yt_dlp/extractor/xuite.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     float_or_none,
   5     get_element_by_attribute,
   6     parse_iso8601,
   7     remove_end,
   8 )
   9
  10
  11 class XuiteIE(InfoExtractor):
  12     IE_DESC = '隨意窩Xuite影音'
  13     _REGEX_BASE64 = r'(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}==|[A-Za-z0-9+/]{3}=)?'
  14     _VALID_URL = r'https?://vlog\.xuite\.net/(?:play|embed)/(?P<id>%s)' % _REGEX_BASE64
  15     _TESTS = [{
  16         # Audio
  17         'url': 'http://vlog.xuite.net/play/RGkzc1ZULTM4NjA5MTQuZmx2',
  18         'md5': 'e79284c87b371424885448d11f6398c8',
  19         'info_dict': {
  20             'id': '3860914',
  21             'ext': 'mp3',
  22             'title': '孤單南半球-歐德陽',
  23             'description': '孤單南半球-歐德陽',
  24             'thumbnail': r're:^https?://.*\.jpg$',
  25             'duration': 247.246,
  26             'timestamp': 1314932940,
  27             'upload_date': '20110902',
  28             'uploader': '阿能',
  29             'uploader_id': '15973816',
  30             'categories': ['個人短片'],
  31         },
  32     }, {
  33         # Video with only one format
  34         'url': 'http://vlog.xuite.net/play/WUxxR2xCLTI1OTI1MDk5LmZsdg==',
  35         'md5': '21f7b39c009b5a4615b4463df6eb7a46',
  36         'info_dict': {
  37             'id': '25925099',
  38             'ext': 'mp4',
  39             'title': 'BigBuckBunny_320x180',
  40             'thumbnail': r're:^https?://.*\.jpg$',
  41             'duration': 596.458,
  42             'timestamp': 1454242500,
  43             'upload_date': '20160131',
  44             'uploader': '屁姥',
  45             'uploader_id': '12158353',
  46             'categories': ['個人短片'],
  47             'description': 'http://download.blender.org/peach/bigbuckbunny_movies/BigBuckBunny_320x180.mp4',
  48         },
  49     }, {
  50         # Video with two formats
  51         'url': 'http://vlog.xuite.net/play/bWo1N1pLLTIxMzAxMTcwLmZsdg==',
  52         'md5': '1166e0f461efe55b62e26a2d2a68e6de',
  53         'info_dict': {
  54             'id': '21301170',
  55             'ext': 'mp4',
  56             'title': '暗殺教室 02',
  57             'description': '字幕:【極影字幕社】',
  58             'thumbnail': r're:^https?://.*\.jpg$',
  59             'duration': 1384.907,
  60             'timestamp': 1421481240,
  61             'upload_date': '20150117',
  62             'uploader': '我只是想認真點',
  63             'uploader_id': '242127761',
  64             'categories': ['電玩動漫'],
  65         },
  66         'skip': 'Video removed',
  67     }, {
  68         # Video with encoded media id
  69         # from http://forgetfulbc.blogspot.com/2016/06/date.html
  70         'url': 'http://vlog.xuite.net/embed/cE1xbENoLTI3NDQ3MzM2LmZsdg==?ar=0&as=0',
  71         'info_dict': {
  72             'id': '27447336',
  73             'ext': 'mp4',
  74             'title': '男女平權只是口號？專家解釋約會時男生是否該幫女生付錢 (中字)',
  75             'description': 'md5:1223810fa123b179083a3aed53574706',
  76             'timestamp': 1466160960,
  77             'upload_date': '20160617',
  78             'uploader': 'B.C. & Lowy',
  79             'uploader_id': '232279340',
  80         },
  81     }, {
  82         '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',
  83         'only_matching': True,
  84     }]
  85
  86     def _real_extract(self, url):
  87         # /play/ URLs provide embedded video URL and more metadata
  88         url = url.replace('/embed/', '/play/')
  89         video_id = self._match_id(url)
  90
  91         webpage = self._download_webpage(url, video_id)
  92
  93         error_msg = self._search_regex(
  94             r'<div id="error-message-content">([^<]+)',
  95             webpage, 'error message', default=None)
  96         if error_msg:
  97             raise ExtractorError(
  98                 '%s returned error: %s' % (self.IE_NAME, error_msg),
  99                 expected=True)
 100
 101         media_info = self._parse_json(self._search_regex(
 102             r'var\s+mediaInfo\s*=\s*({.*});', webpage, 'media info'), video_id)
 103
 104         video_id = media_info['MEDIA_ID']
 105
 106         formats = []
 107         for key in ('html5Url', 'html5HQUrl'):
 108             video_url = media_info.get(key)
 109             if not video_url:
 110                 continue
 111             format_id = self._search_regex(
 112                 r'\bq=(.+?)\b', video_url, 'format id', default=None)
 113             formats.append({
 114                 'url': video_url,
 115                 'ext': 'mp4' if format_id.isnumeric() else format_id,
 116                 'format_id': format_id,
 117                 'height': int(format_id) if format_id.isnumeric() else None,
 118             })
 119
 120         timestamp = media_info.get('PUBLISH_DATETIME')
 121         if timestamp:
 122             timestamp = parse_iso8601(timestamp + ' +0800', ' ')
 123
 124         category = media_info.get('catName')
 125         categories = [category] if category else []
 126
 127         uploader = media_info.get('NICKNAME')
 128         uploader_url = None
 129
 130         author_div = get_element_by_attribute('itemprop', 'author', webpage)
 131         if author_div:
 132             uploader = uploader or self._html_search_meta('name', author_div)
 133             uploader_url = self._html_search_regex(
 134                 r'<link[^>]+itemprop="url"[^>]+href="([^"]+)"', author_div,
 135                 'uploader URL', fatal=False)
 136
 137         return {
 138             'id': video_id,
 139             'title': media_info['TITLE'],
 140             'description': remove_end(media_info.get('metaDesc'), ' (Xuite 影音)'),
 141             'thumbnail': media_info.get('ogImageUrl'),
 142             'timestamp': timestamp,
 143             'uploader': uploader,
 144             'uploader_id': media_info.get('MEMBER_ID'),
 145             'uploader_url': uploader_url,
 146             'duration': float_or_none(media_info.get('MEDIA_DURATION'), 1000000),
 147             'categories': categories,
 148             'formats': formats,
 149         }