mirror of
https://github.com/Mikaela/Limnoria.git
synced 2024-11-30 14:59:34 +01:00
457 lines
16 KiB
Python
Executable File
457 lines
16 KiB
Python
Executable File
#!/usr/bin/env python
|
|
# -*- coding:utf-8 -*-
|
|
|
|
###
|
|
# Copyright (c) 2004, Stéphan Kochen
|
|
# All rights reserved.
|
|
#
|
|
# Redistribution and use in source and binary forms, with or without
|
|
# modification, are permitted provided that the following conditions are met:
|
|
#
|
|
# * Redistributions of source code must retain the above copyright notice,
|
|
# this list of conditions, and the following disclaimer.
|
|
# * Redistributions in binary form must reproduce the above copyright notice,
|
|
# this list of conditions, and the following disclaimer in the
|
|
# documentation and/or other materials provided with the distribution.
|
|
# * Neither the name of the author of this software nor the name of
|
|
# contributors to this software may be used to endorse or promote products
|
|
# derived from this software without specific prior written consent.
|
|
#
|
|
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
|
|
# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
# ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
|
|
# LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
|
|
# CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
|
|
# SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
|
|
# INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
|
|
# CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
|
|
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
|
|
# POSSIBILITY OF SUCH DAMAGE.
|
|
###
|
|
|
|
"""
|
|
A plugin that tries to emulate Infobot somewhat faithfully.
|
|
"""
|
|
|
|
__revision__ = "$Id$"
|
|
|
|
import plugins
|
|
|
|
import re
|
|
import anydbm
|
|
import random
|
|
import os.path
|
|
|
|
import log
|
|
import conf
|
|
import ircmsgs
|
|
import ircutils
|
|
import privmsgs
|
|
import registry
|
|
import callbacks
|
|
|
|
conf.registerPlugin('Infobot')
|
|
|
|
conf.registerGlobalValue(conf.supybot.plugins.Infobot, 'infobotStyleStatus',
|
|
registry.Boolean(False, """Whether to reply to the status command with an
|
|
original Infobot style message or with a short message."""))
|
|
# FIXME: rename; description
|
|
conf.registerChannelValue(conf.supybot.plugins.Infobot,
|
|
'catchWhenNotAddressed', registry.Boolean(True, """Whether to catch
|
|
non-addressed stuff at all."""))
|
|
conf.registerChannelValue(conf.supybot.plugins.Infobot,
|
|
'replyQuestionWhenNotAddressed', registry.Boolean(True, """Whether to
|
|
answer to non-addressed stuff we know about."""))
|
|
conf.registerChannelValue(conf.supybot.plugins.Infobot,
|
|
'replyDontKnowWhenNotAddressed', registry.Boolean(False, """Whether to
|
|
answer to non-addressed stuff we don't know about."""))
|
|
conf.registerChannelValue(conf.supybot.plugins.Infobot,
|
|
'replyWhenNotAddressed', registry.Boolean(False, """Whether to answer to
|
|
non-addressed, non-question stuff."""))
|
|
|
|
# Replies
|
|
# XXX: Move this to a plaintext db of some sort?
|
|
# XXX: random.choice() doesn't like sets? :/
|
|
repliesHello = [
|
|
'Hello', 'Hi',
|
|
'Hey', 'Niihau',
|
|
'Bonjour', 'Hola',
|
|
'Salut', 'Que tal',
|
|
'Privet', 'What\'s up'
|
|
]
|
|
|
|
repliesWelcome = [
|
|
'No problem', 'My pleasure',
|
|
'Sure thing', 'No worries',
|
|
'De nada', 'De rien',
|
|
'Bitte', 'Pas de quoi'
|
|
]
|
|
|
|
repliesDontKnow = [
|
|
'I don\'t know.', 'Wish I knew',
|
|
'I don\'t have a clue.', 'No idea.',
|
|
'Bugger all, I dunno.'
