[vlive] Improve and extract more metadata (Closes #8446)

This commit is contained in:
Sergey M․ 2016-02-07 06:17:40 +06:00
parent 81e0b4f2d1
commit 52f5889f77

View file

@ -1,32 +1,27 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import hmac
from hashlib import sha1
from base64 import b64encode
from time import time
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
ExtractorError, dict_get,
determine_ext, float_or_none,
int_or_none int_or_none,
) )
from ..compat import compat_urllib_parse from ..compat import compat_urllib_parse
class VLiveIE(InfoExtractor): class VLiveIE(InfoExtractor):
IE_NAME = 'vlive' IE_NAME = 'vlive'
# vlive.tv/video/ links redirect to www.vlive.tv/video/ _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
_VALID_URL = r'https?://(?:(www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
_TEST = { _TEST = {
'url': 'http://www.vlive.tv/video/1326', 'url': 'http://www.vlive.tv/video/1326',
'md5': 'cc7314812855ce56de70a06a27314983', 'md5': 'cc7314812855ce56de70a06a27314983',
'info_dict': { 'info_dict': {
'id': '1326', 'id': '1326',
'ext': 'mp4', 'ext': 'mp4',
'title': '[V] Girl\'s Day\'s Broadcast', 'title': "[V] Girl's Day's Broadcast",
'creator': 'Girl\'s Day', 'creator': "Girl's Day",
'view_count': int,
}, },
} }
@ -34,49 +29,60 @@ def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage( webpage = self._download_webpage(
'http://www.vlive.tv/video/%s' % video_id, 'http://www.vlive.tv/video/%s' % video_id, video_id)
video_id, note='Download video page')
long_video_id = self._search_regex( long_video_id = self._search_regex(
r'vlive\.tv\.video\.ajax\.request\.handler\.init\("[0-9]+",\s?"[^"]*",\s?"([^"]+)",\s?"[^"]+",\s?"[^"]*",\s?"[^"]*"\)', webpage, 'long_video_id') r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"([^"]+)"',
webpage, 'long video id')
key = self._search_regex( key = self._search_regex(
r'vlive\.tv\.video\.ajax\.request\.handler\.init\("[0-9]+",\s?"[^"]*",\s?"[^"]+",\s?"([^"]+)",\s?"[^"]*",\s?"[^"]*"\)', webpage, 'key') r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"[^"]+"\s*,\s*"([^"]+)"',
webpage, 'key')
title = self._og_search_title(webpage) title = self._og_search_title(webpage)
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)
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')]
self._sort_formats(formats)
thumbnail = self._og_search_thumbnail(webpage) thumbnail = self._og_search_thumbnail(webpage)
creator = self._html_search_regex( creator = self._html_search_regex(
r'<div class="info_area">\s*<strong[^>]+class="name">([^<>]+)</strong>', webpage, 'creator',fatal=False) r'<div[^>]+class="info_area"[^>]*>\s*<strong[^>]+class="name"[^>]*>([^<]+)</strong>',
webpage, 'creator', fatal=False)
# doct = document type (xml or json), cpt = caption type (vtt or ttml) view_count = int_or_none(playinfo.get('meta', {}).get('count'))
url = "http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?videoId=%s&key=%s&ptc=http&doct=json&cpt=vtt" % (long_video_id, key)
playinfo = self._download_json(url, video_id, 'Downloading video json')
formats = []
for vid in playinfo.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.get('encodingOption', {}).get('name'),
'height': int_or_none(vid.get('encodingOption', {}).get('height')),
'width': int_or_none(vid.get('encodingOption', {}).get('width')),
})
self._sort_formats(formats)
subtitles = {} subtitles = {}
for caption in playinfo.get('captions', {}).get('list', []): for caption in playinfo.get('captions', {}).get('list', []):
subtitles[caption['language']] = [ lang = dict_get(caption, ('language', 'locale', 'country', 'label'))
{'ext': determine_ext(caption['source'], default_ext='vtt'), if lang and caption.get('source'):
'url': caption['source']}] subtitles[lang] = [{
'ext': 'vtt',
'url': caption['source']}]
return { return {
'id': video_id, 'id': video_id,
'title': title, 'title': title,
'creator': creator, 'creator': creator,
'thumbnail': thumbnail, 'thumbnail': thumbnail,
'view_count': view_count,
'formats': formats, 'formats': formats,
'subtitles': subtitles, 'subtitles': subtitles,
} }