yt_dlp/extractor/xtube.py

   1 import itertools
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     js_to_json,
   8     orderedSet,
   9     parse_duration,
  10     sanitized_Request,
  11     str_to_int,
  12     url_or_none,
  13 )
  14
  15
  16 class XTubeIE(InfoExtractor):
  17     _VALID_URL = r'''(?x)
  18                         (?:
  19                             xtube:|
  20                             https?://(?:www\.)?xtube\.com/(?:watch\.php\?.*\bv=|video-watch/(?:embedded/)?(?P<display_id>[^/]+)-)
  21                         )
  22                         (?P<id>[^/?&#]+)
  23                     '''
  24
  25     _TESTS = [{
  26         # old URL schema
  27         'url': 'http://www.xtube.com/watch.php?v=kVTUy_G222_',
  28         'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab',
  29         'info_dict': {
  30             'id': 'kVTUy_G222_',
  31             'ext': 'mp4',
  32             'title': 'strange erotica',
  33             'description': 'contains:an ET kind of thing',
  34             'uploader': 'greenshowers',
  35             'duration': 450,
  36             'view_count': int,
  37             'comment_count': int,
  38             'age_limit': 18,
  39         }
  40     }, {
  41         # new URL schema
  42         'url': 'http://www.xtube.com/video-watch/strange-erotica-625837',
  43         'only_matching': True,
  44     }, {
  45         'url': 'xtube:625837',
  46         'only_matching': True,
  47     }, {
  48         'url': 'xtube:kVTUy_G222_',
  49         'only_matching': True,
  50     }, {
  51         'url': 'https://www.xtube.com/video-watch/embedded/milf-tara-and-teen-shared-and-cum-covered-extreme-bukkake-32203482?embedsize=big',
  52         'only_matching': True,
  53     }]
  54
  55     def _real_extract(self, url):
  56         mobj = self._match_valid_url(url)
  57         video_id = mobj.group('id')
  58         display_id = mobj.group('display_id')
  59
  60         if not display_id:
  61             display_id = video_id
  62
  63         if video_id.isdigit() and len(video_id) < 11:
  64             url_pattern = 'http://www.xtube.com/video-watch/-%s'
  65         else:
  66             url_pattern = 'http://www.xtube.com/watch.php?v=%s'
  67
  68         webpage = self._download_webpage(
  69             url_pattern % video_id, display_id, headers={
  70                 'Cookie': 'age_verified=1; cookiesAccepted=1',
  71             })
  72
  73         title, thumbnail, duration, sources, media_definition = [None] * 5
  74
  75         config = self._parse_json(self._search_regex(
  76             r'playerConf\s*=\s*({.+?})\s*,\s*(?:\n|loaderConf|playerWrapper)', webpage, 'config',
  77             default='{}'), video_id, transform_source=js_to_json, fatal=False)
  78         if config:
  79             config = config.get('mainRoll')
  80             if isinstance(config, dict):
  81                 title = config.get('title')
  82                 thumbnail = config.get('poster')
  83                 duration = int_or_none(config.get('duration'))
  84                 sources = config.get('sources') or config.get('format')
  85                 media_definition = config.get('mediaDefinition')
  86
  87         if not isinstance(sources, dict) and not media_definition:
  88             sources = self._parse_json(self._search_regex(
  89                 r'(["\'])?sources\1?\s*:\s*(?P<sources>{.+?}),',
  90                 webpage, 'sources', group='sources'), video_id,
  91                 transform_source=js_to_json)
  92
  93         formats = []
  94         format_urls = set()
  95
  96         if isinstance(sources, dict):
  97             for format_id, format_url in sources.items():
  98                 format_url = url_or_none(format_url)
  99                 if not format_url:
 100                     continue
 101                 if format_url in format_urls:
 102                     continue
 103                 format_urls.add(format_url)
 104                 formats.append({
 105                     'url': format_url,
 106                     'format_id': format_id,
 107                     'height': int_or_none(format_id),
 108                 })
 109
 110         if isinstance(media_definition, list):
 111             for media in media_definition:
 112                 video_url = url_or_none(media.get('videoUrl'))
 113                 if not video_url:
 114                     continue