|
|
]
|
|
|
|
repliesConfirm = [
|
|
'Gotcha', 'Ok',
|
|
'10-4', 'I hear ya',
|
|
'Got it'
|
|
]
|
|
|
|
repliesStatement = [
|
|
'%(key)s %(verb)s %(value)s',
|
|
'I think %(key)s %(verb)s %(value)s',
|
|
'Hmmm... %(key)s %(verb)s %(value)s',
|
|
'It has been said that %(key)s %(verb)s %(value)s',
|
|
'%(key)s %(verb)s probably %(value)s',
|
|
'Rumour has it %(key)s %(verb)s %(value)s',
|
|
'I heard %(key)s %(overb)s %(value)s',
|
|
'Somebody said %(key)s %(overb)s %(value)s',
|
|
'I guess ${key}s %(verb)s %(value)s',
|
|
'Well, %(key)s %(verb)s %(key)s',
|
|
'%(key)s is, like, %(value)s',
|
|
'I\'m pretty sure %(key)s is %(value)s'
|
|
]
|
|
|
|
# XXX: Most of these are a bunch of regexps taken from Infobot.
|
|
# Hopefully this is compatible with it's license (I think it is), but
|
|
# I'm not a lawyer, so someone should take a look at it:
|
|
# http://www.opensource.org/licenses/artistic-license.php
|
|
# We should atleast leave that link and Infobot's address here I think.
|
|
canonicalizeRe = [(re.compile(r[0], re.I), r[1]) for r in [
|
|
(r"(.*)", r" \1 "),
|
|
(r"\s\s+", r" "),
|
|
|
|
(r" si ", r" is "),
|
|
(r" teh ", r "the "),
|
|
|
|
# where blah is -> where is blah
|
|
# XXX: Why two?
|
|
(r" (where|what|who) (\S+) (is|are) ", r" \1 \3 \2 "),
|
|
# XXX: Non greedy (.*) ?
|
|
(r" (where|what|who) (.*) (is|are) ", r" \1 \3 \2 "),
|
|
|
|
# XXX: Does absolutely nothing?
|
|
#(r"^\s*(.*?)\s*", r"\1"),
|
|
|
|
(r" be tellin'?g? ", r" tell "),
|
|
(r" '?bout ", r" about "),
|
|
|
|
(r",? any(hoo?w?|ways?) ", r" "),
|
|
(r",?\s?(pretty )*please\?? $", r"\?"),
|
|
|
|
# Profanity filters; just delete it.
|
|
(r" th(e|at|is) (((m(o|u)th(a|er) ?)?fuck(in'?g?)?|hell|heck|(god-?)?damn?(ed)?) ?)+ ", r" "),
|
|
(r" wtf ", r" where "),
|
|
(r" this (.*) thingy? ", r" \1 "),
|
|
(r" this thingy?( called)? ", r" "),
|
|
(r" ha(s|ve) (an?y?|some|ne) (idea|clue|guess|seen) ", r" know "),
|
|
(r" does (any|ne|some) ?(1|one|body) know ", r" "),
|
|
(r" do you know ", r" "),
|
|
(r" can (you|u|((any|ne|some) ?(1|one|body)))( please)? tell (me|us|him|her) ", r" "),
|
|
(r" where (\S+) can \S+( (a|an|the))? ", r" "),
|
|
(r" (can|do) (i|you|one|we|he|she) (find|get)( this)? ", r" is "),
|
|
(r" (i|one|we|he|she) can (find|get) ", r" is "),
|
|
(r" (the )?(add?ress?|url) (for|to) ", r" "), # this should be more specific
|
|
(r" (where is )+", r" where is "),
|
|
|
|
# Switch person.
|
|
(" \x00[s'] ", " \x00's "), # fix genitives
|
|
(r" i'm ", " \x00 is "),
|
|
(r" i('?ve| have) ", " \x00 has "),
|
|
(r" i haven'?t ", " \x00 has not "),
|
|
(r" i ", " \x00 "),
|
|
(r" am ", " is "),
|
|
(r" (me|myself) ", " \x00 "),
|
|
(r" my ", " \x00's "), # turn 'my' into name's
|
|
(r" you'?re ", r" you are "),
|
|
(r" were ", r" are "),
|
|
(r" was ", r" is ")
|
|
]]
|
|
|
|
# Regexps which replace with the bot's nickname
|
|
replaceWithSelfRe = [(re.compile(r[0], re.I), r[1]) for r in [
|
|
(r" are you ", " is \x00 "),
|
|
(r" you are ", " \x00 is "),
|
|
(r" you ", " \x00 "),
|
|
(r" your ", " \x00's "),
|
|
(r" yourself ", " \x00 ")
|
|
]]
