plomlombot.py

   1 #!/usr/bin/python3
   2
   3 import argparse
   4 import socket
   5 import datetime
   6 import select
   7 import time
   8 import re
   9 import requests
  10 import bs4
  11 import random
  12 import hashlib
  13 import os
  14 import plomsearch
  15
  16 URLREGEX = "(https?://[^\s>]+)"
  17
  18 # Defaults, may be overwritten by command line arguments.
  19 SERVER = "irc.freenode.net"
  20 PORT = 6667
  21 TIMEOUT = 240
  22 USERNAME = "plomlombot"
  23 NICKNAME = USERNAME
  24
  25
  26 class ExceptionForRestart(Exception):
  27     pass
  28
  29
  30 class IO:
  31
  32     def __init__(self, server, port, timeout):
  33         self.timeout = timeout
  34         self.socket = socket.socket()
  35         self.socket.connect((server, port))
  36         self.socket.setblocking(0)
  37         self.line_buffer = []
  38         self.rune_buffer = ""
  39         self.last_pong = time.time()
  40         self.servername = self.recv_line(send_ping=False).split(" ")[0][1:]
  41
  42     def _pingtest(self, send_ping=True):
  43         if self.last_pong + self.timeout < time.time():
  44             print("SERVER NOT ANSWERING")
  45             raise ExceptionForRestart
  46         if send_ping:
  47             self.send_line("PING " + self.servername)
  48
  49     def send_line(self, msg):
  50         msg = msg.replace("\r", " ")
  51         msg = msg.replace("\n", " ")
  52         if len(msg.encode("utf-8")) > 510:
  53             print("NOT SENT LINE TO SERVER (too long): " + msg)
  54         print("LINE TO SERVER: "
  55               + str(datetime.datetime.now()) + ": " + msg)
  56         msg = msg + "\r\n"
  57         msg_len = len(msg)
  58         total_sent_len = 0
  59         while total_sent_len < msg_len:
  60             sent_len = self.socket.send(bytes(msg[total_sent_len:], "UTF-8"))
  61             if sent_len == 0:
  62                 print("SOCKET CONNECTION BROKEN")
  63                 raise ExceptionForRestart
  64             total_sent_len += sent_len
  65
  66     def _recv_line_wrapped(self, send_ping=True):
  67         if len(self.line_buffer) > 0:
  68             return self.line_buffer.pop(0)
  69         while True:
  70             ready = select.select([self.socket], [], [], int(self.timeout / 2))
  71             if not ready[0]:
  72                 self._pingtest(send_ping)
  73                 return None
  74             self.last_pong = time.time()
  75             received_bytes = self.socket.recv(1024)
  76             try:
  77                 received_runes = received_bytes.decode("UTF-8")
  78             except UnicodeDecodeError:
  79                 received_runes = received_bytes.decode("latin1")
  80             if len(received_runes) == 0:
  81                 print("SOCKET CONNECTION BROKEN")
  82                 raise ExceptionForRestart
  83             self.rune_buffer += received_runes
  84             lines_split = str.split(self.rune_buffer, "\r\n")
  85             self.line_buffer += lines_split[:-1]
  86             self.rune_buffer = lines_split[-1]
  87             if len(self.line_buffer) > 0:
  88                 return self.line_buffer.pop(0)
  89
  90     def recv_line(self, send_ping=True):
  91         line = self._recv_line_wrapped(send_ping)
  92         if line:
  93             print("LINE FROM SERVER " + str(datetime.datetime.now()) + ": " +
  94                   line)
  95         return line
  96
  97
  98 def handle_command(command, argument, notice, target, session):
  99     hash_string = hashlib.md5(target.encode("utf-8")).hexdigest()
 100     quotesfile_name = "quotes_" + hash_string
 101
 102     def addquote():
 103         if not os.access(quotesfile_name, os.F_OK):
 104             quotesfile = open(quotesfile_name, "w")
 105             quotesfile.write("QUOTES FOR " + target + ":\n")
 106             quotesfile.close()
 107         quotesfile = open(quotesfile_name, "a")
 108         quotesfile.write(argument + "\n")
 109         quotesfile.close()
 110         quotesfile = open(quotesfile_name, "r")
 111         lines = quotesfile.readlines()
 112         quotesfile.close()
 113         notice("ADDED QUOTE #" + str(len(lines) - 1))
 114
 115     def quote():
 116
 117         def help():
 118             notice("SYNTAX: !quote [int] OR !quote search QUERY")
 119             notice("QUERY may be a boolean grouping of quoted or unquoted " +
 120                    "search terms, examples:")
 121             notice("!quote search foo")
 122             notice("!quote search foo AND (bar OR NOT baz)")
