yt_dlp/extractor/rumble.py

   1 import itertools
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_HTTPError
   6 from ..utils import (
   7     ExtractorError,
   8     UnsupportedError,
   9     clean_html,
  10     get_element_by_class,
  11     int_or_none,
  12     parse_count,
  13     parse_iso8601,
  14     traverse_obj,
  15     unescapeHTML,
  16 )
  17
  18
  19 class RumbleEmbedIE(InfoExtractor):
  20     _VALID_URL = r'https?://(?:www\.)?rumble\.com/embed/(?:[0-9a-z]+\.)?(?P<id>[0-9a-z]+)'
  21     _EMBED_REGEX = [fr'(?:<(?:script|iframe)[^>]+\bsrc=|["\']embedUrl["\']\s*:\s*)["\'](?P<url>{_VALID_URL})']
  22     _TESTS = [{
  23         'url': 'https://rumble.com/embed/v5pv5f',
  24         'md5': '36a18a049856720189f30977ccbb2c34',
  25         'info_dict': {
  26             'id': 'v5pv5f',
  27             'ext': 'mp4',
  28             'title': 'WMAR 2 News Latest Headlines | October 20, 6pm',
  29             'timestamp': 1571611968,
  30             'upload_date': '20191020',
  31             'channel_url': 'https://rumble.com/c/WMAR',
  32             'channel': 'WMAR',
  33             'thumbnail': 'https://sp.rmbl.ws/s8/1/5/M/z/1/5Mz1a.OvCc-small-WMAR-2-News-Latest-Headline.jpg',
  34             'duration': 234,
  35             'uploader': 'WMAR',
  36             'live_status': 'not_live',
  37         }
  38     }, {
  39         'url': 'https://rumble.com/embed/vslb7v',
  40         'md5': '7418035de1a30a178b8af34dc2b6a52b',
  41         'info_dict': {
  42             'id': 'vslb7v',
  43             'ext': 'mp4',
  44             'title': 'Defense Sec. says US Commitment to NATO Defense \'Ironclad\'',
  45             'timestamp': 1645142135,
  46             'upload_date': '20220217',
  47             'channel_url': 'https://rumble.com/c/CyberTechNews',
  48             'channel': 'CTNews',
  49             'thumbnail': 'https://sp.rmbl.ws/s8/6/7/i/9/h/7i9hd.OvCc.jpg',
  50             'duration': 901,
  51             'uploader': 'CTNews',
  52             'live_status': 'not_live',
  53         }
  54     }, {
  55         'url': 'https://rumble.com/embed/vunh1h',
  56         'info_dict': {
  57             'id': 'vunh1h',
  58             'ext': 'mp4',
  59             'title': '‘Gideon, op zoek naar de waarheid’ including ENG SUBS',
  60             'timestamp': 1647197663,
  61             'upload_date': '20220313',
  62             'channel_url': 'https://rumble.com/user/BLCKBX',
  63             'channel': 'BLCKBX',
  64             'thumbnail': r're:https://.+\.jpg',
  65             'duration': 5069,
  66             'uploader': 'BLCKBX',
  67             'live_status': 'not_live',
  68             'subtitles': {
  69                 'en': [
  70                     {
  71                         'url': r're:https://.+\.vtt',
  72                         'name': 'English',
  73                         'ext': 'vtt'
  74                     }
  75                 ]
  76             },
  77         },
  78         'params': {'skip_download': True}
  79     }, {
  80         'url': 'https://rumble.com/embed/v1essrt',
  81         'info_dict': {
  82             'id': 'v1essrt',
  83             'ext': 'mp4',
  84             'title': 'startswith:lofi hip hop radio - beats to relax/study',
  85             'timestamp': 1661519399,
  86             'upload_date': '20220826',
  87             'channel_url': 'https://rumble.com/c/LofiGirl',
  88             'channel': 'Lofi Girl',
  89             'thumbnail': r're:https://.+\.jpg',
  90             'duration': None,
  91             'uploader': 'Lofi Girl',
  92             'live_status': 'is_live',
  93         },
  94         'params': {'skip_download': True}
  95     }, {
  96         'url': 'https://rumble.com/embed/v1amumr',
  97         'info_dict': {
  98             'id': 'v1amumr',
  99             'ext': 'webm',
 100             'fps': 60,
 101             'title': 'Turning Point USA 2022 Student Action Summit DAY 1  - Rumble Exclusive Live',
 102             'timestamp': 1658518457,
 103             'upload_date': '20220722',
 104             'channel_url': 'https://rumble.com/c/RumbleEvents',
 105             'channel': 'Rumble Events',
 106             'thumbnail': r're:https://.+\.jpg',
 107             'duration': 16427,
 108             'uploader': 'Rumble Events',
 109             'live_status': 'was_live',
 110         },
 111         'params': {'skip_download': True}
 112     }, {
 113         'url': 'https://rumble.com/embed/ufe9n.v5pv5f',
 114         'only_matching': True,
 115     }]
 116
 117     _WEBPAGE_TESTS = [
 118         {
 119             'note': 'Rumble JS embed',
 120             'url': 'https://therightscoop.com/what-does-9-plus-1-plus-1-equal-listen-to-this-audio-of-attempted-kavanaugh-assassins-call-and-youll-get-it',
