X-Git-Url: http://www.chiark.greenend.org.uk/ucgi/~matthewv/git?p=irc.git;a=blobdiff_plain;f=commands.py;h=86d0d2331acd6c8374219c9c740c82b40fe5e2c3;hp=8c22d293a09429aa2fab542f132376c17b1cb4e0;hb=a56e53f21a6bc6b9d3d7f3e56171cfbb110659fd;hpb=87650c67dbdf05e88db791be3df68dc8d3ae58e8 diff --git a/commands.py b/commands.py index 8c22d29..86d0d23 100755 --- a/commands.py +++ b/commands.py @@ -1,5 +1,6 @@ # Part of Acrobat. -import string, cPickle, random, urllib, sys, time, re, os, twitter, subprocess, datetime +import string, cPickle, random, urllib, sys, time, re, os, twitter, subprocess, datetime, urlparse +from collections import defaultdict from irclib import irc_lower, nm_to_n # query karma @@ -61,6 +62,9 @@ def fish_quota(pond): pond.cur_fish=pond.max_fish pond.quotatime=time.time() +# List of things the bot might be called to work round the self-trouting code +synonyms=["itself","the bot","themself"] + # trout someone, or flirt with them def troutq(bot, cmd, nick, conn, public, cfg): fishlist=cfg[0] @@ -86,7 +90,7 @@ def troutq(bot, cmd, nick, conn, public, cfg): trout_msg = random.choice(fishlist) fishpond.last=trout_msg # The bot won't trout or flirt with itself; - if irc_lower(me) == irc_lower(target): + if irc_lower(me) == irc_lower(target) or irc_lower(target) in synonyms: target = nick # There's a chance the game may be given away if the request was not # public... @@ -132,7 +136,7 @@ def slashq(bot, cmd, nick, conn, public, cfg): return # The bot won't slash with itself, instead slashing the requester for n in [0,1]: - if irc_lower(me) == irc_lower(who[n]): + if irc_lower(me) == irc_lower(who[n]) or irc_lower(who[n]) in synonyms: who[n] = nick # Perhaps someone asked to slash themselves with the bot then we get if irc_lower(who[0]) == irc_lower(who[1]): @@ -253,7 +257,7 @@ def currencyq(bot, cmd, nick, conn, public): targ = ("http://www.xe.com/ucc/convert.cgi?From=%s&To=%s" % (args[0], args[1])) try: currencypage = urllib.urlopen(targ).read() - match = re.search(r"(1 %s = [\d\.]+ %s)" % (args[0],args[1]),currencypage,re.MULTILINE) + match = re.search(r"(1 %s = [\d\.]+ %s)" % (args[0].upper(),args[1].upper()),currencypage,re.MULTILINE) if match == None: bot.automsg(public,nick,"Dear Chief Secretary, there is no money.") else: @@ -437,7 +441,6 @@ def urlq(bot, cmd, nick, conn, public,urldb): conn.privmsg(bot.channel,"%s remarks: %s" % (nick," ".join(cmd.split()[1:]))) else: conn.privmsg(bot.channel,"(via %s) %s"%(nick," ".join(cmd.split()[1:]))) - bot.automsg(False,nick,"That URL was unique; I have posted it into IRC") urldb[url]=UrlLog(url,nick) ### Deal with URLs spotted in channel @@ -490,7 +493,7 @@ def nsfwify(match): def twitterq(bot,cmd,nick,conn,public,twitapi): if (not urlre.search(cmd)): - bot.automsg(False,nick,"Please use 'twit' only with http URLs") + bot.automsg(False,nick,"Please use 'twit' only with http or https URLs") return urlstring = urlre.search(cmd).group(1) @@ -498,17 +501,96 @@ def twitterq(bot,cmd,nick,conn,public,twitapi): stringout = getTweet(urlstring,twitapi) bot.automsg(public, nick, stringout) -def getTweet(urlstring,twitapi): +def getTweet(urlstring,twitapi,inclusion=False): + unobfuscate_urls=True + expand_included_tweets=True + parts = string.split(urlstring,'/') tweetID = parts[-1] try: status = twitapi.GetStatus(tweetID) - tweeter_screen = status.user.screen_name.encode('UTF-8', 'replace') - tweeter_name = status.user.name.encode('UTF-8', 'replace') - tweetText = status.text.encode('UTF-8', 'replace') - tweetText = tweetText.replace('\n',' ') + if