Fix buggy HTMLParser.

[plomlombot-irc.git] / plomlombot.py

diff --git a/plomlombot.py b/plomlombot.py
index 0b605e9946b04f79a8f3e843da56eee2b59b0169..9e9d0a44ce21bc1d07204dcb9780c54ac1b31d0e 100755 (executable)
--- a/plomlombot.py
+++ b/plomlombot.py
@@ -9,6 +9,7 @@ import re
 import urllib.request
 import http.client
 import html
+import html.parser
 
 # Defaults, may be overwritten by command line arguments.
 SERVER = "irc.freenode.net"
@@ -18,6 +19,23 @@ USERNAME = "plomlombot"
 NICKNAME = USERNAME
 
 
+class HTMLParser(html.parser.HTMLParser):
+    def __init__(self, html, tag):
+        super().__init__()
+        self._tag_to_check = tag
+        self._tag = ""
+        self.data = ""
+        self.feed(html)
+    def handle_starttag(self, tag, attrs):
+        if self.data == "" and tag == self._tag_to_check:
+            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
 
@@ -122,7 +140,8 @@ def lineparser_loop(io, nickname):
                     continue
                 charset = webpage.info().get_content_charset()
                 if not charset:
-                    charset = "utf-8"
+                    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'):
@@ -130,7 +149,7 @@ def lineparser_loop(io, nickname):
                            + content_type)
                     continue
                 content = webpage.read().decode(charset)
-                title = str(content).split('<title>')[1].split('</title>')[0]
+                title = HTMLParser(content, "title").data
                 title = html.unescape(title)
                 notice("PAGE TITLE FOR URL: " + title)