X-Git-Url: https://plomlompom.com/repos/?p=plomlombot-irc.git;a=blobdiff_plain;f=plomlombot.py;h=412ae2117aeb9fc73df1a105ed70014094e992f7;hp=73533840403165d2251479d6b2c4e3e2239b79f5;hb=ee64c72b59346940c29cf068dd7b1cb9425f332e;hpb=e68573e0f640a77680bd1ab51e18f70d8011caf4 diff --git a/plomlombot.py b/plomlombot.py index 7353384..412ae21 100755 --- a/plomlombot.py +++ b/plomlombot.py @@ -11,7 +11,9 @@ import bs4 import random import hashlib import os +import signal import plomsearch +import irclog # Defaults, may be overwritten by command line arguments. SERVER = "irc.freenode.net" @@ -23,16 +25,45 @@ TWTFILE = "" DBDIR = os.path.expanduser("~/plomlombot_db") +def write_to_file(path, mode, text): + f = open(path, mode) + f.write(text) + f.close() + + class ExceptionForRestart(Exception): pass +class Line: + + def __init__(self, line): + self.line = line + self.tokens = line.split(" ") + self.sender = "" + if self.tokens[0][0] == ":": + for rune in self.tokens[0][1:]: + if rune in {"!", "@"}: + break + self.sender += rune + self.receiver = "" + if len(self.tokens) > 2: + for rune in self.tokens[2]: + if rune in {"!", "@"}: + break + if rune != ":": + self.receiver += rune + + class IO: def __init__(self, server, port, timeout): self.timeout = timeout self.socket = socket.socket() - self.socket.connect((server, port)) + try: + self.socket.connect((server, port)) + except TimeoutError: + raise ExceptionForRestart self.socket.setblocking(0) self.line_buffer = [] self.rune_buffer = "" @@ -99,21 +130,18 @@ def handle_command(command, argument, notice, target, session): def addquote(): if not os.access(session.quotesfile, os.F_OK): - quotesfile = open(session.quotesfile, "w") - quotesfile.write("QUOTES FOR " + target + ":\n") - quotesfile.close() - quotesfile = open(session.quotesfile, "a") - quotesfile.write(argument + "\n") - quotesfile.close() + write_to_file(session.quotesfile, "w", + "QUOTES FOR " + target + ":\n") + write_to_file(session.quotesfile, "a", argument + "\n") quotesfile = open(session.quotesfile, "r") lines = quotesfile.readlines() quotesfile.close() - notice("ADDED QUOTE #" + str(len(lines) - 1)) + notice("added quote #" + str(len(lines) - 1)) def quote(): def help(): - notice("SYNTAX: !quote [int] OR !quote search QUERY") + notice("syntax: !quote [int] OR !quote search QUERY") notice("QUERY may be a boolean grouping of quoted or unquoted " + "search terms, examples:") notice("!quote search foo") @@ -131,7 +159,7 @@ def handle_command(command, argument, notice, target, session): help() return if not os.access(session.quotesfile, os.F_OK): - notice("NO QUOTES AVAILABLE") + notice("no quotes available") return quotesfile = open(session.quotesfile, "r") lines = quotesfile.readlines() @@ -140,7 +168,7 @@ def handle_command(command, argument, notice, target, session): if len(tokens) == 1: i = int(tokens[0]) if i == 0 or i > len(lines): - notice("THERE'S NO QUOTE OF THAT INDEX") + notice("there's no quote of that index") return i = i - 1 elif len(tokens) > 1: @@ -148,23 +176,25 @@ def handle_command(command, argument, notice, target, session): try: results = plomsearch.search(query, lines) except plomsearch.LogicParserError as err: - notice("FAILED QUERY PARSING: " + str(err)) + notice("failed query parsing: " + str(err)) return if len(results) == 0: - notice("NO QUOTES MATCHING QUERY") + notice("no quotes matching query") else: - for result in results: - notice("QUOTE #" + str(result[0] + 1) + " : " - + result[1][-1]) + if len(results) > 3: + notice("showing 3 of " + str(len(results)) + " quotes") + for result in results[:3]: + notice("quote #" + str(result[0] + 1) + ": " + + result[1][:-1]) return else: i = random.randrange(len(lines)) - notice("QUOTE #" + str(i + 1) + ": " + lines[i][:-1]) + notice("quote #" + str(i + 1) + ": " + lines[i][:-1]) def markov(): - from random import choice, shuffle - select_length = 2 - selections = [] + + def help(): + notice("syntax: !markov [integer from 1 to infinite]") def markov(snippet): usable_selections = [] @@ -185,8 +215,26 @@ def handle_command(command, argument, notice, target, session): selection = choice(usable_selections) return selection[select_length] + if "" == argument: + tokens = [] + else: + tokens = argument.split(" ") + if (len(tokens) > 1 or (len(tokens) == 1 and not tokens[0].isdigit())): + help() + return + + from random import choice, shuffle + select_length = 2 + if len(tokens) == 1: + n = int(tokens[0]) + if n > 0: + select_length = n + else: + notice("bad value, using default: " + str(select_length)) + selections = [] + if not