home · contact · privacy
Update package requirements.
[plomlombot-irc.git] / plomlombot.py
index 80def0db6990eba4dd76fd180529e648cf8c3883..0e02f1c6dd426b60a380e8cb14dd02b43e992424 100755 (executable)
@@ -17,6 +17,7 @@ import irclog
 
 # Defaults, may be overwritten by command line arguments.
 SERVER = "irc.freenode.net"
+CHANNEL = "#plomlombot-test"
 PORT = 6667
 TIMEOUT = 240
 USERNAME = "plomlombot"
@@ -198,21 +199,27 @@ 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 \
+           (len(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):
@@ -229,7 +236,12 @@ 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:
@@ -238,11 +250,16 @@ 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))
@@ -454,21 +471,33 @@ 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, sasl, channel, twtfile, dbdir, rmlogs,
+                 markov_input, no_show_page_titles):
+        import base64
         self.io = io
         self.nickname = nickname
         self.users_in_chan = []
         self.twtfile = twtfile
         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)
+        if sasl:
+            self.io.send_line("CAP REQ :sasl")
         self.io.send_line("NICK " + self.nickname)
         self.io.send_line("USER " + username + " 0 * : ")
+        if sasl:
+            self.io.send_line("AUTHENTICATE PLAIN")
+            auth = username + '\0' + username + '\0' + sasl
+            auth_encoded = base64.b64encode(auth.encode()).decode().rstrip()
+            self.io.send_line("AUTHENTICATE " + auth_encoded)
+            self.io.send_line("CAP END")
         self.io.send_line("JOIN " + channel)
         self.io.log = self.log
         self.log.separator_line()
+        self.show_page_titles = not no_show_page_titles
 
     def loop(self):
 
@@ -482,21 +511,23 @@ class Session:
             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 self.show_page_titles:
+                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] 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")
 
         while True:
             self.log.rmlogs()
@@ -533,6 +564,8 @@ def parse_command_line_arguments():
     parser.add_argument("-p, --port", action="store", dest="port", type=int,
                         default=PORT, help="port to connect to (default : "
                         + str(PORT) + ")")
+    parser.add_argument("-c, --channel", action="store", dest="channel",
+                        default=SERVER, help="channel to join")
     parser.add_argument("-w, --wait", action="store", dest="timeout",
                         type=int, default=TIMEOUT,
                         help="timeout in seconds after which to attempt "
@@ -543,6 +576,8 @@ 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("-a, --authenticate", action="store", dest="sasl",
+                        default=None, help="SASL password (default: none)")
     parser.add_argument("-t, --twtxtfile", action="store", dest="twtfile",
                         default=TWTFILE, help="twtxt file to use (default: "
                         + TWTFILE + ")")
@@ -552,7 +587,12 @@ 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("CHANNEL", action="store", help="channel to join")
+    parser.add_argument("-m, --markov_store", action="store_true",
+                        dest="markov_store",
+                        help="log channel discussions for !markov input")
+    parser.add_argument("--no-show-page-titles", action="store_true",
+                        dest="no_show_page_titles",
+                        help="do not show page titles")
     opts, unknown = parser.parse_known_args()
     return opts
 
@@ -562,9 +602,10 @@ 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, dbdir, opts.rmlogs)
+        dbdir = opts.dbdir + "/" + hash_server
+        session = Session(io, opts.username, opts.nickname, opts.sasl, opts.channel,
+                          opts.twtfile, dbdir, opts.rmlogs, opts.markov_store,
+                          opts.no_show_page_titles)
         session.loop()
     except ExceptionForRestart:
         io.socket.close()