mirror of
https://github.com/l1ving/youtube-dl
synced 2025-01-27 13:52:50 +08:00
[naver] extract subtitles and hls formats and reuse code in VLiveIE
This commit is contained in:
remitamine
parent
0a9fad8527
commit
a0b06b344a
@ -84,49 +84,53 @@ class NaverIE(InfoExtractor):
|
||||
},
|
||||
}]
|
||||
|
||||
def _extract_video_formats(self, formats_list):
|
||||
def _extract_video_formats(self, formats_list, vid):
|
||||
formats = []
|
||||
for format_el in formats_list:
|
||||
url = format_el.get('source')
|
||||
if url:
|
||||
encoding_option = format_el.get('encodingOption')
|
||||
bitrate = format_el.get('bitrate')
|
||||
formats.append({
|
||||
'format_id': encoding_option.get('id') or encoding_option.get('name'),
|
||||
'url': format_el['source'],
|
||||
'width': int_or_none(encoding_option.get('width')),
|
||||
'height': int_or_none(encoding_option.get('height')),
|
||||
'vbr': float_or_none(bitrate.get('video')),
|
||||
'abr': float_or_none(bitrate.get('audio')),
|
||||
'filesize': int_or_none(format_el.get('size')),
|
||||
'vcodec': format_el.get('type'),
|
||||
'ext': determine_ext(url, 'mp4'),
|
||||
})
|
||||
if format_el.get('type') == 'HLS':
|
||||
key = format_el.get('key')
|
||||
if key:
|
||||
url += '?%s=%s' % (key['name'], key['value'])
|
||||
formats.extend(self._extract_m3u8_formats(url, vid, 'mp4', m3u8_id='hls'))
|
||||
else:
|
||||
encoding_option = format_el.get('encodingOption')
|
||||
bitrate = format_el.get('bitrate')
|
||||
formats.append({
|
||||
'format_id': encoding_option.get('id') or encoding_option.get('name'),
|
||||
'url': format_el['source'],
|
||||
'width': int_or_none(encoding_option.get('width')),
|
||||
'height': int_or_none(encoding_option.get('height')),
|
||||
'vbr': float_or_none(bitrate.get('video')),
|
||||
'abr': float_or_none(bitrate.get('audio')),
|
||||
'filesize': int_or_none(format_el.get('size')),
|
||||
'vcodec': format_el.get('type'),
|
||||
'ext': determine_ext(url, 'mp4'),
|
||||
})
|
||||
if formats:
|
||||
self._sort_formats(formats)
|
||||
return formats
|
||||
|
||||
def _extract_video_info(self, vid, key):
|
||||
play_data = self._download_json(
|
||||
'http://global.apis.naver.com/linetv/rmcnmv/vod_play_videoInfo.json?' + compat_urllib_parse.urlencode({'videoId': vid, 'key': key}),
|
||||
vid, 'Downloading video info')
|
||||
def _parse_video_info(self, play_data, vid):
|
||||
meta = play_data.get('meta')
|
||||
user = meta.get('user')
|
||||
user = meta.get('user', {})
|
||||
|
||||
thumbnails = []
|
||||
for thumbnail in play_data['thumbnails']['list']:
|
||||
for thumbnail in play_data.get('thumbnails', {}).get('list', []):
|
||||
thumbnails.append({'url': thumbnail['source']})
|
||||
|
||||
formats = self._extract_video_formats(play_data['videos']['list'])
|
||||
if not formats:
|
||||
video_info = self._download_json(
|
||||
'http://serviceapi.rmcnmv.naver.com/mobile/getVideoInfo.nhn?' + compat_urllib_parse.urlencode({'videoId': vid, 'inKey': key, 'protocol': 'http'}),
|
||||
vid, 'Downloading video info')
|
||||
formats = self._extract_video_formats(video_info['videos']['list'])
|
||||
subtitles = {}
|
||||
for caption in play_data.get('captions', {}).get('list', []):
|
||||
subtitles[caption['language']] = [
|
||||
{'ext': determine_ext(caption['source'], default_ext='vtt'),
|
||||
'url': caption['source']}]
|
||||
|
||||
formats = self._extract_video_formats(play_data['videos']['list'] + play_data.get('streams', []), vid)
|
||||
|
||||
return {
|
||||
'id': vid,
|
||||