 121             'md5': '4701209ac99095592e73dbba21889690',
 122             'info_dict': {
 123                 'id': 'v15eqxl',
 124                 'ext': 'mp4',
 125                 'channel': 'Mr Producer Media',
 126                 'duration': 92,
 127                 'title': '911 Audio From The Man Who Wanted To Kill Supreme Court Justice Kavanaugh',
 128                 'channel_url': 'https://rumble.com/c/RichSementa',
 129                 'thumbnail': 'https://sp.rmbl.ws/s8/1/P/j/f/A/PjfAe.OvCc-small-911-Audio-From-The-Man-Who-.jpg',
 130                 'timestamp': 1654892716,
 131                 'uploader': 'Mr Producer Media',
 132                 'upload_date': '20220610',
 133                 'live_status': 'not_live',
 134             }
 135         },
 136     ]
 137
 138     @classmethod
 139     def _extract_embed_urls(cls, url, webpage):
 140         embeds = tuple(super()._extract_embed_urls(url, webpage))
 141         if embeds:
 142             return embeds
 143         return [f'https://rumble.com/embed/{mobj.group("id")}' for mobj in re.finditer(
 144             r'<script>\s*Rumble\(\s*"play"\s*,\s*{\s*[\'"]video[\'"]\s*:\s*[\'"](?P<id>[0-9a-z]+)[\'"]', webpage)]
 145
 146     def _real_extract(self, url):
 147         video_id = self._match_id(url)
 148         video = self._download_json(
 149             'https://rumble.com/embedJS/u3/', video_id,
 150             query={'request': 'video', 'ver': 2, 'v': video_id})
 151
 152         sys_msg = traverse_obj(video, ('sys', 'msg'))
 153         if sys_msg:
 154             self.report_warning(sys_msg, video_id=video_id)
 155
 156         if video.get('live') == 0:
 157             live_status = 'not_live' if video.get('livestream_has_dvr') is None else 'was_live'
 158         elif video.get('live') == 1:
 159             live_status = 'is_upcoming' if video.get('livestream_has_dvr') else 'was_live'
 160         elif video.get('live') == 2:
 161             live_status = 'is_live'
 162         else:
 163             live_status = None
 164
 165         formats = []
 166         for ext, ext_info in (video.get('ua') or {}).items():
 167             for height, video_info in (ext_info or {}).items():
 168                 meta = video_info.get('meta') or {}
 169                 if not video_info.get('url'):
 170                     continue
 171                 if ext == 'hls':
 172                     if meta.get('live') is True and video.get('live') == 1:
 173                         live_status = 'post_live'
 174                     formats.extend(self._extract_m3u8_formats(
 175                         video_info['url'], video_id,
 176                         ext='mp4', m3u8_id='hls', fatal=False, live=live_status == 'is_live'))
 177                     continue
 178                 formats.append({
 179                     'ext': ext,
 180                     'url': video_info['url'],
 181                     'format_id': '%s-%sp' % (ext, height),
 182                     'height': int_or_none(height),
 183                     'fps': video.get('fps'),
 184                     **traverse_obj(meta, {
 185                         'tbr': 'bitrate',
 186                         'filesize': 'size',
 187                         'width': 'w',
 188                         'height': 'h',
 189                     }, expected_type=lambda x: int(x) or None)
 190                 })
 191
 192         subtitles = {
 193             lang: [{
 194                 'url': sub_info['path'],
 195                 'name': sub_info.get('language') or '',
 196             }] for lang, sub_info in (video.get('cc') or {}).items() if sub_info.get('path')
 197         }
 198
 199         author = video.get('author') or {}
 200         thumbnails = traverse_obj(video, ('t', ..., {'url': 'i', 'width': 'w', 'height': 'h'}))
 201         if not thumbnails and video.get('i'):
 202             thumbnails = [{'url': video['i']}]
 203
 204         if live_status in {'is_live', 'post_live'}:
 205             duration = None
 206         else:
 207             duration = int_or_none(video.get('duration'))
 208
 209         return {
 210             'id': video_id,
 211             'title': unescapeHTML(video.get('title')),
 212             'formats': formats,
 213             'subtitles': subtitles,
 214             'thumbnails': thumbnails,
 215             'timestamp': parse_iso8601(video.get('pubDate')),
 216             'channel': author.get('name'),
 217             'channel_url': author.get('url'),
 218             'duration': duration,
 219             'uploader': author.get('name'),
 220             'live_status': live_status,
 221         }
 222
 223
 224 class RumbleIE(InfoExtractor):
