depau/youtube-dl
1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-11-22 10:11:52 +00:00
Code Issues Releases Wiki Activity

[yourporn] added support for posts with multiple videos

Browse source 
closes #27554
This commit is contained in:
schn0sch 2020-12-27 21:48:24 +01:00
parent 794771a164
commit 7907e765bf
1 changed files with 64 additions and 31 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

69
youtube_dl/extractor/yourporn.py
View file

@ -30,38 +30,71 @@ class YourPornIE(InfoExtractor):
}]
def _real_extract(self, url):
video_id = self._match_id(url)
get_duration = lambda webpage: parse_duration(self._search_regex(
r'duration\s*:\s*<[^>]+>([\d:]+)', webpage, 'duration',
default=None))
webpage = self._download_webpage(url, video_id)
# Only for posts containing a single video is the post_id equal to
# video_id. If there a multiple videos there also exists a post with the
# video_id and we use this page to fetch the video title.
post_id = self._match_id(url)
parts = self._parse_json(
webpage = self._download_webpage(url, post_id)
videos = self._parse_json(
self._search_regex(
r'data-vnfo=(["\'])(?P<data>{.+?})\1', webpage, 'data info',
group='data'),
video_id)[video_id].split('/')
post_id)
for video_id in videos:
parts = videos[video_id].split('/')
num = 0
for c in parts[6] + parts[7]:
if c.isnumeric():
num += int(c)
parts[5] = compat_str(int(parts[5]) - num)
parts[1] += '8'
video_url = urljoin(url, '/'.join(parts))
videos[video_id] = urljoin(url, '/'.join(parts))
title = (self._search_regex(
r'<[^>]+\bclass=["\']PostEditTA[^>]+>([^<]+)', webpage, 'title',
default=None) or self._og_search_description(webpage)).strip()
thumbnail = self._og_search_thumbnail(webpage)
duration = parse_duration(self._search_regex(
r'duration\s*:\s*<[^>]+>([\d:]+)', webpage, 'duration',
default=None))
# If there ist only one video in the post (as is most likely) the
# video_id is equal to post_id and we save us from re-fetching the page
# to obtain the meta data.
# This may fail but if needed will be obtained in the next step.
titles = {
post_id: self._og_search_description(webpage, default=None)
}
thumbnails = {
post_id: self._og_search_thumbnail(webpage, default=None)
}
durations = {
post_id: get_duration(webpage)
}
return {
# obtain missing metadata for all videos in the post
for video_id in videos:
if not titles.get(video_id) or not thumbnails.get(video_id) or video_id not in durations:
webpage = self._download_webpage('https://sxyprn.com/post/%s.html' % video_id, video_id)
if not titles.get(video_id):
titles[video_id] = self._og_search_description(webpage)
if not thumbnails.get(video_id):
thumbnails[video_id] = self._og_search_thumbnail(webpage)
if video_id not in durations:
durations[video_id] = get_duration(webpage)
entries = []
for video_id in videos:
entries.append({
'id': video_id,
'url': video_url,
'title': title,
'thumbnail': thumbnail,
'duration': duration,
'url': videos[video_id],
'title': titles[video_id],
'thumbnail': thumbnails[video_id],
'duration': durations[video_id],
'age_limit': 18,
'ext': 'mp4',
}
})
if len(entries) == 1:
return entries[0]
else:
return self.playlist_result(entries, post_id, titles[post_id])