'title': meta['subject'],
|
||||
'title': meta.get('subject'),
|
||||
'formats': formats,
|
||||
'thumbnail': meta.get('cover', {}).get('source'),
|
||||
'thumbnails': thumbnails,
|
||||
@ -135,6 +139,18 @@ class NaverIE(InfoExtractor):
|
||||
'uploader': user.get('name'),
|
||||
}
|
||||
|
||||
def _extract_video_info(self, vid, key):
|
||||
play_data = self._download_json(
|
||||
'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?' + compat_urllib_parse.urlencode({'videoId': vid, 'key': key}),
|
||||
vid, 'Downloading video info')
|
||||
info = self._parse_video_info(play_data, vid)
|
||||
if not info['formats']:
|
||||
play_data = self._download_json(
|
||||
'http://serviceapi.rmcnmv.naver.com/mobile/getVideoInfo.nhn?' + compat_urllib_parse.urlencode({'videoId': vid, 'inKey': key, 'protocol': 'http'}),
|
||||
vid, 'Downloading video info')
|
||||
info['formats'] = self._extract_video_formats(play_data['videos']['list'] + play_data.get('streams', []), vid)
|
||||
return info
|
||||
|
||||
def _extract_id_and_key(self, webpage):
|
||||
m_id = re.search(r'(?s)new\s+nhn.rmcnmv.RMCVideoPlayer\(\s*["\']([^"\']+)["\']\s*,\s*(?:{[^}]*?value[^:]*?:\s*?)?["\']([^"\']+)["\']', webpage)
|
||||
if not m_id:
|
||||
|
||||
@ -6,19 +6,19 @@ from hashlib import sha1
|
||||
from base64 import b64encode
|
||||
from time import time
|
||||
|
||||
from .common import InfoExtractor
|
||||
from .naver import NaverIE
|
||||
from ..utils import (
|
||||
ExtractorError,
|
||||
determine_ext
|
||||
int_or_none,
|
||||
)
|
||||
from ..compat import compat_urllib_parse
|
||||
|
||||
|
||||
class VLiveIE(InfoExtractor):
|
||||
class VLiveIE(NaverIE):
|
||||
IE_NAME = 'vlive'
|
||||
# www.vlive.tv/video/ links redirect to m.vlive.tv/video/ for mobile devices
|
||||
_VALID_URL = r'https?://(?:(www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
|
||||
_TEST = {
|
||||
_TESTS = [{
|
||||
'url': 'http://m.vlive.tv/video/1326',
|
||||
'md5': 'cc7314812855ce56de70a06a27314983',
|
||||
'info_dict': {
|
||||
@ -27,60 +27,55 @@ class VLiveIE(InfoExtractor):
|
||||
'title': '[V] Girl\'s Day\'s Broadcast',
|
||||
'creator': 'Girl\'s Day',
|
||||
},
|
||||
}
|
||||
}]
|
||||
_SECRET = 'rFkwZet6pqk1vQt6SxxUkAHX7YL3lmqzUMrU4IDusTo4jEBdtOhNfT4BYYAdArwH'
|
||||
|
||||
def _real_extract(self, url):
|
||||
video_id = self._match_id(url)
|
||||
|
||||
webpage = self._download_webpage(
|
||||
'http://m.vlive.tv/video/%s' % video_id,
|
||||
video_id, note='Download video page')
|
||||
status = self._download_json(
|
||||
'http://www.vlive.tv/video/status?videoSeq=%s' % video_id,
|
||||
video_id, note='Download status metadata')
|
||||
|
||||
title = self._og_search_title(webpage)
|
||||
thumbnail = self._og_search_thumbnail(webpage)
|
||||
creator = self._html_search_regex(
|
||||
r'<span[^>]+class="name">([^<>]+)</span>', webpage, 'creator')
|
||||
|
||||
url = 'http://global.apis.naver.com/globalV/globalV/vod/%s/playinfo?' % video_id
|
||||
msgpad = '%.0f' % (time() * 1000)
|
||||
md = b64encode(
|
||||
hmac.new(self._SECRET.encode('ascii'),
|
||||
(url[:255] + msgpad).encode('ascii'), sha1).digest()
|
||||
)
|
||||
url += '&' + compat_urllib_parse.urlencode({'msgpad': msgpad, 'md': md})
|
||||
playinfo = self._download_json(url, video_id, 'Downloading video json')
|
||||
|
||||
if playinfo.get('message', '') != 'success':
|
||||
raise ExtractorError(playinfo.get('message', 'JSON request unsuccessful'))
|
||||
|
||||
if not playinfo.get('result'):
|
||||
raise ExtractorError('No videos found.')
