Cancel page title reading on bad charset.

[plomlombot-irc.git] / plomlombot.py

diff --git a/plomlombot.py b/plomlombot.py
index a47158c302a3341cbe3e3777410b6e073c0a485d..8ac8f01c00cf0cc4e00ad6e4a664a743926ac2f6 100755 (executable)
--- a/plomlombot.py
+++ b/plomlombot.py
@@ -7,6 +7,7 @@ import select
 import time
 import re
 import urllib.request
 import time
 import re
 import urllib.request

+import http.client

 import html
 
 # Defaults, may be overwritten by command line arguments.
 import html
 
 # Defaults, may be overwritten by command line arguments.


@@ -66,7 +67,11 @@ class IO:
                 self._pingtest(send_ping)
                 return None
             self.last_pong = time.time()
                 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
             if len(received_runes) == 0:
                 print("SOCKET CONNECTION BROKEN")
                 raise ExceptionForRestart


@@ -99,29 +104,36 @@ def lineparser_loop(io, nickname):
     def act_on_privmsg(tokens):
 
         def url_check(msg):
     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]
             for i in range(len(matches)):
                 url = matches[i]

+                request = urllib.request.Request(url, headers={
+                    "User-Agent": "plomlombot"
+                })

                 try:
                 try:

-                    webpage = urllib.request.urlopen(url, timeout=15)
+                    webpage = urllib.request.urlopen(request, timeout=15)

                 except (urllib.error.HTTPError, urllib.error.URLError,
                 except (urllib.error.HTTPError, urllib.error.URLError,

-                        UnicodeError) as error:
-                    print("TROUBLE FOLLOWING URL: " + str(error))
+                        UnicodeError, http.client.BadStatusLine) as error:
+                    notice("TROUBLE FOLLOWING URL: " + str(error))

                     continue
                 charset = webpage.info().get_content_charset()
                 if not charset:
                     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'):
                 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)
+                    notice("TROUBLE READING PAGE TITLE: bad content type "
+                           + content_type)

                     continue
                 content = webpage.read().decode(charset)
                 title = str(content).split('<title>')[1].split('</title>')[0]
                 title = html.unescape(title)
                     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)
+                notice("PAGE TITLE FOR URL: " + title)

 
         sender = ""
         for rune in tokens[0]:
 
         sender = ""
         for rune in tokens[0]:


@@ -141,7 +153,7 @@ def lineparser_loop(io, nickname):
         msg = str.join(" ", tokens[3:])[1:]
         url_check(msg)
 
         msg = str.join(" ", tokens[3:])[1:]
         url_check(msg)
 

-    while 1:
+    while True:

         line = io.recv_line()
         if not line:
             continue
         line = io.recv_line()
         if not line:
             continue


@@ -177,7 +189,7 @@ def parse_command_line_arguments():
     return opts
 
 opts = parse_command_line_arguments()
     return opts
 
 opts = parse_command_line_arguments()

-while 1:
+while True:

     try:
         io = init_session(opts.server, opts.port, opts.timeout, opts.nickname,
                           opts.username, opts.CHANNEL)
     try:
         io = init_session(opts.server, opts.port, opts.timeout, opts.nickname,
                           opts.username, opts.CHANNEL)