X-Git-Url: https://git.kaliko.me/?a=blobdiff_plain;ds=inline;f=sid%2Ffeeds.py;h=5a33d6d5ccb40dbaa2c1f88d3c5a1795d02e16aa;hb=8d02f7db31a8587f1cd9bb31bcdffd517f6c144e;hp=043c784dcb23f4cf0484bee46badee0f2e370e27;hpb=1d5d7c371bdecaf42cd6785f8c8ab152f0eaec48;p=sid.git diff --git a/sid/feeds.py b/sid/feeds.py index 043c784..5a33d6d 100644 --- a/sid/feeds.py +++ b/sid/feeds.py @@ -17,6 +17,7 @@ import datetime import threading import time +import traceback from feedparser import parse as feed_parse @@ -46,41 +47,34 @@ class FeedMonitor(threading.Thread): threading.Thread.__init__(self) self.feeds_list = plugin.FEEDS self.tempo = plugin.TEMPO - self.bot = plugin.bot + self.plugin = plugin self.last_check = datetime.datetime.utcnow() self.seen = dict() self.thread_killed = False - def send(self, message): - """simple wrapper around bot send_message method""" - self.bot.send_message(mto=self.bot.room, - mbody=message[1], - mhtml=message[0], - mtype='groupchat') - def new_posts(self, feed): """Send new posts in feed""" parsed_feed = feed_parse(feed) # Cannot resolve address if 'status' not in parsed_feed: - self.bot.log.error('Error from "%s": %s.' % - (feed, parsed_feed.bozo_exception.__repr__())) + self.plugin.log.error('Error from "%s": %s.', + feed, parsed_feed.bozo_exception.__repr__()) return # unusual return http code if parsed_feed.status != 200: - self.bot.log.error( - 'Got code %(status)d from "%(href)s" (please update).' % - parsed_feed) + self.plugin.log.warning( + 'Got code %(status)d from "%(href)s" (please update).', + parsed_feed) return feed_updated = parsed_feed.feed.get('updated_parsed', None) # Avoid looping over all posts if possible if feed_updated and strtm_to_dtm(feed_updated) < self.last_check: - self.bot.log.debug('updated : %s' % strtm_to_dtm(feed_updated)) - self.bot.log.debug('last check: %s' % self.last_check) + self.plugin.log.debug('updated : %s', strtm_to_dtm(feed_updated)) + self.plugin.log.debug('last check: %s', self.last_check) return title = '"%s":' % parsed_feed.feed.get('title', 'n/a') @@ -92,34 +86,39 @@ class FeedMonitor(threading.Thread): if not self.seen.setdefault(feed_id): # Fills with post id when first started (prevent from posting all # entries at startup) - self.seen[feed_id] = [post.id for post in parsed_feed.entries] + self.seen[feed_id] = {p.id for p in parsed_feed.entries} return - for post in parsed_feed.entries: - if post.id not in self.seen.get(feed_id): - self.seen[feed_id].append(post.id) - self.bot.log.info(post.title) - - body = '%(title)s %(link)s' % post - text.append(body) - - xpost = dict(**post) - xpost['title'] = html_escape(xpost.get('title', 'n/a')) - xbody = '%(title)s' % xpost - xhtml.append(xbody) - + # Detecting new post + entries = {p.id for p in parsed_feed.entries} + new_entries = [p for p in parsed_feed.entries + if p.id in entries - self.seen.get(feed_id)] + for post in new_entries: + self.plugin.log.info(post.title) + + body = '%(title)s %(link)s' % post + text.append(body) + + xpost = {'title': html_escape(post.get('title', 'n/a'))} + xpost['link'] = html_escape(post.get('link',)) + xbody = '{title}'.format(**xpost) + xhtml.append(xbody) + # Updating self.seen + self.seen[feed_id] = entries if len(text) > 1: - self.send(('
'.join(xhtml), '\n'.join(text))) + self.plugin.send(self.plugin.bot.room, + {'mhtml':'
'.join(xhtml), 'mbody':'\n'.join(text)}, + mtype='groupchat') def run(self): while not self.thread_killed: - self.bot.log.debug('feeds check') + self.plugin.log.debug('feeds check') for feed in self.feeds_list: try: self.new_posts(feed) except Exception as err: - self.bot.log.error('feeds thread crashed') - self.bot.log.error(err) + self.plugin.log.error('feeds thread crashed: %s', err) + self.plugin.log.error(''.join(traceback.format_exc())) self.thread_killed = True self.last_check = datetime.datetime.utcnow() for _ in list(range(self.tempo)): @@ -138,13 +137,13 @@ class Feeds(Plugin): # 'http://www.debian.org/News/news', # DPN in french - 'http://www.debian.org/News/weekly/dwn.fr.rdf', + 'http://www.debian.org/News/weekly/dwn.fr.rdf', # Misc - 'http://rss.gmane.org/topics/excerpts/gmane.linux.debian.devel.announce', - 'http://rss.gmane.org/gmane.linux.debian.user.security.announce', - 'http://planet-fr.debian.net/users/rss20.xml', - 'http://planet.debian.org/atom.xml', + 'http://rss.gmane.org/topics/excerpts/gmane.linux.debian.devel.announce', + 'http://rss.gmane.org/gmane.linux.debian.user.security.announce', + 'http://planet-fr.debian.net/users/rss20.xml', + 'http://planet.debian.org/atom.xml', ] def __init__(self, bot): @@ -157,10 +156,16 @@ class Feeds(Plugin): self.th_mon.thread_killed = True @botcmd - def feeds(self, message, args): + def feeds(self, rcv, args): """feeds monitors debian project related feeds. !feeds : registred feeds list !feeds last : last check time""" if 'last' in args: - return 'Last feeds check: %s' % self.th_mon.last_check - return '\n'.join(Feeds.FEEDS) + self.reply(rcv, 'Last feeds check: %s' % self.th_mon.last_check) + return + html = ['{1}'.format(html_escape(u), + html_escape(u[7:]) + ) for u in Feeds.FEEDS] + msg = {'mbody': 'Feeds:\n' + '\n'.join(Feeds.FEEDS), + 'mhtml': 'Feeds:
' + '
'.join(html),} + self.reply(rcv, msg)