def process_list(url, title, raw_content, depth):
if depth <= 0:
logger.info('Reached depth limit: {0}'.format(title))
return
logger.info('Processing list {0}'.format(title))
content = BeautifulSoup(
bs_preprocess(raw_content), 'html.parser')
url_name = url_name_of(url)
WikiList.objects.get_or_create(
url_name=url_name,
defaults={'title': title})
# remove noisy links
link_list = [a for a in content.find_all('a')]
for f in LINK_FILTERS:
link_list = filter(f, link_list)
# crawl all links of the list
for link in link_list:
abs_url = urljoin(url, link['href'])
crawl_page.delay(abs_url, url_name, depth - 1)
def process_article(url, title, raw_content, parent=None):
logger.info('Processing article {0}'.format(title))
content = BeautifulSoup(
bs_preprocess(raw_content), 'html.parser')
body = BeautifulSoup(raw_content, 'html.parser')
for edits in body.find_all(class_='mw-editsection'):
edits.extract()
body.find(id='jump-to-nav').extract()
body = BeautifulSoup(str(body), 'html.parser')
page, exist = WikiPage.objects.get_or_create(
url_name=url_name_of(url),
defaults={
'title': title,
'origin': url})
page.body = str(body)
page.save()
# add lists
if parent:
wl, exist = WikiList.objects.get_or_create(
url_name=parent)
page.lists.add(wl)
# add categories
for cat in content.find(id='mw-normal-catlinks')\
.find_all('li'):
url_name = url_name_of_cat(cat.find('a')['href'])
if not url_name:
continue
wc, exist = WikiCategory.objects.get_or_create(
url_name=url_name,
defaults={'title': str(cat.text)})
page.categories.add(wc)