depau/youtube-dl
1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-05-18 19:19:32 +00:00
Code Issues Releases Wiki Activity

[pornhub] Improve and simplify (closes #19135)

Browse source
This commit is contained in:
Sergey M․ 2019-02-05 23:06:55 +07:00
parent d2d970d07e
commit 5dda1edef9
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
1 changed files with 6 additions and 11 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

17
youtube_dl/extractor/pornhub.py
View file

@ -302,17 +302,12 @@ class PornHubIE(PornHubBaseIE):
comment_count = self._extract_count(
r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
def _get_items(class_name):
def extract_list(meta_key):
div = self._search_regex(
r'<div class="' + class_name + '">([\S\s]+?)</div>',
webpage, class_name, default=None)
r'(?s)<div[^>]+\bclass=["\'].*?\b%sWrapper[^>]*>(.+?)</div>'
% meta_key, webpage, meta_key, default=None)
if div:
return [a for a in re.findall(r'<a href=[^>]+>([^<]+)', div)]
else:
return None
categories = _get_items('categoriesWrapper')
tags = _get_items('tagsWrapper')
return re.findall(r'<a[^>]+\bhref=[^>]+>([^<]+)', div)
return {
'id': video_id,
@ -327,8 +322,8 @@ class PornHubIE(PornHubBaseIE):
'comment_count': comment_count,
'formats': formats,
'age_limit': 18,
'tags': tags,
'categories': categories,
'tags': extract_list('tags'),
'categories': extract_list('categories'),
'subtitles': subtitles,
}