2 import string, cPickle, random, urllib, sys, time, re, os, twitter, subprocess, datetime, urlparse
3 from collections import defaultdict
4 from irclib import irc_lower, nm_to_n
7 def karmaq(bot, cmd, nick, conn, public, karma):
9 item=cmd.split()[1].lower()
13 bot.automsg(public,nick,"I have karma on %s items." %
15 elif karma.has_key(item):
16 bot.automsg(public,nick,"%s has karma %s."
19 bot.automsg(public,nick, "%s has no karma set." % item)
22 def karmadelq(bot, cmd, nick, conn, public, karma):
24 item=cmd.split()[1].lower()
26 conn.notice(nick, "What should I delete?")
29 conn.notice(nick, "You are not my owner.")
31 if karma.has_key(item):
33 conn.notice(nick, "Item %s deleted."%item)
35 conn.notice(nick, "There is no karma stored for %s."%item)
37 # help - provides the URL of the help file
38 def helpq(bot, cmd, nick, conn, public):
39 bot.automsg(public,nick,
40 "For help see http://www.chiark.greenend.org.uk/~matthewv/irc/servus.html")
44 def infoq(bot, cmd, nick, conn, public, karma):
45 bot.automsg(public,nick,
46 ("I am Acrobat %s, on %s, as nick %s. "+
47 "My owner is %s; I have karma on %s items.") %
48 (bot.revision.split()[1], bot.channel, conn.get_nickname(),
49 bot.owner, len(karma.keys())))
51 # Check on fish stocks
54 if time.time()>=pond.quotatime:
58 if (time.time()-pond.quotatime)>pond.fish_time_inc:
59 pond.cur_fish+=(((time.time()-pond.quotatime)
60 /pond.fish_time_inc)*pond.fish_inc)
61 if pond.cur_fish>pond.max_fish:
62 pond.cur_fish=pond.max_fish
63 pond.quotatime=time.time()
65 # List of things the bot might be called to work round the self-trouting code
66 synonyms=["itself","the bot","themself"]
68 # trout someone, or flirt with them
69 def troutq(bot, cmd, nick, conn, public, cfg):
76 selftroutchance=cfg[6]
80 conn.notice(nick, quietmsg%fishpond.Boring_Git)
82 if fishpond.cur_fish<=0:
83 conn.notice(nick, nofishmsg)
85 target = string.join(cmd.split()[1:])
87 conn.notice(nick, notargetmsg)
89 me = bot.connection.get_nickname()
90 trout_msg = random.choice(fishlist)
91 fishpond.last=trout_msg
92 # The bot won't trout or flirt with itself;
93 if irc_lower(me) == irc_lower(target) or irc_lower(target) in synonyms:
95 # There's a chance the game may be given away if the request was not
98 if random.random()<=selftroutchance:
99 trout_msg=trout_msg+(selftrout%nick)
101 conn.action(bot.channel, trout_msg % target)
105 def slashq(bot, cmd, nick, conn, public, cfg):
112 selfslashchance=cfg[6]
116 conn.notice(nick, quietmsg%fishpond.Boring_Git)
118 if fishpond.cur_fish<=0:
119 conn.notice(nick, nofishmsg)
121 target = string.join(cmd.split()[1:])
122 #who = cmd.split()[1:]
123 who = ' '.join(cmd.split()[1:]).split(' / ')
125 conn.notice(nick, "it takes two to tango!")
128 conn.notice(nick, "we'll have none of that round here")
130 me = bot.connection.get_nickname()
131 slash_msg = random.choice(fishlist)
132 fishpond.last=slash_msg
133 # The bot won't slash people with themselves
134 if irc_lower(who[0]) == irc_lower(who[1]):
135 conn.notice(nick, "oooooh no missus!")
137 # The bot won't slash with itself, instead slashing the requester
139 if irc_lower(me) == irc_lower(who[n]) or irc_lower(who[n]) in synonyms:
141 # Perhaps someone asked to slash themselves with the bot then we get
142 if irc_lower(who[0]) == irc_lower(who[1]):
143 conn.notice(nick, "you wish!")
145 # There's a chance the game may be given away if the request was not
148 if random.random()<=selfslashchance:
149 slash_msg=slash_msg+(selfslash%nick)
151 conn.action(bot.channel, slash_msg % (who[0], who[1]))
155 def unitq(bot, cmd, nick, conn, public):
156 args = ' '.join(cmd.split()[1:]).split(' as ')
158 args = ' '.join(cmd.split()[1:]).split(' / ')
160 conn.notice(nick, "syntax: units arg1 as arg2")
163 sin,sout=os.popen4(["units","--verbose","--",args[0]],"r")
165 sin,sout=os.popen4(["units","--verbose","--",args[0],args[1]],"r")
168 #popen2 doesn't clean up the child properly. Do this by hand
170 if os.WEXITSTATUS(child[1])==0:
171 bot.automsg(public,nick,res[0].strip())
173 conn.notice(nick,'; '.join(map(lambda x: x.strip(),res)))
175 # Shut up trouting for a minute
176 def nofishq(bot, cmd, nick, conn, public, fish):
180 fish.quotatime=time.time()
181 fish.quotatime+=fish.nofish_time
182 conn.notice(nick, "Fish stocks depleted, as you wish.")
185 def reloadq(bot, cmd, nick, conn, public):
186 if not public and irc_lower(nick) == irc_lower(bot.owner):
189 conn.notice(nick, "Config reloaded.")
191 conn.notice(nick, "Config reloading failed!")
193 bot.automsg(public,nick,
194 "Configuration can only be reloaded by my owner, by /msg.")
197 def quitq(bot, cmd, nick, conn, public):
198 if irc_lower(nick) == irc_lower(bot.owner):
199 bot.die(msg = "I have been chosen!")
201 conn.notice(nick, "Such aggression in public!")
203 conn.notice(nick, "You're not my owner.")
205 # google for something
206 def googleq(bot, cmd, nick, conn, public):
207 cmdrest = string.join(cmd.split()[1:])
208 # "I'm Feeling Lucky" rather than try and parse the html
209 targ = ("http://www.google.com/search?q=%s&btnI=I'm+Feeling+Lucky"
210 % urllib.quote_plus(cmdrest))
212 # get redirected and grab the resulting url for returning
213 gsearch = urllib.urlopen(targ).geturl()
214 if gsearch != targ: # we've found something
215 bot.automsg(public,nick,str(gsearch))
216 else: # we haven't found anything.
217 bot.automsg(public,nick,"No pages found.")
218 except IOError: # if the connection times out. This blocks. :(
219 bot.automsg(public,nick,"The web's broken. Waah!")
221 # Look up the definition of something using google
222 def defineq(bot, cmd, nick, conn, public):
223 #this doesn't work any more
224 bot.automsg(public,nick,"'define' is broken because google are bastards :(")
226 cmdrest = string.join(cmd.split()[1:])
227 targ = ("http://www.google.co.uk/search?q=define%%3A%s&ie=utf-8&oe=utf-8"
228 % urllib.quote_plus(cmdrest))
230 # Just slurp everything into a string
231 defnpage = urllib.urlopen(targ).read()
232 # For definitions we really do have to parse the HTML, sadly.
233 # This is of course going to be a bit fragile. We first look for
234 # 'Definitions of %s on the Web' -- if this isn't present we
235 # assume we have the 'no definitions found page'.
236 # The first defn starts after the following <p> tag, but as the
237 # first <li> in a <ul type="disc" class=std>
238 # Following that we assume that each definition is all the non-markup
239 # before a <br> tag. Currently we just dump out the first definition.
240 match = re.search(r"Definitions of <b>.*?</b> on the Web.*?<li>\s*([^>]*)((<br>)|(<li>))",defnpage,re.MULTILINE)
242 bot.automsg(public,nick,"Some things defy definition.")
244 # We assume google has truncated the definition for us so this
245 # won't flood the channel with text...
246 defn = " ".join(match.group(1).split("\n"))
247 bot.automsg(public,nick,defn)
248 except IOError: # if the connection times out. This blocks. :(
249 bot.automsg(public,nick,"The web's broken. Waah!")
251 # Look up a currency conversion via xe.com
252 def currencyq(bot, cmd, nick, conn, public):
253 args = ' '.join(cmd.split()[1:]).split(' as ')
254 if len(args) != 2 or len(args[0]) != 3 or len(args[1]) != 3:
255 conn.notice(nick, "syntax: currency arg1 as arg2")
257 targ = ("http://www.xe.com/ucc/convert.cgi?From=%s&To=%s" % (args[0], args[1]))
259 currencypage = urllib.urlopen(targ).read()
260 match = re.search(r"(1 %s = [\d\.]+ %s)" % (args[0].upper(),args[1].upper()),currencypage,re.MULTILINE)
262 bot.automsg(public,nick,"Dear Chief Secretary, there is no money.")