os.access(session.markovfile, os.F_OK): - notice("NOT ENOUGH TEXT TO MARKOV.") + notice("not enough text to markov for selection length") return # Lowercase incoming lines, ensure they end in a sentence end mark. @@ -200,8 +248,8 @@ def handle_command(command, argument, notice, target, session): if line[-1] not in sentence_end_markers: line += "." tokens += line.split() - if len(tokens) <= select_length: - notice("NOT ENOUGH TEXT TO MARKOV.") + if len(tokens) - 1 <= select_length: + notice("not enough text to markov") return # Replace URLs with escape string for now, so that the Markov selector @@ -236,8 +284,8 @@ def handle_command(command, argument, notice, target, session): shuffle(selections) for i in range(len(selections)): if selections[i][0][-1] in sentence_end_markers: - for i in range(select_length): - snippet[i] = selections[i][i + 1] + for j in range(select_length): + snippet[j] = selections[j][j + 1] break msg = "" malkovich = "malkovich" @@ -269,7 +317,7 @@ def handle_command(command, argument, notice, target, session): try: twtfile = open(session.twtfile, mode) except (PermissionError, FileNotFoundError) as err: - notice("CAN'T ACCESS OR CREATE TWT FILE: " + str(err)) + notice("can't access or create twt file: " + str(err)) return None return twtfile @@ -284,7 +332,7 @@ def handle_command(command, argument, notice, target, session): return twtfile.write(datetime.utcnow().isoformat() + "\t" + argument + "\n") twtfile.close() - notice("WROTE TWT.") + notice("wrote twt.") if "addquote" == command: addquote() @@ -309,127 +357,144 @@ def handle_url(url, notice, show_url=False): handle_url(url, notice, True) return True + class TimeOut(Exception): + pass + + def timeout_handler(ignore1, ignore2): + raise TimeOut("timeout") + + signal.signal(signal.SIGALRM, timeout_handler) + signal.alarm(15) try: - r = requests.get(url, timeout=15) + r = requests.get(url, headers = {'User-Agent': 'plomlombot'}, stream=True) + r.raw.decode_content = True + text = r.raw.read(10000000+1) + if len(text) > 10000000: + raise ValueError('Too large a response') except (requests.exceptions.TooManyRedirects, requests.exceptions.ConnectionError, requests.exceptions.InvalidURL, + TimeOut, UnicodeError, + ValueError, requests.exceptions.InvalidSchema) as error: - notice("TROUBLE FOLLOWING URL: " + str(error)) - return + signal.alarm(0) + notice("trouble following url: " + str(error)) + return False + signal.alarm(0) if mobile_twitter_hack(url): - return - title = bs4.BeautifulSoup(r.text, "html5lib").title + return True + title = bs4.BeautifulSoup(text, "html5lib").title if title and title.string: - prefix = "PAGE TITLE: " + prefix = "page title: " if show_url: - prefix = "PAGE TITLE FOR <" + url + ">: " + prefix = "page title for <" + url + ">: " notice(prefix + title.string.strip()) else: - notice("PAGE HAS NO TITLE TAG") + notice("page has no title tag") + return True class Session: - def __init__(self, io, username, nickname, channel, twtfile, dbdir): + def __init__(self, io, username, nickname, channel, twtfile, dbdir, rmlogs): self.io = io self.nickname = nickname + self.username = username self.channel = channel self.users_in_chan = [] self.twtfile = twtfile self.dbdir = dbdir + self.rmlogs = rmlogs self.io.send_line("NICK " + self.nickname) - self.io.send_line("USER " + username + " 0 * : ") + self.io.send_line("USER " + self.username + " 0 * : ") self.io.send_line("JOIN " + self.channel) hash_channel = hashlib.md5(self.channel.encode("utf-8")).hexdigest() self.chandir = self.dbdir + "/" + hash_channel + "/" + self.rawlogdir = self.chandir + "raw_logs/" self.logdir = self.chandir + "logs/" if not os.path.exists(self.logdir): os.makedirs(self.logdir) + if not os.path.exists(self.rawlogdir): + os.makedirs(self.rawlogdir) self.markovfile = self.chandir + "markovfeed" self.quotesfile = self.chandir + "quotes" def loop(self): def log(line): + if type(line) == str: + line = Line(":" + self.nickname + "!