378c5e092cf064a78a2137d82350c7..3e4734fa2ec2403aa093768623e1ca9b189d5cd8 100755 (executable)
--- a/plomlombot.py
+++ b/plomlombot.py
@@ -6,10 +6,8 @@ import datetime
 import select
 import time
 import re
-import urllib.request
-import http.client
-import html
-import html.parser
+import requests
+import bs4
 
 # Defaults, may be overwritten by command line arguments.
 SERVER = "irc.freenode.net"
@@ -19,22 +17,6 @@ USERNAME = "plomlombot"
 NICKNAME = USERNAME
 
 
-class HTMLParser(html.parser.HTMLParser):
-    def __init__(self, html, tag):
-        super().__init__()
-        self._tag = ""
-        self.data = ""
-        self.feed(html)
-    def handle_starttag(self, tag, attrs):
-        if self.data == "":
-            self._tag = tag
-    def handle_endtag(self, tag):
-        self._tag = ""
-    def handle_data(self, data):
-        if self._tag != "":
-            self.data = data
-
-
 class ExceptionForRestart(Exception):
     pass
 
@@ -128,29 +110,18 @@ def lineparser_loop(io, nickname):
             matches = re.findall("(https?://[^\s>]+)", msg)
             for i in range(len(matches)):
                 url = matches[i]
-                request = urllib.request.Request(url, headers={
-                    "User-Agent": "plomlombot"
-                })
                 try:
-                    webpage = urllib.request.urlopen(request, timeout=15)
-                except (urllib.error.HTTPError, urllib.error.URLError,
-                        UnicodeError, http.client.BadStatusLine) as error:
+                    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
-                charset = webpage.info().get_content_charset()
-                if not charset:
-                    notice("TROUBLE READING PAGE TITLE: no charset in header")
-                    continue
-                content_type = webpage.info().get_content_type()
-                if content_type not in ('text/html', 'text/xml',
-                                        'application/xhtml+xml'):
-                    notice("TROUBLE READING PAGE TITLE: bad content type "
-                           + content_type)
-                    continue
-                content = webpage.read().decode(charset)
-                title = HTMLParser(content, "title").data
-                title = html.unescape(title)
-                notice("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]: