yt_dlp/extractor/gab.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     clean_html,
   6     int_or_none,
   7     parse_codecs,
   8     parse_duration,
   9     str_to_int,
  10     unified_timestamp
  11 )
  12
  13
  14 class GabTVIE(InfoExtractor):
  15     _VALID_URL = r'https?://tv\.gab\.com/channel/[^/]+/view/(?P<id>[a-z0-9-]+)'
  16     _TESTS = [{
  17         'url': 'https://tv.gab.com/channel/wurzelroot/view/why-was-america-in-afghanistan-61217eacea5665de450d0488',
  18         'info_dict': {
  19             'id': '61217eacea5665de450d0488',
  20             'ext': 'mp4',
  21             'title': 'WHY WAS AMERICA IN AFGHANISTAN - AMERICA FIRST AGAINST AMERICAN OLIGARCHY',
  22             'description': None,
  23             'uploader': 'Wurzelroot',
  24             'uploader_id': '608fb0a85738fd1974984f7d',
  25             'thumbnail': 'https://tv.gab.com/image/61217eacea5665de450d0488',
  26         }
  27     }]
  28
  29     def _real_extract(self, url):
  30         id = self._match_id(url).split('-')[-1]
  31         webpage = self._download_webpage(url, id)
  32         channel_id = self._search_regex(r'data-channel-id=\"(?P<channel_id>[^\"]+)', webpage, 'channel_id')
  33         channel_name = self._search_regex(r'data-channel-name=\"(?P<channel_id>[^\"]+)', webpage, 'channel_name')
  34         title = self._search_regex(r'data-episode-title=\"(?P<channel_id>[^\"]+)', webpage, 'title')
  35         view_key = self._search_regex(r'data-view-key=\"(?P<channel_id>[^\"]+)', webpage, 'view_key')
  36         description = clean_html(
  37             self._html_search_regex(self._meta_regex('description'), webpage, 'description', group='content')) or None
  38         available_resolutions = re.findall(r'<a\ data-episode-id=\"%s\"\ data-resolution=\"(?P<resolution>[^\"]+)' % id,
  39                                            webpage)
  40
  41         formats = []
  42         for resolution in available_resolutions:
  43             frmt = {
  44                 'url': f'https://tv.gab.com/media/{id}?viewKey={view_key}&r={resolution}',
  45                 'format_id': resolution,
  46                 'vcodec': 'h264',
  47                 'acodec': 'aac',
  48                 'ext': 'mp4'
  49             }
  50             if 'audio-' in resolution:
  51                 frmt['abr'] = str_to_int(resolution.replace('audio-', ''))
  52                 frmt['height'] = 144
  53                 frmt['quality'] = -10
  54             else:
  55                 frmt['height'] = str_to_int(resolution.replace('p', ''))
  56             formats.append(frmt)
  57         self._sort_formats(formats)
  58
  59         return {
  60             'id': id,
  61             'title': title,
  62             'formats': formats,
  63             'description': description,
  64             'uploader': channel_name,
  65             'uploader_id': channel_id,
  66             'thumbnail': f'https://tv.gab.com/image/{id}',
  67         }
  68
  69
  70 class GabIE(InfoExtractor):
  71     _VALID_URL = r'https?://(?:www\.)?gab\.com/[^/]+/posts/(?P<id>\d+)'
  72     _TESTS = [{
  73         'url': 'https://gab.com/SomeBitchIKnow/posts/107163961867310434',
  74         'md5': '8ca34fb00f1e1033b5c5988d79ec531d',
  75         'info_dict': {
  76             'id': '107163961867310434-0',
  77             'ext': 'mp4',
  78             'title': 'L on Gab',
  79             'uploader_id': '946600',
  80             'uploader': 'SomeBitchIKnow',
  81             'description': 'md5:204055fafd5e1a519f5d6db953567ca3',
  82             'timestamp': 1635192289,
  83             'upload_date': '20211025',
  84         }
  85     }, {
  86         'url': 'https://gab.com/TheLonelyProud/posts/107045884469287653',
  87         'md5': 'f9cefcfdff6418e392611a828d47839d',
  88         'info_dict': {
  89             'id': '107045884469287653-0',
  90             'ext': 'mp4',
  91             'title': 'Jody Sadowski on Gab',
  92             'uploader_id': '1390705',
  93             'timestamp': 1633390571,
  94             'upload_date': '20211004',
  95             'uploader': 'TheLonelyProud',
  96         }
  97     }]
  98
  99     def _real_extract(self, url):
 100         post_id = self._match_id(url)
 101         json_data = self._download_json(f'https://gab.com/api/v1/statuses/{post_id}', post_id)
 102
 103         entries = []
 104         for idx, media in enumerate(json_data['media_attachments']):
 105             if media.get('type') not in ('video', 'gifv'):
 106                 continue
 107             metadata = media['meta']
 108             format_metadata = {
 109                 'acodec': parse_codecs(metadata.get('audio_encode')).get('acodec'),
 110                 'asr': int_or_none((metadata.get('audio_bitrate') or '').split(' ')[0]),
 111                 'fps': metadata.get('fps'),
 112             }
 113
 114             formats = [{
 115                 'url': url,
 116                 'width': f.get('width'),
 117                 'height': f.get('height'),
 118                 'tbr': int_or_none(f.get('bitrate'), scale=1000),
 119                 **format_metadata,
 120             } for url, f in ((media.get('url'), metadata.get('original') or {}),
 121                              (media.get('source_mp4'), metadata.get('playable') or {})) if url]
 122
 123             self._sort_formats(formats)
 124
 125             author = json_data.get('account') or {}
 126             entries.append({
 127                 'id': f'{post_id}-{idx}',
 128                 'title': f'{json_data["account"]["display_name"]} on Gab',
 129                 'timestamp': unified_timestamp(json_data.get('created_at')),
 130                 'formats': formats,
 131                 'description': clean_html(json_data.get('content')),
 132                 'duration': metadata.get('duration') or parse_duration(metadata.get('length')),
 133                 'like_count': json_data.get('favourites_count'),
 134                 'comment_count': json_data.get('replies_count'),
 135                 'repost_count': json_data.get('reblogs_count'),
 136                 'uploader': author.get('username'),
 137                 'uploader_id': author.get('id'),
 138                 'uploader_url': author.get('url'),
 139             })
 140
 141         if len(entries) > 1:
 142             return self.playlist_result(entries, post_id)
 143
 144         return entries[0]