Add some more checks against evil URLs.

[plomlombot-irc.git] / plomlombot.py

diff --git a/plomlombot.py b/plomlombot.py
index 747ab077700fa772d1c54cc62b61290c1a7bcad1..7fa9194af8236e27fd7d547843fc1a1e3577dc8b 100644 (file)
--- a/plomlombot.py
+++ b/plomlombot.py
@@ -1,18 +1,21 @@
 import socket
 import socket

-import datetime 
+import datetime

 import select
 import time
 import re
 import urllib.request
 import html
 
 import select
 import time
 import re
 import urllib.request
 import html
 

+servernet = "irc.freenode.net"
+port = 6667

 servername = ""
 servername = ""

-timeout = 480
+timeout = 240

 username = "plomlombot"
 nickname = username
 username = "plomlombot"
 nickname = username

-channel = "#zrolaps"
+channel = "#zrolaps-test"

 
 class IO:
 
 class IO:

+

     def __init__(self, server, port):
         self.socket = socket.socket()
         self.socket.connect((server, port))
     def __init__(self, server, port):
         self.socket = socket.socket()
         self.socket.connect((server, port))


@@ -20,12 +23,20 @@ class IO:
         self.line_buffer = []
         self.rune_buffer = ""
         self.last_pong = time.time()
         self.line_buffer = []
         self.rune_buffer = ""
         self.last_pong = time.time()

+

     def _pingtest(self):
         if self.last_pong + timeout < time.time():
             raise RuntimeError("server not answering")
         self.send_line("PING " + nickname + " " + servername)
     def _pingtest(self):
         if self.last_pong + timeout < time.time():
             raise RuntimeError("server not answering")
         self.send_line("PING " + nickname + " " + servername)

-    def send_line(self, msg_orig):
-        msg = msg_orig + "\r\n"
+
+    def send_line(self, msg):
+        msg = msg.replace("\r", " ")
+        msg = msg.replace("\n", " ")
+        if len(msg.encode("utf-8")) > 510:
+            print("NOT SENT LINE TO SERVER (too long): " + msg)
+        print("LINE TO SERVER: "
+            + str(datetime.datetime.now()) + ": " + msg)
+        msg = msg + "\r\n"

         msg_len = len(msg)
         total_sent_len = 0
         while total_sent_len < msg_len:
         msg_len = len(msg)
         total_sent_len = 0
         while total_sent_len < msg_len:


@@ -33,8 +44,7 @@ class IO:
             if sent_len == 0:
                 raise RuntimeError("socket connection broken")
             total_sent_len += sent_len
             if sent_len == 0:
                 raise RuntimeError("socket connection broken")
             total_sent_len += sent_len

-        print("LINE TO SERVER: "
-            + str(datetime.datetime.now()) + ": " + msg_orig)
+

     def recv_line_wrapped(self):
         if len(self.line_buffer) > 0:
             return self.line_buffer.pop(0)
     def recv_line_wrapped(self):
         if len(self.line_buffer) > 0:
             return self.line_buffer.pop(0)


@@ -53,6 +63,7 @@ class IO:
             self.rune_buffer = lines_split[-1]
             if len(self.line_buffer) > 0:
                 return self.line_buffer.pop(0)
             self.rune_buffer = lines_split[-1]
             if len(self.line_buffer) > 0:
                 return self.line_buffer.pop(0)

+

     def recv_line(self):
         line = self.recv_line_wrapped()
         if line:
     def recv_line(self):
         line = self.recv_line_wrapped()
         if line:


@@ -60,7 +71,7 @@ class IO:
             line)
         return line
 
             line)
         return line
 

-io = IO("irc.freenode.net", 6667)
+io = IO(servernet, port)

 io.send_line("NICK " + nickname)
 io.send_line("USER " + username + " 0 * : ")
 io.send_line("JOIN " + channel)
 io.send_line("NICK " + nickname)
 io.send_line("USER " + username + " 0 * : ")
 io.send_line("JOIN " + channel)


@@ -92,7 +103,11 @@ while 1:
             for i in range(len(matches)):
                 url = matches[i]
                 webpage = urllib.request.urlopen(url)
             for i in range(len(matches)):
                 url = matches[i]
                 webpage = urllib.request.urlopen(url)

+                content_type = webpage.info().get_content_type()

                 charset = webpage.info().get_content_charset()
                 charset = webpage.info().get_content_charset()

+                if not charset or not content_type in ('text/html', 'text/xml',
+                    'application/xhtml+xml'):
+                    continue

                 content = webpage.read().decode(charset)
                 title = str(content).split('<title>')[1].split('</title>')[0]
                 title = html.unescape(title)
                 content = webpage.read().decode(charset)
                 title = str(content).split('<title>')[1].split('</title>')[0]
                 title = html.unescape(title)