diff options
Diffstat (limited to 'pyaggr3g470r')
| -rw-r--r-- | pyaggr3g470r/lib/crawler.py | 3 | ||||
| -rw-r--r-- | pyaggr3g470r/lib/utils.py | 32 |
2 files changed, 24 insertions, 11 deletions
diff --git a/pyaggr3g470r/lib/crawler.py b/pyaggr3g470r/lib/crawler.py index 2ba5403a..45b1acde 100644 --- a/pyaggr3g470r/lib/crawler.py +++ b/pyaggr3g470r/lib/crawler.py @@ -193,8 +193,7 @@ class PyAggUpdater(AbstractCrawler): 'last_modified': self.headers.get('last-modified', strftime('%a, %d %b %Y %X %Z', gmtime()))} fresh_feed = construct_feed_from(url=self.feed['link'], - fp_parsed=self.parsed_feed, - feed=self.feed) + fp_parsed=self.parsed_feed) for key in ('description', 'site_link', 'icon'): if fresh_feed.get(key) and fresh_feed[key] != self.feed.get(key): up_feed[key] = fresh_feed[key] diff --git a/pyaggr3g470r/lib/utils.py b/pyaggr3g470r/lib/utils.py index a0154b7f..041a2d29 100644 --- a/pyaggr3g470r/lib/utils.py +++ b/pyaggr3g470r/lib/utils.py @@ -37,6 +37,13 @@ def rebuild_url(url, base_split): return urllib.parse.urlunsplit(new_split) +def try_splits(url, *splits): + for split in splits: + if requests.get(rebuild_url(url, split), verify=False).ok: + return rebuild_url(url, split) + return None + + def construct_feed_from(url=None, fp_parsed=None, feed=None, query_site=True): if url is None and fp_parsed is not None: url = fp_parsed.get('url') @@ -45,7 +52,7 @@ def construct_feed_from(url=None, fp_parsed=None, feed=None, query_site=True): fp_parsed = feedparser.parse(response.content) assert url is not None and fp_parsed is not None feed = feed or {} - split = urllib.parse.urlsplit(url) + feed_split = urllib.parse.urlsplit(url) if not fp_parsed['bozo']: feed['link'] = url feed['site_link'] = try_keys(fp_parsed['feed'], 'href', 'link') @@ -56,11 +63,13 @@ def construct_feed_from(url=None, fp_parsed=None, feed=None, query_site=True): feed['site_link'] = url if feed.get('site_link'): - feed['site_link'] = rebuild_url(feed['site_link'], split) - split = urllib.parse.urlsplit(feed['site_link']) + feed['site_link'] = rebuild_url(feed['site_link'], feed_split) + site_split = urllib.parse.urlsplit(feed['site_link']) if feed.get('icon'): - feed['icon'] = rebuild_url(feed['icon'], split) + feed['icon'] = try_splits(feed['icon'], site_split, feed_split) + if feed['icon'] is None: + del feed['icon'] if not feed.get('site_link') or not query_site \ or all(bool(feed.get(key)) for key in ('link', 'title', 'icon')): @@ -91,11 +100,16 @@ def construct_feed_from(url=None, fp_parsed=None, feed=None, query_site=True): if not len(icons): icons = bs_parsed.find_all(check_keys(rel=['icon'])) if len(icons) >= 1: - feed['icon'] = rebuild_url(icons[0].attrs['href'], split) - else: # trying the default one - icon = rebuild_url('/favicon.ico', split) - if requests.get(icon, verify=False).ok: - feed['icon'] = icon + for icon in icons: + feed['icon'] = try_splits(icon.attrs['href'], + site_split, feed_split) + if feed['icon'] is not None: + break + + if feed['icon'] is None: + feed['icon'] = try_splits('/favicon.ico', site_split, feed_split) + if feed['icon'] is None: + del feed['icon'] if not feed.get('link'): alternate = bs_parsed.find_all(check_keys(rel=['alternate'], |