yt_dlp/extractor/archiveorg.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6
   7 from .common import InfoExtractor
   8 from .youtube import YoutubeIE
   9 from ..compat import (
  10     compat_urllib_parse_unquote,
  11     compat_urllib_parse_unquote_plus,
  12     compat_HTTPError
  13 )
  14 from ..utils import (
  15     clean_html,
  16     determine_ext,
  17     dict_get,
  18     extract_attributes,
  19     ExtractorError,
  20     HEADRequest,
  21     int_or_none,
  22     KNOWN_EXTENSIONS,
  23     merge_dicts,
  24     mimetype2ext,
  25     parse_duration,
  26     parse_qs,
  27     RegexNotFoundError,
  28     str_to_int,
  29     str_or_none,
  30     try_get,
  31     unified_strdate,
  32     unified_timestamp,
  33 )
  34
  35
  36 class ArchiveOrgIE(InfoExtractor):
  37     IE_NAME = 'archive.org'
  38     IE_DESC = 'archive.org video and audio'
  39     _VALID_URL = r'https?://(?:www\.)?archive\.org/(?:details|embed)/(?P<id>[^?#]+)(?:[?].*)?$'
  40     _TESTS = [{
  41         'url': 'http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect',
  42         'md5': '8af1d4cf447933ed3c7f4871162602db',
  43         'info_dict': {
  44             'id': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect',
  45             'ext': 'ogv',
  46             'title': '1968 Demo - FJCC Conference Presentation Reel #1',
  47             'description': 'md5:da45c349df039f1cc8075268eb1b5c25',
  48             'release_date': '19681210',
  49             'timestamp': 1268695290,
  50             'upload_date': '20100315',
  51             'creator': 'SRI International',
  52             'uploader': 'laura@archive.org',
  53         },
  54     }, {
  55         'url': 'https://archive.org/details/Cops1922',
  56         'md5': '0869000b4ce265e8ca62738b336b268a',
  57         'info_dict': {
  58             'id': 'Cops1922',
  59             'ext': 'mp4',
  60             'title': 'Buster Keaton\'s "Cops" (1922)',
  61             'description': 'md5:43a603fd6c5b4b90d12a96b921212b9c',
  62             'uploader': 'yorkmba99@hotmail.com',
  63             'timestamp': 1387699629,
  64             'upload_date': "20131222",
  65         },
  66     }, {
  67         'url': 'http://archive.org/embed/XD300-23_68HighlightsAResearchCntAugHumanIntellect',
  68         'only_matching': True,
  69     }, {
  70         'url': 'https://archive.org/details/Election_Ads',
  71         'md5': '284180e857160cf866358700bab668a3',
  72         'info_dict': {
  73             'id': 'Election_Ads/Commercial-JFK1960ElectionAdCampaignJingle.mpg',
  74             'title': 'Commercial-JFK1960ElectionAdCampaignJingle.mpg',
  75             'ext': 'mp4',
  76         },
  77     }, {
  78         'url': 'https://archive.org/details/Election_Ads/Commercial-Nixon1960ElectionAdToughonDefense.mpg',
  79         'md5': '7915213ef02559b5501fe630e1a53f59',
  80         'info_dict': {
  81             'id': 'Election_Ads/Commercial-Nixon1960ElectionAdToughonDefense.mpg',
  82             'title': 'Commercial-Nixon1960ElectionAdToughonDefense.mpg',
  83             'ext': 'mp4',
  84             'timestamp': 1205588045,
  85             'uploader': 'mikedavisstripmaster@yahoo.com',
  86             'description': '1960 Presidential Campaign Election Commercials John F Kennedy, Richard M Nixon',
  87             'upload_date': '20080315',
  88         },
  89     }, {
  90         'url': 'https://archive.org/details/gd1977-05-08.shure57.stevenson.29303.flac16',