status == {}: + return "twitapi.GetStatus returned nothing :-(" + if status.user == None and status.text == None: + return "Empty status object returned :(" + if status.retweeted_status and status.retweeted_status.text: + status = status.retweeted_status + if status.user is not None: + tweeter_screen = status.user.screen_name #.encode('UTF-8', 'replace') + tweeter_name = status.user.name #.encode('UTF-8', 'replace') + else: + tweeter_screen = "[not returned]" ; tweeter_name = "[not returned]" + tweeter_name = tweeter_name + " RTing " + status.user.name #.encode('UTF-8', 'replace') + tweetText = status.full_text + if status.media: + replacements = defaultdict( list ) + for medium in status.media: + replacements[medium.url].append(medium.media_url_https) + + for k,v in replacements.items(): + + v = [re.sub(r"/tweet_video_thumb/([\w\-]+).jpg", r"/tweet_video/\1.mp4", link) for link in v] + if len(v) > 1: + replacementstring = "[" + " ; ".join(v) +"]" + else: + replacementstring = v[0] + tweetText = tweetText.replace(k, replacementstring) + + for url in status.urls: + toReplace = url.expanded_url + + if unobfuscate_urls: + import urllib + rv = urlparse.urlparse(toReplace) + if rv.hostname in { + # sourced from http://bit.do/list-of-url-shorteners.php + "bit.do", "t.co", "lnkd.in", "db.tt", "qr.ae", "adf.ly", + "goo.gl", "bitly.com", "cur.lv", "tinyurl.com", "ow.ly", + "bit.ly", "adcrun.ch", "ity.im", "q.gs", "viralurl.com", + "is.gd", "po.st", "vur.me", "bc.vc", "twitthis.com", "u.to", + "j.mp", "buzurl.com", "cutt.us", "u.bb", "yourls.org", + "crisco.com", "x.co", "prettylinkpro.com", "viralurl.biz", + "adcraft.co", "virl.ws", "scrnch.me", "filoops.info", "vurl.bz", + "vzturl.com", "lemde.fr", "qr.net", "1url.com", "tweez.me", + "7vd.cn", "v.gd", "dft.ba", "aka.gr", "tr.im", + # added by ASB: + "trib.al", "dlvr.it" + }: + #expand list as needed. + response = urllib.urlopen('http://urlex.org/txt/' + toReplace) + resptext = response.read() + if resptext.startswith('http'): # ie it looks urlish (http or https) + if resptext != toReplace: + toReplace = resptext + # maybe make a note of the domain of the original URL to compile list of shortenable domains? + + # remove tracking utm_ query parameters, for privacy and brevity + # code snippet from https://gist.github.com/lepture/5997883 + rv = urlparse.urlparse(toReplace) + if rv.query: + query = re.sub(r'utm_\w+=[^&]+&?', '', rv.query) + if query: + toReplace = '%s://%s%s?%s' % (rv.scheme, rv.hostname, rv.path, query) + else: + toReplace = '%s://%s%s' % (rv.scheme, rv.hostname, rv.path) # leave off the final '?' + + if expand_included_tweets and not inclusion: + if rv.hostname == 'twitter.com' and re.search(r'status/\d+',rv.path): + quotedtweet = getTweet(toReplace, twitapi, inclusion=True) # inclusion parameter limits recursion. + tweetText += " Q{" + quotedtweet + "}" + tweetText = tweetText.replace(url.url, toReplace) + + tweetText = tweetText.replace(">",">") + tweetText = tweetText.replace("<","<") + tweetText = tweetText.replace("&","&") + tweetText = tweetText.replace("\n"," ") stringout = "tweet by %s (%s): %s" %(tweeter_screen,tweeter_name,tweetText) except twitter.TwitterError: terror = sys.exc_info() stringout = "Twitter error: %s" % terror[1].__str__() - return stringout + except Exception: + terror = sys.exc_info() + stringout = "Error: %s" % terror[1].__str__() + if inclusion: + return stringout # don't want to double-encode it, so just pass it on for now and encode later + return stringout.encode('UTF-8', 'replace')