X-Git-Url: https://plomlompom.com/repos/?p=plomlombot-irc.git;a=blobdiff_plain;f=plomlombot.py;h=412ae2117aeb9fc73df1a105ed70014094e992f7;hp=aa7061cac656f3f6b5f768e40787ccb6fbc475c4;hb=ee64c72b59346940c29cf068dd7b1cb9425f332e;hpb=42c69ab892b5a9d2dc2c68a059e7b506a00197a0 diff --git a/plomlombot.py b/plomlombot.py index aa7061c..412ae21 100755 --- a/plomlombot.py +++ b/plomlombot.py @@ -11,6 +11,7 @@ import bs4 import random import hashlib import os +import signal import plomsearch import irclog @@ -135,12 +136,12 @@ def handle_command(command, argument, notice, target, session): 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") @@ -158,7 +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() @@ -167,7 +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: @@ -175,25 +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: if len(results) > 3: - notice("SHOWING 3 OF " + str(len(results)) + " QUOTES") + notice("showing 3 of " + str(len(results)) + " quotes") for result in results[:3]: - notice("QUOTE #" + str(result[0] + 1) + ": " + 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 = [] @@ -214,8 +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. @@ -229,8 +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 @@ -265,8 +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" @@ -298,7 +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 @@ -313,7 +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() @@ -338,8 +357,16 @@ 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=5, stream=True) + 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: @@ -347,21 +374,25 @@ def handle_url(url, notice, show_url=False): 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 + 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: @@ -416,8 +447,14 @@ class Session: 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)): - handle_url(matches[i], notice) + 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:])