|
|
|
|
# Check if we're addressed like: "No, supybot, bla is..."
|
|
# XXX: We assume the nickname is followed by punctuation here.
|
|
addressedRe = re.compile('^(no)[, ]+\x00\\s*([-:,]+)', re.I)
|
|
|
|
# XXX: Kinda messy
|
|
def canonicalSentence(s, myNick, hisNick, addressed):
|
|
# Because we can't alter the pattern in compiled regexps to match a
|
|
# nickname, we substitute it with \x00 here, just search for that
|
|
# instead and at the end reverse it back to normal nicks.
|
|
nickReplaceRe = re.compile(re.escape(hisNick), re.I)
|
|
s = nickReplaceRe.sub('\x00', s)
|
|
for (r, subst) in canonicalizeRe:
|
|
s = r.sub(subst, s)
|
|
s = s.replace('\x00', hisNick)
|
|
nickReplaceRe = re.compile(re.escape(myNick), re.I)
|
|
s = nickReplaceRe.sub('\x00', s)
|
|
(s, n) = addressedRe.subn(r'\1\2', s)
|
|
if n:
|
|
addressed = True
|
|
if addressed:
|
|
for (r, subst) in replaceWithSelfRe:
|
|
s = r.sub(subst, s)
|
|
s = s.replace('\x00', myNick)
|
|
return (s.strip(), addressed)
|
|
|
|
|
|
# FIXME: This should become a subclass of the database backend later on
|
|
|
|
# Factoids get stored in the db as db['is/are key'] = 'keyvalue'
|
|
# This is because the actual key in the database is lowercase so we can do
|
|
# lowercase searches, but get the actual string from the value.
|
|
class InfobotDatabase(object):
|
|
def __init__(self, filename=None):
|
|
if not filename:
|
|
filename = 'Infobot.db'
|
|
filename = os.path.join(conf.supybot.directories.data(), filename)
|
|
self._db = anydbm.open(filename, 'c')
|
|
|
|
def die(self):
|
|
self._db.close()
|
|
|
|
# Hardly any try-statements here. db[] raises our exceptions for us.
|
|
def getFactoid(self, verb, key):
|
|
factoid = {}
|
|
verb = verb.lower()
|
|
key = key.lower()
|
|
dbKey = '%s %s' % (verb, key)
|
|
factoid['key'] = self._db[dbKey][:len(key)]
|
|
factoid['verb'] = verb
|
|
if verb == 'are':
|
|
factoid['overb'] = 'were'
|
|
else:
|
|
factoid['overb'] = 'was'
|
|
factoid['value'] = self._db[dbKey][len(key):]
|
|
return factoid
|
|
|
|
def hasFactoid(self, verb, key):
|
|
verb = verb.lower()
|
|
dbKey = key.lower()
|
|
dbKey = '%s %s' % (verb, dbKey)
|
|
try:
|
|
self._db[dbKey]
|
|
return True
|
|
except KeyError:
|
|
return False
|
|
|
|
# FIXME: debugs, baleet
|
|
def insertFactoid(self, key, verb, value):
|
|
verb = verb.lower()
|
|
dbKey = key.lower()
|
|
dbKey = '%s %s' % (verb, dbKey)
|
|
try:
|
|
self._db[dbKey]
|
|
except KeyError:
|
|
log.info('ins: %s =%s= %s' % (key, verb, value))
|
|
self._db[dbKey] = '%s%s' % (key, value)
|
|
return
|
|
log.info('dup: %s !%s! %s' % (key, verb, value))
|
|
raise KeyError, key
|
|
|
|
def setFactoid(self, key, verb, value):
|
|
verb = verb.lower()
|
|
dbKey = key.lower()
|
|
dbKey = '%s %s' % (verb, dbKey)
|
|
self._db[dbKey] = '%s%s' % (key, value)
|
|
log.info('set: %s => %s' % (key, value))
|
|
|
|
def addFactoid(self, key, verb, value):
|
|
verb = verb.lower()
|
|
dbKey = key.lower()
|
|
dbKey = '%s %s' % (verb, dbKey)