 123             notice("!quote search \"foo\\\"bar\" AND ('NOT\"' AND \"'foo'\"" +
 124                    " OR 'bar\\'baz')")
 125
 126         if "" == argument:
 127             tokens = []
 128         else:
 129             tokens = argument.split(" ")
 130         if (len(tokens) > 1 and tokens[0] != "search") or \
 131             (len(tokens) == 1 and
 132                 (tokens[0] == "search" or not tokens[0].isdigit())):
 133             help()
 134             return
 135         if not os.access(quotesfile_name, os.F_OK):
 136             notice("NO QUOTES AVAILABLE")
 137             return
 138         quotesfile = open(quotesfile_name, "r")
 139         lines = quotesfile.readlines()
 140         quotesfile.close()
 141         lines = lines[1:]
 142         if len(tokens) == 1:
 143             i = int(tokens[0])
 144             if i == 0 or i > len(lines):
 145                 notice("THERE'S NO QUOTE OF THAT INDEX")
 146                 return
 147             i = i - 1
 148         elif len(tokens) > 1:
 149             query = str.join(" ", tokens[1:])
 150             try:
 151                 results = plomsearch.search(query, lines)
 152             except plomsearch.LogicParserError as err:
 153                 notice("FAILED QUERY PARSING: " + str(err))
 154                 return
 155             if len(results) == 0:
 156                 notice("NO QUOTES MATCHING QUERY")
 157             else:
 158                 for result in results:
 159                     notice("QUOTE #" + str(result[0] + 1) + " : " + result[1])
 160             return
 161         else:
 162             i = random.randrange(len(lines))
 163         notice("QUOTE #" + str(i + 1) + ": " + lines[i])
 164
 165     def markov():
 166         from random import shuffle
 167         select_length = 2
 168         selections = []
 169
 170         def markov(snippet):
 171             usable_selections = []
 172             for i in range(select_length, 0, -1):
 173                 for selection in selections:
 174                     add = True
 175                     for j in range(i):
 176                         if snippet[j] != selection[j]:
 177                             add = False
 178                             break
 179                     if add:
 180                         usable_selections += [selection]
 181                 if [] != usable_selections:
 182                     break
 183             if [] == usable_selections:
 184                 usable_selections = selections
 185             shuffle(usable_selections)
 186             return usable_selections[0][select_length]
 187
 188         def purge_undesired(tokens):
 189             for token in tokens:
 190                 if None != re.match("^" + URLREGEX, token):
 191                     del(tokens[tokens.index(token)])
 192             for name in session.uses_in_chan:
 193                 while True:
 194                     try:
 195                         del(tokens[tokens.index(name.lower())])
 196                     except ValueError:
 197                         break
 198             return tokens
 199
 200         hash_string = hashlib.md5(target.encode("utf-8")).hexdigest()
 201         markovfeed_name = "markovfeed_" + hash_string
 202         if not os.access(markovfeed_name, os.F_OK):
 203             notice("NOT ENOUGH TEXT TO MARKOV.")
 204             return
 205         file = open(markovfeed_name, "r")
 206         lines = file.readlines()
 207         file.close()
 208         tokens = []
 209         for line in lines:
 210             line = line.replace("\n", "").lower()
 211             tokens += line.split()
 212         tokens = purge_undesired(tokens)
 213         if len(tokens) <= select_length:
 214             notice("NOT ENOUGH TEXT TO MARKOV.")
 215             return
 216         for i in range(len(tokens) - select_length):
 217             token_list = []
 218             for j in range(select_length + 1):
 219                 token_list += [tokens[i + j]]
 220             selections += [token_list]
 221         snippet = []
 222         for i in range(select_length):
 223             snippet += [""]
 224         msg = ""
 225         while 1:
 226             new_end = markov(snippet)
 227             if len(msg) + len(new_end) > 200:
 228                 break
 229             msg += new_end + " "
 230             for i in range(select_length - 1):
 231                 snippet[i] = snippet[i + 1]
 232             snippet[select_length - 1] = new_end
 233         notice(msg + "malkovich.")