 225     _VALID_URL = r'https?://(?:www\.)?rumble\.com/(?P<id>v(?!ideos)[\w.-]+)[^/]*$'
 226     _EMBED_REGEX = [r'<a class=video-item--a href=(?P<url>/v[\w.-]+\.html)>']
 227     _TESTS = [{
 228         'add_ie': ['RumbleEmbed'],
 229         'url': 'https://rumble.com/vdmum1-moose-the-dog-helps-girls-dig-a-snow-fort.html',
 230         'md5': '53af34098a7f92c4e51cf0bd1c33f009',
 231         'info_dict': {
 232             'id': 'vb0ofn',
 233             'ext': 'mp4',
 234             'timestamp': 1612662578,
 235             'uploader': 'LovingMontana',
 236             'channel': 'LovingMontana',
 237             'upload_date': '20210207',
 238             'title': 'Winter-loving dog helps girls dig a snow fort ',
 239             'description': 'Moose the dog is more than happy to help with digging out this epic snow fort. Great job, Moose!',
 240             'channel_url': 'https://rumble.com/c/c-546523',
 241             'thumbnail': r're:https://.+\.jpg',
 242             'duration': 103,
 243             'like_count': int,
 244             'view_count': int,
 245             'live_status': 'not_live',
 246         }
 247     }, {
 248         'url': 'http://www.rumble.com/vDMUM1?key=value',
 249         'only_matching': True,
 250     }]
 251
 252     _WEBPAGE_TESTS = [{
 253         'url': 'https://rumble.com/videos?page=2',
 254         'playlist_count': 25,
 255         'info_dict': {
 256             'id': 'videos?page=2',
 257             'title': 'All videos',
 258             'description': 'Browse videos uploaded to Rumble.com',
 259             'age_limit': 0,
 260         },
 261     }, {
 262         'url': 'https://rumble.com/live-videos',
 263         'playlist_mincount': 19,
 264         'info_dict': {
 265             'id': 'live-videos',
 266             'title': 'Live Videos',
 267             'description': 'Live videos on Rumble.com',
 268             'age_limit': 0,
 269         },
 270     }, {
 271         'url': 'https://rumble.com/search/video?q=rumble&sort=views',
 272         'playlist_count': 24,
 273         'info_dict': {
 274             'id': 'video?q=rumble&sort=views',
 275             'title': 'Search results for: rumble',
 276             'age_limit': 0,
 277         },
 278     }]
 279
 280     def _real_extract(self, url):
 281         page_id = self._match_id(url)
 282         webpage = self._download_webpage(url, page_id)
 283         url_info = next(RumbleEmbedIE.extract_from_webpage(self._downloader, url, webpage), None)
 284         if not url_info:
 285             raise UnsupportedError(url)
 286
 287         release_ts_str = self._search_regex(
 288             r'(?:Livestream begins|Streamed on):\s+<time datetime="([^"]+)',
 289             webpage, 'release date', fatal=False, default=None)
 290         view_count_str = self._search_regex(r'<span class="media-heading-info">([\d,]+) Views',
 291                                             webpage, 'view count', fatal=False, default=None)
 292
 293         return self.url_result(
 294             url_info['url'], ie_key=url_info['ie_key'], url_transparent=True,
 295             view_count=parse_count(view_count_str),
 296             release_timestamp=parse_iso8601(release_ts_str),
 297             like_count=parse_count(get_element_by_class('rumbles-count', webpage)),
 298             description=clean_html(get_element_by_class('media-description', webpage)),
 299         )
 300
 301
 302 class RumbleChannelIE(InfoExtractor):
 303     _VALID_URL = r'(?P<url>https?://(?:www\.)?rumble\.com/(?:c|user)/(?P<id>[^&?#$/]+))'
 304
 305     _TESTS = [{
 306         'url': 'https://rumble.com/c/Styxhexenhammer666',
 307         'playlist_mincount': 1160,
 308         'info_dict': {
 309             'id': 'Styxhexenhammer666',
 310         },
 311     }, {
 312         'url': 'https://rumble.com/user/goldenpoodleharleyeuna',
 313         'playlist_mincount': 4,
 314         'info_dict': {
 315             'id': 'goldenpoodleharleyeuna',
 316         },
 317     }]
 318
 319     def entries(self, url, playlist_id):
 320         for page in itertools.count(1):
 321             try:
 322                 webpage = self._download_webpage(f'{url}?page={page}', playlist_id, note='Downloading page %d' % page)
 323             except ExtractorError as e:
 324                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 325                     break
 326                 raise
 327             for video_url in re.findall(r'class=video-item--a\s?href=([^>]+\.html)', webpage):
 328                 yield self.url_result('https://rumble.com' + video_url)
 329
 330     def _real_extract(self, url):
 331         url, playlist_id = self._match_valid_url(url).groups()
 332         return self.playlist_result(self.entries(url, playlist_id), playlist_id=playlist_id)