def _parse_comment(soup):
c = {
'author': '',
'authorLink': '',
'score': 0,
'date': 0,
'text': '',
'comments': [],
}
if soup.find('link'):
title = _soup_get_text(soup.find('link'))
if title and 'By:' in title:
c['author'] = title.strip('By:').strip()
if soup.find('dc:creator'):
c['author'] = _soup_get_text(soup.find('dc:creator'))
if soup.find('link'):
c['authorLink'] = _soup_get_text(soup.find('link'))
if soup.find('description'):
c['text'] = clean(_soup_get_text(soup.find('description')))
if soup.find('pubdate'):
c['date'] = unix(soup.find('pubdate').text)
elif soup.find('pubDate'):
c['date'] = unix(soup.find('pubDate').text)
return c
def feed(self):
too_old = datetime.now().timestamp() - settings.MAX_STORY_AGE
stories = json(lambda x: api_stories(x, self.BASE_DOMAIN),
headers={'Referer': self.BASE_DOMAIN})
if not stories: return []
stories = list(
filter(None, [
i if i.get("audience") == "everyone" else None for i in stories
]))
stories = list(
filter(None, [
i if unix(i.get('post_date')) > too_old else None
for i in stories
]))
stories.sort(key=lambda a: unix(a.get('post_date')), reverse=True)
return [self.ref_prefix(str(i.get("id"))) for i in stories or []]
def _filter_links(links, excludes=None):
too_old = datetime.now().timestamp() - settings.MAX_STORY_AGE
links = list(
filter(None, [a if _get_sitemap_date(a) else None for a in links]))
links = list(
filter(None, [
a if unix(_get_sitemap_date(a)) > too_old else None for a in links
]))
links.sort(key=lambda a: unix(_get_sitemap_date(a)), reverse=True)
links = [x.find('loc').text for x in links] or []
links = list(set(links))
if excludes:
links = list(
filter(None, [
None if any(e in link for e in excludes) else link
for link in links
]))
return links
def story(self, ref, urlref, is_manual=False):
if urlref is None:
return False
markup = xml(lambda x: urlref)
if not markup:
return False
s = {}
s['author'] = ''
s['author_link'] = ''
s['score'] = 0
s['comments'] = []
s['num_comments'] = 0
s['link'] = urlref
s['url'] = urlref
s['date'] = 0
s['title'] = ''
icons = get_icons(markup, url=urlref)
if icons:
s['icon'] = icons[0]
try:
data = extruct.extract(markup)
s = parse_extruct(s, data)
except KeyboardInterrupt:
raise
except Exception as e:
logging.error(e)
if s['title']:
s['title'] = clean(s['title'])
if s['date']:
s['date'] = unix(s['date'], tz=self.tz)
if 'disqus' in markup:
try:
s['comments'] = headless.get_comments(urlref)
except KeyboardInterrupt:
raise
except Exception as e:
logging.error(e)
if urlref.startswith('https://www.stuff.co.nz'):
s['comments'] = stuff.get_json_comments(urlref, markup)
if s['comments']:
s['comments'] = [clean_comment(c) for c in s['comments']]
s['comments'] = list(filter(bool, s['comments']))
s['num_comments'] = comment_count(s) - 1
if not is_manual and not s['date']:
return False
return s
def feed(self):
too_old = datetime.now().timestamp() - settings.MAX_STORY_AGE
stories = json(SUBSTACK_API_TOP_POSTS,
headers={'Referer': SUBSTACK_REFERER})
if not stories: return []
stories = list(
filter(None, [
i if i.get("audience") == "everyone" else None for i in stories
]))
stories = list(
filter(None, [
i if unix(i.get('post_date')) > too_old else None
for i in stories
]))
stories.sort(key=lambda a: unix(a.get('post_date')), reverse=True)
stories = [
self.ref_prefix(str(i.get("pub").get("base_url")),
str(i.get("id"))) for i in stories
]
return stories
def story(self, ref):
ref = self.strip_ref_prefix(ref)
stories = json(lambda x: api_stories(x, self.BASE_DOMAIN),
headers={'Referer': self.BASE_DOMAIN})
if not stories: return False
stories = list(
filter(None, [
i if i.get("audience") == "everyone" else None for i in stories
]))
stories = list(
filter(None,
[i if str(i.get('id')) == ref else None for i in stories]))
if len(stories) == 0:
return False
r = stories[0]
if not r:
return False
s = {}
s['author'] = ''
s['author_link'] = ''
s['date'] = unix(r.get('post_date'))
s['score'] = r.get('reactions').get('❤')
s['title'] = r.get('title', '')
s['link'] = r.get('canonical_url', '')
s['url'] = r.get('canonical_url', '')
comments = json(lambda x: api_comments(x, self.BASE_DOMAIN),
r.get('id'),
headers={'Referer': self.BASE_DOMAIN})
s['comments'] = [] if not comments else [
comment(i) for i in comments.get('comments')
]
s['comments'] = list(filter(bool, s['comments']))
s['num_comments'] = r.get('comment_count', 0)
authors = list(
filter(None, [
self._bylines(byline) for byline in r.get('publishedBylines')
]))
if len(authors):
s['author'] = authors[0].get('name')
s['author_link'] = authors[0].get('link')
markup = xml(lambda x: s['link'])
if markup:
icons = get_icons(markup, url=s['link'])
if icons:
s['icon'] = icons[0]
return s
def comment(i):
if 'body' not in i:
return False
c = {}
c['date'] = unix(i.get('date'))
c['author'] = i.get('name', '')
c['score'] = i.get('reactions').get('❤')
c['text'] = clean(i.get('body', '') or '')
c['comments'] = [comment(j) for j in i['children']]
c['comments'] = list(filter(bool, c['comments']))
return c
def story(self, ref):
ref = self.strip_ref_prefix(ref)
stories = json(SUBSTACK_API_TOP_POSTS,
headers={'Referer': SUBSTACK_REFERER})
if not stories: return False
stories = list(
filter(None, [
i if i.get("audience") == "everyone" else None for i in stories
]))
stories = list(
filter(None,
[i if str(i.get('id')) == ref else None for i in stories]))
if len(stories) == 0:
return False
r = stories[0]
if not r:
return False
s = {}
pub = r.get('pub')
base_url = pub.get('base_url')
s['author'] = pub.get('author_name')
s['author_link'] = author_link(pub.get('author_id'), base_url)
s['date'] = unix(r.get('post_date'))
s['score'] = r.get('score')
s['title'] = r.get('title', '')
s['link'] = r.get('canonical_url', '')
s['url'] = r.get('canonical_url', '')
comments = json(lambda x: api_comments(x, base_url),
r.get('id'),
headers={'Referer': base_url})
s['comments'] = [] if not comments else [
comment(i) for i in comments.get('comments')
]
s['comments'] = list(filter(bool, s['comments']))
s['num_comments'] = r.get('comment_count', 0)
return s
def clean_comment(comment):
comment['text'] = clean(comment['text'])
if isinstance(comment['date'], str):
comment['date'] = unix(comment['date'])
comment['comments'] = [clean_comment(c) for c in comment['comments']]
return comment
