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

flatten html links before searching for things to embed

Browse source
This commit is contained in:
Kyle Mahan 2015-12-07 09:03:23 -08:00
parent 6e5038e282
commit 474fcf07ba
1 changed files with 9 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

14
woodwind/views.py
View file

@ -573,20 +573,26 @@ def add_preview(content):
# don't add a preview to a post that already has one # don't add a preview to a post that already has one
return content 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()
instagram_regex = r'https?://(?:www\.)?instagram.com/p/[\w\-]+/?' instagram_regex = r'https?://(?:www\.)?instagram.com/p/[\w\-]+/?'
vimeo_regex = r'https?://(?:www\.)?vimeo.com/(\d+)/?' vimeo_regex = r'https?://(?:www\.)?vimeo.com/(\d+)/?'
youtube_regex = r'https?://(?:www\.)?youtube.com/watch\?v=([\w\-]+)' youtube_regex = r'https?://(?:www\.)?youtube.com/watch\?v=([\w\-]+)'
youtube_short_regex = r'https://youtu.be/([\w\-]+)' youtube_short_regex = r'https://youtu.be/([\w\-]+)'
twitter_regex = r'https?://(?:www\.)?twitter.com/(\w+)/status/(\d+)' twitter_regex = r'https?://(?:www\.)?twitter.com/(\w+)/status/(\d+)'
m = re.search(instagram_regex, content) m = re.search(instagram_regex, flat)
if m: if m:
ig_url = m.group(0) ig_url = m.group(0)
media_url = urllib.parse.urljoin(ig_url, 'media/?size=l') media_url = urllib.parse.urljoin(ig_url, 'media/?size=l')
return '{}<a href="{}"><img src="{}" /></a>'.format( return '{}<a href="{}"><img src="{}" /></a>'.format(
content, ig_url, media_url) content, ig_url, media_url)
m = re.search(vimeo_regex, content) m = re.search(vimeo_regex, flat)
if m: if m:
# vimeo_url = m.group(0) # vimeo_url = m.group(0)
vimeo_id = m.group(1) vimeo_id = m.group(1)
@ -596,7 +602,7 @@ def add_preview(content):
'mozallowfullscreen allowfullscreen></iframe>' 'mozallowfullscreen allowfullscreen></iframe>'
).format(content, vimeo_id) ).format(content, vimeo_id)
m = re.search(youtube_regex, content) m = re.search(youtube_regex, flat)
if not m: if not m:
m = re.search(youtube_short_regex, content) m = re.search(youtube_short_regex, content)
@ -608,8 +614,6 @@ def add_preview(content):
'allowfullscreen></iframe>' 'allowfullscreen></iframe>'
).format(content, youtube_id) ).format(content, youtube_id)
# flatten links
flat = re.sub(r'<a[^>]+href="([^"]*)"[^>]*>[^<]*</a>', r'\1', content)
m = re.search(twitter_regex + '$', flat) m = re.search(twitter_regex + '$', flat)
if m: if m:
tweet_url = m.group() tweet_url = m.group()