yt_dlp/extractor/nitter.py

   1 from .common import InfoExtractor
   2 from ..compat import compat_urlparse
   3 from ..utils import (
   4     parse_count,
   5     unified_timestamp,
   6     remove_end,
   7     determine_ext,
   8 )
   9 import re
  10 import random
  11
  12
  13 class NitterIE(InfoExtractor):
  14     # Taken from https://github.com/zedeus/nitter/wiki/Instances
  15
  16     NON_HTTP_INSTANCES = (
  17         '3nzoldnxplag42gqjs23xvghtzf6t6yzssrtytnntc6ppc7xxuoneoad.onion',
  18         'nitter.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd.onion',
  19         'nitter7bryz3jv7e3uekphigvmoyoem4al3fynerxkj22dmoxoq553qd.onion',
  20         'npf37k3mtzwxreiw52ccs5ay4e6qt2fkcs2ndieurdyn2cuzzsfyfvid.onion',
  21         'nitter.v6vgyqpa7yefkorazmg5d5fimstmvm2vtbirt6676mt7qmllrcnwycqd.onion',
  22         'i23nv6w3juvzlw32xzoxcqzktegd4i4fu3nmnc2ewv4ggiu4ledwklad.onion',
  23         '26oq3gioiwcmfojub37nz5gzbkdiqp7fue5kvye7d4txv4ny6fb4wwid.onion',
  24         'vfaomgh4jxphpbdfizkm5gbtjahmei234giqj4facbwhrfjtcldauqad.onion',
  25         'iwgu3cv7ywf3gssed5iqtavmrlszgsxazkmwwnt4h2kdait75thdyrqd.onion',
  26         'erpnncl5nhyji3c32dcfmztujtl3xaddqb457jsbkulq24zqq7ifdgad.onion',
  27         'ckzuw5misyahmg7j5t5xwwuj3bwy62jfolxyux4brfflramzsvvd3syd.onion',
  28         'jebqj47jgxleaiosfcxfibx2xdahjettuydlxbg64azd4khsxv6kawid.onion',
  29         'nttr2iupbb6fazdpr2rgbooon2tzbbsvvkagkgkwohhodjzj43stxhad.onion',
  30         'nitraeju2mipeziu2wtcrqsxg7h62v5y4eqgwi75uprynkj74gevvuqd.onion',
  31         'nitter.lqs5fjmajyp7rvp4qvyubwofzi6d4imua7vs237rkc4m5qogitqwrgyd.onion',
  32         'ibsboeui2im5o7dxnik3s5yghufumgy5abevtij5nbizequfpu4qi4ad.onion',
  33         'ec5nvbycpfa5k6ro77blxgkyrzbkv7uy6r5cngcbkadtjj2733nm3uyd.onion',
  34
  35         'nitter.i2p',
  36         'u6ikd6zndl3c4dsdq4mmujpntgeevdk5qzkfb57r4tnfeccrn2qa.b32.i2p',
  37
  38         'nitterlgj3n5fgwesu3vxc5h67ruku33nqaoeoocae2mvlzhsu6k7fqd.onion',
  39     )
  40
  41     HTTP_INSTANCES = (
  42         'nitter.42l.fr',
  43         'nitter.pussthecat.org',
  44         'nitter.nixnet.services',
  45         'nitter.fdn.fr',
  46         'nitter.1d4.us',
  47         'nitter.kavin.rocks',
  48         'nitter.unixfox.eu',
  49         'nitter.domain.glass',
  50         'nitter.eu',
  51         'nitter.namazso.eu',
  52         'nitter.actionsack.com',
  53         'birdsite.xanny.family',
  54         'nitter.hu',
  55         'twitr.gq',
  56         'nitter.moomoo.me',
  57         'nittereu.moomoo.me',
  58         'bird.from.tf',
  59         'nitter.it',
  60         'twitter.censors.us',
  61         'twitter.grimneko.de',
  62         'nitter.alefvanoon.xyz',
  63         'n.hyperborea.cloud',
  64         'nitter.ca',
  65         'twitter.076.ne.jp',
  66         'twitter.mstdn.social',
  67         'nitter.fly.dev',
  68         'notabird.site',
  69         'nitter.weiler.rocks',
  70         'nitter.silkky.cloud',
  71         'nitter.sethforprivacy.com',
  72         'nttr.stream',
  73         'nitter.cutelab.space',
  74         'nitter.nl',
  75         'nitter.mint.lgbt',
  76         'nitter.bus-hit.me',
  77         'fuckthesacklers.network',
  78         'nitter.govt.land',
  79         'nitter.datatunnel.xyz',
  80         'nitter.esmailelbob.xyz',
  81         'tw.artemislena.eu',
  82         'de.nttr.stream',
  83         'nitter.winscloud.net',
  84         'nitter.tiekoetter.com',
  85         'nitter.spaceint.fr',
  86         'twtr.bch.bar',
  87         'nitter.exonip.de',
  88         'nitter.mastodon.pro',
  89         'nitter.notraxx.ch',
  90
  91
  92         # not in the list anymore
  93         'nitter.skrep.in',
  94         'nitter.snopyta.org',
  95     )
  96
  97     DEAD_INSTANCES = (
  98         # maintenance
  99         'nitter.ethibox.fr',
 100
 101         # official, rate limited
 102         'nitter.net',
 103         # offline
 104         'is-nitter.resolv.ee',
