yt_dlp/extractor/imdb.py

   1 import base64
   2 import json
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     determine_ext,
   8     int_or_none,
   9     mimetype2ext,
  10     qualities,
  11     traverse_obj,
  12     try_get,
  13     url_or_none,
  14 )
  15
  16
  17 class ImdbIE(InfoExtractor):
  18     IE_NAME = 'imdb'
  19     IE_DESC = 'Internet Movie Database trailers'
  20     _VALID_URL = r'https?://(?:www|m)\.imdb\.com/(?:video|title|list).*?[/-]vi(?P<id>\d+)'
  21
  22     _TESTS = [{
  23         'url': 'http://www.imdb.com/video/imdb/vi2524815897',
  24         'info_dict': {
  25             'id': '2524815897',
  26             'ext': 'mp4',
  27             'title': 'No. 2',
  28             'description': 'md5:87bd0bdc61e351f21f20d2d7441cb4e7',
  29             'duration': 152,
  30             'thumbnail': r're:^https?://.+\.jpg',
  31         },
  32     }, {
  33         'url': 'https://www.imdb.com/video/vi3516832537',
  34         'info_dict': {
  35             'id': '3516832537',
  36             'ext': 'mp4',
  37             'title': 'Paul: U.S. Trailer #1',
  38             'description': 'md5:17fcc4fe11ec29b4399be9d4c5ef126c',
  39             'duration': 153,
  40             'thumbnail': r're:^https?://.+\.jpg',
  41         },
  42     }, {
  43         'url': 'http://www.imdb.com/video/_/vi2524815897',
  44         'only_matching': True,
  45     }, {
  46         'url': 'http://www.imdb.com/title/tt1667889/?ref_=ext_shr_eml_vi#lb-vi2524815897',
  47         'only_matching': True,
  48     }, {
  49         'url': 'http://www.imdb.com/title/tt1667889/#lb-vi2524815897',
  50         'only_matching': True,
  51     }, {
  52         'url': 'http://www.imdb.com/videoplayer/vi1562949145',
  53         'only_matching': True,
  54     }, {
  55         'url': 'http://www.imdb.com/title/tt4218696/videoplayer/vi2608641561',
  56         'only_matching': True,
  57     }, {
  58         'url': 'https://www.imdb.com/list/ls009921623/videoplayer/vi260482329',
  59         'only_matching': True,
  60     }]
  61
  62     def _real_extract(self, url):
  63         video_id = self._match_id(url)
  64         webpage = self._download_webpage(f'https://www.imdb.com/video/vi{video_id}', video_id)
  65         info = self._search_nextjs_data(webpage, video_id)
  66         video_info = traverse_obj(info, ('props', 'pageProps', 'videoPlaybackData', 'video'), default={})
  67         title = (traverse_obj(video_info, ('name', 'value'), ('primaryTitle', 'titleText', 'text'))
  68                  or self._html_search_meta(('og:title', 'twitter:title'), webpage, default=None)
  69                  or self._html_extract_title(webpage))
  70         data = video_info.get('playbackURLs') or try_get(self._download_json(
  71             'https://www.imdb.com/ve/data/VIDEO_PLAYBACK_DATA', video_id,
  72             query={
  73                 'key': base64.b64encode(json.dumps({
  74                     'type': 'VIDEO_PLAYER',
  75                     'subType': 'FORCE_LEGACY',
  76                     'id': f'vi{video_id}',
  77                 }).encode()).decode(),
  78             }), lambda x: x[0]['videoLegacyEncodings'])
  79         quality = qualities(('SD', '480p', '720p', '1080p'))
  80         formats, subtitles = [], {}
  81         for encoding in data:
  82             if not encoding or not isinstance(encoding, dict):
  83                 continue
  84             video_url = url_or_none(encoding.get('url'))
  85             if not video_url:
  86                 continue
  87             ext = mimetype2ext(encoding.get(
  88                 'mimeType')) or determine_ext(video_url)
  89             if ext == 'm3u8':
  90                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
  91                     video_url, video_id, 'mp4', entry_protocol='m3u8_native',
  92                     preference=1, m3u8_id='hls', fatal=False)
  93                 subtitles = self._merge_subtitles(subtitles, subs)
  94                 formats.extend(fmts)
  95                 continue
  96             format_id = traverse_obj(encoding, ('displayName', 'value'), 'definition')
  97             formats.append({
  98                 'format_id': format_id,
  99                 'url': video_url,
 100                 'ext': ext,
 101                 'quality': quality(format_id),
 102             })
 103
 104         return {
 105             'id': video_id,
 106             'title': title,
 107             'alt_title': info.get('videoSubTitle'),
 108             'formats': formats,
 109             'description': try_get(video_info, lambda x: x['description']['value']),
 110             'thumbnail': url_or_none(try_get(video_info, lambda x: x['thumbnail']['url'])),
 111             'duration': int_or_none(try_get(video_info, lambda x: x['runtime']['value'])),
 112             'subtitles': subtitles,
 113         }
 114
 115
 116 class ImdbListIE(InfoExtractor):
 117     IE_NAME = 'imdb:list'
 118     IE_DESC = 'Internet Movie Database lists'
 119     _VALID_URL = r'https?://(?:www\.)?imdb\.com/list/ls(?P<id>\d{9})(?!/videoplayer/vi\d+)'
 120     _TEST = {
 121         'url': 'https://www.imdb.com/list/ls009921623/',
 122         'info_dict': {
 123             'id': '009921623',
 124             'title': 'The Bourne Legacy',
 125             'description': 'A list of trailers, clips, and more from The Bourne Legacy, starring Jeremy Renner and Rachel Weisz.',
 126         },
 127         'playlist_count': 8,
 128     }
 129
 130     def _real_extract(self, url):
 131         list_id = self._match_id(url)
 132         webpage = self._download_webpage(url, list_id)
 133         entries = [
 134             self.url_result('http://www.imdb.com' + m, 'Imdb')
 135             for m in re.findall(rf'href="(/list/ls{list_id}/videoplayer/vi[^"]+)"', webpage)]
 136
 137         list_title = self._html_search_regex(
 138             r'<h1[^>]+class="[^"]*header[^"]*"[^>]*>(.*?)</h1>',
 139             webpage, 'list title')
 140         list_description = self._html_search_regex(
 141             r'<div[^>]+class="[^"]*list-description[^"]*"[^>]*><p>(.*?)</p>',
 142             webpage, 'list description')
 143
 144         return self.playlist_result(entries, list_id, list_title, list_description)