X-Git-Url: https://plomlompom.com/repos/?p=plomlombot-irc.git;a=blobdiff_plain;f=plomlombot.py;h=f0f387e45a3d9c583efc23d8509d454291ac46b8;hp=b5db868f7bcba32a561a9f63fdcfe9dfd0f091ac;hb=4d3d76d71ea68d969322b69491fc1433ab1f556e;hpb=40440315a169a69f79ef2eccaca6e407f9a2929a diff --git a/plomlombot.py b/plomlombot.py index b5db868..f0f387e 100755 --- a/plomlombot.py +++ b/plomlombot.py @@ -55,9 +55,59 @@ class Line: self.receiver += rune +class Log: + + def __init__(self, chandir, nickname, username, channel, rmlogs): + self.nickname = nickname + self.username = username + self.channel = channel + self.chandir = chandir + self.rmlogcycle = rmlogs + self.rawlogdir = chandir + "raw_logs/" + self.logdir = chandir + "logs/" + if not os.path.exists(self.logdir): + os.makedirs(self.logdir) + if not os.path.exists(self.rawlogdir): + os.makedirs(self.rawlogdir) + + def log(self, line, sent=False): + identity = "" + separator = " > " + if sent: + separator = " " + line = Line("< " + line) + line.sender = self.nickname + identity = self.username + "@localhost" + else: + if type(line) == str: + line = Line(line) + now = datetime.datetime.utcnow() + form = "%Y-%m-%d %H:%M:%S UTC" + write_to_file(self.rawlogdir + now.strftime("%Y-%m-%d") + ".txt", "a", + now.strftime(form) + separator + line.line + "\n") + to_log = irclog.format_logline(line, self.channel, identity) + if to_log != None: + write_to_file(self.logdir + now.strftime("%Y-%m-%d") + ".txt", "a", + now.strftime(form) + " " + to_log + "\n") + + def rmlogs(self): + if self.rmlogcycle > 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.rmlogcycle: + os.remove(f) + + def separator_line(self): + now = datetime.datetime.utcnow() + write_to_file(self.logdir + now.strftime("%Y-%m-%d") + ".txt", "a", + "-----------------------\n") + + class IO: def __init__(self, server, port, timeout): + self.log = None self.timeout = timeout self.socket = socket.socket() try: @@ -68,7 +118,10 @@ class IO: self.line_buffer = [] self.rune_buffer = "" self.last_pong = time.time() - self.servername = self.recv_line(send_ping=False).split(" ")[0][1:] + line = self.recv_line(send_ping=False) + if not line or len(line) < 1: + raise ExceptionForRestart + self.servername = line.split(" ")[0][1:] def _pingtest(self, send_ping=True): if self.last_pong + self.timeout < time.time(): @@ -84,6 +137,8 @@ class IO: print("NOT SENT LINE TO SERVER (too long): " + msg) print("LINE TO SERVER: " + str(datetime.datetime.now()) + ": " + msg) + if self.log != None: + self.log.log(msg, True) msg = msg + "\r\n" msg_len = len(msg) total_sent_len = 0 @@ -121,6 +176,8 @@ class IO: def recv_line(self, send_ping=True): line = self._recv_line_wrapped(send_ping) if line: + if self.log != None: + self.log.log(line) print("LINE FROM SERVER " + str(datetime.datetime.now()) + ": " + line) return line @@ -141,21 +198,27 @@ def handle_command(command, argument, notice, target, session): def quote(): def help(): - notice("syntax: !quote [int] OR !quote search QUERY") + notice("syntax: !quote [int] OR !quote search QUERY " + "OR !quote offset-search [int] QUERY") notice("QUERY may be a boolean grouping of quoted or unquoted " + "search terms, examples:") notice("!quote search foo") notice("!quote search foo AND (bar OR NOT baz)") notice("!quote search \"foo\\\"bar\" AND ('NOT\"' AND \"'foo'\"" + " OR 'bar\\'baz')") + notice("The offset-search int argument defines how many matches " + "to skip (useful if results are above maximum number to " + "display).") if "" == argument: tokens = [] else: tokens = argument.split(" ") - if (len(tokens) > 1 and tokens[0] != "search") or \ - (len(tokens) == 1 and - (tokens[0] == "search" or not tokens[0].isdigit())): + if (len(tokens) == 1 and not tokens[0].isdigit()) or \ + (tokens > 1 and + tokens[0] not in {"search", "offset-search"} or + (tokens[0] == "offset-search" and + ((not