1
1
mirror of https://github.com/ytdl-org/youtube-dl synced 2024-12-05 03:02:56 +01:00
youtube-dl/youtube_dl/extractor/tudou.py

95 lines
3.4 KiB
Python
Raw Normal View History

# coding: utf-8
2014-09-01 00:16:26 +02:00
from __future__ import unicode_literals
2013-06-25 19:48:08 +02:00
from .common import InfoExtractor
from ..compat import compat_str
2013-06-25 19:48:08 +02:00
class TudouIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?tudou\.com/(?:listplay|programs(?:/view)?|albumplay)/([^/]+/)*(?P<id>[^/?#]+?)(?:\.html)?/?(?:$|[?#])'
_TESTS = [{
2014-09-01 00:16:26 +02:00
'url': 'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
'md5': '140a49ed444bd22f93330985d8475fcb',
'info_dict': {
'id': '159448201',
'ext': 'f4v',
'title': '卡马乔国足开大脚长传冲吊集锦',
'thumbnail': 're:^https?://.*\.jpg$',
2013-06-27 20:46:46 +02:00
}
2014-09-01 00:20:12 +02:00
}, {
'url': 'http://www.tudou.com/programs/view/ajX3gyhL0pc/',
'info_dict': {
'id': '117049447',
'ext': 'f4v',
'title': 'La Sylphide-Bolshoi-Ekaterina Krysanova & Vyacheslav Lopatin 2012',
'thumbnail': 're:^https?://.*\.jpg$',
}
2015-09-12 16:42:57 +02:00
}, {
'url': 'http://www.tudou.com/albumplay/cJAHGih4yYg.html',
'only_matching': True,
}]
2013-06-25 19:48:08 +02:00
2015-08-05 12:22:25 +02:00
_PLAYER_URL = 'http://js.tudouui.com/bin/lingtong/PortalPlayer_177.swf'
2015-09-12 16:52:51 +02:00
def _url_for_id(self, video_id, quality=None):
info_url = 'http://v2.tudou.com/f?id=' + compat_str(video_id)
if quality:
info_url += '&hd' + quality
2015-09-12 20:36:51 +02:00
xml_data = self._download_xml(info_url, video_id, "Opening the info XML page")
final_url = xml_data.text
return final_url
2013-06-25 19:48:08 +02:00
def _real_extract(self, url):
2015-01-08 17:50:46 +01:00
video_id = self._match_id(url)
2013-06-25 19:48:08 +02:00
webpage = self._download_webpage(url, video_id)
2015-09-12 16:51:49 +02:00
youku_vcode = self._search_regex(
2015-09-12 20:51:20 +02:00
r'vcode\s*:\s*[\'"]([^\'"]*)[\'"]', webpage, 'youku vcode', default=None)
2015-09-12 16:51:49 +02:00
if youku_vcode:
return self.url_result('youku:' + youku_vcode, ie='Youku')
2013-10-18 11:16:11 +02:00
title = self._search_regex(
2015-09-12 20:51:20 +02:00
r',kw\s*:\s*[\'"]([^\'"]+)[\'"]', webpage, 'title')
thumbnail_url = self._search_regex(
2015-09-12 20:51:20 +02:00
r',pic\s*:\s*[\'"]([^\'"]+)[\'"]', webpage, 'thumbnail URL', fatal=False)
player_url = self._search_regex(
2015-09-12 20:51:20 +02:00
r'playerUrl\s*:\s*[\'"]([^\'"]+\.swf)[\'"]',
webpage, 'player URL', default=self._PLAYER_URL)
2015-09-12 16:51:49 +02:00
segments = self._parse_json(self._search_regex(
2015-09-12 20:51:20 +02:00
r'segs: \'([^\']+)\'', webpage, 'segments'), video_id)
# It looks like the keys are the arguments that have to be passed as
# the hd field in the request url, we pick the higher
# Also, filter non-number qualities (see issue #3643).
quality = sorted(filter(lambda k: k.isdigit(), segments.keys()),
key=lambda k: int(k))[-1]
parts = segments[quality]
result = []
len_parts = len(parts)
if len_parts > 1:
2014-11-26 13:06:02 +01:00
self.to_screen('%s: found %s parts' % (video_id, len_parts))
for part in parts:
part_id = part['k']
final_url = self._url_for_id(part_id, quality)
ext = (final_url.split('?')[0]).split('.')[-1]
2014-09-01 00:16:26 +02:00
part_info = {
'id': '%s' % part_id,
'url': final_url,
'ext': ext,
'title': title,
'thumbnail': thumbnail_url,
2015-08-05 12:22:25 +02:00
'http_headers': {
'Referer': player_url,
2015-08-05 12:22:25 +02:00
},
2014-09-01 00:16:26 +02:00
}
result.append(part_info)
2015-01-08 17:50:46 +01:00
return {
'_type': 'multi_video',
'entries': result,
'id': video_id,
'title': title,
}