|
||||
|
||||
formats = []
|
||||
for vid in playinfo['result'].get('videos', {}).get('list', []):
|
||||
formats.append({
|
||||
'url': vid['source'],
|
||||
'ext': 'mp4',
|
||||
'abr': vid.get('bitrate', {}).get('audio'),
|
||||
'vbr': vid.get('bitrate', {}).get('video'),
|
||||
'format_id': vid['encodingOption']['name'],
|
||||
'height': vid.get('height'),
|
||||
'width': vid.get('width'),
|
||||
vid = status.get('vodId')
|
||||
if vid:
|
||||
key = status.get('vodInKey')
|
||||
if not key:
|
||||
key = self._download_webpage('http://www.vlive.tv/video/inkey?vodId=%s' % vid, video_id)
|
||||
if key:
|
||||
video_info = self._extract_video_info(vid, key)
|
||||
elif status['status'] not in ('CANCELED', 'COMING_SOON', 'NOT_FOUND'):
|
||||
webpage = self._download_webpage(
|
||||
'http://m.vlive.tv/video/%s' % video_id,
|
||||
video_id, note='Download video page')
|
||||
title = self._og_search_title(webpage)
|
||||
thumbnail = self._og_search_thumbnail(webpage)
|
||||
creator = self._html_search_regex(
|
||||
r'<span[^>]+class="name">([^<>]+)</span>', webpage, 'creator')
|
||||
url = 'http://global.apis.naver.com/globalV/globalV/vod/%s/playinfo?' % video_id
|
||||
msgpad = '%.0f' % (time() * 1000)
|
||||
md = b64encode(
|
||||
hmac.new(self._SECRET.encode('ascii'),
|
||||
(url[:255] + msgpad).encode('ascii'), sha1).digest()
|
||||
)
|
||||
url += '&' + compat_urllib_parse.urlencode({'msgpad': msgpad, 'md': md})
|
||||
playinfo = self._download_json(url, video_id, 'Downloading video json')
|
||||
if playinfo.get('message', '') != 'success':
|
||||
raise ExtractorError(playinfo.get('message', 'JSON request unsuccessful'))
|
||||
result = playinfo.get('result')
|
||||
if not result:
|
||||
raise ExtractorError('No videos found.')
|
||||
video_info = self._parse_video_info(result, video_id)
|
||||
video_info.update({
|
||||
'title': title,
|
||||
'thumbnail': thumbnail,
|
||||
'creator': creator,
|
||||
})
|
||||
self._sort_formats(formats)
|
||||
|
||||
subtitles = {}
|
||||
for caption in playinfo['result'].get('captions', {}).get('list', []):
|
||||
subtitles[caption['language']] = [
|
||||
{'ext': determine_ext(caption['source'], default_ext='vtt'),
|
||||
'url': caption['source']}]
|
||||
|
||||
return {
|
||||
'id': video_id,
|
||||
'title': title,
|
||||
'creator': creator,
|
||||
'thumbnail': thumbnail,
|
||||
'formats': formats,
|
||||
'subtitles': subtitles,
|
||||
}
|
||||
if video_info:
|
||||
video_info.update({
|
||||
'id': video_id,
|
||||
'view_count': int_or_none(status.get('playCount')),
|
||||
'likes': int_or_none(status.get('likeCount')),
|
||||
})
|
||||
return video_info
|
||||
raise ExtractorError(status['status'])
|
||||
|
||||
Loading…
Reference in New Issue
Block a user