X-Git-Url: https://plomlompom.com/repos/?p=plomlombot-irc.git;a=blobdiff_plain;f=plomlombot.py;h=7f412ea63d5505e11f68b90893c515cbdbe129e4;hp=b95bb42499f2f99f3c701b869d4ac84670efcc8b;hb=e7fab400e3f2bfa5ea5320a7ce8eebf8a01afcdd;hpb=031eeac526e6aad183d35a929680148afbb4a94e diff --git a/plomlombot.py b/plomlombot.py index b95bb42..7f412ea 100755 --- a/plomlombot.py +++ b/plomlombot.py @@ -20,6 +20,7 @@ TIMEOUT = 240 USERNAME = "plomlombot" NICKNAME = USERNAME TWTFILE = "" +DBDIR = os.path.expanduser("~/plomlombot_db") class ExceptionForRestart(Exception): @@ -96,7 +97,7 @@ class IO: def handle_command(command, argument, notice, target, session): hash_string = hashlib.md5(target.encode("utf-8")).hexdigest() - quotesfile_name = "quotes_" + hash_string + quotesfile_name = session.dbdir + "/quotes_" + hash_string def addquote(): if not os.access(quotesfile_name, os.F_OK): @@ -162,7 +163,7 @@ def handle_command(command, argument, notice, target, session): notice("QUOTE #" + str(i + 1) + ": " + lines[i]) def markov(): - from random import choice + from random import choice, shuffle select_length = 2 selections = [] @@ -186,20 +187,28 @@ def handle_command(command, argument, notice, target, session): return selection[select_length] hash_string = hashlib.md5(target.encode("utf-8")).hexdigest() - markovfeed_name = "markovfeed_" + hash_string + markovfeed_name = session.dbdir + "/markovfeed_" + hash_string if not os.access(markovfeed_name, os.F_OK): notice("NOT ENOUGH TEXT TO MARKOV.") return + + # Lowercase incoming lines, ensure they end in a sentence end mark. file = open(markovfeed_name, "r") lines = file.readlines() file.close() tokens = [] + sentence_end_markers = ".!?)(" for line in lines: - line = line.replace("\n", "").lower() + line = line.lower().replace("\n", "") + if line[-1] not in sentence_end_markers: + line += "." tokens += line.split() if len(tokens) <= select_length: notice("NOT ENOUGH TEXT TO MARKOV.") return + + # Replace URLs with escape string for now, so that the Markov selector + # won't see them as different strings. Stash replaced URLs in urls. urls = [] url_escape = "\nURL" url_starts = ["http://", "https://", " 200: break @@ -236,17 +256,21 @@ def handle_command(command, argument, notice, target, session): for i in range(select_length - 1): snippet[i] = snippet[i + 1] snippet[select_length - 1] = new_end + + # Replace occurences of url escape string with random choice from urls. while True: index = msg.find(url_escape) if index < 0: break msg = msg.replace(url_escape, choice(urls), 1) - notice(msg + "malkovich.") + + # More meaningful ways to randomly end sentences. + notice(msg + malkovich + ".") def twt(): def try_open(mode): try: - twtfile = open(session.twtfile, "w") + twtfile = open(session.twtfile, mode) except (PermissionError, FileNotFoundError) as err: notice("CAN'T ACCESS OR CREATE TWT FILE: " + str(err)) return None @@ -293,13 +317,14 @@ def handle_url(url, notice, show_url=False): except (requests.exceptions.TooManyRedirects, requests.exceptions.ConnectionError, requests.exceptions.InvalidURL, + UnicodeError, requests.exceptions.InvalidSchema) as error: notice("TROUBLE FOLLOWING URL: " + str(error)) return if mobile_twitter_hack(url): return - title = bs4.BeautifulSoup(r.text, "html.parser").title - if title: + title = bs4.BeautifulSoup(r.text, "html5lib").title + if title and title.string: prefix = "PAGE TITLE: " if show_url: prefix = "PAGE TITLE FOR <" + url + ">: " @@ -310,18 +335,30 @@ def handle_url(url, notice, show_url=False): class Session: - def __init__(self, io, username, nickname, channel, twtfile): + def __init__(self, io, username, nickname, channel, twtfile, dbdir): self.io = io self.nickname = nickname self.channel = channel self.users_in_chan = [] self.twtfile = twtfile + self.dbdir = dbdir self.io.send_line("NICK " + self.nickname) self.io.send_line("USER " + username + " 0 * : ") self.io.send_line("JOIN " + self.channel) + hash_string = hashlib.md5(self.channel.encode("utf-8")).hexdigest() + self.logdir = self.dbdir + "/irclogs_" + hash_string + "/" + if not os.path.exists(self.logdir): + os.makedirs(self.logdir) def loop(self): + def log(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): @@ -338,7 +375,7 @@ class Session: handle_command(tokens[0], argument, notice, target, self) return hash_string = hashlib.md5(target.encode("utf-8")).hexdigest() - markovfeed_name = "markovfeed_" + hash_string + markovfeed_name = self.dbdir + "/markovfeed_" + hash_string file = open(markovfeed_name, "a") file.write(msg + "\n") file.close() @@ -359,6 +396,8 @@ class Session: if receiver != self.nickname: target = receiver msg = str.join(" ", tokens[3:])[1:] + if target == self.channel: + log("<" + sender + "> " + msg) handle_input(msg, target) def name_from_join_or_part(tokens): @@ -382,14 +421,22 @@ class Session: elif tokens[1] == "353": names = tokens[5:] names[0] = names[0][1:] + for i in range(len(names)): + names[i] = names[i].replace("@", "").replace("+", "") self.users_in_chan += names + log(line) elif tokens[1] == "JOIN": name = name_from_join_or_part(tokens) if name != self.nickname: self.users_in_chan += [name] + log(line) elif tokens[1] == "PART": name = name_from_join_or_part(tokens) del(self.users_in_chan[self.users_in_chan.index(name)]) + log(line) + else: + log(line) + def parse_command_line_arguments(): parser = argparse.ArgumentParser() @@ -410,9 +457,11 @@ def parse_command_line_arguments(): parser.add_argument("-n, --nickname", action="store", dest="nickname", default=NICKNAME, help="nickname to use (default: " + NICKNAME + ")") - parser.add_argument("-t, --twtfile", action="store", dest="twtfile", - default=TWTFILE, help="twtfile to use (default: " + parser.add_argument("-t, --twtxtfile", action="store", dest="twtfile", + default=TWTFILE, help="twtxt file to use (default: " + TWTFILE + ")") + parser.add_argument("-d, --dbdir", action="store", dest="dbdir", + default=DBDIR, help="directory to store DB files in") parser.add_argument("CHANNEL", action="store", help="channel to join") opts, unknown = parser.parse_known_args() return opts @@ -423,7 +472,7 @@ while True: try: io = IO(opts.server, opts.port, opts.timeout) session = Session(io, opts.username, opts.nickname, opts.CHANNEL, - opts.twtfile) + opts.twtfile, opts.dbdir) session.loop() except ExceptionForRestart: io.socket.close()