 105         'lu-nitter.resolv.ee',
 106         'nitter.13ad.de',
 107         'nitter.40two.app',
 108         'nitter.cattube.org',
 109         'nitter.cc',
 110         'nitter.dark.fail',
 111         'nitter.himiko.cloud',
 112         'nitter.koyu.space',
 113         'nitter.mailstation.de',
 114         'nitter.mastodont.cat',
 115         'nitter.tedomum.net',
 116         'nitter.tokhmi.xyz',
 117         'nitter.weaponizedhumiliation.com',
 118         'nitter.vxempire.xyz',
 119         'tweet.lambda.dance',
 120     )
 121
 122     INSTANCES = NON_HTTP_INSTANCES + HTTP_INSTANCES + DEAD_INSTANCES
 123
 124     _INSTANCES_RE = f'(?:{"|".join(map(re.escape, INSTANCES))})'
 125     _VALID_URL = fr'https?://{_INSTANCES_RE}/(?P<uploader_id>.+)/status/(?P<id>[0-9]+)(#.)?'
 126     current_instance = random.choice(HTTP_INSTANCES)
 127
 128     _TESTS = [
 129         {
 130             # GIF (wrapped in mp4)
 131             'url': f'https://{current_instance}/firefox/status/1314279897502629888#m',
 132             'info_dict': {
 133                 'id': '1314279897502629888',
 134                 'ext': 'mp4',
 135                 'title': 'md5:7890a9277da4639ab624dd899424c5d8',
 136                 'description': 'md5:5fea96a4d3716c350f8b95b21b3111fe',
 137                 'thumbnail': r're:^https?://.*\.jpg$',
 138                 'uploader': 'Firefox 🔥',
 139                 'uploader_id': 'firefox',
 140                 'uploader_url': f'https://{current_instance}/firefox',
 141                 'upload_date': '20201008',
 142                 'timestamp': 1602183720,
 143                 'like_count': int,
 144                 'repost_count': int,
 145                 'comment_count': int,
 146             },
 147         }, {  # normal video
 148             'url': f'https://{current_instance}/Le___Doc/status/1299715685392756737#m',
 149             'info_dict': {
 150                 'id': '1299715685392756737',
 151                 'ext': 'mp4',
 152                 'title': 're:^.* - "Je ne prédis jamais rien"\nD Raoult, Août 2020...',
 153                 'description': '"Je ne prédis jamais rien"\nD Raoult, Août 2020...',
 154                 'thumbnail': r're:^https?://.*\.jpg$',
 155                 'uploader': 're:^Le *Doc',
 156                 'uploader_id': 'Le___Doc',
 157                 'uploader_url': f'https://{current_instance}/Le___Doc',
 158                 'upload_date': '20200829',
 159                 'timestamp': 1598711340,
 160                 'view_count': int,
 161                 'like_count': int,
 162                 'repost_count': int,
 163                 'comment_count': int,
 164             },
 165         }, {  # video embed in a "Streaming Political Ads" box
 166             'url': f'https://{current_instance}/mozilla/status/1321147074491092994#m',
 167             'info_dict': {
 168                 'id': '1321147074491092994',
 169                 'ext': 'mp4',
 170                 'title': 'md5:8290664aabb43b9189145c008386bf12',
 171                 'description': 'md5:9cf2762d49674bc416a191a689fb2aaa',
 172                 'thumbnail': r're:^https?://.*\.jpg$',
 173                 'uploader': 'Mozilla',
 174                 'uploader_id': 'mozilla',
 175                 'uploader_url': f'https://{current_instance}/mozilla',
 176                 'upload_date': '20201027',
 177                 'timestamp': 1603820940,
 178                 'view_count': int,
 179                 'like_count': int,
 180                 'repost_count': int,
 181                 'comment_count': int,
 182             },
 183             'expected_warnings': ['Ignoring subtitle tracks found in the HLS manifest'],
 184         }, {  # not the first tweet but main-tweet
 185             'url': f'https://{current_instance}/firefox/status/1354848277481414657#m',
 186             'info_dict': {
 187                 'id': '1354848277481414657',
 188                 'ext': 'mp4',