 115                 if video_url in format_urls:
 116                     continue
 117                 format_urls.add(video_url)
 118                 format_id = media.get('format')
 119                 if format_id == 'hls':
 120                     formats.extend(self._extract_m3u8_formats(
 121                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 122                         m3u8_id='hls', fatal=False))
 123                 elif format_id == 'mp4':
 124                     height = int_or_none(media.get('quality'))
 125                     formats.append({
 126                         'url': video_url,
 127                         'format_id': '%s-%d' % (format_id, height) if height else format_id,
 128                         'height': height,
 129                     })
 130
 131         self._remove_duplicate_formats(formats)
 132         self._sort_formats(formats)
 133
 134         if not title:
 135             title = self._search_regex(
 136                 (r'<h1>\s*(?P<title>[^<]+?)\s*</h1>', r'videoTitle\s*:\s*(["\'])(?P<title>.+?)\1'),
 137                 webpage, 'title', group='title')
 138         description = self._og_search_description(
 139             webpage, default=None) or self._html_search_meta(
 140             'twitter:description', webpage, default=None) or self._search_regex(
 141             r'</h1>\s*<p>([^<]+)', webpage, 'description', fatal=False)
 142         uploader = self._search_regex(
 143             (r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"',
 144              r'<span[^>]+class="nickname"[^>]*>([^<]+)'),
 145             webpage, 'uploader', fatal=False)
 146         if not duration:
 147             duration = parse_duration(self._search_regex(
 148                 r'<dt>Runtime:?</dt>\s*<dd>([^<]+)</dd>',
 149                 webpage, 'duration', fatal=False))
 150         view_count = str_to_int(self._search_regex(
 151             (r'["\']viewsCount["\'][^>]*>(\d+)\s+views',
 152              r'<dt>Views:?</dt>\s*<dd>([\d,\.]+)</dd>'),
 153             webpage, 'view count', fatal=False))
 154         comment_count = str_to_int(self._html_search_regex(
 155             r'>Comments? \(([\d,\.]+)\)<',
 156             webpage, 'comment count', fatal=False))
 157
 158         return {
 159             'id': video_id,
 160             'display_id': display_id,
 161             'title': title,
 162             'description': description,
 163             'thumbnail': thumbnail,
 164             'uploader': uploader,
 165             'duration': duration,
 166             'view_count': view_count,
 167             'comment_count': comment_count,
 168             'age_limit': 18,
 169             'formats': formats,
 170         }
 171
 172
 173 class XTubeUserIE(InfoExtractor):
 174     IE_DESC = 'XTube user profile'
 175     _VALID_URL = r'https?://(?:www\.)?xtube\.com/profile/(?P<id>[^/]+-\d+)'
 176     _TEST = {
 177         'url': 'http://www.xtube.com/profile/greenshowers-4056496',
 178         'info_dict': {
 179             'id': 'greenshowers-4056496',
 180             'age_limit': 18,
 181         },
 182         'playlist_mincount': 154,
 183     }
 184
 185     def _real_extract(self, url):
 186         user_id = self._match_id(url)
 187
 188         entries = []
 189         for pagenum in itertools.count(1):
 190             request = sanitized_Request(
 191                 'http://www.xtube.com/profile/%s/videos/%d' % (user_id, pagenum),
 192                 headers={
 193                     'Cookie': 'popunder=4',
 194                     'X-Requested-With': 'XMLHttpRequest',
 195                     'Referer': url,
 196                 })
 197
 198             page = self._download_json(
 199                 request, user_id, 'Downloading videos JSON page %d' % pagenum)
 200
 201             html = page.get('html')
 202             if not html:
 203                 break
 204
 205             for video_id in orderedSet([video_id for _, video_id in re.findall(
 206                     r'data-plid=(["\'])(.+?)\1', html)]):
 207                 entries.append(self.url_result('xtube:%s' % video_id, XTubeIE.ie_key()))
 208
 209             page_count = int_or_none(page.get('pageCount'))
 210             if not page_count or pagenum == page_count:
 211                 break
 212
 213         playlist = self.playlist_result(entries, user_id)
 214         playlist['age_limit'] = 18
 215         return playlist