|
|
# XXX: should we create new entries here too
|
|
# if the factoid doesn't exist already?
|
|
currentValue = self._db[dbKey]
|
|
newValue = '%s, or %s' % (currentValue, value)
|
|
self._db[dbKey] = newValue
|
|
log.info('add: %s +%s+ %s' % (key, verb, value))
|
|
|
|
def deleteFactoid(self, verb, key):
|
|
verb = verb.lower()
|
|
dbKey = key.lower()
|
|
dbKey = '%s %s' % (verb, dbKey)
|
|
del db[dbKey]
|
|
log.info('forgot: %s' % key)
|
|
|
|
def getNumberOfFactoids(self):
|
|
return len(self._db)
|
|
|
|
|
|
class Infobot(callbacks.Privmsg):
|
|
def __init__(self):
|
|
callbacks.Privmsg.__init__(self)
|
|
self.db = InfobotDatabase()
|
|
|
|
# Patterns for processing private messages.
|
|
statementRe = re.compile(r'^(no[-:, ]+)?(.+?)\s+(is|are|was|were)'
|
|
r'\s+(also\s+)?(.+)$', re.I)
|
|
questionRe = re.compile(r'^(?:what|where|when)\s+(is|are|was|were)'
|
|
r'\s+(.*)$', re.I)
|
|
# FIXME: Matches everything with a question mark stuck to the end. o_O
|
|
shortQuestionRe = re.compile(r'^(.+)$')
|
|
# XXX: If possible, extend this, because people tend to use
|
|
# periods inside sentences as well, not just at the end.
|
|
# (indicating a pause with triple dots for example)
|
|
splitRe = re.compile(r'\s*([^?!.]+)([?!.]*)\s*')
|
|
def doPrivmsg(self, irc, msg):
|
|
channel = privmsgs.getChannel(msg, None, raiseError=False)
|
|
message = callbacks.addressed(irc.nick, msg)
|
|
addressed = bool(message)
|
|
if not addressed:
|
|
message = msg.args[1]
|
|
message = ircutils.stripFormatting(message)
|
|
for m in self.splitRe.finditer(message):
|
|
(s, ending) = m.groups()
|
|
question = '?' in ending
|
|
(s, addressed) = canonicalSentence(s, irc.nick, msg.nick,
|
|
addressed)
|
|
log.debug('canonicalSentence(): %s' % s)
|
|
if not (addressed or self.registryValue('catchWhenNotAddressed')):
|
|
continue
|
|
# FIXME: This is a friggin mess.
|
|
# XXX: PrivmsgCommandAndRegexp should take care of this?
|
|
# (probably not)
|
|
match = self.questionRe.search(s)
|
|
proxy = callbacks.IrcObjectProxyRegexp(irc, msg)
|
|
if match:
|
|
self.question(proxy, match, addressed)
|
|
elif not question:
|
|
match = self.statementRe.search(s)
|
|
if match:
|
|
self.statement(proxy, match, addressed)
|
|
if not match and question:
|
|
match = self.shortQuestionRe.search(s)
|
|
if match:
|
|
self.shortQuestion(proxy, match, addressed)
|
|
|
|
def statement(self, irc, match, addressed):
|
|
(correction, key, verb, addition, value) = match.groups()
|
|
if self.db.hasFactoid(verb, key):
|
|
if correction:
|
|
self.db.setFactoid(key, verb, value)
|
|
elif addition:
|
|
self.db.addFactoid(key, verb, value)
|
|
elif addressed or self.registryValue('replyWhenNotAddressed'):
|
|
factoid = self.db.getFactoid(verb, key)
|
|
if factoid['value'].lower() == value.lower():
|
|
irc.reply('I already had it like that.')
