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

[baidu] improve info extraction

Browse source
This commit is contained in:
remitamine 2015-10-30 13:56:21 +01:00
parent 721f5a277c
commit 957e0db1d2
1 changed files with 16 additions and 29 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

45
youtube_dl/extractor/baidu.py
View file

@ -14,8 +14,8 @@ class BaiduVideoIE(InfoExtractor):
'url': 'http://v.baidu.com/comic/1069.htm?frp=bdbrand&q=%E4%B8%AD%E5%8D%8E%E5%B0%8F%E5%BD%93%E5%AE%B6', 'url': 'http://v.baidu.com/comic/1069.htm?frp=bdbrand&q=%E4%B8%AD%E5%8D%8E%E5%B0%8F%E5%BD%93%E5%AE%B6',
'info_dict': { 'info_dict': {
'id': '1069', 'id': '1069',
'title': '中华小当家 TV版 (全52集)', 'title': '中华小当家 TV版国语',
'description': 'md5:395a419e41215e531c857bb037bbaf80', 'description': 'md5:40a9c1b1c7f4e05d642e7bb1c84eeda0',
}, },
'playlist_count': 52, 'playlist_count': 52,
}, { }, {
@ -25,45 +25,32 @@ class BaiduVideoIE(InfoExtractor):
'title': 're:^奔跑吧兄弟', 'title': 're:^奔跑吧兄弟',
'description': 'md5:1bf88bad6d850930f542d51547c089b8', 'description': 'md5:1bf88bad6d850930f542d51547c089b8',
}, },
'playlist_mincount': 3, 'playlist_mincount': 12,
}] }]
def _call_api(self, path, category, playlist_id):
return self._download_json('http://app.video.baidu.com/%s/?worktype=adnative%s&id=%s' % (path, category, playlist_id), playlist_id)
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) category, playlist_id = re.match(self._VALID_URL, url).groups()
playlist_id = mobj.group('id')
category = category2 = mobj.group('type')
if category == 'show': if category == 'show':
category2 = 'tvshow' category = 'tvshow'
if category == 'tv':
category = 'tvplay'
webpage = self._download_webpage(url, playlist_id) playlist_detail = self._call_api('xqinfo', category, playlist_id)
playlist_title = self._html_search_regex( playlist_title = playlist_detail['title']
r'title\s*:\s*(["\'])(?P<title>[^\']+)\1', webpage, playlist_description = playlist_detail.get('intro')
'playlist title', group='title')
playlist_description = self._html_search_regex(
r'<input[^>]+class="j-data-intro"[^>]+value="([^"]+)"/>', webpage,
playlist_id, 'playlist description')
site = self._html_search_regex( episodes_detail = self._call_api('xqsingle', category, playlist_id)
r'filterSite\s*:\s*["\']([^"]*)["\']', webpage,
'primary provider site')
api_result = self._download_json(
'http://v.baidu.com/%s_intro/?dtype=%sPlayUrl&id=%s&site=%s' % (
category, category2, playlist_id, site),
playlist_id, 'Get playlist links')
entries = [] entries = []
for episode in api_result[0]['episodes']: for episode in episodes_detail['videos']:
episode_id = '%s_%s' % (playlist_id, episode['episode']) episode_id = '%s_%s' % (playlist_id, episode['episode'])
redirect_page = self._download_webpage(
compat_urlparse.urljoin(url, episode['url']), episode_id,
note='Download Baidu redirect page')
real_url = self._html_search_regex(
r'location\.replace\("([^"]+)"\)', redirect_page, 'real URL')
entries.append(self.url_result( entries.append(self.url_result(
real_url, video_title=episode['single_title'])) episode['url'], video_title=episode['title']))
return self.playlist_result( return self.playlist_result(
entries, playlist_id, playlist_title, playlist_description) entries, playlist_id, playlist_title, playlist_description)