264 conversion = match.group(1);
265 conversion = conversion.replace(' ',' ');
266 bot.automsg(public,nick,conversion + " (from xe.com)")
267 except IOError: # if the connection times out. This blocks. :(
268 bot.automsg(public,nick,"The web's broken. Waah!")
271 ### extract the commit message and timestamp for commit
272 def __getcommitinfo(commit):
273 cmd=["git","log","-n","1","--pretty=format:%ct|%s",commit]
274 x=subprocess.Popen(cmd,
275 stdout=subprocess.PIPE,stderr=subprocess.PIPE)
276 out,err=x.communicate()
281 ts,mes=out.split('|')
282 when=datetime.date.fromtimestamp(float(ts))
283 return mes.strip(), when
285 ###Return an array of commit messages and timestamps for lines in db that match what
286 def __getcommits(db,keys,what):
290 ret=__getcommitinfo(db[k])
291 if len(ret)==1: #error message
292 return ["Error message from git blame: %s" % ret]
294 ans.append( (k,ret[0],ret[1]) )
297 ###search all three databases for what
298 def __getall(tdb,tdbk,fdb,fdbk,sdb,sdbk,what):
301 tans=__getcommits(tdb,tdbk,what)
302 fans=__getcommits(fdb,fdbk,what)
303 sans=__getcommits(sdb,sdbk,what)
304 return tans+fans+sans
306 def blameq(bot,cmd,nick,conn,public,fish,tdb,tdbk,fdb,fdbk,sdb,sdbk):
309 bot.automsg(public,nick,"Who or what do you want to blame?")
311 cwhat=' '.join(clist[2:])
312 if clist[1]=="#last":
313 ans=__getall(tdb,tdbk,fdb,fdbk,sdb,sdbk,fish.last)
314 elif clist[1]=="#trouts" or clist[1]=="#trout":
315 ans=__getcommits(tdb,tdbk,cwhat)
316 elif clist[1]=="#flirts" or clist[1]=="#flirt":
317 ans=__getcommits(fdb,fdbk,cwhat)
318 elif clist[1]=="#slashes" or clist[1]=="#slash":
319 ans=__getcommits(sdb,sdbk,cwhat)
321 cwhat=' '.join(clist[1:])
322 ans=__getall(tdb,tdbk,fdb,fdbk,sdb,sdbk,cwhat)
324 bot.automsg(public,nick,"No match found")
327 bot.automsg(public,nick,ans[0])
329 bot.automsg(public,nick,"Modified %s: %s" % (ans[0][2].isoformat(),ans[0][1]))
331 bot.automsg(public,nick,"I found %d matches, which is too many. Please be more specific!" % (len(ans)) )
335 bot.automsg(public,nick,a)
337 bot.automsg(public,nick,"'%s' modified on %s: %s" % (a[0],a[2].isoformat(),a[1]))
339 ### say to msg/channel
340 def sayq(bot, cmd, nick, conn, public):
341 if irc_lower(nick) == irc_lower(bot.owner):
342 conn.privmsg(bot.channel, string.join(cmd.split()[1:]))
345 conn.notice(nick, "You're not my owner!")
347 ### action to msg/channel
348 def doq(bot, cmd, nick, conn, public):
349 sys.stderr.write(irc_lower(bot.owner))
350 sys.stderr.write(irc_lower(nick))
352 if irc_lower(nick) == irc_lower(bot.owner):
353 conn.action(bot.channel, string.join(cmd.split()[1:]))
355 conn.notice(nick, "You're not my owner!")
358 def disconnq(bot, cmd, nick, conn, public):
359 if cmd == "disconnect": # hop off for 60s
360 bot.disconnect(msg="Be right back.")
362 ### list keys of a dictionary
363 def listkeysq(bot, cmd, nick, conn, public, dict, sort=False):
367 bot.automsg(public,nick,string.join(d))
369 ### rot13 text (yes, I could have typed out the letters....)
