yt_dlp/extractor/imdb.py

   1 from __future__ import unicode_literals
   2
   3 import base64
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     determine_ext,
  10     int_or_none,
  11     mimetype2ext,
  12     qualities,
  13     traverse_obj,
  14     try_get,
  15     url_or_none,
  16 )
  17
  18
  19 class ImdbIE(InfoExtractor):
  20     IE_NAME = 'imdb'
  21     IE_DESC = 'Internet Movie Database trailers'
  22     _VALID_URL = r'https?://(?:www|m)\.imdb\.com/(?:video|title|list).*?[/-]vi(?P<id>\d+)'
  23
  24     _TESTS = [{
  25         'url': 'http://www.imdb.com/video/imdb/vi2524815897',
  26         'info_dict': {
  27             'id': '2524815897',
  28             'ext': 'mp4',
  29             'title': 'No. 2',
  30             'description': 'md5:87bd0bdc61e351f21f20d2d7441cb4e7',
  31             'duration': 152,
  32             'thumbnail': r're:^https?://.+\.jpg',
  33         }
  34     }, {
  35         'url': 'https://www.imdb.com/video/vi3516832537',
  36         'info_dict': {
  37             'id': '3516832537',
  38             'ext': 'mp4',
  39             'title': 'Paul: U.S. Trailer #1',
  40             'description': 'md5:17fcc4fe11ec29b4399be9d4c5ef126c',
  41             'duration': 153,
  42             'thumbnail': r're:^https?://.+\.jpg',
  43         }
  44     }, {
  45         'url': 'http://www.imdb.com/video/_/vi2524815897',
  46         'only_matching': True,
  47     }, {
  48         'url': 'http://www.imdb.com/title/tt1667889/?ref_=ext_shr_eml_vi#lb-vi2524815897',
  49         'only_matching': True,
  50     }, {
  51         'url': 'http://www.imdb.com/title/tt1667889/#lb-vi2524815897',
  52         'only_matching': True,
  53     }, {
  54         'url': 'http://www.imdb.com/videoplayer/vi1562949145',
  55         'only_matching': True,
  56     }, {
  57         'url': 'http://www.imdb.com/title/tt4218696/videoplayer/vi2608641561',
  58         'only_matching': True,
  59     }, {
  60         'url': 'https://www.imdb.com/list/ls009921623/videoplayer/vi260482329',
  61         'only_matching': True,
  62     }]
  63
  64     def _real_extract(self, url):
  65         video_id = self._match_id(url)
  66         webpage = self._download_webpage(f'https://www.imdb.com/video/vi{video_id}', video_id)
  67         info = self._search_nextjs_data(webpage, video_id)
  68         video_info = traverse_obj(info, ('props', 'pageProps', 'videoPlaybackData', 'video'), default={})
  69         title = (traverse_obj(video_info, ('name', 'value'), ('primaryTitle', 'titleText', 'text'))
  70                  or self._html_search_meta(('og:title', 'twitter:title'), webpage, default=None)
  71                  or self._html_extract_title(webpage))
  72         data = video_info.get('playbackURLs') or try_get(self._download_json(
  73             'https://www.imdb.com/ve/data/VIDEO_PLAYBACK_DATA', video_id,
  74             query={
  75                 'key': base64.b64encode(json.dumps({
  76                     'type': 'VIDEO_PLAYER',
  77                     'subType': 'FORCE_LEGACY',
  78                     'id': 'vi%s' % video_id,
  79                 }).encode()).decode(),
  80             }), lambda x: x[0]['videoLegacyEncodings'])
  81         quality = qualities(('SD', '480p', '720p', '1080p'))
  82         formats, subtitles = [], {}
  83         for encoding in data:
  84             if not encoding or not isinstance(encoding, dict):
  85                 continue
  86             video_url = url_or_none(encoding.get('url'))
  87             if not video_url:
  88                 continue
  89             ext = mimetype2ext(encoding.get(
  90                 'mimeType')) or determine_ext(video_url)
  91             if ext == 'm3u8':
  92                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
  93                     video_url, video_id, 'mp4', entry_protocol='m3u8_native',
  94                     preference=1, m3u8_id='hls', fatal=False)
  95                 subtitles = self._merge_subtitles(subtitles, subs)
  96                 formats.extend(fmts)
  97                 continue
  98             format_id = traverse_obj(encoding, ('displayName', 'value'), 'definition')
  99             formats.append({
 100                 'format_id': format_id,
 101                 'url': video_url,
 102                 'ext': ext,
 103                 'quality': quality(format_id),
 104             })
 105         self._sort_formats(formats)
 106
 107         return {
 108             'id': video_id,
 109             'title': title,
 110             'alt_title': info.get('videoSubTitle'),
 111             'formats': formats,
 112             'description': try_get(video_info, lambda x: x['description']['value']),
 113             'thumbnail': url_or_none(try_get(video_info, lambda x: x['thumbnail']['url'])),
 114             'duration': int_or_none(try_get(video_info, lambda x: x['runtime']['value'])),
 115             'subtitles': subtitles,
 116         }
 117
 118
 119 class ImdbListIE(InfoExtractor):
 120     IE_NAME = 'imdb:list'
 121     IE_DESC = 'Internet Movie Database lists'
 122     _VALID_URL = r'https?://(?:www\.)?imdb\.com/list/ls(?P<id>\d{9})(?!/videoplayer/vi\d+)'
 123     _TEST = {
 124         'url': 'https://www.imdb.com/list/ls009921623/',
 125         'info_dict': {
 126             'id': '009921623',
 127             'title': 'The Bourne Legacy',
 128             'description': 'A list of trailers, clips, and more from The Bourne Legacy, starring Jeremy Renner and Rachel Weisz.',
 129         },
 130         'playlist_count': 8,
 131     }
 132
 133     def _real_extract(self, url):
 134         list_id = self._match_id(url)
 135         webpage = self._download_webpage(url, list_id)
 136         entries = [
 137             self.url_result('http://www.imdb.com' + m, 'Imdb')
 138             for m in re.findall(r'href="(/list/ls%s/videoplayer/vi[^"]+)"' % list_id, webpage)]
 139
 140         list_title = self._html_search_regex(
 141             r'<h1[^>]+class="[^"]*header[^"]*"[^>]*>(.*?)</h1>',
 142             webpage, 'list title')
 143         list_description = self._html_search_regex(
 144             r'<div[^>]+class="[^"]*list-description[^"]*"[^>]*><p>(.*?)</p>',
 145             webpage, 'list description')
 146
 147         return self.playlist_result(entries, list_id, list_title, list_description)