X-Git-Url: https://projects.mako.cc/source/iron-blogger/blobdiff_plain/13bce463950da66febe7acbd3fef70fd62e12535..4441af62e955d91c3c00713dd34c5e7b9f81b6b5:/scan-feeds.py?ds=inline diff --git a/scan-feeds.py b/scan-feeds.py index 9801762..0cf58f1 100755 --- a/scan-feeds.py +++ b/scan-feeds.py @@ -9,23 +9,30 @@ import dateutil.tz as tz with open('bloggers.yml') as f: users = yaml.safe_load(f.read()) -log = {} +try: + with open('out/report.yml') as f: + log = yaml.safe_load(f.read()) +except IOError: + log = {} -START = datetime.datetime(2009, 12, 21, 6) +START = datetime.datetime(2011, 10, 24, 6) def parse_published(pub): return parse(pub).astimezone(tz.tzlocal()).replace(tzinfo=None) def get_date(post): - if 'published' in post: - return post.published - return post.updated + for k in ('published', 'created', 'updated'): + if k in post: + return post[k] def get_link(post): return post.link def parse_feeds(weeks, uri): feed = feedparser.parse(uri) + + print >>sys.stderr, "Parsing: %s" % uri + if not feed.entries: print >>sys.stderr, "WARN: no entries for ", uri for post in feed.entries: @@ -37,19 +44,23 @@ def parse_feeds(weeks, uri): while len(weeks) <= wn: weeks.append([]) - weeks[wn].append(dict( - date=date, - title=post.title, - url=get_link(post))) - -for (username, u) in users.items(): - weeks = [] - print "[%s]" % (username) - for l in u['links']: - parse_feeds(weeks, l[2]) - log[username] = weeks - for (i, w) in enumerate(weeks): - print " [%d]: %s" % (i, w) + + post = dict(date=date, + title=post.title, + url=get_link(post)) + if post['url'] not in [p['url'] for p in weeks[wn]]: + weeks[wn].append(post) + +if len(sys.argv) > 1: + for username in sys.argv[1:]: + weeks = log.setdefault(username, []) + for l in users[username]['links']: + parse_feeds(weeks, l[2]) +else: + for (username, u) in users.items(): + weeks = log.setdefault(username, []) + for l in u['links']: + parse_feeds(weeks, l[2]) with open('out/report.yml', 'w') as f: yaml.safe_dump(log, f)