1 # -*- coding: utf-8 -*-
2 from __future__ import unicode_literals
5 from .common import InfoExtractor
8 compat_urllib_parse_unquote,
17 # ref: http://stackoverflow.com/questions/475074/regex-to-parse-or-validate-base64-data
18 REGEX_BASE64 = r'(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}==|[A-Za-z0-9+/]{3}=)?'
20 CST_ZONE = +8 # China Standard Time
23 class XuiteIE(InfoExtractor):
24 _VALID_URL = r'http://vlog.xuite.net/play/(?P<id>%s)(/.*)?' % REGEX_BASE64
27 'url': 'http://vlog.xuite.net/play/RGkzc1ZULTM4NjA5MTQuZmx2',
28 'md5': '63a42c705772aa53fd4c1a0027f86adf',
30 'id': 'RGkzc1ZULTM4NjA5MTQuZmx2',
32 'upload_date': '20110902',
33 'uploader_id': '15973816',
35 'timestamp': 1314932940,
39 # Audio with alternative form of url
40 'url': 'http://vlog.xuite.net/play/S1dDUjdyLTMyOTc3NjcuZmx2/%E5%AD%AB%E7%87%95%E5%A7%BF-%E7%9C%BC%E6%B7%9A%E6%88%90%E8%A9%A9',
41 'md5': 'c91645f324de53d82ebb80930e1a73d2',
43 'id': 'S1dDUjdyLTMyOTc3NjcuZmx2',
45 'upload_date': '20101226',
46 'uploader_id': '10102699',
48 'timestamp': 1293367080,
52 # Video with only one format
53 'url': 'http://vlog.xuite.net/play/TkRZNjhULTM0NDE2MjkuZmx2',
54 'md5': 'c45737fc8ac5dc8ac2f92ecbcecf505e',
56 'id': 'TkRZNjhULTM0NDE2MjkuZmx2',
58 'upload_date': '20110306',
59 'uploader_id': '10400126',
61 'timestamp': 1299383640,
62 'title': '孫燕姿 - 眼淚成詩',
65 # Video with two formats
66 'url': 'http://vlog.xuite.net/play/bWo1N1pLLTIxMzAxMTcwLmZsdg==',
67 'md5': '1166e0f461efe55b62e26a2d2a68e6de',
69 'id': 'bWo1N1pLLTIxMzAxMTcwLmZsdg==',
71 'upload_date': '20150117',
72 'uploader_id': '242127761',
73 'uploader': '我只是想認真點',
74 'timestamp': 1421481240,
79 def _flv_config(self, media_id):
80 base64_media_id = base64.b64encode(media_id.encode('utf-8')).decode('utf-8')
81 flv_config_url = 'http://vlog.xuite.net/flash/player?media=' + base64_media_id
82 flv_config = self._download_xml(flv_config_url, 'flv config')
85 for prop in flv_config.findall('./property'):
86 prop_id = base64.b64decode(prop.attrib['id']).decode('utf-8')
89 continue # CDATA may be empty in flv config
91 encoded_content = base64.b64decode(prop.text).decode('utf-8')
92 prop_dict[prop_id] = compat_urllib_parse_unquote(encoded_content)
96 def _type_string(self, media_url):
97 query_string = compat_urlparse.urlparse(media_url).query
98 type_string = compat_parse_qs(query_string)['q'][0]
101 def _guess_ext(self, media_url):
102 type_string = self._type_string(media_url)
103 if type_string == 'mp3':
105 elif type_string == '360' or type_string == '720':
108 raise ExtractorError('Unknown type string %s' % type_string)
110 def _real_extract(self, url):
111 video_id = self._match_id(url)
113 page = self._download_webpage(url, video_id)
114 media_id = self._html_search_regex(r'data-mediaid="(\d+)"', page, 'media id')
115 flv_config = self._flv_config(media_id)
117 timestamp_local = parse_iso8601(flv_config['publish_datetime'], ' ')
118 timestamp_gmt = timestamp_local - CST_ZONE * 3600
122 'title': flv_config['title'],
123 'thumbnail': flv_config['thumb'],
124 'uploader': flv_config['author_name'],
125 'timestamp': timestamp_gmt,
126 'uploader_id': flv_config['author_id'],
127 'duration': parse_duration(flv_config['duration']),
128 'categories': [flv_config['category']]
131 if 'hq_src' in flv_config:
132 src = flv_config['src']
133 src_hq = flv_config['hq_src']
134 ret_attrs['formats'] = [{
136 'ext': self._guess_ext(src),
137 'format_id': self._type_string(src)
140 'ext': self._guess_ext(src_hq),
141 'format_id': self._type_string(src_hq)
144 ret_attrs['url'] = flv_config['src']
145 ret_attrs['ext'] = self._guess_ext(flv_config['src'])