  91         'md5': '7d07ffb42aba6537c28e053efa4b54c9',
  92         'info_dict': {
  93             'id': 'gd1977-05-08.shure57.stevenson.29303.flac16/gd1977-05-08d01t01.flac',
  94             'title': 'Turning',
  95             'ext': 'flac',
  96         },
  97     }, {
  98         'url': 'https://archive.org/details/gd1977-05-08.shure57.stevenson.29303.flac16/gd1977-05-08d01t07.flac',
  99         'md5': 'a07cd8c6ab4ee1560f8a0021717130f3',
 100         'info_dict': {
 101             'id': 'gd1977-05-08.shure57.stevenson.29303.flac16/gd1977-05-08d01t07.flac',
 102             'title': 'Deal',
 103             'ext': 'flac',
 104             'timestamp': 1205895624,
 105             'uploader': 'mvernon54@yahoo.com',
 106             'description': 'md5:6a31f1996db0aa0fc9da6d6e708a1bb0',
 107             'upload_date': '20080319',
 108             'location': 'Barton Hall - Cornell University',
 109         },
 110     }, {
 111         'url': 'https://archive.org/details/lp_the-music-of-russia_various-artists-a-askaryan-alexander-melik',
 112         'md5': '7cb019baa9b332e82ea7c10403acd180',
 113         'info_dict': {
 114             'id': 'lp_the-music-of-russia_various-artists-a-askaryan-alexander-melik/disc1/01.01. Bells Of Rostov.mp3',
 115             'title': 'Bells Of Rostov',
 116             'ext': 'mp3',
 117         },
 118     }, {
 119         'url': 'https://archive.org/details/lp_the-music-of-russia_various-artists-a-askaryan-alexander-melik/disc1/02.02.+Song+And+Chorus+In+The+Polovetsian+Camp+From+%22Prince+Igor%22+(Act+2%2C+Scene+1).mp3',
 120         'md5': '1d0aabe03edca83ca58d9ed3b493a3c3',
 121         'info_dict': {
 122             'id': 'lp_the-music-of-russia_various-artists-a-askaryan-alexander-melik/disc1/02.02. Song And Chorus In The Polovetsian Camp From "Prince Igor" (Act 2, Scene 1).mp3',
 123             'title': 'Song And Chorus In The Polovetsian Camp From "Prince Igor" (Act 2, Scene 1)',
 124             'ext': 'mp3',
 125             'timestamp': 1569662587,
 126             'uploader': 'associate-joygen-odiongan@archive.org',
 127             'description': 'md5:012b2d668ae753be36896f343d12a236',
 128             'upload_date': '20190928',
 129         },
 130     }]
 131
 132     @staticmethod
 133     def _playlist_data(webpage):
 134         element = re.findall(r'''(?xs)
 135             <input
 136             (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
 137             \s+class=['"]?js-play8-playlist['"]?
 138             (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
 139             \s*/>
 140         ''', webpage)[0]
 141
 142         return json.loads(extract_attributes(element)['value'])
 143
 144     def _real_extract(self, url):
 145         video_id = compat_urllib_parse_unquote_plus(self._match_id(url))
 146         identifier, entry_id = (video_id.split('/', 1) + [None])[:2]
 147
 148         # Archive.org metadata API doesn't clearly demarcate playlist entries
 149         # or subtitle tracks, so we get them from the embeddable player.
 150         embed_page = self._download_webpage(
 151             'https://archive.org/embed/' + identifier, identifier)
 152         playlist = self._playlist_data(embed_page)
 153
 154         entries = {}
 155         for p in playlist:
 156             # If the user specified a playlist entry in the URL, ignore the
 157             # rest of the playlist.
