yt_dlp/extractor/gab.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     clean_html,
   9     int_or_none,
  10     parse_codecs,
  11     parse_duration,
  12     str_to_int,
  13     unified_timestamp
  14 )
  15
  16
  17 class GabTVIE(InfoExtractor):
  18     _VALID_URL = r'https?://tv\.gab\.com/channel/[^/]+/view/(?P<id>[a-z0-9-]+)'
  19     _TESTS = [{
  20         'url': 'https://tv.gab.com/channel/wurzelroot/view/why-was-america-in-afghanistan-61217eacea5665de450d0488',
  21         'info_dict': {
  22             'id': '61217eacea5665de450d0488',
  23             'ext': 'mp4',
  24             'title': 'WHY WAS AMERICA IN AFGHANISTAN - AMERICA FIRST AGAINST AMERICAN OLIGARCHY',
  25             'description': None,
  26             'uploader': 'Wurzelroot',
  27             'uploader_id': '608fb0a85738fd1974984f7d',
  28             'thumbnail': 'https://tv.gab.com/image/61217eacea5665de450d0488',
  29         }
  30     }]
  31
  32     def _real_extract(self, url):
  33         id = self._match_id(url).split('-')[-1]
  34         webpage = self._download_webpage(url, id)
  35         channel_id = self._search_regex(r'data-channel-id=\"(?P<channel_id>[^\"]+)', webpage, 'channel_id')
  36         channel_name = self._search_regex(r'data-channel-name=\"(?P<channel_id>[^\"]+)', webpage, 'channel_name')
  37         title = self._search_regex(r'data-episode-title=\"(?P<channel_id>[^\"]+)', webpage, 'title')
  38         view_key = self._search_regex(r'data-view-key=\"(?P<channel_id>[^\"]+)', webpage, 'view_key')
  39         description = clean_html(
  40             self._html_search_regex(self._meta_regex('description'), webpage, 'description', group='content')) or None
  41         available_resolutions = re.findall(r'<a\ data-episode-id=\"%s\"\ data-resolution=\"(?P<resolution>[^\"]+)' % id,
  42                                            webpage)
  43
  44         formats = []
  45         for resolution in available_resolutions:
  46             frmt = {
  47                 'url': f'https://tv.gab.com/media/{id}?viewKey={view_key}&r={resolution}',
  48                 'format_id': resolution,
  49                 'vcodec': 'h264',
  50                 'acodec': 'aac',
  51                 'ext': 'mp4'
  52             }
  53             if 'audio-' in resolution:
  54                 frmt['abr'] = str_to_int(resolution.replace('audio-', ''))
  55                 frmt['height'] = 144
  56                 frmt['quality'] = -10
  57             else:
  58                 frmt['height'] = str_to_int(resolution.replace('p', ''))
  59             formats.append(frmt)
  60         self._sort_formats(formats)
  61
  62         return {
  63             'id': id,
  64             'title': title,
  65             'formats': formats,
  66             'description': description,
  67             'uploader': channel_name,
  68             'uploader_id': channel_id,
  69             'thumbnail': f'https://tv.gab.com/image/{id}',
  70         }
  71
  72
  73 class GabIE(InfoExtractor):
  74     _VALID_URL = r'https?://(?:www\.)?gab\.com/[^/]+/posts/(?P<id>\d+)'
  75     _TESTS = [{
  76         'url': 'https://gab.com/SomeBitchIKnow/posts/107163961867310434',
  77         'md5': '8ca34fb00f1e1033b5c5988d79ec531d',
  78         'info_dict': {
  79             'id': '107163961867310434-0',
  80             'ext': 'mp4',
  81             'title': 'L on Gab',
  82             'uploader_id': '946600',
  83             'uploader': 'SomeBitchIKnow',
  84             'description': 'md5:204055fafd5e1a519f5d6db953567ca3',
  85             'timestamp': 1635192289,
  86             'upload_date': '20211025',
  87         }
  88     }, {
  89         'url': 'https://gab.com/TheLonelyProud/posts/107045884469287653',
  90         'md5': 'f9cefcfdff6418e392611a828d47839d',
  91         'info_dict': {
  92             'id': '107045884469287653-0',
  93             'ext': 'mp4',
  94             'title': 'Jody Sadowski on Gab',
  95             'uploader_id': '1390705',
  96             'timestamp': 1633390571,
  97             'upload_date': '20211004',
  98             'uploader': 'TheLonelyProud',
  99         }
 100     }]
 101
 102     def _real_extract(self, url):
 103         post_id = self._match_id(url)
 104         json_data = self._download_json(f'https://gab.com/api/v1/statuses/{post_id}', post_id)
 105
 106         entries = []
 107         for idx, media in enumerate(json_data['media_attachments']):
 108             if media.get('type') not in ('video', 'gifv'):
 109                 continue
 110             metadata = media['meta']
 111             format_metadata = {
 112                 'acodec': parse_codecs(metadata.get('audio_encode')).get('acodec'),
 113                 'asr': int_or_none((metadata.get('audio_bitrate') or '').split(' ')[0]),
 114                 'fps': metadata.get('fps'),
 115             }
 116
 117             formats = [{
 118                 'url': url,
 119                 'width': f.get('width'),
 120                 'height': f.get('height'),
 121                 'tbr': int_or_none(f.get('bitrate'), scale=1000),
 122                 **format_metadata,
 123             } for url, f in ((media.get('url'), metadata.get('original') or {}),
 124                              (media.get('source_mp4'), metadata.get('playable') or {})) if url]
 125
 126             self._sort_formats(formats)
 127
 128             author = json_data.get('account') or {}
 129             entries.append({
 130                 'id': f'{post_id}-{idx}',
 131                 'title': f'{json_data["account"]["display_name"]} on Gab',
 132                 'timestamp': unified_timestamp(json_data.get('created_at')),
 133                 'formats': formats,
 134                 'description': clean_html(json_data.get('content')),
 135                 'duration': metadata.get('duration') or parse_duration(metadata.get('length')),
 136                 'like_count': json_data.get('favourites_count'),
 137                 'comment_count': json_data.get('replies_count'),
 138                 'repost_count': json_data.get('reblogs_count'),
 139                 'uploader': author.get('username'),
 140                 'uploader_id': author.get('id'),
 141                 'uploader_url': author.get('url'),
 142             })
 143
 144         if len(entries) > 1:
 145             return self.playlist_result(entries, post_id)
 146
 147         return entries[0]