jeena/woodwind
1
0
Fork 0
Code Issues Pull requests Projects Packages Wiki Activity Actions

use regex instead of beautiful soup to clean tags before looking for links to embed

Browse source
This commit is contained in:
Kyle Mahan 2015-12-07 09:11:53 -08:00
parent 474fcf07ba
commit 486e3f5827
1 changed files with 5 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

10
woodwind/views.py
View file

@ -573,11 +573,11 @@ def add_preview(content):
# don't add a preview to a post that already has one
return content
# flatten links
soup = bs4.BeautifulSoup(content)
for link in soup.find_all('a'):
link.replace_with(link.get('href'))
flat = soup.get_text().strip()
# flatten links and strip tags
flat = content
flat = re.sub(r'<a [^>]*href="([^"]+)"[^>]*>[^<]*</a>', r'\1', flat)
flat = re.sub(r'</?\w+[^>]*>', '', flat)
flat = flat.strip()
instagram_regex = r'https?://(?:www\.)?instagram.com/p/[\w\-]+/?'
vimeo_regex = r'https?://(?:www\.)?vimeo.com/(\d+)/?'