yt_dlp/extractor/xtube.py

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