 234
 235     if "addquote" == command:
 236         addquote()
 237     elif "quote" == command:
 238         quote()
 239     elif "markov" == command:
 240         markov()
 241
 242
 243 def handle_url(url, notice, show_url=False):
 244
 245     def mobile_twitter_hack(url):
 246         re1 = 'https?://(mobile.twitter.com/)[^/]+(/status/)'
 247         re2 = 'https?://mobile.twitter.com/([^/]+)/status/([^\?/]+)'
 248         m = re.search(re1, url)
 249         if m and m.group(1) == 'mobile.twitter.com/' \
 250                 and m.group(2) == '/status/':
 251             m = re.search(re2, url)
 252             url = 'https://twitter.com/' + m.group(1) + '/status/' + m.group(2)
 253             handle_url(url, notice, True)
 254             return True
 255
 256     try:
 257         r = requests.get(url, timeout=15)
 258     except (requests.exceptions.TooManyRedirects,
 259             requests.exceptions.ConnectionError,
 260             requests.exceptions.InvalidURL,
 261             requests.exceptions.InvalidSchema) as error:
 262         notice("TROUBLE FOLLOWING URL: " + str(error))
 263         return
 264     if mobile_twitter_hack(url):
 265         return
 266     title = bs4.BeautifulSoup(r.text, "html.parser").title
 267     if title:
 268         prefix = "PAGE TITLE: "
 269         if show_url:
 270             prefix = "PAGE TITLE FOR <" + url + ">: "
 271         notice(prefix + title.string.strip())
 272     else:
 273         notice("PAGE HAS NO TITLE TAG")
 274
 275
 276 class Session:
 277
 278     def __init__(self, io, username, nickname, channel):
 279         self.io = io
 280         self.nickname = nickname
 281         self.channel = channel
 282         self.uses_in_chan = []
 283         self.io.send_line("NICK " + self.nickname)
 284         self.io.send_line("USER " + username + " 0 * : ")
 285         self.io.send_line("JOIN " + self.channel)
 286
 287     def loop(self):
 288
 289         def handle_privmsg(tokens):
 290
 291             def handle_input(msg, target):
 292
 293                 def notice(msg):
 294                     self.io.send_line("NOTICE " + target + " :" + msg)
 295
 296                 matches = re.findall(URLREGEX, msg)
 297                 for i in range(len(matches)):
 298                     handle_url(matches[i], notice)
 299                 if "!" == msg[0]:
 300                     tokens = msg[1:].split()
 301                     argument = str.join(" ", tokens[1:])
 302                     handle_command(tokens[0], argument, notice, target, self)
 303                     return
 304                 hash_string = hashlib.md5(target.encode("utf-8")).hexdigest()
 305                 markovfeed_name = "markovfeed_" + hash_string
 306                 file = open(markovfeed_name, "a")
 307                 file.write(msg + "\n")
 308                 file.close()
 309
 310             sender = ""
 311             for rune in tokens[0]:
 312                 if rune == "!":
 313                     break
 314                 if rune != ":":
 315                     sender += rune
 316             receiver = ""
 317             for rune in tokens[2]:
 318                 if rune == "!":
 319                     break
 320                 if rune != ":":
 321                     receiver += rune
 322             target = sender
 323             if receiver != self.nickname:
 324                 target = receiver
 325             msg = str.join(" ", tokens[3:])[1:]
 326             handle_input(msg, target)
 327
 328         def name_from_join_or_part(tokens):
 329             token = tokens[0][1:]
 330             index_cut = token.find("@")
 331             index_ex = token.find("!")
 332             if index_ex > 0 and index_ex < index_cut:
 333                 index_cut = index_ex
 334             return token[:index_cut]
 335
 336         while True:
 337             line = self.io.recv_line()
 338             if not line:
 339                 continue
 340             tokens = line.split(" ")
 341             if len(tokens) > 1:
 342                 if tokens[0] == "PING":
 343                     self.io.send_line("PONG " + tokens[1])
 344                 elif tokens[1] == "PRIVMSG":
 345                     handle_privmsg(tokens)
 346                 elif tokens[1] == "353":
 347                     names = tokens[5:]
 348                     names[0] = names[0][1:]
 349                     self.uses_in_chan += names
 350                 elif tokens[1] == "JOIN":
 351                     name = name_from_join_or_part(tokens)
 352                     if name != self.nickname:
 353                         self.uses_in_chan += [name]
 354                 elif tokens[1] == "PART":
 355                     name = name_from_join_or_part(tokens)
 356                     del(self.uses_in_chan[self.uses_in_chan.index(name)])
 357
 358 def parse_command_line_arguments():
 359     parser = argparse.ArgumentParser()
 360     parser.add_argument("-s, --server", action="store", dest="server",
 361                         default=SERVER,
 362                         help="server or server net to connect to (default: "
 363                         + SERVER + ")")
 364     parser.add_argument("-p, --port", action="store", dest="port", type=int,
 365                         default=PORT, help="port to connect to (default : "
 366                         + str(PORT) + ")")
 367     parser.add_argument("-t, --timeout", action="store", dest="timeout",
 368                         type=int, default=TIMEOUT,
 369                         help="timeout in seconds after which to attempt " +
 370                         "reconnect (default: " + str(TIMEOUT) + ")")
 371     parser.add_argument("-u, --username", action="store", dest="username",
 372                         default=USERNAME, help="username to use (default: "
 373                         + USERNAME + ")")
 374     parser.add_argument("-n, --nickname", action="store", dest="nickname",
 375                         default=NICKNAME, help="nickname to use (default: "
 376                         + NICKNAME + ")")
 377     parser.add_argument("CHANNEL", action="store", help="channel to join")
 378     opts, unknown = parser.parse_known_args()
 379     return opts
 380
 381
 382 opts = parse_command_line_arguments()
 383 while True:
 384     try:
 385         io = IO(opts.server, opts.port, opts.timeout)
 386         session = Session(io, opts.username, opts.nickname, opts.CHANNEL)
 387         session.loop()
 388     except ExceptionForRestart:
 389         io.socket.close()
 390         continue