len(tokens) > 2) or (not tokens[1].isdigit())))): help() return if not os.access(session.quotesfile, os.F_OK): @@ -172,7 +235,12 @@ def handle_command(command, argument, notice, target, session): return i = i - 1 elif len(tokens) > 1: - query = str.join(" ", tokens[1:]) + to_skip = 0 + if tokens[0] == "search": + query = str.join(" ", tokens[1:]) + elif tokens[0] == "offset-search": + to_skip = int(tokens[1]) + query = str.join(" ", tokens[2:]) try: results = plomsearch.search(query, lines) except plomsearch.LogicParserError as err: @@ -181,20 +249,25 @@ def handle_command(command, argument, notice, target, session): if len(results) == 0: notice("no quotes matching query") else: - 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]) + if to_skip >= len(results): + notice("skipped all quotes matching query") + else: + notice("found %s matches, showing max. 3, skipping %s" + % (len(results), to_skip)) + for i in range(len(results)): + if i >= to_skip and i < to_skip + 3: + result = results[i] + notice("quote #" + str(result[0] + 1) + ": " + + result[1][:-1]) return else: i = random.randrange(len(lines)) 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 = [] @@ -215,8 +288,26 @@ 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. @@ -230,7 +321,7 @@ 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: + if len(tokens) - 1 <= select_length: notice("not enough text to markov") return @@ -266,8 +357,8 @@ 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" @@ -379,50 +470,31 @@ def handle_url(url, notice, show_url=False): class Session: - def __init__(self, io, username, nickname, channel, twtfile, dbdir, rmlogs): + def __init__(self, io, username, nickname, channel, twtfile, dbdir, rmlogs, + markov_input): 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 + hash_channel = hashlib.md5(channel.encode("utf-8")).hexdigest() + chandir = dbdir + "/" + hash_channel + "/" + self.markov_input = markov_input + self.markovfile = chandir + "markovfeed" + self.quotesfile = chandir + "quotes" + self.log = Log(chandir, self.nickname, username, channel, rmlogs) self.io.send_line("NICK " + self.nickname) - 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" + self.io.send_line("USER " + username + " 0 * : ") + self.io.send_line("JOIN " + channel) + self.io.log = self.log + self.log.separator_line() def loop(self): - def log(line): - if type(line) == str: - line = Line(":" + self.nickname + "!~" + self.username + - "@localhost" + " " + line) - now = datetime.datetime.utcnow() - form = "%Y-%m-%d %H:%M:%S UTC\t" - 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: @@ -437,28 +509,20 @@ class Session: notice("maximum number of urls to parse per message " "reached") break - if "!" == msg[0]: + if "!" == msg[0] and len(msg) > 1: 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") + if self.markov_input: + 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) + self.log.rmlogs() line = self.io.recv_line() if not line: continue line = Line(line) - log(line) if len(line.tokens) > 1: if line.tokens[0] == "PING": self.io.send_line("PONG " + line.tokens[1]) @@ -507,6 +571,9 @@ def parse_command_line_arguments(): type=int, default=0, help="maximum age in seconds for logfiles in logs/ " "(0 means: never delete, and is default)") + parser.add_argument("-m, --markov_store", action="store_true", + dest="markov_store", + help="log channel discussions for !markov input") parser.add_argument("CHANNEL", action="store", help="channel to join") opts, unknown = parser.parse_known_args() return opts @@ -517,9 +584,9 @@ 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 + dbdir = opts.dbdir + "/" + hash_server session = Session(io, opts.username, opts.nickname, opts.CHANNEL, - opts.twtfile, dbdir, opts.rmlogs) + opts.twtfile, dbdir, opts.rmlogs, opts.markov_store) session.loop() except ExceptionForRestart: io.socket.close()