w/youtube-dl
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

[vk] Add support for running live streams (Closes #10799)

Browse source
This commit is contained in:
Sergey M․ 2016-09-29 23:21:39 +07:00
parent 93aa0b6318
commit 475f8a4580
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
1 changed files with 33 additions and 14 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

47
youtube_dl/extractor/vk.py
View file

@ -253,6 +253,12 @@ class VKIE(VKBaseIE):
'duration': 5239, 'duration': 5239,
}, },
}, },
{
# live stream, hls and rtmp links,most likely already finished live
# stream by the time you are reading this comment
'url': 'https://vk.com/video-140332_456239111',
'only_matching': True,
},
{ {
# removed video, just testing that we match the pattern # removed video, just testing that we match the pattern
'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a', 'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
@ -361,6 +367,11 @@ class VKIE(VKBaseIE):
data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars') data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars')
data = json.loads(data_json) data = json.loads(data_json)
title = unescapeHTML(data['md_title'])
if data.get('live') == 2:
title = self._live_title(title)
# Extract upload date # Extract upload date
upload_date = None upload_date = None
mobj = re.search(r'id="mv_date(?:_views)?_wrap"[^>]*>([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page) mobj = re.search(r'id="mv_date(?:_views)?_wrap"[^>]*>([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
@ -377,25 +388,33 @@ class VKIE(VKBaseIE):
r'([\d,.]+)', views, 'view count', fatal=False)) r'([\d,.]+)', views, 'view count', fatal=False))
formats = [] formats = []
for k, v in data.items(): for format_id, format_url in data.items():
if (not k.startswith('url') and not k.startswith('cache') if not isinstance(format_url, compat_str) or not format_url.startswith(('http', '//', 'rtmp')):
and k not in ('extra_data', 'live_mp4')):
continue continue
if not isinstance(v, compat_str) or not v.startswith('http'): if format_id.startswith(('url', 'cache')) or format_id in ('extra_data', 'live_mp4'):
continue height = int_or_none(self._search_regex(
height = int_or_none(self._search_regex( r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
r'^(?:url|cache)(\d+)', k, 'height', default=None)) formats.append({
formats.append({ 'format_id': format_id,
'format_id': k, 'url': format_url,
'url': v, 'height': height,
'height': height, })
}) elif format_id == 'hls':
formats.extend(self._extract_m3u8_formats(
format_url, video_id, 'mp4', m3u8_id=format_id,
fatal=False, live=True))
elif format_id == 'rtmp':
formats.append({
'format_id': format_id,
'url': format_url,
'ext': 'flv',
})
self._sort_formats(formats) self._sort_formats(formats)
return { return {
'id': compat_str(data['vid']), 'id': compat_str(data.get('vid') or video_id),
'formats': formats, 'formats': formats,
'title': unescapeHTML(data['md_title']), 'title': title,
'thumbnail': data.get('jpg'), 'thumbnail': data.get('jpg'),
'uploader': data.get('md_author'), 'uploader': data.get('md_author'),
'duration': data.get('duration'), 'duration': data.get('duration'),