370 ### also "foo".encode('rot13') would have worked
371 def rot13q(bot, cmd, nick, conn, public):
372 a=''.join(map(chr,range((ord('a')),(ord('z')+1))))
374 trans=string.maketrans(a+a.upper(),b+b.upper())
375 conn.notice(nick, string.join(cmd.split()[1:]).translate(trans))
377 ### URL-tracking stuff
379 ### return a easy-to-read approximation of a time period
380 def nicetime(tempus):
382 tm="%d seconds ago"%int(tempus)
384 tm="%d minutes ago"%int(tempus/60)
386 tm="%d hours ago"%int(tempus/3600)
389 ### class to store URL data
391 "contains meta-data about a URL seen on-channel"
392 def __init__(self,url,nick):
395 self.first=time.time()
397 self.lastseen=time.time()
398 self.lastasked=time.time()
399 def recenttime(self):
400 return max(self.lastseen,self.lastasked)
402 return nicetime(time.time()-self.first)
404 z=min(len(urlinfos)-1, self.count-1)
407 #(?:) is a regexp that doesn't group
408 urlre = re.compile(r"((?:(?:http)|(?:nsfw))s?://[^ ]+)( |$)")
409 hturlre= re.compile(r"(http)(s?://[^ ]+)( |$)")
410 #matches \bre\:?\s+ before a regexp; (?i)==case insensitive match
411 shibboleth = re.compile(r"(?i)\bre\:?\s+((?:(?:http)|(?:nsfw))s?://[^ ]+)( |$)")
412 #How long (in s) to wait since the most recent mention before commenting
413 url_repeat_time = 300
419 ### Deal with /msg bot url or ~url in channel
420 def urlq(bot, cmd, nick, conn, public,urldb):
421 if (not urlre.search(cmd)):
422 bot.automsg(False,nick,"Please use 'url' only with http, https, nsfw, or nsfws URLs")
425 urlstring=urlre.search(cmd).group(1)
426 url=canonical_url(urlstring)
429 complaint="That's %s URL that was first mentioned %s by %s" % \
430 (T.urltype(),T.firstmen(),T.nick)
432 complaint=complaint+". Furthermore it defeats the point of this command to use it other than via /msg."
434 bot.automsg(False,nick,complaint)
435 T.lastasked=time.time()
436 #URL suppressed, so mention in #urls
437 if urlstring != cmd.split()[1]: #first argument to URL was not the url
438 conn.privmsg("#urls","%s remarks: %s" % (nick," ".join(cmd.split()[1:])))
440 conn.privmsg("#urls","(via %s) %s"%(nick," ".join(cmd.split()[1:])))
443 bot.automsg(False,nick,"That URL was unique. There is little point in using !url out loud; please use it via /msg")
445 if urlstring != cmd.split()[1]: #first argument to URL was not the url
446 conn.privmsg(bot.channel,"%s remarks: %s" % (nick," ".join(cmd.split()[1:])))
448 conn.privmsg(bot.channel,"(via %s) %s"%(nick," ".join(cmd.split()[1:])))
449 urldb[url]=UrlLog(url,nick)
451 ### Deal with URLs spotted in channel
452 def dourl(bot,conn,nick,command,urldb):
453 urlstring=urlre.search(command).group(1)
454 urlstring=canonical_url(urlstring)
456 if urlstring in urldb:
458 message="observes %s URL, first mentioned %s by %s" % \
459 (T.urltype(),T.firstmen(),T.nick)
460 if shibboleth.search(command)==None and \
461 time.time() - T.lastseen > url_repeat_time:
462 conn.action(bot.channel, message)
463 T.lastseen=time.time()
466 urldb[urlstring]=UrlLog(urlstring,nick)
469 def urlexpire(urldb,expire):
472 if time.time() - urldb[u].recenttime() > expire:
475 # canonicalise BBC URLs (internal use only)
476 def canonical_url(urlstring):
477 if "nsfw://" in urlstring or "nsfws://" in urlstring:
478 urlstring=urlstring.replace("nsfw","http",1)
479 if (urlstring.find("news.bbc.co.uk") != -1):
480 for middle in ("/low/","/mobile/"):
481 x = urlstring.find(middle)
483 urlstring.replace(middle,"/hi/")
486 # automatically make nsfw urls for you and pass them on to url
487 def nsfwq(bot,cmd,nick,conn,public,urldb):
488 if (not hturlre.search(cmd)):
489 bot.automsg(False,nick,"Please use 'nsfw' only with http or https URLs")
491 newcmd=hturlre.sub(nsfwify,cmd)
492 urlq(bot,newcmd,nick,conn,public,urldb)
499 def twitterq(bot,cmd,nick,conn,public,twitapi):
501 if (not urlre.search(cmd)):
502 bot.automsg(False,nick,"Please use 'twit' only with http or https URLs")
505 urlstring = urlre.search(cmd).group(1)