 158             if entry_id and p['orig'] != entry_id:
 159                 continue
 160
 161             entries[p['orig']] = {
 162                 'formats': [],
 163                 'thumbnails': [],
 164                 'artist': p.get('artist'),
 165                 'track': p.get('title'),
 166                 'subtitles': {}}
 167
 168             for track in p.get('tracks', []):
 169                 if track['kind'] != 'subtitles':
 170                     continue
 171
 172                 entries[p['orig']][track['label']] = {
 173                     'url': 'https://archive.org/' + track['file'].lstrip('/')}
 174
 175         metadata = self._download_json(
 176             'http://archive.org/metadata/' + identifier, identifier)
 177         m = metadata['metadata']
 178         identifier = m['identifier']
 179
 180         info = {
 181             'id': identifier,
 182             'title': m['title'],
 183             'description': clean_html(m.get('description')),
 184             'uploader': dict_get(m, ['uploader', 'adder']),
 185             'creator': m.get('creator'),
 186             'license': m.get('licenseurl'),
 187             'release_date': unified_strdate(m.get('date')),
 188             'timestamp': unified_timestamp(dict_get(m, ['publicdate', 'addeddate'])),
 189             'webpage_url': 'https://archive.org/details/' + identifier,
 190             'location': m.get('venue'),
 191             'release_year': int_or_none(m.get('year'))}
 192
 193         for f in metadata['files']:
 194             if f['name'] in entries:
 195                 entries[f['name']] = merge_dicts(entries[f['name']], {
 196                     'id': identifier + '/' + f['name'],
 197                     'title': f.get('title') or f['name'],
 198                     'display_id': f['name'],
 199                     'description': clean_html(f.get('description')),
 200                     'creator': f.get('creator'),
 201                     'duration': parse_duration(f.get('length')),
 202                     'track_number': int_or_none(f.get('track')),
 203                     'album': f.get('album'),
 204                     'discnumber': int_or_none(f.get('disc')),
 205                     'release_year': int_or_none(f.get('year'))})
 206                 entry = entries[f['name']]
 207             elif f.get('original') in entries:
 208                 entry = entries[f['original']]
 209             else:
 210                 continue
 211
 212             if f.get('format') == 'Thumbnail':
 213                 entry['thumbnails'].append({
 214                     'id': f['name'],
 215                     'url': 'https://archive.org/download/' + identifier + '/' + f['name'],
 216                     'width': int_or_none(f.get('width')),
 217                     'height': int_or_none(f.get('width')),
 218                     'filesize': int_or_none(f.get('size'))})
 219
 220             extension = (f['name'].rsplit('.', 1) + [None])[1]
 221             if extension in KNOWN_EXTENSIONS:
 222                 entry['formats'].append({
 223                     'url': 'https://archive.org/download/' + identifier + '/' + f['name'],
 224                     'format': f.get('format'),
 225                     'width': int_or_none(f.get('width')),
 226                     'height': int_or_none(f.get('height')),
 227                     'filesize': int_or_none(f.get('size')),
 228                     'protocol': 'https'})
 229
 230         # Sort available formats by filesize
 231         for entry in entries.values():
 232             entry['formats'] = list(sorted(entry['formats'], key=lambda x: x.get('filesize', -1)))
 233
 234         if len(entries) == 1:
 235             # If there's only one item, use it as the main info dict
 236             only_video = entries[list(entries.keys())[0]]
 237             if entry_id:
 238                 info = merge_dicts(only_video, info)
 239             else:
 240                 info = merge_dicts(info, only_video)
 241         else:
 242             # Otherwise, we have a playlist.
 243             info['_type'] = 'playlist'
 244             info['entries'] = list(entries.values())
 245
 246         if metadata.get('reviews'):
 247             info['comments'] = []
 248             for review in metadata['reviews']:
 249                 info['comments'].append({
 250                     'id': review.get('review_id'),
 251                     'author': review.get('reviewer'),
 252                     'text': str_or_none(review.get('reviewtitle'), '') + '\n\n' + review.get('reviewbody'),
 253                     'timestamp': unified_timestamp(review.get('createdate')),
 254                     'parent': 'root'})
 255
 256         return info
 257
 258
 259 class YoutubeWebArchiveIE(InfoExtractor):
 260     IE_NAME = 'web.archive:youtube'
 261     IE_DESC = 'web.archive.org saved youtube videos'
 262     _VALID_URL = r"""(?x)^
 263                 (?:https?://)?web\.archive\.org/
 264                     (?:web/)?