|
|
else:
|
|
irc.reply('...but %s %s %s...' % (key, verb, value))
|
|
return
|
|
else:
|
|
self.db.insertFactoid(key, verb, value)
|
|
if addressed or self.registryValue('replyWhenNotAddressed'):
|
|
irc.reply(random.choice(repliesConfirm))
|
|
|
|
def question(self, irc, match, addressed):
|
|
(verb, key) = match.groups()
|
|
try:
|
|
factoid = self.db.getFactoid(verb, key)
|
|
except KeyError:
|
|
if addressed or
|
|
self.registryValue('replyDontKnowWhenNotAddressed'):
|
|
irc.reply(random.choice(repliesDontKnow))
|
|
return
|
|
if addressed or self.registryValue('replyQuestionWhenNotAddressed'):
|
|
irc.reply(random.choice(repliesStatement) % factoid)
|
|
|
|
def shortQuestion(self, irc, match, addressed):
|
|
# FIXME: clean up that regexp first.
|
|
return
|
|
key = match.group(1)
|
|
try:
|
|
factoid = self.db.getFactoid('is', key)
|
|
except KeyError:
|
|
try:
|
|
factoid = self.db.getFactoid('are', key)
|
|
except KeyError:
|
|
if addressed or
|
|
self.registryValue('replyDontKnowWhenNotAddressed'):
|
|
irc.reply(random.choice(repliesDontKnow))
|
|
return
|
|
if addressed or self.registryValue('replyQuestionWhenNotAddressed'):
|
|
irc.reply(random.choice(repliesStatement) % factoid)
|
|
|
|
def forget(self, irc, msg, args):
|
|
"""<factoid>
|
|
|
|
Make the bot forget about <factoid>."""
|
|
key = privmsgs.getArgs()
|
|
try:
|
|
self.db.deleteFactoid('is', key)
|
|
except KeyError:
|
|
try:
|
|
self.db.deleteFactoid('are', key)
|
|
except KeyError:
|
|
irc.reply('I didn\'t know about that in the first place.')
|
|
return
|
|
irc.reply('I forgot %s' % key)
|
|
|
|
def whatis(self, irc, msg, args):
|
|
"""<factoid>
|
|
|
|
Explain what <factoid> is."""
|
|
key = privmsgs.getArgs()
|
|
try:
|
|
factoid = self.db.getFactoid('is', key)
|
|
except KeyError:
|
|
try:
|
|
factoid = self.db.getFactoid('are', key)
|
|
except KeyError:
|
|
irc.reply('I don\'t know anything about %s' % key)
|
|
return
|
|
irc.reply(random.choice(repliesStatement) % factoid)
|
|
|
|
def stats(self, irc, msg, args):
|
|
"""<requires no arguments>
|
|
|
|
Display some statistics on the infobot database."""
|
|
num = self.db.getNumberOfFactoids()
|
|
if self.registryValue('infobotStyleStatus'):
|
|
# FIXME: Original infobot style status message
|
|
# <Inf0bot> Since Sat Apr 17 19:31:21 2004, there have been 0
|
|
# modifications and 0 questions. I have been awake for
|
|
# 59 seconds this session, and currently reference
|
|
# 20 factoids. Addressing is in optional mode.
|
|
pass
|
|
else:
|
|
irc.reply('I know about %s factoids.' % num)
|
|
|
|
|
|
Class = Infobot
|
|
|
|
if __name__ == '__main__':
|
|
import sys
|
|
if len(sys.argv) < 2 and sys.argv[1] not in ('is', 'are'):
|
|
print 'Usage: %s <is|are> <factpack> [<factpack> ...]' % sys.argv[0]
|
|
sys.exit(-1)
|
|
r = re.compile(r'\s+=>\s+')
|
|
db = InfobotDatabase()
|
|
for filename in sys.argv[2:]:
|
|
fd = file(filename)
|
|
for line in fd:
|
|
line = line.strip()
|
|
if not line or line[0] in ('*', '#'):
|
|
continue
|
|
else:
|
|
try:
|
|
(key, value) = r.split(line, 1)
|
|
db.setFactoid(key, sys.argv[1], value)
|
|
except Exception, e:
|
|
print 'Invalid line (%s): %r' %(utils.exnToString(e),line)
|
|
|
|
|
|
# vim:set shiftwidth=4 tabstop=8 expandtab textwidth=78:
|