X-Git-Url: https://plomlompom.com/repos/?p=plomlombot-irc.git;a=blobdiff_plain;f=plomlombot.py;h=73533840403165d2251479d6b2c4e3e2239b79f5;hp=7bf2208ab94a71774b4f1b7c6a49cf0e58661ad0;hb=e68573e0f640a77680bd1ab51e18f70d8011caf4;hpb=3c640fc4f1d19ac34cf871eb32369b90d9381a8c diff --git a/plomlombot.py b/plomlombot.py index 7bf2208..7353384 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): @@ -95,18 +96,16 @@ class IO: def handle_command(command, argument, notice, target, session): - hash_string = hashlib.md5(target.encode("utf-8")).hexdigest() - quotesfile_name = "quotes_" + hash_string def addquote(): - if not os.access(quotesfile_name, os.F_OK): - quotesfile = open(quotesfile_name, "w") + if not os.access(session.quotesfile, os.F_OK): + quotesfile = open(session.quotesfile, "w") quotesfile.write("QUOTES FOR " + target + ":\n") quotesfile.close() - quotesfile = open(quotesfile_name, "a") + quotesfile = open(session.quotesfile, "a") quotesfile.write(argument + "\n") quotesfile.close() - quotesfile = open(quotesfile_name, "r") + quotesfile = open(session.quotesfile, "r") lines = quotesfile.readlines() quotesfile.close() notice("ADDED QUOTE #" + str(len(lines) - 1)) @@ -131,10 +130,10 @@ def handle_command(command, argument, notice, target, session): (tokens[0] == "search" or not tokens[0].isdigit())): help() return - if not os.access(quotesfile_name, os.F_OK): + if not os.access(session.quotesfile, os.F_OK): notice("NO QUOTES AVAILABLE") return - quotesfile = open(quotesfile_name, "r") + quotesfile = open(session.quotesfile, "r") lines = quotesfile.readlines() quotesfile.close() lines = lines[1:] @@ -155,14 +154,15 @@ def handle_command(command, argument, notice, target, session): notice("NO QUOTES MATCHING QUERY") else: for result in results: - notice("QUOTE #" + str(result[0] + 1) + " : " + result[1]) + notice("QUOTE #" + str(result[0] + 1) + " : " + + result[1][-1]) return else: i = random.randrange(len(lines)) - notice("QUOTE #" + str(i + 1) + ": " + lines[i]) + notice("QUOTE #" + str(i + 1) + ": " + lines[i][:-1]) def markov(): - from random import choice + from random import choice, shuffle select_length = 2 selections = [] @@ -185,20 +185,19 @@ def handle_command(command, argument, notice, target, session): selection = choice(usable_selections) return selection[select_length] - hash_string = hashlib.md5(target.encode("utf-8")).hexdigest() - markovfeed_name = "markovfeed_" + hash_string - if not os.access(markovfeed_name, os.F_OK): + if not os.access(session.markovfile, 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") + file = open(session.markovfile, "r") lines = file.readlines() file.close() tokens = [] + sentence_end_markers = ".!?)(" for line in lines: line = line.lower().replace("\n", "") - if line[-1] not in ".!?": + if line[-1] not in sentence_end_markers: line += "." tokens += line.split() if len(tokens) <= select_length: @@ -225,6 +224,7 @@ def handle_command(command, argument, notice, target, session): # For each snippet of select_length, use markov() to find continuation # token from selections. Replace present users' names with malkovich. + # Start snippets with the beginning of a sentence, if possible. for i in range(len(tokens) - select_length): token_list = [] for j in range(select_length + 1): @@ -233,6 +233,12 @@ def handle_command(command, argument, notice, target, session): snippet = [] for i in range(select_length): snippet += [""] + 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] + break msg = "" malkovich = "malkovich" while 1: @@ -314,8 +320,8 @@ def handle_url(url, notice, show_url=False): 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 + ">: " @@ -326,24 +332,41 @@ 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_channel = hashlib.md5(self.channel.encode("utf-8")).hexdigest() + self.chandir = self.dbdir + "/" + hash_channel + "/" + self.logdir = self.chandir + "logs/" + if not os.path.exists(self.logdir): + os.makedirs(self.logdir) + self.markovfile = self.chandir + "markovfeed" + self.quotesfile = self.chandir + "quotes" 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): def notice(msg): - self.io.send_line("NOTICE " + target + " :" + msg) + line = "NOTICE " + target + " :" + msg + self.io.send_line(line) + log(line) matches = re.findall("(https?://[^\s>]+)", msg) for i in range(len(matches)): @@ -353,9 +376,7 @@ class Session: argument = str.join(" ", tokens[1:]) handle_command(tokens[0], argument, notice, target, self) return - hash_string = hashlib.md5(target.encode("utf-8")).hexdigest() - markovfeed_name = "markovfeed_" + hash_string - file = open(markovfeed_name, "a") + file = open(self.markovfile, "a") file.write(msg + "\n") file.close() @@ -389,6 +410,7 @@ class Session: line = self.io.recv_line() if not line: continue + log(line) tokens = line.split(" ") if len(tokens) > 1: if tokens[0] == "PING": @@ -398,6 +420,8 @@ 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 elif tokens[1] == "JOIN": name = name_from_join_or_part(tokens) @@ -407,6 +431,7 @@ class Session: name = name_from_join_or_part(tokens) del(self.users_in_chan[self.users_in_chan.index(name)]) + def parse_command_line_arguments(): parser = argparse.ArgumentParser() parser.add_argument("-s, --server", action="store", dest="server", @@ -426,9 +451,11 @@ 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 @@ -438,8 +465,10 @@ opts = parse_command_line_arguments() while True: try: io = IO(opts.server, opts.port, opts.timeout) + 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) + opts.twtfile, dbdir) session.loop() except ExceptionForRestart: io.socket.close()