 265                     (?:[0-9A-Za-z_*]+/)?  # /web and the version index is optional
 266
 267                 (?:https?(?::|%3[Aa])//)?
 268                 (?:
 269                     (?:\w+\.)?youtube\.com/watch(?:\?|%3[fF])(?:[^\#]+(?:&|%26))?v(?:=|%3[dD])  # Youtube URL
 270                     |(wayback-fakeurl\.archive\.org/yt/)  # Or the internal fake url
 271                 )
 272                 (?P<id>[0-9A-Za-z_-]{11})(?:%26|\#|&|$)
 273                 """
 274
 275     _TESTS = [
 276         {
 277             'url': 'https://web.archive.org/web/20150415002341/https://www.youtube.com/watch?v=aYAGB11YrSs',
 278             'info_dict': {
 279                 'id': 'aYAGB11YrSs',
 280                 'ext': 'webm',
 281                 'title': 'Team Fortress 2 - Sandviches!'
 282             }
 283         },
 284         {
 285             # Internal link
 286             'url': 'https://web.archive.org/web/2oe/http://wayback-fakeurl.archive.org/yt/97t7Xj_iBv0',
 287             'info_dict': {
 288                 'id': '97t7Xj_iBv0',
 289                 'ext': 'mp4',
 290                 'title': 'How Flexible Machines Could Save The World'
 291             }
 292         },
 293         {
 294             # Video from 2012, webm format itag 45.
 295             'url': 'https://web.archive.org/web/20120712231619/http://www.youtube.com/watch?v=AkhihxRKcrs&gl=US&hl=en',
 296             'info_dict': {
 297                 'id': 'AkhihxRKcrs',
 298                 'ext': 'webm',
 299                 'title': 'Limited Run: Mondo\'s Modern Classic 1 of 3 (SDCC 2012)'
 300             }
 301         },
 302         {
 303             # Old flash-only video. Webpage title starts with "YouTube - ".
 304             'url': 'https://web.archive.org/web/20081211103536/http://www.youtube.com/watch?v=jNQXAC9IVRw',
 305             'info_dict': {
 306                 'id': 'jNQXAC9IVRw',
 307                 'ext': 'unknown_video',
 308                 'title': 'Me at the zoo'
 309             }
 310         },
 311         {
 312             # Flash video with .flv extension (itag 34). Title has prefix "YouTube         -"
 313             # Title has some weird unicode characters too.
 314             'url': 'https://web.archive.org/web/20110712231407/http://www.youtube.com/watch?v=lTx3G6h2xyA',
 315             'info_dict': {
 316                 'id': 'lTx3G6h2xyA',
 317                 'ext': 'flv',
 318                 'title': '‪Madeon - Pop Culture (live mashup)‬‏'
 319             }
 320         },
 321         {   # Some versions of Youtube have have "YouTube" as page title in html (and later rewritten by js).
 322             'url': 'https://web.archive.org/web/http://www.youtube.com/watch?v=kH-G_aIBlFw',
 323             'info_dict': {
 324                 'id': 'kH-G_aIBlFw',
 325                 'ext': 'mp4',
 326                 'title': 'kH-G_aIBlFw'
 327             },
 328             'expected_warnings': [
 329                 'unable to extract title',
 330             ]
 331         },
 332         {
 333             # First capture is a 302 redirect intermediary page.