 189                 'title': 'md5:bef647f03bd1c6b15b687ea70dfc9700',
 190                 'description': 'md5:5efba25e2f9dac85ebcd21160cb4341f',
 191                 'thumbnail': r're:^https?://.*\.jpg$',
 192                 'uploader': 'Firefox 🔥',
 193                 'uploader_id': 'firefox',
 194                 'uploader_url': f'https://{current_instance}/firefox',
 195                 'upload_date': '20210128',
 196                 'timestamp': 1611855960,
 197                 'view_count': int,
 198                 'like_count': int,
 199                 'repost_count': int,
 200                 'comment_count': int,
 201             }
 202         }
 203     ]
 204
 205     def _real_extract(self, url):
 206         video_id, uploader_id = self._match_valid_url(url).group('id', 'uploader_id')
 207         parsed_url = compat_urlparse.urlparse(url)
 208         base_url = f'{parsed_url.scheme}://{parsed_url.netloc}'
 209
 210         self._set_cookie(parsed_url.netloc, 'hlsPlayback', 'on')
 211         full_webpage = webpage = self._download_webpage(url, video_id)
 212
 213         main_tweet_start = full_webpage.find('class="main-tweet"')
 214         if main_tweet_start > 0:
 215             webpage = full_webpage[main_tweet_start:]
 216
 217         video_url = '%s%s' % (base_url, self._html_search_regex(
 218             r'(?:<video[^>]+data-url|<source[^>]+src)="([^"]+)"', webpage, 'video url'))
 219         ext = determine_ext(video_url)
 220
 221         if ext == 'unknown_video':
 222             formats = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
 223         else:
 224             formats = [{
 225                 'url': video_url,
 226                 'ext': ext
 227             }]
 228
 229         title = description = self._og_search_description(full_webpage) or self._html_search_regex(
 230             r'<div class="tweet-content[^>]+>([^<]+)</div>', webpage, 'title', fatal=False)
 231
 232         uploader_id = self._html_search_regex(
 233             r'<a class="username"[^>]+title="@([^"]+)"', webpage, 'uploader id', fatal=False) or uploader_id
 234
 235         uploader = self._html_search_regex(
 236             r'<a class="fullname"[^>]+title="([^"]+)"', webpage, 'uploader name', fatal=False)
 237         if uploader:
 238             title = f'{uploader} - {title}'
 239
 240         counts = {
 241             f'{x[0]}_count': self._html_search_regex(
 242                 fr'<span[^>]+class="icon-{x[1]}[^>]*></span>([^<]*)</div>',
 243                 webpage, f'{x[0]} count', fatal=False)
 244             for x in (('view', 'play'), ('like', 'heart'), ('repost', 'retweet'), ('comment', 'comment'))
 245         }
 246         counts = {field: 0 if count == '' else parse_count(count) for field, count in counts.items()}
 247
 248         thumbnail = (
 249             self._html_search_meta('og:image', full_webpage, 'thumbnail url')
 250             or remove_end('%s%s' % (base_url, self._html_search_regex(
 251                 r'<video[^>]+poster="([^"]+)"', webpage, 'thumbnail url', fatal=False)), '%3Asmall'))
 252
 253         thumbnails = [
 254             {'id': id, 'url': f'{thumbnail}%3A{id}'}
 255             for id in ('thumb', 'small', 'large', 'medium', 'orig')
 256         ]
 257
 258         date = self._html_search_regex(
 259             r'<span[^>]+class="tweet-date"[^>]*><a[^>]+title="([^"]+)"',
 260             webpage, 'upload date', default='').replace('·', '')
 261
 262         return {
 263             'id': video_id,
 264             'title': title,
 265             'description': description,
 266             'uploader': uploader,
 267             'timestamp': unified_timestamp(date),
 268             'uploader_id': uploader_id,
 269             'uploader_url': f'{base_url}/{uploader_id}',
 270             'formats': formats,
 271             'thumbnails': thumbnails,
 272             'thumbnail': thumbnail,
 273             **counts,
 274         }