506 if (urlstring.find("twitter.com") !=-1):
507 stringsout = getTweet(urlstring,twitapi)
508 for stringout in stringsout:
509 bot.automsg(public, nick, stringout)
511 def getTweet(urlstring,twitapi,inclusion=False,recurlvl=0):
512 unobfuscate_urls=True
513 expand_included_tweets=True
516 parts = string.split(urlstring,'/')
519 status = twitapi.GetStatus(tweetID)
521 return "twitapi.GetStatus returned nothing :-("
522 if status.user == None and status.text == None:
523 return "Empty status object returned :("
524 if status.retweeted_status and status.retweeted_status.text:
525 status = status.retweeted_status
526 if status.user is not None:
527 tweeter_screen = status.user.screen_name #.encode('UTF-8', 'replace')
528 tweeter_name = status.user.name #.encode('UTF-8', 'replace')
530 tweeter_screen = "[not returned]" ; tweeter_name = "[not returned]"
531 tweeter_name = tweeter_name + " RTing " + status.user.name #.encode('UTF-8', 'replace')
532 tweetText = status.full_text
534 replacements = defaultdict( list )
535 for medium in status.media:
536 replacements[medium.url].append(medium.media_url_https)
538 for k,v in replacements.items():
540 v = [re.sub(r"/tweet_video_thumb/([\w\-]+).jpg", r"/tweet_video/\1.mp4", link) for link in v]
542 replacementstring = "[" + " ; ".join(v) +"]"
544 replacementstring = v[0]
545 tweetText = tweetText.replace(k, replacementstring)
547 for url in status.urls:
548 toReplace = url.expanded_url
552 rv = urlparse.urlparse(toReplace)
554 # sourced from http://bit.do/list-of-url-shorteners.php
555 "bit.do", "t.co", "lnkd.in", "db.tt", "qr.ae", "adf.ly",
556 "goo.gl", "bitly.com", "cur.lv", "tinyurl.com", "ow.ly",
557 "bit.ly", "adcrun.ch", "ity.im", "q.gs", "viralurl.com",
558 "is.gd", "po.st", "vur.me", "bc.vc", "twitthis.com", "u.to",
559 "j.mp", "buzurl.com", "cutt.us", "u.bb", "yourls.org",
560 "crisco.com", "x.co", "prettylinkpro.com", "viralurl.biz",
561 "adcraft.co", "virl.ws", "scrnch.me", "filoops.info", "vurl.bz",
562 "vzturl.com", "lemde.fr", "qr.net", "1url.com", "tweez.me",
563 "7vd.cn", "v.gd", "dft.ba", "aka.gr", "tr.im",
567 #expand list as needed.
568 response = urllib.urlopen('http://urlex.org/txt/' + toReplace)
569 resptext = response.read()
570 if resptext.startswith('http'): # ie it looks urlish (http or https)
571 if resptext != toReplace:
573 # maybe make a note of the domain of the original URL to compile list of shortenable domains?
575 # remove tracking utm_ query parameters, for privacy and brevity
576 # code snippet from https://gist.github.com/lepture/5997883
577 rv = urlparse.urlparse(toReplace)
579 query = re.sub(r'utm_\w+=[^&]+&?', '', rv.query)
581 toReplace = '%s://%s%s?%s' % (rv.scheme, rv.hostname, rv.path, query)
583 toReplace = '%s://%s%s' % (rv.scheme, rv.hostname, rv.path) # leave off the final '?'
585 if expand_included_tweets:
586 if rv.hostname == 'twitter.com' and re.search(r'status/\d+',rv.path):
588 stringsout = [ "{{ Recursion level too high }}" ] + stringsout
590 quotedtweet = getTweet(toReplace, twitapi, inclusion=True, recurlvl=recurlvl+1) # inclusion parameter limits recursion.
593 quotedtweet[0] = "Q{ " + quotedtweet[0]
594 quotedtweet[-1] += " }"
595 stringsout = quotedtweet + stringsout
597 tweetText = tweetText.replace(url.url, toReplace)
599 tweetText = tweetText.replace(">",">")
600 tweetText = tweetText.replace("<","<")
601 tweetText = tweetText.replace("&","&")
602 tweetText = tweetText.replace("\n"," ")
603 stringout = "tweet by %s (%s): %s" %(tweeter_screen,tweeter_name,tweetText)
604 except twitter.TwitterError:
605 terror = sys.exc_info()
606 stringout = "Twitter error: %s" % terror[1].__str__()
608 terror = sys.exc_info()
609 stringout = "Error: %s" % terror[1].__str__()
610 stringsout = [stringout] + stringsout
612 return stringsout # don't want to double-encode it, so just pass it on for now and encode later
614 return map(lambda x: x.encode('UTF-8', 'replace'), stringsout)