 334             'url': 'https://web.archive.org/web/20050214000000/http://www.youtube.com/watch?v=0altSZ96U4M',
 335             'info_dict': {
 336                 'id': '0altSZ96U4M',
 337                 'ext': 'mp4',
 338                 'title': '0altSZ96U4M'
 339             },
 340             'expected_warnings': [
 341                 'unable to extract title',
 342             ]
 343         },
 344         {
 345             # Video not archived, only capture is unavailable video page
 346             'url': 'https://web.archive.org/web/20210530071008/https://www.youtube.com/watch?v=lHJTf93HL1s&spfreload=10',
 347             'only_matching': True,
 348         },
 349         {   # Encoded url
 350             'url': 'https://web.archive.org/web/20120712231619/http%3A//www.youtube.com/watch%3Fgl%3DUS%26v%3DAkhihxRKcrs%26hl%3Den',
 351             'only_matching': True,
 352         },
 353         {
 354             'url': 'https://web.archive.org/web/20120712231619/http%3A//www.youtube.com/watch%3Fv%3DAkhihxRKcrs%26gl%3DUS%26hl%3Den',
 355             'only_matching': True,
 356         }
 357     ]
 358
 359     def _real_extract(self, url):
 360         video_id = self._match_id(url)
 361         title = video_id  # if we are not able get a title
 362
 363         def _extract_title(webpage):
 364             page_title = self._html_search_regex(
 365                 r'<title>([^<]*)</title>', webpage, 'title', fatal=False) or ''
 366             # YouTube video pages appear to always have either 'YouTube -' as suffix or '- YouTube' as prefix.
 367             try:
 368                 page_title = self._html_search_regex(
 369                     r'(?:YouTube\s*-\s*(.*)$)|(?:(.*)\s*-\s*YouTube$)',
 370                     page_title, 'title', default='')
 371             except RegexNotFoundError:
 372                 page_title = None
 373
 374             if not page_title:
 375                 self.report_warning('unable to extract title', video_id=video_id)
 376                 return
 377             return page_title
 378
 379         # If the video is no longer available, the oldest capture may be one before it was removed.
 380         # Setting the capture date in url to early date seems to redirect to earliest capture.
 381         webpage = self._download_webpage(
 382             'https://web.archive.org/web/20050214000000/http://www.youtube.com/watch?v=%s' % video_id,
 383             video_id=video_id, fatal=False, errnote='unable to download video webpage (probably not archived).')
 384         if webpage:
 385             title = _extract_title(webpage) or title
 386
 387         # Use link translator mentioned in https://github.com/ytdl-org/youtube-dl/issues/13655
 388         internal_fake_url = 'https://web.archive.org/web/2oe_/http://wayback-fakeurl.archive.org/yt/%s' % video_id
 389         try:
 390             video_file_webpage = self._request_webpage(
 391                 HEADRequest(internal_fake_url), video_id,
 392                 note='Fetching video file url', expected_status=True)
 393         except ExtractorError as e:
 394             # HTTP Error 404 is expected if the video is not saved.
 395             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 396                 raise ExtractorError(
 397                     'HTTP Error %s. Most likely the video is not archived or issue with web.archive.org.' % e.cause.code,
 398                     expected=True)
 399             raise
 400         video_file_url = compat_urllib_parse_unquote(video_file_webpage.url)
 401         video_file_url_qs = parse_qs(video_file_url)
 402
 403         # Attempt to recover any ext & format info from playback url
 404         format = {'url': video_file_url}
 405         itag = try_get(video_file_url_qs, lambda x: x['itag'][0])
 406         if itag and itag in YoutubeIE._formats:  # Naughty access but it works
 407             format.update(YoutubeIE._formats[itag])
 408             format.update({'format_id': itag})
 409         else:
 410             mime = try_get(video_file_url_qs, lambda x: x['mime'][0])
 411             ext = mimetype2ext(mime) or determine_ext(video_file_url)
 412             format.update({'ext': ext})
 413         return {
 414             'id': video_id,
 415             'title': title,
 416             'formats': [format],
 417             'duration': str_to_int(try_get(video_file_url_qs, lambda x: x['dur'][0]))
 418         }