1
0
Fork 0
mirror of https://gitlab.com/dstftw/youtube-dl.git synced 2020-11-16 09:42:26 +00:00
youtube-dl/youtube_dl/extractor/vlive.py

130 lines
4.8 KiB
Python
Raw Normal View History

2015-08-20 04:56:11 +00:00
# coding: utf-8
from __future__ import unicode_literals
2015-08-20 04:56:11 +00:00
2016-04-26 14:30:24 +00:00
import re
2016-04-29 11:29:00 +00:00
2015-08-20 04:56:11 +00:00
from .common import InfoExtractor
from ..utils import (
dict_get,
2016-04-29 11:29:00 +00:00
ExtractorError,
float_or_none,
int_or_none,
remove_start,
2015-08-20 04:56:11 +00:00
)
from ..compat import compat_urllib_parse_urlencode
2015-08-20 04:56:11 +00:00
class VLiveIE(InfoExtractor):
IE_NAME = 'vlive'
_VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
2015-08-20 04:56:11 +00:00
_TEST = {
'url': 'http://www.vlive.tv/video/1326',
2015-08-20 04:56:11 +00:00
'md5': 'cc7314812855ce56de70a06a27314983',
'info_dict': {
'id': '1326',
'ext': 'mp4',
'title': "[V LIVE] Girl's Day's Broadcast",
'creator': "Girl's Day",
'view_count': int,
2015-08-20 04:56:11 +00:00
},
}
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(
'http://www.vlive.tv/video/%s' % video_id, video_id)
2015-08-20 04:56:11 +00:00
2016-04-26 14:30:24 +00:00
video_params = self._search_regex(
r'\bvlive\.video\.init\(([^)]+)\)',
2016-04-26 14:30:24 +00:00
webpage, 'video params')
status, _, _, live_params, long_video_id, key = re.split(
r'"\s*,\s*"', video_params)[2:8]
status = remove_start(status, 'PRODUCT_')
2016-04-26 14:30:24 +00:00
if status == 'LIVE_ON_AIR' or status == 'BIG_EVENT_ON_AIR':
live_params = self._parse_json('"%s"' % live_params, video_id)
live_params = self._parse_json(live_params, video_id)
return self._live(video_id, webpage, live_params)
elif status == 'VOD_ON_AIR' or status == 'BIG_EVENT_INTRO':
if long_video_id and key:
return self._replay(video_id, webpage, long_video_id, key)
else:
status = 'COMING_SOON'
if status == 'LIVE_END':
raise ExtractorError('Uploading for replay. Please wait...',
expected=True)
elif status == 'COMING_SOON':
raise ExtractorError('Coming soon!', expected=True)
2016-04-26 14:30:24 +00:00
elif status == 'CANCELED':
raise ExtractorError('We are sorry, '
'but the live broadcast has been canceled.',
expected=True)
else:
raise ExtractorError('Unknown status %s' % status)
def _get_common_fields(self, webpage):
2015-08-20 04:56:11 +00:00
title = self._og_search_title(webpage)
2016-04-26 14:30:24 +00:00
creator = self._html_search_regex(
r'<div[^>]+class="info_area"[^>]*>\s*<a\s+[^>]*>([^<]+)',
webpage, 'creator', fatal=False)
thumbnail = self._og_search_thumbnail(webpage)
return {
'title': title,
'creator': creator,
'thumbnail': thumbnail,
}
2015-09-05 07:25:19 +00:00
2016-04-26 14:30:24 +00:00
def _live(self, video_id, webpage, live_params):
formats = []
for vid in live_params.get('resolutions', []):
formats.extend(self._extract_m3u8_formats(
vid['cdnUrl'], video_id, 'mp4',
m3u8_id=vid.get('name'),
fatal=False, live=True))
self._sort_formats(formats)
return dict(self._get_common_fields(webpage),
2016-04-29 11:29:00 +00:00
id=video_id,
formats=formats,
is_live=True)
2016-04-26 14:30:24 +00:00
def _replay(self, video_id, webpage, long_video_id, key):
playinfo = self._download_json(
'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?%s'
% compat_urllib_parse_urlencode({
'videoId': long_video_id,
'key': key,
'ptc': 'http',
'doct': 'json', # document type (xml or json)
'cpt': 'vtt', # captions type (vtt or ttml)
}), video_id)
2015-08-20 04:56:11 +00:00
formats = [{
'url': vid['source'],
'format_id': vid.get('encodingOption', {}).get('name'),
'abr': float_or_none(vid.get('bitrate', {}).get('audio')),
'vbr': float_or_none(vid.get('bitrate', {}).get('video')),
'width': int_or_none(vid.get('encodingOption', {}).get('width')),
'height': int_or_none(vid.get('encodingOption', {}).get('height')),
'filesize': int_or_none(vid.get('size')),
} for vid in playinfo.get('videos', {}).get('list', []) if vid.get('source')]
2015-08-20 04:56:11 +00:00
self._sort_formats(formats)
view_count = int_or_none(playinfo.get('meta', {}).get('count'))
2015-08-20 04:56:11 +00:00
subtitles = {}
for caption in playinfo.get('captions', {}).get('list', []):
lang = dict_get(caption, ('language', 'locale', 'country', 'label'))
if lang and caption.get('source'):
subtitles[lang] = [{
'ext': 'vtt',
'url': caption['source']}]
2015-08-20 04:56:11 +00:00
2016-04-26 14:30:24 +00:00
return dict(self._get_common_fields(webpage),
2016-04-29 11:29:00 +00:00
id=video_id,
formats=formats,
view_count=view_count,
subtitles=subtitles)