~" + self.username + + "@localhost" + " " + line) now = datetime.datetime.utcnow() - logfile = open(self.logdir + now.strftime("%Y-%m-%d") + ".txt", "a") form = "%Y-%m-%d %H:%M:%S UTC\t" - logfile.write(now.strftime(form) + " " + line + "\n") - logfile.close() - - def handle_privmsg(tokens): - - def handle_input(msg, target): - - def notice(msg): - line = "NOTICE " + target + " :" + msg - self.io.send_line(line) - log(line) - - matches = re.findall("(https?://[^\s>]+)", msg) - for i in range(len(matches)): - handle_url(matches[i], notice) - if "!" == msg[0]: - tokens = msg[1:].split() - argument = str.join(" ", tokens[1:]) - handle_command(tokens[0], argument, notice, target, self) - return - file = open(self.markovfile, "a") - file.write(msg + "\n") - file.close() - - sender = "" - for rune in tokens[0]: - if rune == "!": - break - if rune != ":": - sender += rune - receiver = "" - for rune in tokens[2]: - if rune == "!": - break - if rune != ":": - receiver += rune - target = sender - if receiver != self.nickname: - target = receiver - msg = str.join(" ", tokens[3:])[1:] - handle_input(msg, target) - - def name_from_join_or_part(tokens): - token = tokens[0][1:] - index_cut = token.find("@") - index_ex = token.find("!") - if index_ex > 0 and index_ex < index_cut: - index_cut = index_ex - return token[:index_cut] + write_to_file(self.rawlogdir + now.strftime("%Y-%m-%d") + ".txt", + "a", now.strftime(form) + " " + line.line + "\n") + to_log = irclog.format_logline(line, self.channel) + if to_log != None: + write_to_file(self.logdir + now.strftime("%Y-%m-%d") + ".txt", + "a", now.strftime(form) + " " + to_log + "\n") + + def handle_privmsg(line): + + def notice(msg): + line = "NOTICE " + target + " :" + msg + self.io.send_line(line) + log(line) + + target = line.sender + if line.receiver != self.nickname: + target = line.receiver + msg = str.join(" ", line.tokens[3:])[1:] + matches = re.findall("(https?://[^\s>]+)", msg) + url_count = 0 + for i in range(len(matches)): + if handle_url(matches[i], notice): + url_count += 1 + if url_count == 3: + notice("maximum number of urls to parse per message " + "reached") + break + if "!" == msg[0]: + tokens = msg[1:].split() + argument = str.join(" ", tokens[1:]) + handle_command(tokens[0], argument, notice, target, self) + return + write_to_file(self.markovfile, "a", msg + "\n") + now = datetime.datetime.utcnow() + write_to_file(self.logdir + now.strftime("%Y-%m-%d") + ".txt", "a", + "-----------------------\n") while True: + if self.rmlogs > 0: + for f in os.listdir(self.logdir): + f = os.path.join(self.logdir, f) + if os.path.isfile(f) and \ + os.stat(f).st_mtime < time.time() - self.rmlogs: + os.remove(f) line = self.io.recv_line() if not line: continue + line = Line(line) log(line) - tokens = line.split(" ") - if len(tokens) > 1: - if tokens[0] == "PING": - self.io.send_line("PONG " + tokens[1]) - elif tokens[1] == "PRIVMSG": - handle_privmsg(tokens) - elif tokens[1] == "353": - names = tokens[5:] + if len(line.tokens) > 1: + if line.tokens[0] == "PING": + self.io.send_line("PONG " + line.tokens[1]) + elif line.tokens[1] == "PRIVMSG": + handle_privmsg(line) + elif line.tokens[1] == "353": + names = line.tokens[5:] names[0] = names[0][1:] for i in range(len(names)): names[i] = names[i].replace("@", "").replace("+", "") self.users_in_chan += names - elif tokens[1] == "JOIN": - name = name_from_join_or_part(tokens) - if name != self.nickname: - self.users_in_chan += [name] - elif tokens[1] == "PART": - name = name_from_join_or_part(tokens) - del(self.users_in_chan[self.users_in_chan.index(name)]) + elif line.tokens[1] == "JOIN" and line.sender != self.nickname: + self.users_in_chan += [line.sender] + elif line.tokens[1] == "PART": + del(self.users_in_chan[self.users_in_chan.index(line.sender)]) + elif line.tokens[1] == "NICK": + del(self.users_in_chan[self.users_in_chan.index(line.sender)]) + self.users_in_chan += [line.receiver] def parse_command_line_arguments(): @@ -443,7 +508,7 @@ def parse_command_line_arguments(): + str(PORT) + ")") parser.add_argument("-w, --wait", action="store", dest="timeout", type=int, default=TIMEOUT, - help="timeout in seconds after which to attempt " + + help="timeout in seconds after which to attempt " "reconnect (default: " + str(TIMEOUT) + ")") parser.add_argument("-u, --username", action="store", dest="username", default=USERNAME, help="username to use (default: " @@ -456,6 +521,10 @@ def parse_command_line_arguments(): + TWTFILE + ")") parser.add_argument("-d, --dbdir", action="store", dest="dbdir", default=DBDIR, help="directory to store DB files in") + parser.add_argument("-r, --rmlogs", action="store", dest="rmlogs", + type=int, default=0, + help="maximum age in seconds for logfiles in logs/ " + "(0 means: never delete, and is default)") parser.add_argument("CHANNEL", action="store", help="channel to join") opts, unknown = parser.parse_known_args() return opts @@ -468,7 +537,7 @@ while True: hash_server = hashlib.md5(opts.server.encode("utf-8")).hexdigest() dbdir = opts.dbdir + "/" + hash_server session = Session(io, opts.username, opts.nickname, opts.CHANNEL, - opts.twtfile, dbdir) + opts.twtfile, dbdir, opts.rmlogs) session.loop() except ExceptionForRestart: io.socket.close()